geoloop 0.0.1__py3-none-any.whl → 1.0.0__py3-none-any.whl

geoloop/plotting/create_plots.py

@@ -0,0 +1,1142 @@
+import re
+from pathlib import Path
+
+import matplotlib.animation as animation
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+import pygfunction as gt
+import seaborn as sns
+import xarray as xr
+from PIL import Image
+
+from geoloop.bin.SingleRunSim import SingleRun
+from geoloop.constants import format_dict, units_dict
+from geoloop.geoloopcore.strat_interpolator import StratInterpolator
+from geoloop.utils.helpers import get_param_names
+
+
+class PlotResults:
+    """
+    A class with functions for generating plots of simulation results. Different plots can be created for different
+    types of simulations and models (e.g. single simulations, stochastic simulation, FINVOL model, ANALYTICAL model).
+
+    """
+
+    @staticmethod
+    def create_scatterplots(
+        results_df: pd.DataFrame,
+        params_df: pd.DataFrame,
+        y_variable: str,
+        out_path: Path,
+    ) -> None:
+        """
+        Only compatible with results and inputs of stochastic (MC) simulations.
+        Generates and saves scatter plots for all parameters in the simulations results, with `y_variable` on the y-axis.
+        The parameters in the list of 'y_variable' are plotted against all other parameters in the results and input
+        dataframes.
+
+        Parameters
+        ----------
+        results_df : pd.DataFrame
+            DataFrame containing simulation result parameters as columns.
+        params_df : pd.DataFrame
+            DataFrame containing simulation input parameters as columns.
+        y_variable : str
+            The variable parameter to be plotted on the y-axis.
+        out_path : Path
+            Path to the directory where the plots will be saved.
+
+        Raises
+        ------
+        ValueError
+            If `y_variable` is not found in either `results_df` or `param_df`.
+
+        Returns
+        -------
+        None
+        """
+        if y_variable in results_df.columns:
+            y_data = results_df
+        elif y_variable in params_df.columns:
+            y_data = params_df
+        else:
+            raise ValueError(
+                f"{y_variable} not found in either results_df or param_df."
+            )
+
+        # Get the list of all variables excluding the y_variable
+        result_variables = [
+            "Q_b",
+            "flowrate",
+            "T_fi",
+            "T_fo",
+            "T_bave",
+            "dploop",
+            "qloop",
+            "T_b",
+            "qzb",
+            "T_f",
+            "Re_in",
+            "Re_out",
+            "k_s_res",
+            "k_g_res",
+        ]
+        variable_param_names, locked_param_names = get_param_names()
+
+        x_variables = result_variables + variable_param_names
+
+        # Loop through variables and create scatterplots
+        for x_variable in x_variables:
+            if x_variable != y_variable and x_variable in results_df.columns:
+                # only plot the variables that are truly variable (stdev != 0)
+                if results_df[x_variable].std().round(7) != 0:
+                    title = f"{format_dict[x_variable]} ({units_dict[x_variable]}) after {results_df.loc[0, 'hours']} hours production vs. {format_dict[y_variable]} ({units_dict[y_variable]})"
+                    g = sns.JointGrid(
+                        results_df, x=x_variable, y=y_data[y_variable], hue="file_name"
+                    )
+                    g.plot_joint(
+                        sns.scatterplot, alpha=0.7, edgecolor=".2", linewidth=0.5
+                    )
+                    g.plot_marginals(
+                        sns.boxplot,
+                        linewidth=0.5,
+                        linecolor=".2",
+                        boxprops=dict(alpha=0.9),
+                    )
+                    sns.set_style("whitegrid")
+                    g.ax_joint.legend(bbox_to_anchor=(0.63, -0.15))
+
+                    # Reverse y-axis if y_variable is "H"
+                    if y_variable == "H":
+                        g.ax_joint.invert_yaxis()
+
+                    # Plot vertical line with turbulent Re value if x_variable is either 'Re_in' or 'Re_out'
+                    if x_variable in ["Re_in", "Re_out"]:
+                        g.ax_joint.axvline(
+                            4000, color="red", linestyle="--", label="Re turbulent"
+                        )
+                        g.ax_joint.legend()  # Ensure the legend is updated
+                    g.set_axis_labels(
+                        xlabel=f"{format_dict[x_variable]} ({units_dict[x_variable]})",
+                        ylabel=f"{format_dict[y_variable]} ({units_dict[y_variable]})",
+                    )
+                    g.fig.suptitle(title)
+                    g.fig.tight_layout()
