metradar 0.1.0__py3-none-any.whl

metradar/decode_pup_rose.py

@@ -0,0 +1,1993 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+'''
+@File    :   decode_pup_rose.py
+@Time    :   2023/04/27 23:12:54
+@Author  :   Wenjian Zhu 
+@Version :   1.0
+@Email   :   kevin2075@163.com
+'''
+
+'''
+解析pup_rose文件
+
+'''
+
+# %%
+import os
+import numpy as np
+import geo_transforms_pyart as geotrans
+from datetime import datetime,timedelta
+# from rose_structer import _unpack_from_buf,_structure_size,SIT_HEADER_BLOCK,GENERIC_HEADER,\
+#                 SITE_CONFIG,TASK_CONFIG,SCAN_CONFIG,PRODUCT_HEADER_BLOCK,PRODUCT_ID_NAME_MAP,\
+#                 PRODUCT_DEPENDENT_PARAMETER,STORM_MOTION_BLOCK,STORM_FST_HIS_NUM,STORM_FST_HIS_BLOCK, \
+#                 STORM_PROPERTY,STORM_TRACK_PARAM,STORM_COMPONENT
+import xarray as xr
+import rose_structer as rs
+
+class READ_ROSE(object):
+
+
+    def __init__(self,):
+        self.stiinfo = None
+        self.mesoinfo = None
+        self.hailinfo = None
+        self.tvsinfo = None
+        self.ssinfo = None
+        pass
+    
+    # 将数字风暴序号转为字母数字组合
+
+    def get_id_char(self,id_num):
+        '''
+        该函数的算法由张持岸提供
+        '''
+        if not isinstance(id_num,int):
+            print('id_num is not int')
+            return None
+        if id_num > 0:
+            num = (id_num-1) // 26
+            tail = (id_num-1) % 26
+            newid = '%s%d'%(chr(65+tail),num)
+        else:
+            newid = '0'
+        return newid
+
+    #将数字的风向转换为文字
+    def get_wind_dir_name(self,wdir):
+        wdir_cn = '未知'
+        if not isinstance(wdir,int) and not isinstance(wdir,float):
+            return wdir_cn
+        if (wdir >= 348.76 and wdir <= 360) or (wdir >= 0 and wdir <= 11.25):
+            wdir_cn = '北'
+        elif wdir >= 11.26 and wdir <=33.75:
+            wdir_cn = '北东北'
+        elif wdir >= 33.76 and wdir <=56.25:
+            wdir_cn = '东北'
+        elif wdir >= 56.26 and wdir <=78.75:
+            wdir_cn = '东东北'
+        elif wdir >= 78.76 and wdir <101.25:
+            wdir_cn = '东'
+        elif wdir >= 101.26 and wdir <123.75:
+            wdir_cn = '东东南'
+        elif wdir >= 123.76 and wdir <146.25:
+            wdir_cn = '东南'
+        elif wdir >= 146.26 and wdir <168.75:
+            wdir_cn = '南东南'
+        elif wdir >= 168.76 and wdir <191.25:
+            wdir_cn = '南'
+        elif wdir >= 191.26 and wdir <213.75: 
+            wdir_cn = '南西南'
+        elif wdir >= 213.76 and wdir <236.25:
+            wdir_cn = '西南'
+        elif wdir >= 236.26 and wdir <258.75:
+            wdir_cn = '西西南'
+        elif wdir >= 258.76 and wdir <281.25:
+            wdir_cn = '西'
+        elif wdir >= 281.26 and wdir <303.75:
+            wdir_cn = '西西北'
+        elif wdir >= 303.76 and wdir <326.25:
+            wdir_cn = '西北'
+        elif wdir >= 326.26 and wdir <348.75:
+            wdir_cn = '北西北'
+
+        return wdir_cn
+    
+    def get_mda_rank(self,range_from_radar,shear_value):
+
+        '''
+        range_from_radar: km
+        shear_value : m/s
+        '''
+        knot2ms = 0.514444445
+        nmi     = 1.852 
+
+        #x = np.array([0,30,60,90,120,150,180,210]) # x取值
+        x  = np.arange(0,250) # x取值
+        y1 = -1.*knot2ms*x/(24*nmi)+25*knot2ms
+        y2 = -1.*knot2ms*x/(18*nmi)+35*knot2ms
+        y3 = -1.*knot2ms*x/(14*nmi)+45*knot2ms
+
+        curx = int(range_from_radar)
+        if curx >=250 or curx <=0:
+            print('range from radar should be larger than 0 and less than 250km ')
+            return None
+        
+        cury = shear_value
+        if cury < y1[curx]:
+            return 1
+        elif cury <y2[curx]:
+            return 2
+        elif cury < y3[curx]:
+            return 3
+        else:
+            return 4
+
+    def get_tvs_rank(self,lldv_value):
+
+        # the "19" icon indicates that LLDV was at least 190 kts.
+
+        ms2knot = 1.943844490
+
+        value = lldv_value * ms2knot
+
+        index = value // 10
+        if index > 19:
+            index=19
+
+        return index+1
+    
+    # 解析风暴追踪产品
+    def read_sti(self,filepath,filename):
+        '''
+        解析sti产品文件
+        '''
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        # pprint(prod_type)
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        pos = pos + 64 # 产品参数长度固定为64个
+
+
+        # 获取产品数据块
+
+        #获取STI头信息
+        dic_sti_header = rs._unpack_from_buf(buf, pos, rs.SIT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.SIT_HEADER_BLOCK)
+
+        # 风暴追踪信息块
+        # 风暴移动信息
+        storm_motion_block=[]
+        for nn in np.arange(dic_sti_header['storm_number']):
+            storm_motion_block.append(rs._unpack_from_buf(buf, pos, rs.STORM_MOTION_BLOCK))
+            pos = pos + rs._structure_size(rs.STORM_MOTION_BLOCK)
+
+        # 风暴预报信息
+        all_storm_fst_block=[]
+        for nn in np.arange(dic_sti_header['storm_number']):
+            dic_storm_fst_num = rs._unpack_from_buf(buf, pos, rs.STORM_FST_HIS_NUM)
+            pos = pos + rs._structure_size(rs.STORM_FST_HIS_NUM)
+
+            storm_fst_block=[]
+            for nn in np.arange(dic_storm_fst_num['position_number']):
+                storm_fst_block.append(rs._unpack_from_buf(buf, pos, rs.STORM_FST_HIS_BLOCK))
+                pos = pos + rs._structure_size(rs.STORM_FST_HIS_BLOCK)
+            all_storm_fst_block.append(storm_fst_block)
+
+        # 风暴历史信息
+        all_storm_his_block=[]
+        for nn in np.arange(dic_sti_header['storm_number']):
+            dic_storm_his_num = rs._unpack_from_buf(buf, pos, rs.STORM_FST_HIS_NUM)
+            pos = pos + rs._structure_size(rs.STORM_FST_HIS_NUM)
+
+            storm_his_block=[]
+            for nn in np.arange(dic_storm_his_num['position_number']):
+                storm_his_block.append(rs._unpack_from_buf(buf, pos, rs.STORM_FST_HIS_BLOCK))
+                pos = pos + rs._structure_size(rs.STORM_FST_HIS_BLOCK)
+            all_storm_his_block.append(storm_his_block)
+
+        # 风暴属性表块数据
+        # 风暴属性
+        all_storm_prop=[]
+        for nn in np.arange(dic_sti_header['storm_number']):
+            dic_storm_prop = rs._unpack_from_buf(buf, pos, rs.STORM_PROPERTY)
+            pos = pos + rs._structure_size(rs.STORM_PROPERTY)
+
+            # 将风暴数字ID转换为字符串ID
+            dic_storm_prop['id_char'] = self.get_id_char(dic_storm_prop['id'])
+
+            # 将风暴移动信息融合进来
+            dic_storm_prop['mv_spd'] = storm_motion_block[nn]['mv_spd']
+            dic_storm_prop['mv_dir'] = storm_motion_block[nn]['mv_dir']
+
+            all_storm_prop.append(dic_storm_prop)
+
+        # 风暴构成表
+        all_storm_comp=[]
+        for nn in np.arange(dic_sti_header['storm_number']):
+            dic_storm_comp = rs._unpack_from_buf(buf, pos, rs.STORM_COMPONENT)
+            pos = pos + rs._structure_size(rs.STORM_COMPONENT)
+            all_storm_comp.append(dic_storm_comp)
+
+        # 风暴追踪适配数据
+        dic_storm_track_param = rs._unpack_from_buf(buf, pos, rs.STORM_TRACK_PARAM)
+        pos = pos + rs._structure_size(rs.STORM_TRACK_PARAM)
+
+
+        allresult={}
+        allresult['dic_gh'] = dic_gh
+        allresult['dic_scfg'] = dic_scfg
+        allresult['dic_tcfg'] = dic_tcfg
+        allresult['cutinfo'] = cutinfo
+        allresult['dic_prod_header'] = dic_prod_header
+        allresult['dic_prod_param'] = dic_prod_param
+        allresult['dic_sti_header'] = dic_sti_header
+        allresult['storm_motion_block'] = storm_motion_block
+        allresult['all_storm_fst_block'] = all_storm_fst_block
+        allresult['all_storm_his_block'] = all_storm_his_block
+        allresult['all_storm_prop'] = all_storm_prop
+        allresult['all_storm_comp'] = all_storm_comp
+        allresult['dic_storm_track_param'] = dic_storm_track_param
+
+        # 将所有的方位角和距离转换成经纬度
+        for nn in np.arange(dic_sti_header['storm_number']):
+            pass
+            x,y,z = geotrans.antenna_to_cartesian(storm_motion_block[nn]['range']/1000.0,storm_motion_block[nn]['azi'],0)