+                    save_path = out_path.with_name(
+                        out_path.name + f"_{y_variable}vs{x_variable}_scat.png"
+                    )
+                    plt.savefig(save_path)
+                    plt.close()
+
+            elif (
+                x_variable != y_variable
+                and x_variable
+                in params_df.select_dtypes(include=["float64", "int64"]).columns
+            ):
+                if params_df[x_variable].std().round(7) != 0:
+                    title = f"{format_dict[x_variable]} ({units_dict[x_variable]}) after {params_df.loc[0, 'nyear']} year production vs. {format_dict[y_variable]} ({units_dict[y_variable]})"
+                    g = sns.JointGrid(
+                        params_df, x=x_variable, y=y_data[y_variable], hue="file_name"
+                    )
+                    g.plot_joint(
+                        sns.scatterplot, alpha=0.7, edgecolor=".2", linewidth=0.5
+                    )
+                    g.plot_marginals(
+                        sns.boxplot,
+                        linewidth=0.5,
+                        linecolor=".2",
+                        boxprops=dict(alpha=0.9),
+                    )
+                    sns.set_style("whitegrid")
+                    g.ax_joint.legend(bbox_to_anchor=(0.63, -0.15))
+
+                    # Reverse y-axis if y_variable is "H"
+                    if y_variable == "H":
+                        g.ax_joint.invert_yaxis()
+
+                    g.set_axis_labels(
+                        xlabel=f"{format_dict[x_variable]} ({units_dict[x_variable]})",
+                        ylabel=f"{format_dict[y_variable]} ({units_dict[y_variable]})",
+                    )
+                    g.fig.suptitle(title)
+                    g.fig.tight_layout()
+                    save_path = out_path.with_name(
+                        out_path.name + f"_{y_variable}vs{x_variable}_scat.png"
+                    )
+                    plt.savefig(save_path)
+                    plt.close()
+
+    @staticmethod
+    def create_timeseriesplot(
+        results_dfs: pd.DataFrame | list[pd.DataFrame],
+        out_path: Path,
+        plot_parameters: list,
+    ):
+        """
+        Only compatible with (multiple) single simulations.
+        Generates and saves a timeseries plot for various simulation results over time. Results of multiple
+        single simulations can be plotted together.
+
+        Plot shows the generated power, flowrate, fluid inlet en outlet temperatures, depth-average borehole wall temperature,
+        pumping pressure and required pumping power over time.
+
+        Parameters
+        ----------
+        results_dfs : Union[pd.DataFrame, List[pd.DataFrame]]
+            DataFrame(s) containing simulation results.
+        out_path : Path
+            Directory where plots are saved.
+        plot_parameters : List[str]
+            List of variables to plot.
+
+        Raises
+        ------
+        ValueError
+            If any required parameter is missing from a DataFrame in `results_dfs`.
+
+        Returns
+        -------
+        None
+        """
+        out_path_prefix = out_path.with_name(out_path.name + "_timeplot")
+
+        # Ensure results_dfs is a list of dataframes
+        if isinstance(results_dfs, pd.DataFrame):
+            results_dfs = [results_dfs]
+
+        figsize = (11, 4)
+        plt.rcParams.update({"font.size": 12})
+
+        # Define labels for each parameter
+        labels = {}
+        for idx, df in enumerate(results_dfs):
+            file_name = df["file_name"].iloc[0]
+            labels[idx] = {
+                "T_fi": f"{file_name}, $T_{{in}}$ [°C]",
+                "T_fo": f"{file_name}, $T_{{out}}$ [°C]",
+                "T_bave": f"{file_name}, $T_{{b,ave}}$ [°C]",
+                "Delta_T": f"{file_name}, Δ Temperature (Tout - Tin)",
+                "Q_b": f"{file_name}, Heat Load [W]",
+                "dploop": f"{file_name}, Pump Pressure [bar]",
+                "qloop": f"{file_name}, Pump Power [W]",
+                "flowrate": f"{file_name}, Flowrate [kg/s]",
+                "COP": f"{file_name}, Coefficient of Performance",
+            }
+
+        # Create a global color palette for consistency across dataframes
+        line_colors = sns.color_palette(
+            "colorblind", n_colors=len(results_dfs) * len(plot_parameters) * 3
+        )
+        color_iter = iter(line_colors)
+
+        # Check if any dataframe needs COP or Delta_T calculated
+        for df in results_dfs:
+            if (
+                "COP" in plot_parameters
+                and "Q_b" in df.columns
+                and "qloop" in df.columns
+            ):
+                df['COP'] = abs(df['Q_b']) / df['qloop']
+            if (
+                "Delta_T" in plot_parameters
+                and "T_fi" in df.columns
+                and "T_fo" in df.columns
+            ):
+                df["Delta_T"] = df["T_fo"] - df["T_fi"]
+
+        plots = [
+            ("Q_b", "Heat Load [W]", "Q_b", None, None),
+            ("flowrate", "Flowrate [kg/s]", "flowrate", None, None),
+            ("COP", "Coefficient of Performance", "COP", None, None)
+        ]
+
+        plotted_params = []
+        extra_artists = []
+        # Plot dploop and qloop, even if only one is present
+        if "dploop" in plot_parameters or "qloop" in plot_parameters:
+            fig, ax = plt.subplots(figsize=figsize)
+            ax.set_xlabel(r"$t$ [hours]")
+            primary_axis_used = False
+
+            if "dploop" in plot_parameters:
+                ax.set_ylabel("Pump Pressure [bar]")
+                for idx, df in enumerate(results_dfs):
+                    if "dploop" in df.columns:
+                        sns.lineplot(
+                            x="time",
+                            y="dploop",
+                            data=df,
+                            ax=ax,
+                            label=labels[idx]["dploop"],
+                            color=next(color_iter),
+                        )
+                        primary_axis_used = True
+                        plotted_params.append("dploop")
+
+                ax.legend(loc=(-0.07, -0.35))
+
+            for idx, df in enumerate(results_dfs):
+                if "qloop" in plot_parameters and "qloop" in df.columns:
+
+                    ax_twin = ax.twinx()
+                    ax_twin.set_ylabel("Pump Power [W]")
+
+                    color = next(color_iter)
+
+                    # Plot the line
+                    sns.lineplot(
+                        x="time",
+                        y="qloop",
+                        data=df,
+                        ax=ax_twin,
+                        label=labels[idx]["qloop"],
+                        linestyle="dotted",
+                        color=color,
+                    )
+
+                    # Match axis color to line color
+                    ax_twin.spines["right"].set_color(color)
+                    ax_twin.yaxis.label.set_color(color)
+                    ax_twin.tick_params(axis="y", colors=color)
+
+                    plotted_params.append("qploop")
+                    ax_twin.legend(loc=(0.53, -0.15 - (len(results_dfs) * 0.1)))
+                    extra_artists.append(ax_twin.legend_)
+                if primary_axis_used:
+                    ax.legend(loc=(-0.07, -0.15 - (len(results_dfs) * 0.1)))
+                    extra_artists.append(ax.legend_)
+
+            ax.grid()
+            if not primary_axis_used:
+                ax.set_yticklabels([])
+                ax.set_ylabel("")
+                ax.grid(False, axis="y")
+                ax_twin.grid(axis="both")
+
+            file_name = out_path.with_name(
+                out_path.name + f"_timeplot_{'_'.join(sorted(set(plotted_params)))}.png"
+            )
+
+            fig.tight_layout()
+            plt.savefig(
+                file_name,
+                dpi=300,
+                bbox_extra_artists=(extra_artists),
+                bbox_inches="tight",
+            )
+            plt.close()
+
+        # Plot T_fi, T_fo, T_bave together, and Delta_T on the secondary axis if applicable
+        plotted_params = []
+        extra_artists = []
+        if (
+            any(param in plot_parameters for param in ["T_fi", "T_fo", "T_bave"])
+            or "Delta_T" in plot_parameters
+        ):
+            fig, ax = plt.subplots(figsize=figsize)
+            ax.set_xlabel(r"$t$ [hours]")
+            primary_axis_used = False
+
+            if any(param in plot_parameters for param in ["T_fi", "T_fo", "T_bave"]):
+                ax.set_ylabel("Temperature [°C]")
+                for idx, df in enumerate(results_dfs):
+                    if "T_fi" in plot_parameters and "T_fi" in df.columns:
+                        sns.lineplot(
+                            x="time",
+                            y="T_fi",
+                            data=df,
+                            ax=ax,
+                            label=labels[idx]["T_fi"],
+                            color=next(color_iter),
+                            linestyle="-",
+                        )
+                        primary_axis_used = True
+                        plotted_params.append("T_fi")
+                    if "T_fo" in plot_parameters and "T_fo" in df.columns:
+                        sns.lineplot(
+                            x="time",
+                            y="T_fo",
+                            data=df,
+                            ax=ax,
+                            label=labels[idx]["T_fo"],
+                            color=next(color_iter),
+                            linestyle="--",
+                        )
+                        primary_axis_used = True
+                        plotted_params.append("T_fo")