+            clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+            storm_motion_block[nn]['lon'] = clon[0]
+            storm_motion_block[nn]['lat'] = clat[0]
+            all_storm_prop[nn]['lon'] = clon[0]
+            all_storm_prop[nn]['lat'] = clat[0]
+            for blk in all_storm_fst_block[nn]:
+                x,y,z = geotrans.antenna_to_cartesian(blk['range']/1000.0,blk['azi'],0)
+                clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+                blk['lon'] = clon[0]
+                blk['lat'] = clat[0]
+            for blk in all_storm_his_block[nn]:
+                x,y,z = geotrans.antenna_to_cartesian(blk['range']/1000.0,blk['azi'],0)
+                clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+                blk['lon'] = clon[0]
+                blk['lat'] = clat[0]
+
+            pass
+
+        # 整理成track
+        alltrackinfo=[]
+        all_past_position=[] # 历史位置
+        all_fst_position=[] # 预报位置
+        all_current_position=[] # 当前位置
+        for nn in np.arange(dic_sti_header['storm_number']):
+            pass
+            
+            if all_storm_prop[nn]['type']==0: # 连续风暴
+                trackinfo=[]
+                
+                # 先添加历史位置
+                for blk in all_storm_his_block[nn][::-1]:
+                    trackinfo.append([blk['lat'],blk['lon']])
+                    all_past_position.append([blk['lat'],blk['lon']])
+
+                # 添加当前风暴位置
+                trackinfo.append([storm_motion_block[nn]['lat'],storm_motion_block[nn]['lon']])
+                all_current_position.append([storm_motion_block[nn]['lat'],storm_motion_block[nn]['lon']])
+
+                # 再添加预报位置
+                for blk in all_storm_fst_block[nn]:
+                    trackinfo.append([blk['lat'],blk['lon']])
+                    all_fst_position.append([blk['lat'],blk['lon']])
+
+                alltrackinfo.append(trackinfo)
+
+        # allresult={}
+        allresult['track'] = alltrackinfo
+        allresult['marker_past'] = all_past_position
+        allresult['marker_current'] = all_current_position
+        allresult['marker_fst'] = all_fst_position
+
+        self.stiinfo = allresult
+        return allresult
+
+    # 解析风暴结构文本产品
+    def read_ss(self,filepath,filename):
+        '''
+        解析ss产品文件
+        '''
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        # pprint(prod_type)
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        pos = pos + 64 # 产品参数长度固定为64个
+
+
+        # 获取产品数据块
+
+        #获取SS头信息
+        dic_ss_header = rs._unpack_from_buf(buf, pos, rs.SS_HEAD_BLOCK)
+        pos = pos + rs._structure_size(rs.SS_HEAD_BLOCK)
+
+        # SS结构信息
+        ss_tab=[]
+        for nn in np.arange(dic_ss_header['storm_number']):
+            tmp_tab = rs._unpack_from_buf(buf, pos, rs.SS_TAB)
+            pos = pos + rs._structure_size(rs.SS_TAB)
+            # 将风暴数字ID转换为字符串ID
+            tmp_tab['id_char'] = self.get_id_char(tmp_tab['storm_id'])
+            
+            ss_tab.append(tmp_tab)
+
+
+        # 风暴趋势信息
+        cell_trend=[]
+        for nn in np.arange(dic_ss_header['storm_number']):
+            cur_cellinfo={}
+            cell_info=rs._unpack_from_buf(buf, pos, rs.CELL_TREND)
+            pos = pos + rs._structure_size(rs.CELL_TREND)
+            cur_cellinfo['head_info'] = cell_info
+            # 历史体扫信息
+            tmpcell_his=[]
+            for im in np.arange(cell_info['his_vol_num']):
+                tmpcell_his.append(rs._unpack_from_buf(buf, pos, rs.HIS_VOL))
+                pos = pos + rs._structure_size(rs.HIS_VOL)
+            cur_cellinfo['cell_info'] = tmpcell_his
+
+            cell_trend.append(cur_cellinfo)
+
+        # 风暴段适配数据
+        seg_adapt = rs._unpack_from_buf(buf, pos, rs.SEG_ADAPT)
+        pos = pos + rs._structure_size(rs.SEG_ADAPT)
+
+        # 风暴质心适配数据
+        centroid_adapt = rs._unpack_from_buf(buf, pos, rs.CENTROIDS_ADAPT)
+        pos = pos + rs._structure_size(rs.CENTROIDS_ADAPT)
+
+        # 风暴追踪适配数据
+        storm_track_adapt = rs._unpack_from_buf(buf, pos, rs.STORM_TRACK_PARAM)
+        pos = pos + rs._structure_size(rs.STORM_TRACK_PARAM)
+        
+
+        # 将所有的方位角和距离转换成经纬度
+        for nn in np.arange(dic_ss_header['storm_number']):
+            pass
+            x,y,z = geotrans.antenna_to_cartesian(ss_tab[nn]['range']/1000.0,ss_tab[nn]['azi'],0)
+            clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+            ss_tab[nn]['lon'] = clon[0]
+            ss_tab[nn]['lat'] = clat[0]
+        
+
+        allresult={}
+        allresult['ss'] = ss_tab
+       
+        self.ssinfo = allresult
+        return allresult
+    
+    # 解析中气旋产品
+    def read_mda(self,filepath,filename):
+        '''
+        解析meso产品文件
+        '''
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        # pprint(prod_type)
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 获取产品数据块
+
+        #获取中气旋头信息
+        dic_meso_header = rs._unpack_from_buf(buf, pos, rs.MESO_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.MESO_HEADER_BLOCK)
+
+        # 中气旋表块
+        meso_tab=[]
+        for nn in np.arange(dic_meso_header['meso_number']):
+            meso_tab.append(rs._unpack_from_buf(buf, pos, rs.MESO_TABLE))
+            pos = pos + rs._structure_size(rs.MESO_TABLE)
+
+        # 中气旋特征表
+        meso_feature_tab=[]
+        for nn in np.arange(dic_meso_header['feature_number']):
+            meso_feature_tab.append(rs._unpack_from_buf(buf, pos, rs.MESO_FEATURE_TAB))
+            pos = pos + rs._structure_size(rs.MESO_FEATURE_TAB)
+
+        # 中气旋适配数据
+        dic_meso_adapt_param = rs._unpack_from_buf(buf, pos, rs.MESO_ADAPTATION_DATA)
+        pos = pos + rs._structure_size(rs.MESO_ADAPTATION_DATA)
+
+
+        
+        for nn in np.arange(dic_meso_header['meso_number']):
+            # 将所有的方位角和距离转换成经纬度
+            x,y,z = geotrans.antenna_to_cartesian(meso_tab[nn]['range']/1000.0,meso_tab[nn]['azi'],0)
+            clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+            meso_tab[nn]['lon'] = clon[0]
+            meso_tab[nn]['lat'] = clat[0]
+            
+            # 将数字序号转换成字母数字组合
+            meso_tab[nn]['storm_id_char'] = self.get_id_char(meso_tab[nn]['storm_id'])
+            meso_tab[nn]['feature_id_char'] = self.get_id_char(meso_tab[nn]['feature_id'])
+            meso_tab[nn]['feature_type'] = meso_feature_tab[nn]['feature_type']
+            
+        for nn in np.arange(dic_meso_header['feature_number']):
+            # 将所有的方位角和距离转换成经纬度
+            x,y,z = geotrans.antenna_to_cartesian(meso_feature_tab[nn]['range']/1000.0,meso_feature_tab[nn]['azi'],0)
+            clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+            meso_feature_tab[nn]['lon'] = clon[0]
+            meso_feature_tab[nn]['lat'] = clat[0]
+            
+            # 将数字序号转换成字母数字组合
+            meso_feature_tab[nn]['storm_id_char'] = self.get_id_char(meso_feature_tab[nn]['storm_id'])
+            meso_feature_tab[nn]['feature_id_char'] = self.get_id_char(meso_feature_tab[nn]['feature_id'])
+
+
+        allresult={}
+        allresult['meso'] = meso_tab
+        allresult['feature'] = meso_feature_tab
+        
+        self.mesoinfo = allresult
+        return allresult
+
+    # 解析龙卷涡旋特征产品
+    def read_tvs(self,filepath,filename):
+        '''
+        解析tvs产品文件
+        '''
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        # pprint(prod_type)
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 获取产品数据块
+
+        #获取TVS头信息
+        dic_tvs_header = rs._unpack_from_buf(buf, pos, rs.TVS_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.TVS_HEADER_BLOCK)
+
+        # TVS表块
+        tvs_tab=[]
+        for nn in np.arange(dic_tvs_header['tvs_number']):
+            tvs_tab.append(rs._unpack_from_buf(buf, pos, rs.TVS_TAB))
+            pos = pos + rs._structure_size(rs.TVS_TAB)
+
+        # TVS适配数据
+        dic_tvs_adapt_param = rs._unpack_from_buf(buf, pos, rs.TVS_ADAPTATION_DATA)
+        pos = pos + rs._structure_size(rs.TVS_ADAPTATION_DATA)
+
+
+        # 将所有的方位角和距离转换成经纬度
+        for nn in np.arange(dic_tvs_header['tvs_number']):
+            pass
+            x,y,z = geotrans.antenna_to_cartesian(tvs_tab[nn]['range']/1000.0,tvs_tab[nn]['azi'],0)
+            clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+            tvs_tab[nn]['lon'] = clon[0]
+            tvs_tab[nn]['lat'] = clat[0]
+            
+            
+            # 将数字序号转换成字母数字组合
+            tvs_tab[nn]['storm_id_char'] = self.get_id_char(tvs_tab[nn]['storm_id'])
+
+
+
+        allresult={}
+        allresult['tvs'] = tvs_tab
+        
+        self.tvsinfo = allresult
+        return allresult
+    
+    # 解析ppi产品文件
+    def read_ppi(self,filepath,filename):
+        '''
+        解析ppi产品文件
+        '''
+        pass
+
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+            return None
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 读取产品径向头信息
+        dic_radial_header = rs._unpack_from_buf(buf, pos, rs.RADIAL_HEADER)
+        pos = pos + rs._structure_size(rs.RADIAL_HEADER)
+
+        print('数据类型: %d-%s'%(dic_radial_header['data_type'],rs.PRODUCT_DATA_TYPE[dic_radial_header['data_type']]))
+        # Value=(Code-Offset)/Scale
+        # 读取径向数据RADIAL_DATA
+        data=[]
+        data_azi=[]
+        binnum = None
+        for nn in np.arange(dic_radial_header['radial_number']):
+            radial_data_info = rs._unpack_from_buf(buf, pos, rs.RADIAL_DATA)
+            pos = pos + rs._structure_size(rs.RADIAL_DATA)
+            binnum = radial_data_info['num_bins']
+            rdata = buf[pos:pos+dic_radial_header['bin_length']*radial_data_info['num_bins']]
+            data.extend(rdata)
+            pos = pos + dic_radial_header['bin_length']*radial_data_info['num_bins']
+            # print(radial_data_info['start_azi'])
+            data_azi.append(radial_data_info['start_azi'])
+            pass
+        data = np.array(data)
+       
+        
+        # data = data.astype('uint8')
+        data = data.astype('float32')
+        nx = dic_radial_header['radial_number']
+        ny = binnum
+        print(nx,ny)
+        # 将data转成numpy数组
+        data = data.reshape((nx,ny))
+
+        # decode data
+        # ppiarray = (ppiarray - dic_radial_header['offset']) / dic_radial_header['scale'] 
+
+        
+        # 将方位角坐标转换成经纬度坐标
+        sweep_azimuths =np.array(data_azi)
+        ele = dic_prod_param['ele']
+        range_reso = dic_radial_header['resolution']
+        # 输出的距离库只保留一半
+        ngates = int(ny//2*2) #//2*2 
+        grid_x = np.arange(-1*ngates,ngates+1,1)
+        grid_y = np.arange(-1*ngates,ngates+1,1)
+        total_gates = len(grid_x)
+
+        aa = np.meshgrid(grid_x,grid_y)
+        azi_grid = np.arctan2(aa[0],aa[1])*180/np.pi
+        azi_grid[azi_grid<0]+=360
+
+        azi_reso = 360/len(sweep_azimuths)
+
+        # 求方位角索引
+        new_azi = azi_grid.flatten()
+        t = new_azi-sweep_azimuths[0]
+        t[t<0]+=360
+        ray_number = np.round(t/azi_reso,0).astype(int)
+        ray_number[ray_number==len(sweep_azimuths)]=0
+        ray_number = np.reshape(ray_number,(total_gates,total_gates))
+
+        # 求距离索引
+        dis_grid = np.sqrt(aa[0]**2 + aa[1]**2)
+        dis_grid = np.round(dis_grid.flatten(),0).astype(int)
+        dis_grid = np.reshape(dis_grid,(total_gates,total_gates))
+
+        # 对数据进行截断，在径向方向上
+        data = data[:,0:ngates]
+
+        # data_grid = np.zeros((total_gates,total_gates),dtype='uint8') + 255
+        data_grid = np.zeros((total_gates,total_gates),dtype='float32')
+        new_data = data_grid.flatten()
+        new_spdata = data.flatten()
+        
+        pos_out = [i+j*total_gates for i in range(total_gates) for j in range(total_gates) if dis_grid[i,j] < ngates]
+        d_out = [rn*ngates+dg for rn,dg in zip(ray_number.flatten(),dis_grid.flatten()) if dg < ngates]
+
+        new_data[pos_out]=new_spdata[d_out]
+       
+        data_grid = np.reshape(new_data,(total_gates,total_gates))
+       
+
+        if total_gates % 2 == 0:
+            out_grid = np.arange(int(-total_gates/2),int(total_gates/2))
+        else:
+            out_grid = np.arange(int(-(total_gates-1)/2),int((total_gates-1)/2)+1)
+        # 将ougrid转换为经纬度坐标
+        out_lon,out_lat = geotrans.cartesian_to_geographic_aeqd(out_grid*range_reso,out_grid*range_reso,dic_scfg['lon'],dic_scfg['lat'])
+        outdata = (data_grid - dic_radial_header['offset']) / dic_radial_header['scale'] 
+        minvalue = (dic_radial_header['min_value'] - dic_radial_header['offset']) / dic_radial_header['scale']
+        maxvalue = (dic_radial_header['max_value'] - dic_radial_header['offset']) / dic_radial_header['scale']
+        # outdata[outdata > 90] = -32
+        # outdata[outdata < -10] = -32
+        outdata[outdata >= maxvalue] = -9999
+        outdata[outdata <= minvalue] = -9999
+        varname = 'ref'
+        units = 'dBZ'
+        if dic_radial_header['data_type'] == 7:
+            varname = 'zdr'
+            units = 'dB'
+        elif dic_radial_header['data_type'] == 9:
+            varname = 'cc'
+            units = '%'
+        elif dic_radial_header['data_type'] == 11:
+            varname = 'kdp'
+            units = 'deg/km'
+            pass
+        data = xr.DataArray(np.array(outdata.transpose()),coords=[out_lat,out_lon],dims=['lat','lon'],name=varname)
+        data.attrs['units'] = units
+        # data.attrs['standard_name'] = 'equivalent_reflectivity_factor'
+        # data.attrs['long_name'] = 'equivalent_reflectivity_factor'
+        data.attrs['varname'] = varname
+        data.attrs['radar_lat'] = dic_scfg['lat']
+        data.attrs['radar_lon'] = dic_scfg['lon']
+        data.attrs['ana_height'] = dic_scfg['ana_height']
+        data.attrs['grid_num'] = total_gates
+        data.attrs['grid_reso'] = range_reso
+        data.attrs['elevation'] = ele
+        data.attrs['obs_range'] = int(range_reso * (total_gates-1)/2)
+        data.attrs['distance_unit'] = 'meter'
+        data.attrs['missing_value'] = -9999
+        data.attrs['datatype'] = 'float32'
+        # data.attrs['decode_method'] = 'dbz = (data - %d) / %d'%(dic_radial_header['offset'],dic_radial_header['scale'])
+        data.attrs['task_name'] = dic_tcfg['task_name'].decode('utf-8').strip('\x00')
+        data.attrs['radar_type'] = dic_scfg['radar_type']
+        data.attrs['scan_time'] = datetime.fromtimestamp(dic_tcfg['scan_stime']).strftime('%Y-%m-%d %H:%M:%S')
+        try:
+            data.attrs['site_name'] = dic_scfg['site_name'].decode('utf-8').strip('\x00')
+        except:
+            data.attrs['site_name'] = 'Unknown'
+        data.attrs['site_id'] = dic_scfg['site_code'].decode('utf-8').strip('\x00')
+        # data.attrs['offset'] = dic_radial_header['offset']
+        # data.attrs['scale'] = dic_radial_header['scale']
+
+        return data
+
+    # 解析cr产品文件，栅格数据
+    def read_cr(self,filepath,filename):
+        '''
+        解析cr产品文件
+        '''
+        pass
+
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+            return None
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        print('产品类型: %d-%s'%(dic_prod_header['product_type'],prod_type))
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 读取产品径向头信息
+        dic_grid_header = rs._unpack_from_buf(buf, pos, rs.GRID_HEADER)
+        pos = pos + rs._structure_size(rs.GRID_HEADER)
+        print('数据类型: %d-%s'%(dic_grid_header['data_type'],rs.PRODUCT_DATA_TYPE[dic_grid_header['data_type']]))
+        # Value=(Code-Offset)/Scale
+        # 读取栅格数据GRID_DATA
+        
+        if dic_grid_header['bin_length'] == 1:
+            data1 = np.frombuffer(buf[pos:pos + dic_grid_header['row_side_len']*dic_grid_header['col_side_len']*dic_grid_header['bin_length']], '>u1')
+            data1 = data1.astype('uint8')
+        data2 = np.reshape(data1,(dic_grid_header['row_side_len'],dic_grid_header['col_side_len']))