+                    if "T_bave" in plot_parameters and "T_bave" in df.columns:
+                        sns.lineplot(
+                            x="time",
+                            y="T_bave",
+                            data=df,
+                            ax=ax,
+                            label=labels[idx]["T_bave"],
+                            color=next(color_iter),
+                            linestyle=":",
+                        )
+                        primary_axis_used = True
+                        plotted_params.append("T_bave")
+                    ax.legend(
+                        loc="lower left",
+                        bbox_to_anchor=(
+                            0.04,
+                            0.02
+                            - (0.05 * 1.25 * len(plotted_params) / len(results_dfs)),
+                        ),
+                        bbox_transform=fig.transFigure,
+                    )
+                    extra_artists.append(ax.legend_)
+
+            if "Delta_T" in plot_parameters:
+                color = next(color_iter)
+
+                ax_twin = ax.twinx()
+                ax_twin.set_ylabel("Δ Temperature (Tout - Tin)")
+                for idx, df in enumerate(results_dfs):
+                    sns.lineplot(
+                        x="time",
+                        y="Delta_T",
+                        data=df,
+                        ax=ax_twin,
+                        label=labels[idx]["Delta_T"],
+                        linestyle="dashdot",
+                        color=color,
+                    )
+                    plotted_params.append("Delta_T")
+
+                # Match axis color to line color
+                ax_twin.spines["right"].set_color(color)
+                ax_twin.yaxis.label.set_color(color)
+                ax_twin.tick_params(axis="y", colors=color)
+
+                ax_twin.legend(
+                    loc="lower right",
+                    bbox_to_anchor=(1.1, -0.25 - (0.08 * len(results_dfs))),
+                )
+                extra_artists.append(ax_twin.legend_)
+                if primary_axis_used:
+                    ax.legend(
+                        loc="lower left",
+                        bbox_to_anchor=(
+                            0.03,
+                            0.1 - (0.02 * len(plotted_params) / len(results_dfs)),
+                        ),
+                        bbox_transform=fig.transFigure,
+                    )
+                    extra_artists.append(ax.legend_)
+
+            ax.grid()
+
+            if not primary_axis_used:
+                ax.set_yticklabels([])
+                ax.set_ylabel("")
+                ax.grid(False, axis="y")
+                ax_twin.grid(axis="both")
+
+            file_name = out_path.with_name(
+                out_path.name
+                + "_timeplot_"
+                + "_".join(sorted(set(plotted_params)))
+                + ".png"
+            )
+
+            fig.tight_layout()
+            plt.savefig(
+                file_name,
+                dpi=300,
+                bbox_extra_artists=(extra_artists),
+                bbox_inches="tight",
+            )
+            plt.close()
+
+        # Plot remaining parameters
+        for param, ylabel, filename, secondary_param, secondary_ylabel in plots:
+            if param not in plot_parameters:
+                continue
+
+            fig, ax = plt.subplots(figsize=figsize)
+            ax.set_xlabel(r"$t$ [hours]")
+            ax.set_ylabel(ylabel)
+
+            for idx, df in enumerate(results_dfs):
+                if param in df.columns:
+                    sns.lineplot(
+                        x="time",
+                        y=param,
+                        data=df,
+                        ax=ax,
+                        label=labels[idx][param],
+                        color=next(color_iter),
+                    )
+                    if param=="COP":
+                        if param == "COP":
+                            ax.text(
+                                0.7, 0.9,  # 2% from left and bottom of the axes
+                                f"mean COP: {df['COP'].mean():.2f}",
+                                transform=ax.transAxes,
+                                bbox=dict(boxstyle="round,pad=0.3", facecolor="white", alpha=0.7),
+                                fontsize=12
+                            )
+
+            ax.legend(loc=(-0.07, -0.17 - (0.08 * len(results_dfs))))
+            ax.grid()
+            fig.tight_layout()
+            plt.savefig(f"{out_path_prefix}_{filename}.png", dpi=300)
+            plt.close()
+
+    @staticmethod
+    def create_depthplot(
+        plot_parameters: list,
+        times: list,
+        params_ds: xr.Dataset | list[xr.Dataset],
+        results_ds: xr.Dataset | list[xr.Dataset],
+        out_path: Path,
+    ):
+        """
+        Only compatible with (multiple) single simulations.