+        # data2[data2<dic_grid_header['min_value']]=255
+        # 根据offset和scale进行解码
+        data2 = (data2 - dic_grid_header['offset']) / dic_grid_header['scale']
+        data2[data2 > 90] = -32
+        data2[data2 < -10] = -32
+        nlat = dic_grid_header['row_side_len']
+        nlon = dic_grid_header['col_side_len']
+        # print(nlat,nlon)
+        lat_reso = dic_grid_header['row_resolution']
+        lon_reso = dic_grid_header['col_resolution']
+
+        # 将data由list转成二维numpy数组
+
+        
+        # 将直角坐标转换成经纬度坐标
+       
+        if nlat % 2 == 0:
+            out_grid_lat = np.arange(int(-nlat/2),int(nlat/2))
+        else:
+            out_grid_lat = np.arange(int(-(nlat-1)/2),int((nlat-1)/2)+1)
+        
+        if nlon % 2 == 0:
+            out_grid_lon = np.arange(int(-nlon/2),int(nlon/2))
+        else:
+            out_grid_lon = np.arange(int(-(nlon-1)/2),int((nlon-1)/2)+1)
+
+        # 将ougrid转换为经纬度坐标
+        out_lon,out_lat = geotrans.cartesian_to_geographic_aeqd(out_grid_lat*lat_reso,out_grid_lon*lon_reso,dic_scfg['lon'],dic_scfg['lat'])
+
+        data = xr.DataArray(np.flipud(data2),coords=[out_lat,out_lon],dims=['lat','lon'],name='cref')
+        data.attrs['units'] = 'dBZ'
+        data.attrs['standard_name'] = 'composite_reflectivity_factor'
+        data.attrs['long_name'] = 'composite_reflectivity_factor'
+        data.attrs['radar_lat'] = dic_scfg['lat']
+        data.attrs['radar_lon'] = dic_scfg['lon']
+        data.attrs['ana_height'] = dic_scfg['ana_height']
+        data.attrs['lat_grid_num'] = nlat
+        data.attrs['lon_grid_num'] = nlon
+        data.attrs['lat_grid_reso'] = dic_grid_header['row_resolution']
+        data.attrs['lon_grid_reso'] = dic_grid_header['col_resolution']
+        data.attrs['distance_unit'] = 'meter'
+        data.attrs['missing_value'] = -32
+        data.attrs['datatype'] = 'float32'
+        # data.attrs['decode_method'] = 'dbz = (data - %d) / %d'%(dic_grid_header['offset'],dic_grid_header['scale'])
+        data.attrs['task_name'] = dic_tcfg['task_name'].decode('utf-8').strip('\x00')
+        data.attrs['radar_type'] = dic_scfg['radar_type']
+        data.attrs['scan_time'] = datetime.fromtimestamp(dic_tcfg['scan_stime']).strftime('%Y-%m-%d %H:%M:%S')
+        try:
+            data.attrs['site_name'] = dic_scfg['site_name'].decode('utf-8').strip('\x00')
+        except:
+            data.attrs['site_name'] = 'Unknown'
+        data.attrs['site_id'] = dic_scfg['site_code'].decode('utf-8').strip('\x00')
+
+        return data
+
+    # 解析vil产品文件，栅格数据
+    def read_vil(self,filepath,filename):
+        '''
+        解析vil产品文件
+        '''
+        pass
+
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        print('产品类型: %d-%s'%(dic_prod_header['product_type'],prod_type))
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 读取产品径向头信息
+        dic_grid_header = rs._unpack_from_buf(buf, pos, rs.GRID_HEADER)
+        pos = pos + rs._structure_size(rs.GRID_HEADER)
+        print('数据类型: %d-%s'%(dic_grid_header['data_type'],rs.PRODUCT_DATA_TYPE[dic_grid_header['data_type']]))
+        # Value=(Code-Offset)/Scale
+        # 读取栅格数据GRID_DATA
+        
+        if dic_grid_header['bin_length'] == 1:
+            data1 = np.frombuffer(buf[pos:pos + dic_grid_header['row_side_len']*dic_grid_header['col_side_len']*dic_grid_header['bin_length']], '>u1')
+            data1 = data1.astype('uint8')
+        data2 = np.reshape(data1,(dic_grid_header['row_side_len'],dic_grid_header['col_side_len']))
+        # data2[data2<dic_grid_header['min_value']]=255
+        # 根据offset和scale进行解码
+        data2 = (data2 - dic_grid_header['offset']) / dic_grid_header['scale']
+        data2[data2 > 120] =  0
+        data2[data2 < 0] = 0
+        nlat = dic_grid_header['row_side_len']
+        nlon = dic_grid_header['col_side_len']
+        # print(nlat,nlon)
+        lat_reso = dic_grid_header['row_resolution']
+        lon_reso = dic_grid_header['col_resolution']
+
+        # 将data由list转成二维numpy数组
+
+        
+        # 将直角坐标转换成经纬度坐标
+       
+        if nlat % 2 == 0:
+            out_grid_lat = np.arange(int(-nlat/2),int(nlat/2))
+        else:
+            out_grid_lat = np.arange(int(-(nlat-1)/2),int((nlat-1)/2)+1)
+        
+        if nlon % 2 == 0:
+            out_grid_lon = np.arange(int(-nlon/2),int(nlon/2))
+        else:
+            out_grid_lon = np.arange(int(-(nlon-1)/2),int((nlon-1)/2)+1)
+
+        # 将ougrid转换为经纬度坐标
+        out_lon,out_lat = geotrans.cartesian_to_geographic_aeqd(out_grid_lat*lat_reso,out_grid_lon*lon_reso,dic_scfg['lon'],dic_scfg['lat'])
+
+        data = xr.DataArray(np.flipud(data2),coords=[out_lat,out_lon],dims=['lat','lon'],name='vil')
+        data.attrs['units'] = 'kg/m^2'
+        data.attrs['standard_name'] = 'vertically_integrated_liquid_water'
+        data.attrs['long_name'] = 'vertically_integrated_liquid_water'
+        data.attrs['radar_lat'] = dic_scfg['lat']
+        data.attrs['radar_lon'] = dic_scfg['lon']
+        data.attrs['ana_height'] = dic_scfg['ana_height']
+        data.attrs['lat_grid_num'] = nlat
+        data.attrs['lon_grid_num'] = nlon
+        data.attrs['lat_grid_reso'] = dic_grid_header['row_resolution']
+        data.attrs['lon_grid_reso'] = dic_grid_header['col_resolution']
+        data.attrs['distance_unit'] = 'meter'
+        data.attrs['missing_value'] = 0
+        data.attrs['datatype'] = 'float32'
+        # data.attrs['decode_method'] = 'dbz = (data - %d) / %d'%(dic_grid_header['offset'],dic_grid_header['scale'])
+        data.attrs['task_name'] = dic_tcfg['task_name'].decode('utf-8').strip('\x00')
+        data.attrs['radar_type'] = dic_scfg['radar_type']
+        data.attrs['scan_time'] = datetime.fromtimestamp(dic_tcfg['scan_stime']).strftime('%Y-%m-%d %H:%M:%S')
+        try:
+            data.attrs['site_name'] = dic_scfg['site_name'].decode('utf-8').strip('\x00')
+        except:
+            data.attrs['site_name'] = 'Unknown'
+        data.attrs['site_id'] = dic_scfg['site_code'].decode('utf-8').strip('\x00')
+
+        return data
+    
+    # 解析tops产品文件，栅格数据，回波顶高
+    def read_tops(self,filepath,filename):
+        '''
+        解析tops产品文件
+        '''
+        pass
+
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        print('产品类型: %d-%s'%(dic_prod_header['product_type'],prod_type))
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 读取产品径向头信息
+        dic_grid_header = rs._unpack_from_buf(buf, pos, rs.GRID_HEADER)
+        pos = pos + rs._structure_size(rs.GRID_HEADER)
+        print('数据类型: %d-%s'%(dic_grid_header['data_type'],rs.PRODUCT_DATA_TYPE[dic_grid_header['data_type']]))
+        # Value=(Code-Offset)/Scale
+        # 读取栅格数据GRID_DATA
+        
+        if dic_grid_header['bin_length'] == 1:
+            data1 = np.frombuffer(buf[pos:pos + dic_grid_header['row_side_len']*dic_grid_header['col_side_len']*dic_grid_header['bin_length']], '>u1')
+            data1 = data1.astype('uint8')
+        data2 = np.reshape(data1,(dic_grid_header['row_side_len'],dic_grid_header['col_side_len']))
+        # data2[data2<dic_grid_header['min_value']]=255
+        # 根据offset和scale进行解码
+        
+        
+        
+        data2[data2 >  dic_grid_header['max_value']] =  0
+        data2 = (data2 - dic_grid_header['offset']) / dic_grid_header['scale']
+        maxv = (dic_grid_header['max_value'] - dic_grid_header['offset']) / dic_grid_header['scale']
+        data2[data2 > maxv] = 0
+
+        nlat = dic_grid_header['row_side_len']
+        nlon = dic_grid_header['col_side_len']
+        # print(nlat,nlon)
+        lat_reso = dic_grid_header['row_resolution']
+        lon_reso = dic_grid_header['col_resolution']
+
+        # 将data由list转成二维numpy数组
+
+        
+        # 将直角坐标转换成经纬度坐标
+       
+        if nlat % 2 == 0:
+            out_grid_lat = np.arange(int(-nlat/2),int(nlat/2))
+        else:
+            out_grid_lat = np.arange(int(-(nlat-1)/2),int((nlat-1)/2)+1)
+        
+        if nlon % 2 == 0:
+            out_grid_lon = np.arange(int(-nlon/2),int(nlon/2))
+        else:
+            out_grid_lon = np.arange(int(-(nlon-1)/2),int((nlon-1)/2)+1)
+
+        # 将ougrid转换为经纬度坐标
+        out_lon,out_lat = geotrans.cartesian_to_geographic_aeqd(out_grid_lat*lat_reso,out_grid_lon*lon_reso,dic_scfg['lon'],dic_scfg['lat'])
+
+        data = xr.DataArray(np.flipud(data2),coords=[out_lat,out_lon],dims=['lat','lon'],name='et')
+        data.attrs['units'] = 'km'
+        data.attrs['standard_name'] = 'echo_top'
+        data.attrs['long_name'] = 'echo_top'
+        data.attrs['radar_lat'] = dic_scfg['lat']
+        data.attrs['radar_lon'] = dic_scfg['lon']
+        data.attrs['ana_height'] = dic_scfg['ana_height']
+        data.attrs['lat_grid_num'] = nlat
+        data.attrs['lon_grid_num'] = nlon
+        data.attrs['lat_grid_reso'] = dic_grid_header['row_resolution']
+        data.attrs['lon_grid_reso'] = dic_grid_header['col_resolution']
+        data.attrs['distance_unit'] = 'kilometer'
+        data.attrs['missing_value'] = 0
+        data.attrs['datatype'] = 'float32'
+        # data.attrs['decode_method'] = 'dbz = (data - %d) / %d'%(dic_grid_header['offset'],dic_grid_header['scale'])
+        data.attrs['task_name'] = dic_tcfg['task_name'].decode('utf-8').strip('\x00')
+        data.attrs['radar_type'] = dic_scfg['radar_type']
+        data.attrs['scan_time'] = datetime.fromtimestamp(dic_tcfg['scan_stime']).strftime('%Y-%m-%d %H:%M:%S')
+        try:
+            data.attrs['site_name'] = dic_scfg['site_name'].decode('utf-8').strip('\x00')
+        except:
+            data.attrs['site_name'] = 'Unknown'
+        data.attrs['site_id'] = dic_scfg['site_code'].decode('utf-8').strip('\x00')
+
+        return data
+    
+
+    # 解析stp产品文件，栅格数据
+    def read_stp(self,filepath,filename):
+        '''
+        解析stp产品文件
+        '''
+        pass
+
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        print('产品类型: %d-%s'%(dic_prod_header['product_type'],prod_type))
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 读取产品径向头信息
+        dic_radial_header = rs._unpack_from_buf(buf, pos, rs.RADIAL_HEADER)
+        pos = pos + rs._structure_size(rs.RADIAL_HEADER)
+        print('数据类型: %d-%s'%(dic_radial_header['data_type'],rs.PRODUCT_DATA_TYPE[dic_radial_header['data_type']]))
+        # Value=(Code-Offset)/Scale
+        # 读取径向数据RADIAL_DATA
+        data=[]
+        data_azi=[]
+        binnum = None
+        for nn in np.arange(dic_radial_header['radial_number']):
+            radial_data_info = rs._unpack_from_buf(buf, pos, rs.RADIAL_DATA)
+            pos = pos + rs._structure_size(rs.RADIAL_DATA)
+            binnum = radial_data_info['num_bins']
+            # rdata = buf[pos:pos+dic_radial_header['bin_length']*radial_data_info['num_bins']]
+            rdata = np.frombuffer(buf[pos:pos+dic_radial_header['bin_length']*radial_data_info['num_bins']], '<u2')
+            data.extend(rdata)
+            pos = pos + dic_radial_header['bin_length']*radial_data_info['num_bins']
+            # print(radial_data_info['start_azi'])
+            data_azi.append(radial_data_info['start_azi'])
+            pass
+        data = np.array(data)
+        # data[data > dic_radial_header['max_value']] = 0
+        # data[data < dic_radial_header['min_value']] = 0
+        
+        data = data.astype('uint16')
+        nx = dic_radial_header['radial_number']
+        ny = binnum
+        print(nx,ny)
+        # 将data转成numpy数组
+        data = data.reshape((nx,ny))
+
+        # 将方位角坐标转换成经纬度坐标
+        sweep_azimuths =np.array(data_azi)
+
+        range_reso = dic_radial_header['resolution']
+        # 输出的距离库只保留一半
+        ngates = int(ny//2*2) #//2*2 
+        grid_x = np.arange(-1*ngates,ngates+1,1)
+        grid_y = np.arange(-1*ngates,ngates+1,1)
+        total_gates = len(grid_x)
+
+        aa = np.meshgrid(grid_x,grid_y)
+        azi_grid = np.arctan2(aa[0],aa[1])*180/np.pi
+        azi_grid[azi_grid<0]+=360
+
+        azi_reso = 360/len(sweep_azimuths)
+
+        # 求方位角索引
+        new_azi = azi_grid.flatten()
+        t = new_azi-sweep_azimuths[0]
+        t[t<0]+=360
+        ray_number = np.round(t/azi_reso,0).astype(int)
+        ray_number[ray_number==len(sweep_azimuths)]=0
+        ray_number = np.reshape(ray_number,(total_gates,total_gates))
+
+        # 求距离索引
+        dis_grid = np.sqrt(aa[0]**2 + aa[1]**2)
+        dis_grid = np.round(dis_grid.flatten(),0).astype(int)
+        dis_grid = np.reshape(dis_grid,(total_gates,total_gates))
+
+        # 对数据进行截断，在径向方向上
+        data = data[:,0:ngates]
+
+        data_grid = np.zeros((total_gates,total_gates),dtype='uint16') + 0
+        new_data = data_grid.flatten()
+        new_spdata = data.flatten()
+        # decode data
+        new_spdata = (new_spdata - dic_radial_header['offset']) / dic_radial_header['scale'] 
+        pos_out = [i+j*total_gates for i in range(total_gates) for j in range(total_gates) if dis_grid[i,j] < ngates]
+        d_out = [rn*ngates+dg for rn,dg in zip(ray_number.flatten(),dis_grid.flatten()) if dg < ngates]
+
+        new_data[pos_out]=new_spdata[d_out]
+       
+        data_grid = np.reshape(new_data,(total_gates,total_gates))
+
+
+        if total_gates % 2 == 0:
+            out_grid = np.arange(int(-total_gates/2),int(total_gates/2))
+        else:
+            out_grid = np.arange(int(-(total_gates-1)/2),int((total_gates-1)/2)+1)
+        # 将ougrid转换为经纬度坐标
+        out_lon,out_lat = geotrans.cartesian_to_geographic_aeqd(out_grid*range_reso,out_grid*range_reso,dic_scfg['lon'],dic_scfg['lat'])
+
+        data = xr.DataArray(np.array(data_grid.transpose()),coords=[out_lat,out_lon],dims=['lat','lon'],name='stp')
+        data.attrs['units'] = 'mm'
+        data.attrs['standard_name'] = 'storm total precipitation'
+        data.attrs['long_name'] = 'storm total precipitation'
+        data.attrs['radar_lat'] = dic_scfg['lat']
+        data.attrs['radar_lon'] = dic_scfg['lon']
+        data.attrs['ana_height'] = dic_scfg['ana_height']
+        data.attrs['grid_num'] = total_gates
+        data.attrs['grid_reso'] = range_reso
+        data.attrs['obs_range'] = int(range_reso * (total_gates-1)/2)
+        data.attrs['distance_unit'] = 'meter'
+        data.attrs['missing_value'] = 0
+        data.attrs['datatype'] = 'uint16'
+        # data.attrs['decode_method'] = 'qpe = (data - %d) / %d'%(dic_radial_header['offset'],dic_radial_header['scale'])
+        data.attrs['task_name'] = dic_tcfg['task_name'].decode('utf-8').strip('\x00')
+        data.attrs['radar_type'] = dic_scfg['radar_type']
+        data.attrs['scan_time'] = datetime.fromtimestamp(dic_tcfg['scan_stime']).strftime('%Y-%m-%d %H:%M:%S')
+        try:
+            data.attrs['site_name'] = dic_scfg['site_name'].decode('utf-8').strip('\x00')
+        except:
+            data.attrs['site_name'] = 'Unknown'
+        data.attrs['site_id'] = dic_scfg['site_code'].decode('utf-8').strip('\x00')
+        # data.attrs['offset'] = dic_radial_header['offset']
+        # data.attrs['scale'] = dic_radial_header['scale']
+
+        return data
+
+    # 解析ohp产品文件，栅格数据
+    def read_ohp(self,filepath,filename):
+        '''
+        解析ohp产品文件
+        '''
+        pass
+
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+            return None
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        print('产品类型: %d-%s'%(dic_prod_header['product_type'],prod_type))
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 读取产品径向头信息
+        dic_radial_header = rs._unpack_from_buf(buf, pos, rs.RADIAL_HEADER)
+        pos = pos + rs._structure_size(rs.RADIAL_HEADER)
+        print('数据类型: %d-%s'%(dic_radial_header['data_type'],rs.PRODUCT_DATA_TYPE[dic_radial_header['data_type']]))
+        # Value=(Code-Offset)/Scale
+        # 读取径向数据RADIAL_DATA
+        data=[]
+        data_azi=[]
+        binnum = None
+        for nn in np.arange(dic_radial_header['radial_number']):
+            radial_data_info = rs._unpack_from_buf(buf, pos, rs.RADIAL_DATA)
+            pos = pos + rs._structure_size(rs.RADIAL_DATA)
+            binnum = radial_data_info['num_bins']
+            # rdata = buf[pos:pos+dic_radial_header['bin_length']*radial_data_info['num_bins']]
+            rdata = np.frombuffer(buf[pos:pos+dic_radial_header['bin_length']*radial_data_info['num_bins']], '<u2')
+            data.extend(rdata)
+            pos = pos + dic_radial_header['bin_length']*radial_data_info['num_bins']
+            # print(radial_data_info['start_azi'])
+            data_azi.append(radial_data_info['start_azi'])
+            pass
+        data = np.array(data)
+        data[data > dic_radial_header['max_value']] = 0
+        data[data < dic_radial_header['min_value']] = 0
+        
+        data = data.astype('uint16')
+        nx = dic_radial_header['radial_number']
+        ny = binnum
+        print(nx,ny)
+        # 将data转成numpy数组
+        data = data.reshape((nx,ny))
+
+        # decode data
+        # ppiarray = (ppiarray - dic_radial_header['offset']) / dic_radial_header['scale'] 
+
+        
+        # 将方位角坐标转换成经纬度坐标
+        sweep_azimuths =np.array(data_azi)
+
+        range_reso = dic_radial_header['resolution']
+        # 输出的距离库只保留一半
+        ngates = int(ny//2*2) #//2*2 
+        grid_x = np.arange(-1*ngates,ngates+1,1)
+        grid_y = np.arange(-1*ngates,ngates+1,1)
+        total_gates = len(grid_x)
+
+        aa = np.meshgrid(grid_x,grid_y)
+        azi_grid = np.arctan2(aa[0],aa[1])*180/np.pi
+        azi_grid[azi_grid<0]+=360
+
+        azi_reso = 360/len(sweep_azimuths)
+
+        # 求方位角索引
+        new_azi = azi_grid.flatten()
+        t = new_azi-sweep_azimuths[0]
+        t[t<0]+=360
+        ray_number = np.round(t/azi_reso,0).astype(int)
+        ray_number[ray_number==len(sweep_azimuths)]=0
+        ray_number = np.reshape(ray_number,(total_gates,total_gates))
+
+        # 求距离索引
+        dis_grid = np.sqrt(aa[0]**2 + aa[1]**2)
+        dis_grid = np.round(dis_grid.flatten(),0).astype(int)
+        dis_grid = np.reshape(dis_grid,(total_gates,total_gates))
+
+        # 对数据进行截断，在径向方向上
+        data = data[:,0:ngates]
+
+        data_grid = np.zeros((total_gates,total_gates),dtype='uint16') + 0
+        new_data = data_grid.flatten()
+        new_spdata = data.flatten()
+
+        new_spdata = (new_spdata - dic_radial_header['offset']) / dic_radial_header['scale'] 
+
+        pos_out = [i+j*total_gates for i in range(total_gates) for j in range(total_gates) if dis_grid[i,j] < ngates]
+        d_out = [rn*ngates+dg for rn,dg in zip(ray_number.flatten(),dis_grid.flatten()) if dg < ngates]
+
+        new_data[pos_out]=new_spdata[d_out]
+       
+        data_grid = np.reshape(new_data,(total_gates,total_gates))
+       
+
+        if total_gates % 2 == 0:
+            out_grid = np.arange(int(-total_gates/2),int(total_gates/2))
+        else:
+            out_grid = np.arange(int(-(total_gates-1)/2),int((total_gates-1)/2)+1)
+        # 将ougrid转换为经纬度坐标
+        out_lon,out_lat = geotrans.cartesian_to_geographic_aeqd(out_grid*range_reso,out_grid*range_reso,dic_scfg['lon'],dic_scfg['lat'])
+
+        data = xr.DataArray(np.array(data_grid.transpose()),coords=[out_lat,out_lon],dims=['lat','lon'],name='ohp')
+        data.attrs['units'] = 'mm'
+        data.attrs['standard_name'] = 'one hour precipitation'
+        data.attrs['long_name'] = 'one hour precipitation'
+        data.attrs['radar_lat'] = dic_scfg['lat']
+        data.attrs['radar_lon'] = dic_scfg['lon']
+        data.attrs['ana_height'] = dic_scfg['ana_height']
+        data.attrs['grid_num'] = total_gates
+        data.attrs['grid_reso'] = range_reso
+        data.attrs['obs_range'] = int(range_reso * (total_gates-1)/2)
+        data.attrs['distance_unit'] = 'meter'
+        data.attrs['missing_value'] = 0
+        data.attrs['datatype'] = 'uint16'
+        # data.attrs['decode_method'] = 'qpe = (data - %d) / %d'%(dic_radial_header['offset'],dic_radial_header['scale'])
+        data.attrs['task_name'] = dic_tcfg['task_name'].decode('utf-8').strip('\x00')
+        data.attrs['radar_type'] = dic_scfg['radar_type']
+        data.attrs['scan_time'] = datetime.fromtimestamp(dic_tcfg['scan_stime']).strftime('%Y-%m-%d %H:%M:%S')
+        try:
+            data.attrs['site_name'] = dic_scfg['site_name'].decode('utf-8').strip('\x00')
+        except:
+            data.attrs['site_name'] = 'Unknown'
+        data.attrs['site_id'] = dic_scfg['site_code'].decode('utf-8').strip('\x00')
+        # data.attrs['offset'] = dic_radial_header['offset']
+        # data.attrs['scale'] = dic_radial_header['scale']
+
+        return data
+
+    
+    # 解析冰雹指数产品
+    def read_hda(self,filepath,filename):
+        '''
+        解析hail产品文件
+        '''
+        fin = open(filepath + os.sep + filename,'rb')
+        buf = fin.read()
+        fin.close()
+        buf_length = len(buf)
+        pos = 0
+
+        # 获取通用信息头
+        dic_gh = rs._unpack_from_buf(buf, pos, rs.GENERIC_HEADER)
+        # pprint(dic_gh)
+        if dic_gh['magic_number'] != 0x4D545352:
+            print('源数据格式错误！')
+        pos = pos + rs._structure_size(rs.GENERIC_HEADER)
+
+        # 获取站点信息
+        dic_scfg = rs._unpack_from_buf(buf, pos, rs.SITE_CONFIG)
+        pos = pos + rs._structure_size(rs.SITE_CONFIG)
+
+        # 获取任务信息
+        dic_tcfg = rs._unpack_from_buf(buf, pos, rs.TASK_CONFIG)
+        pos = pos + rs._structure_size(rs.TASK_CONFIG)
+
+        # 获取扫描信息
+        cutinfo = []
+        for im in np.arange(dic_tcfg['cut_number']):
+            dic_cutcfg = rs._unpack_from_buf(buf, pos, rs.SCAN_CONFIG)
+            cutinfo.append(dic_cutcfg)
+            pos = pos + rs._structure_size(rs.SCAN_CONFIG)
+
+        # 获取产品头信息
+        dic_prod_header = rs._unpack_from_buf(buf, pos, rs.PRODUCT_HEADER_BLOCK)
+        pos = pos + rs._structure_size(rs.PRODUCT_HEADER_BLOCK)
+
+        # 获取产品参数信息
+        prod_type = rs.PRODUCT_ID_NAME_MAP[dic_prod_header['product_type']]
+        # pprint(prod_type)
+        dic_prod_param = rs._unpack_from_buf(buf, pos, rs.PRODUCT_DEPENDENT_PARAMETER[prod_type])
+        pos = pos + 64 # 产品参数长度固定为64个
+
+        # 获取产品数据块
+        # 冰雹个数
+        hail_number = rs._unpack_from_buf(buf, pos, rs.HAIL_NUMBER)
+        pos = pos + rs._structure_size(rs.HAIL_NUMBER)
+
+        # TVS表块
+        hail_tab=[]
+        for nn in np.arange(hail_number['hail_number']):
+            hail_tab.append(rs._unpack_from_buf(buf, pos, rs.HAIL_TAB))
+            pos = pos + rs._structure_size(rs.HAIL_TAB)
+
+        # TVS适配数据
+        dic_hail_adapt_param = rs._unpack_from_buf(buf, pos, rs.HAIL_ADAPTATION_DATA)
+        pos = pos + rs._structure_size(rs.HAIL_ADAPTATION_DATA)
+
+        # 将所有的方位角和距离转换成经纬度
+        for nn in np.arange(hail_number['hail_number']):
+            pass
+            x,y,z = geotrans.antenna_to_cartesian(hail_tab[nn]['range']/1000.0,hail_tab[nn]['azi'],0)
+            clon,clat = geotrans.cartesian_to_geographic_aeqd(x,y,dic_scfg['lon'],dic_scfg['lat'])
+            hail_tab[nn]['lon'] = clon[0]
+            hail_tab[nn]['lat'] = clat[0]
+
+            # 将数字序号转换成字母数字组合
+            hail_tab[nn]['hail_id_char'] = self.get_id_char(hail_tab[nn]['hail_id'])
+
+
+        allresult={}
+        allresult['hail'] = hail_tab
+
+        self.hailinfo = allresult
+        return allresult
+    
+    # 将风暴追踪产品转换为gr2格式
+    def trans_sti_gr2(self,outpath_gr2=None,outname_gr2=None,m_pret=None,time_type='UTC',bhistory=False,bdebug=False):
+        
+        if self.stiinfo is None:
+            print('mesoinfo is None, please run read_sti first!')
+            return False
+
+        # try:
+        # duration 代表该时次产品的可视时间段
+        if not os.path.exists(outpath_gr2):
+            os.makedirs(outpath_gr2)
+
+
+        if outpath_gr2 is None or outname_gr2 is None:
+            print('outpath_gr2 and outname_gr2 params should be set')
+            return False
+
+        if not os.path.exists(outpath_gr2):
+            try:
+                os.makedirs(outpath_gr2)
+            except:
+                print(outpath_gr2 + ' not exist, and fail to create this path!')
+                return False
+
+        of = open(outpath_gr2 + os.sep + outname_gr2,'wt',encoding='GBK')
+
+        # 单个文件就不用信息头了
+        if bdebug and not bhistory:
+            of.write('Title: ROSE2风暴追踪信息\n')
+            of.write('Refreshseconds: 30\n')
+            of.write('Threshold: 500\n')
+            of.write('Font: 1, 20, 1, "Courier New"\n')
+            # of.write('IconFile:  1, 18, 18, 9, 9, 2016_144dpi.png\n')
+            of.write('IconFile:  1, 24, 24, 12, 12, 2016_192dpi.png\n')
+
+        # 这里直接用m_pret 可以保证不会出现重叠的记录
+        if not bdebug:
+            if not m_pret is None and bhistory:
+                pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                endtime = m_pret + timedelta(minutes=1*int(6/2))
+            elif not bhistory:
+                pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                endtime = m_pret + timedelta(minutes=30)
+                pass
+            
+            
+            timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+
+            # 实时模式下，改时间一定要放到timerange后面
+            if time_type == "BJT":
+                pre_time = pre_time + timedelta(hours=8)
+                endtime = endtime + timedelta(hours=8)
+
+            if bhistory:
+                timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+                
+            of.write(timerangestr + '\n')
+
+        for dd in self.stiinfo['track']:
+            if len(dd)  > 1:
+                of.write('Line: 2,0\n')     
+                for nn in dd:
+                    of.write('%.4f,%.4f\n'%(nn[0],nn[1])) 
+                of.write('End:\n\n')
+
+        for dd in self.stiinfo['marker_past']:
+            of.write('Icon: %.4f,%.4f,0,1,88\n'%(dd[0],dd[1]))
+        for dd in self.stiinfo['marker_fst']:
+            of.write('Icon: %.4f,%.4f,0,1,91\n'%(dd[0],dd[1]))
+
+        for nn in np.arange(len(self.stiinfo['all_storm_prop'])):
+
+            curstorm = self.stiinfo['all_storm_prop'][nn]
+            clon = curstorm['lon']
+            clat = curstorm['lat']
+
+            lbstr=r'数据来源: ROSE风暴识别产品\n'
+            lbstr += r'风暴编号: %s\n'%curstorm['id_char']
+            lbstr += r'方位角/距离: %.1f度/%.1f公里\n'%(curstorm['azi'],curstorm['range']/1000.0)
+            lbstr += r'风暴移动方向 : %s\n'%(self.get_wind_dir_name((curstorm['mv_dir']+180)%360))
+            lbstr += r'风暴移动速度 : %.1f 公里/小时\n'%(curstorm['mv_spd']*3.6)
+            
+            lbstr += r'风暴低高: %.1f 公里\n'%(curstorm['base_height']/1000)
+            lbstr += r'风暴顶高: %.1f 公里\n'%(curstorm['top_height']/1000)
+            lbstr += r'风暴体VIL: %.1f 千克/平方米\n'%(curstorm['vil'])
+            lbstr += r'最大反射率: %d dBZ\n'%(curstorm['max_ref'])
+            lbstr += r'最大反射率高度 : %.1f 公里\n'%(curstorm['max_ref_height']/1000)        
+            
+            of.write('Icon: %.4f,%.4f,0,1,40,"%s"\n'%(clat,clon,lbstr))
+
+            
+        of.write('End:\n\n')
+
+        of.close()
+        print(outpath_gr2 + os.sep + outname_gr2 + ' saved!')
+        return outpath_gr2 + os.sep + outname_gr2
+        # except:
+        #     return False
+    
+
+    # 将中气旋的信息转换为gr2格式
+    def trans_mda_gr2(self,outpath_gr2=None,outname_gr2=None,m_pret=None,time_type='UTC',bhistory=False,bdebug=False):
+        if self.mesoinfo is None:
+            print('mesoinfo is None, please run read_mda first!')
+            return False
+        try:
+            # duration 代表该时次产品的可视时间段
+            if not os.path.exists(outpath_gr2):
+                os.makedirs(outpath_gr2)
+
+
+            if outpath_gr2 is None or outname_gr2 is None:
+                print('outpath_gr2 and outname_gr2 params should be set')
+                return False
+
+            if not os.path.exists(outpath_gr2):
+                try:
+                    os.makedirs(outpath_gr2)
+                except:
+                    print(outpath_gr2 + ' not exist, and fail to create this path!')
+                    return False
+
+            of = open(outpath_gr2 + os.sep + outname_gr2,'wt',encoding='GBK')
+
+            # 单个文件就不用信息头了
+            if bdebug and not bhistory:
+                of.write('Title: ROSE2中气旋识别产品\n')
+                of.write('Refreshseconds: 30\n')
+                of.write('Threshold: 500\n')
+                of.write('Font: 1, 20, 1, "Courier New"\n')
+                # of.write('IconFile:  1, 47, 47, 23, 23, mda_icons_144.png\n')
+                of.write('IconFile:  1, 63, 63, 31, 31, mda_icons_192.png\n')
+
+            # 这里直接用m_pret 可以保证不会出现重叠的记录
+            if not bdebug:
+                if not m_pret is None and bhistory:
+                    pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                    endtime = m_pret + timedelta(minutes=1*int(6/2))
+                elif not bhistory:
+                    pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                    endtime = m_pret + timedelta(minutes=30)
+                    pass
+                
+                
+                timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+
+                # 实时模式下，改时间一定要放到timerange后面
+                if time_type == "BJT":
+                    pre_time = pre_time + timedelta(hours=8)
+                    endtime = endtime + timedelta(hours=8)
+
+                if bhistory:
+                    timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+                    
+                of.write(timerangestr + '\n')
+
+            for nn in range(len(self.mesoinfo['meso'])):
+                curitem = self.mesoinfo['meso'][nn]
+                clon = curitem['lon']
+                clat = curitem['lat']
+                lbstr=''
+                icon_index=0
+                baseindex=0
+                if curitem['base'] > 1500: # 米
+                    baseindex = 0
+                else:
+                    baseindex = 1*4
+                '''
+                关于风切变的单位换算问题
+                https://www.theweatherprediction.com/habyhints/193/
+                The wind speed is 20 knots at 850 mb while the wind speed is 30 knots at 700 mb. The distance between 850 and 700 mb is 1,600 meters. What is the value of wind shear?
+
+                The change in wind speed is 30 - 20 = 10 knots = 5 m/s
+                The change is distance is 1,600 m
+
+                The change in wind speed over distance = 5/1,600 s^-1 = 0.003125 s^-1
+
+                '''
+                shear_ms = curitem['ave_shear'] * (curitem['azi_diam'] + curitem['radial_diam'])/2/1000.0/2
+                mrank = self.get_mda_rank(curitem['range']/1000,shear_ms)
+
+                if mrank is None:
+                    continue
+                else:
+                    icon_index = baseindex + mrank
+
+                mlevel =''
+                if mrank==1:
+                    mlevel = '弱切变'
+                elif mrank == 2:
+                    mlevel = '弱中气旋'
+                elif mrank == 3:
+                    mlevel = '中等强度中气旋'
+                elif mrank == 4:
+                    mlevel = '强中气旋'
+
+                mtype='未知'
+                if curitem['feature_type'] == 1:
+                    mtype = '中气旋'
+                elif curitem['feature_type'] == 2:
+                    mtype = '3D相关切变'
+                elif curitem['feature_type'] == 3:
+                    mtype = '非相关切变'
+
+                lbstr=''
+                lbstr += r'数据来源: ROSE2中气旋识别产品\n'
+                lbstr += r'所属风暴编号: %s\n'%curitem['storm_id_char']
+                lbstr += r'中气旋编号: %s\n'%curitem['feature_id_char']
+                lbstr += r'类型: %s\n'%mtype
+                lbstr += r'中气旋所在方位/距离: %.1f度 / %.1f 公里\n'%(curitem['azi'],curitem['range']/1000.0)
+                lbstr += r'强度等级: %s\n'%mlevel
+                lbstr += r'中气旋底高: %.1f 公里\n'%(curitem['base']/1000.0)
+                lbstr += r'中气旋顶高: %.1f 公里\n'%(curitem['top']/1000.0)
+                lbstr += r'中气旋沿径向直径: %.1f 公里\n'%(curitem['radial_diam']/1000.0)
+                lbstr += r'中气旋沿方位直径: %.1f 公里\n'%(curitem['azi_diam']/1000.0)
+                lbstr += r'平均切变值: %.1f (E-3/S)\n'%(curitem['ave_shear'])
+
+                if shear_ms > 25:
+                    lbstr += r'===============提醒============\n换算后的切变值超过了25米/秒\n可能是很强的中气旋或者是错误数据\n可能是存在速度模糊对PUP算法的影响。\n'
+
+                of.write('Icon: %.4f,%.4f,0,1,%d,"%s"\n'%(clat,clon,icon_index, lbstr))
+                pass
+            of.write('End:\n\n')
+
+            of.close()
+            print(outpath_gr2 + os.sep + outname_gr2 + ' saved!')
+            return outpath_gr2 + os.sep + outname_gr2
+        except:
+            return False
+        
+
+    # 将tvs的信息转换为gr2格式
+    def trans_tvs_gr2(self,outpath_gr2=None,outname_gr2=None,m_pret=None,time_type='UTC',bhistory=False,bdebug=False):
+        
+        if self.tvsinfo is None:
+            print('tvsinfo is None, please run read_tvs  first')
+            return False
+        
+        try:
+            # duration 代表该时次产品的可视时间段
+            if not os.path.exists(outpath_gr2):
+                os.makedirs(outpath_gr2)
+
+
+            if outpath_gr2 is None or outname_gr2 is None:
+                print('outpath_gr2 and outname_gr2 params should be set')
+                return False
+
+            if not os.path.exists(outpath_gr2):
+                try:
+                    os.makedirs(outpath_gr2)
+                except:
+                    print(outpath_gr2 + ' not exist, and fail to create this path!')
+                    return False
+
+            of = open(outpath_gr2 + os.sep + outname_gr2,'wt',encoding='GBK')
+
+            # 单个文件就不用信息头了
+            if bdebug and not bhistory:
+                of.write('Title: ROSE2龙卷涡旋特征识别产品\n')
+                of.write('Refreshseconds: 30\n')
+                of.write('Threshold: 500\n')
+                of.write('Font: 1, 20, 1, "Courier New"\n')
+                of.write('IconFile:  1, 51, 51, 26, 26, lsr_icons_192.png\n')
+
+            # 这里直接用m_pret 可以保证不会出现重叠的记录
+            if not bdebug:
+                if not m_pret is None and bhistory:
+                    pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                    endtime = m_pret + timedelta(minutes=1*int(6/2))
+                elif not bhistory:
+                    pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                    endtime = m_pret + timedelta(minutes=30)
+                    pass
+                
+                
+                timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+
+                # 实时模式下，改时间一定要放到timerange后面
+                if time_type == "BJT":
+                    pre_time = pre_time + timedelta(hours=8)
+                    endtime = endtime + timedelta(hours=8)
+
+                if bhistory:
+                    timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+                    
+                of.write(timerangestr + '\n')
+
+            for nn in range(len(self.tvsinfo['tvs'])):
+                curitem = self.tvsinfo['tvs'][nn]
+                clon = curitem['lon']
+                clat = curitem['lat']
+                curtype = '未知'
+                if curitem['type'] == 1:
+                    curtype = 'TVS'
+                elif curitem['type'] == 2:
+                    curtype = 'ETVS'
+                icon_index=1
+                
+                lbstr=''
+                lbstr += r'数据来源: ROSE2龙卷涡旋识别产品\n'
+                lbstr += r'所属风暴编号: %s\n'%curitem['storm_id_char']
+                lbstr += r'类型: %s\n'%curtype
+                lbstr += r'TVS所在方位/距离: %.1f度 / %.1f 公里\n'%(curitem['azi'],curitem['range']/1000.0)
+
+                lbstr += r'TVS底高: %.1f 公里\n'%(curitem['base']/1000.0)
+                lbstr += r'TVS顶高: %.1f 公里\n'%(curitem['top']/1000.0)
+                lbstr += r'TVS厚度: %.1f 公里\n'%(curitem['depth']/1000.0)
+                lbstr += r'低层速度差LLDV: %.1f 米/秒\n'%(curitem['lldv']) # *1.609344/3.6
+                lbstr += r'最大速度差MXDV: %.1f 米/秒\n'%(curitem['mxdv']) # *1.609344/3.6
+                lbstr += r'最大速度差所在高度: %.1f 公里\n'%(curitem['hmdv']/1000.0)
+
+                lbstr += r'最大切变值MASHR: %.1f (E-3/S)\n'%(curitem['max_shear'])
+                lbstr += r'最大切变值所在高度: %.1f 公里\n'%(curitem['h_max_shear']/1000.0)
+            
+                if curtype=='ETVS':
+                    icon_index=20
+                    of.write('Icon: %.4f,%.4f,0,1,%d,"%s"\n'%(clat,clon,icon_index, lbstr))
+                else:
+                    icon_index=11
+                    of.write('Icon: %.4f,%.4f,0,1,%d,"%s"\n'%(clat,clon,icon_index, lbstr))
+
+            of.write('End:\n\n')
+
+            of.close()
+            print(outpath_gr2 + os.sep + outname_gr2 + ' saved!')
+            return outpath_gr2 + os.sep + outname_gr2
+        except:
+            return False
+        
+
+    # 将hail的信息转换为gr2格式
+    def trans_hda_gr2(self,outpath_gr2=None,outname_gr2=None,m_pret=None,time_type='UTC',bhistory=False,bdebug=False):
+
+        if self.hailinfo is None:
+            print('hailinfo is None, please run read_hda first!')
+            return False
+
+        try:
+            # duration 代表该时次产品的可视时间段
+            if not os.path.exists(outpath_gr2):
+                os.makedirs(outpath_gr2)
+
+
+            if outpath_gr2 is None or outname_gr2 is None:
+                print('outpath_gr2 and outname_gr2 params should be set')
+                return False
+
+            if not os.path.exists(outpath_gr2):
+                try:
+                    os.makedirs(outpath_gr2)
+                except:
+                    print(outpath_gr2 + ' not exist, and fail to create this path!')
+                    return False
+
+            of = open(outpath_gr2 + os.sep + outname_gr2,'wt',encoding='GBK')
+
+            # 单个文件就不用信息头了
+            if bdebug and not bhistory:
+                of.write('Title: ROSE2冰雹指数产品\n')
+                of.write('Refreshseconds: 30\n')
+                of.write('Threshold: 500\n')
+                of.write('Font: 1, 20, 1, "Courier New"\n')
+                # of.write('IconFile:  1, 37, 37, 18, 18, hail_icons_144.png\n')
+                of.write('IconFile:  1, 51, 51, 26, 26, hail_icons_192.png\n')
+
+            # 这里直接用m_pret 可以保证不会出现重叠的记录
+            if not bdebug:
+                if not m_pret is None and bhistory:
+                    pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                    endtime = m_pret + timedelta(minutes=1*int(6/2))
+                elif not bhistory:
+                    pre_time = m_pret + timedelta(minutes=-1*int(6/2))
+                    endtime = m_pret + timedelta(minutes=30)
+                    pass
+                
+                
+                timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+
+                # 实时模式下，改时间一定要放到timerange后面
+                if time_type == "BJT":
+                    pre_time = pre_time + timedelta(hours=8)
+                    endtime = endtime + timedelta(hours=8)
+
+                if bhistory:
+                    timerangestr = 'TimeRange: ' + pre_time.strftime('%Y-%m-%dT%H:%M:%S') + ' ' + endtime.strftime('%Y-%m-%dT%H:%M:%S')
+                    
+                of.write(timerangestr + '\n')
+
+            for nn in range(len(self.hailinfo['hail'])):
+                curitem = self.hailinfo['hail'][nn]
+                clon = curitem['lon']
+                clat = curitem['lat']
+                
+                lbstr=''
+                lbstr=r'数据来源: ROSE2冰雹指数产品\n'
+                lbstr += r'冰雹编号: %s\n'%curitem['hail_id_char']
+                lbstr += r'大冰雹概率: %d%%\n'%curitem['psh']
+                lbstr += r'冰雹概率: %d%%\n'%curitem['ph']
+                lbstr += r'最大冰雹尺寸估计: %.1f 厘米 或 %.1f 英寸\n'%(curitem['hsize'],curitem['hsize']/2.54,)
+                
+                icon_index=0
+                hsize_inch = curitem['hsize']/2.54
+                
+                if hsize_inch < 1:
+                    icon_index = 1
+                elif hsize_inch < 2:
+                    icon_index = 5
+                elif hsize_inch < 3:
+                    icon_index = 9
+                elif hsize_inch < 4:
+                    icon_index = 13
+                elif hsize_inch >= 4:
+                    icon_index = 17
+                else:
+                    icon_index = 1
+                if curitem['ph'] >= 30:
+                    of.write('Icon: %.4f,%.4f,0,1,%d,"%s"\n'%(clat,clon,icon_index, lbstr))
+
+            of.write('End:\n\n')
+
+            of.close()
+            print(outpath_gr2 + os.sep + outname_gr2 + ' saved!')
+            return outpath_gr2 + os.sep + outname_gr2
+        except:
+            return False
+        
+if __name__ == "__main__":
+    rootpath = '/Users/wenjianzhu/Library/CloudStorage/OneDrive-个人/PythonCode/MyWork/nmc_met_radar/testdata/rose_product'
+    staid = 'Z9852'
+    # filepath = 'D:\Downloads\CMADAAS\pup\Z9852\STI'
+    outpath_gr2 = 'output/gr2data/sti' + os.sep + staid
+    if not os.path.exists(outpath_gr2):
+        os.makedirs(outpath_gr2)
+
+    decoder = READ_ROSE()
+
+    # # ppi zdr
+    # filepath_ppi = '/Users/wenjianzhu/Downloads/CMADAAS/pup/Z9852/SS'
+
+    # # ss
+    # filepath_ss = rootpath + os.sep + staid + os.sep + 'SS'
+    # filename_ss = 'Z_RADR_I_Z9852_20230421164954_P_DOR_CDD_SS_NUL_300_NUL_FMT.bin'
+    # allresult_ss = decoder.read_ss(filepath_ss, filename_ss)
+    
+
+    # sti
+    filepath_sti = rootpath + os.sep + staid + os.sep + 'STI'
+    filename_sti = 'Z_RADR_I_Z9852_20230421164954_P_DOR_CDD_STI_NUL_300_NUL_FMT.bin'
+    allresult_sti = decoder.read_sti(filepath_sti, filename_sti)
+    outname_sti = 'pls_sti.txt'
+    decoder.trans_sti_gr2(outpath_gr2,outname_sti,time_type='BJT',bdebug = True)
+    
+    
+    # meso
+    filepath_meso = rootpath + os.sep + staid + os.sep + 'M'
+    filename_meso = 'Z_RADR_I_Z9852_20230421135042_P_DOR_CDD_M_NUL_200_NUL_FMT.bin'
+    allresult_meso = decoder.read_mda(filepath_meso, filename_meso)
+    outname_meso = 'pls_meso.txt'
+    decoder.trans_mda_gr2(outpath_gr2,outname_meso,time_type='BJT',bdebug = True)
+
+    # tvs
+    filepath_tvs = rootpath + os.sep + staid + os.sep + 'TVS'
+    filename_tvs = 'Z_RADR_I_Z9852_20230421201451_P_DOR_CDD_TVS_NUL_200_NUL_FMT.bin'
+    allresult_tvs = decoder.read_tvs(filepath_tvs, filename_tvs)
+    outname_tvs = 'pls_tvs.txt'
+    decoder.trans_tvs_gr2(outpath_gr2,outname_tvs,time_type='BJT',bdebug = True)
+
+
+    # hail
+    filepath_hail = rootpath + os.sep + staid + os.sep + 'HI'
+    filename_hail = 'Z_RADR_I_Z9852_20230421181642_P_DOR_CDD_HI_NUL_200_NUL_FMT.bin'
+    allresult_hail = decoder.read_hda(filepath_hail, filename_hail)
+    outname_hail = 'pls_hail.txt'
+    decoder.trans_hda_gr2(outpath_gr2,outname_hail,time_type='BJT',bdebug = True)
+
+# %%