+        Generates and saves depth-profiles at different timesteps in the simulation results, for fluid temperatures,
+        borehole wall temperature, subsurface temperature, subsurface heat flow and subsurface bulk thermal conductivity.
+
+        Parameters
+        ----------
+        plot_parameters : List[str]
+            List of parameters to plot.
+        times : List[float]
+            Simulation timesteps for which profiles are plotted.
+        params_ds : Union[xr.Dataset, List[xr.Dataset]]
+            Dataset(s) containing simulation input parameters.
+        results_ds : Union[xr.Dataset, List[xr.Dataset]]
+            Dataset(s) containing simulation results.
+        out_path : Path
+            Directory to save plots.
+
+        Raises
+        ------
+        ValueError
+            If any required parameter is missing from a DataFrame in `results_ds` or `params_ds`.
+
+        Returns
+        -------
+        None
+        """
+        method = "nearest"
+        ax_colors = sns.color_palette("pastel", n_colors=10)
+        axtwin_colors = sns.color_palette("dark", n_colors=10)
+        plt.rcParams.update(
+            {"font.size": 16, "xtick.labelsize": 16, "ytick.labelsize": 16}
+        )
+
+        if not isinstance(params_ds, list):
+            params_ds = [params_ds]
+            results_ds = [results_ds]
+
+        for time in times:
+            fig1, ax1 = plt.subplots(
+                figsize=(8, 10)
+            )  # First plot: Temperature profiles
+            fig2, ax2 = plt.subplots(
+                figsize=(8, 10)
+            )  # Second plot: Heat flow & conductivity
+            ax2twin = ax2.twiny()
+
+            ax1.set_xlabel("Temperature [°C]")
+            ax1.set_ylabel("Depth from borehole head [m]")
+            ax1.grid()
+
+            ax2.set_xlabel("Heat flow [W/m]")
+            ax2.set_ylabel("Depth from borehole head [m]")
+            ax2.grid()
+            ax2twin.set_xlabel("Thermal conductivity [W/mK]")
+
+            color_iter_ax2 = iter(ax_colors)
+            color_iter_ax2twin = iter(axtwin_colors)
+
+            for param, results in zip(params_ds, results_ds):
+                file_name = str(
+                    param["file_name"]
+                    .isel(nPipes=0, layer_k_s=0, layer_k_g=0, layer_Tg=0)
+                    .item()
+                )
+
+                ptemp = results["T_f"].sel(time=time, method=method)
+                Tb = results["T_b"].sel(time=time, method=method)
+                Tg = results.get("Tg", None)
+                qzb = results["qzb"].sel(time=time, method=method)
+                k_s = results["k_s"].values
+                zp = -results["z"]
+                zseg = -results["zseg"]
+
+                # Check if the parameter is in the list before plotting
+                if plot_parameters == False:
+                    plot_parameters = ["T_f", "Tg", "T_b"]
+                else:
+                    plot_parameters = plot_parameters
+
+                model_type = param["model_type"].item()
+                if "T_f" in plot_parameters:
+                    for i in range(len(ptemp[0])):
+                        ax1.plot(
+                            ptemp[:, i], zp, label=f"{file_name}: Fluid temp Pipe {i}"
+                        )
+                        deltaT = ptemp[:, -1] - ptemp[:, 0]
+                if "Delta_T" in plot_parameters:
+                    ax1twin = ax1.twiny()
+                    (deltaT_line,) = ax1twin.plot(deltaT, zp, label="Delta T")
+
+                    # Get the color of the Delta T line
+                    deltaT_color = deltaT_line.get_color()
+
+                    # Set axis label and tick color to match the line
+                    ax1twin.set_xlabel(
+                        "Temperature diff. outlet-inlet [\u00b0C]", color=deltaT_color
+                    )
+                    ax1twin.tick_params(axis="x", colors=deltaT_color)
+                    ax1twin.spines["top"].set_color(
+                        deltaT_color
+                    )  # For the twiny axis, 'top' is used
+
+                if "T_b" in plot_parameters and len(Tb) == len(zseg):
+                    if model_type in ["ANALYTICAL", "FINVOL"]:
+                        ax1.plot(Tb, zseg, "k--", label=f"{file_name}: $T_b$")
+                    elif model_type in ["PYG", "PYGFIELD"]:
+                        Tb_plot = Tb * np.ones(len(zp))
+                        ax1.plot(Tb_plot, zp, "k--", label=f"{file_name}: $T_b$")
+
+                if "Tg" in plot_parameters and Tg is not None and len(Tg) == len(zseg):
+                    if model_type in ["ANALYTICAL", "FINVOL"]:
+                        ax1.plot(
+                            Tg, zseg, ":", label=f"{file_name}: $T_g$", color="red"
+                        )
+                    elif model_type in ["PYG", "PYGFIELD"]:
+                        Tg_plot = Tg * np.ones(len(zp))
+                        ax1.plot(
+                            Tg_plot, zp, ":", label=f"{file_name}: $T_g$", color="red"
+                        )
+
+                dz = zp[1] - zp[0]
+                if model_type in ["ANALYTICAL"]:
+                    qbzm = qzb / dz
+                    ax2.plot(
+                        qbzm,
+                        zseg,
+                        label=f"{file_name}: Heat Flow",
+                        color=next(color_iter_ax2),
+                    )
+                elif model_type in ["FINVOL"]:
+                    qbzm = qzb / dz
+                    qbzm[0] *= 2
+                    qbzm[-1] *= 2
+                    ax2.plot(
+                        -qbzm,
+                        zp,
+                        label=f"{file_name}: Heat Flow",
+                        color=next(color_iter_ax2),
+                    )
+                elif model_type in ["PYG", "PYGFIELD"]:
+                    # only single value for Tb, Qb etc, do not plot heat flow
+                    qbzm = qzb
+                    q_plot = qbzm * np.ones(len(zp))
+                    # because of UBWT condition the heat flow is not linear
+                    # ax2.plot(q_plot, zp, label=f'{file_name}: Heat Flow', color=next(color_iter_ax2))
+                else:
+                    print(f"Unrecognized model type: {model_type}")
+
+                # zseg is depth of mid point of segments
+                zseg = results.zseg.values
+                zval = k_s
+
+                if param["model_type"] == "PYG" or param["model_type"] == "PYGTILTED":
+                    zp_plot = np.array([zseg[0], zseg[-1]])
+                    k_s_plot = k_s * np.ones(2)
+                else:
+                    interp_obj = StratInterpolator(zseg, zval)
+
+                    zz = np.linspace(
+                        int(param.D),
+                        int(param.D) + int(param.H),
+                        int(param.nsegments) + 1,
+                    )
+
+                    zp = interp_obj.zp
+                    # Interpolate basecase thermal conductivity values
+                    k_s_interpolated = interp_obj.interp_plot(zz[0:-1], zz[1:])
+
+                    # select interpolated depth values for which the thermal conductivities are
+                    zp_plot = zp[(zp >= zz[0])]
+                    start_index_zp_plot = np.where(zp >= zz[0])[0][0]
+                    k_s_plot = k_s_interpolated[start_index_zp_plot:]
+
+                # Plot thermal conductivity on ax2twin
+                if (
+                    (param["model_type"].item() == "ANALYTICAL")
+                    or (param["model_type"].item() == "PYG")
+                    or (param["model_type"].item() == "PYGFIELD")
+                    or (param["model_type"].item() == "FINVOL")
+                ):
+                    ax2twin.plot(
+                        k_s_plot,
+                        -zp_plot,
+                        label=f"{file_name}: Thermal conductivity",
+                        color=next(color_iter_ax2twin),
+                    )
+
+            legend1 = ax1.legend(bbox_to_anchor=(1, 1))
+            # combine the legends from the second plot primary and secondary axis
+            handles2, labels2 = ax2.get_legend_handles_labels()
+            handles3, labels3 = ax2twin.get_legend_handles_labels()
+            all_handles = handles2 + handles3
+            all_labels = labels2 + labels3
+            # Create a single combined legend and define legend location
+            if len(params_ds) > 1:
+                legend_loc = ((1.8 - (len(params_ds) * 0.4)), 1)
+            else:
+                legend_loc = (1.9, 1)
+            combined_legend = ax2.legend(
+                all_handles, all_labels, bbox_to_anchor=legend_loc
+            )
+
+            plt.figure(fig1.number)  # Make fig1 the current active figure
+            plt.savefig(
+                f"{out_path}_temperature_depth_{time}.png",
+                dpi=300,
+                bbox_extra_artists=[legend1],
+                bbox_inches="tight",
+            )
+
+            plt.figure(fig2.number)  # Make fig2 the current active figure
+            plt.savefig(
+                f"{out_path}_heatflow_depth_{time}.png",
+                dpi=300,
+                bbox_extra_artists=[combined_legend],
+                bbox_inches="tight",
+            )
+
+            plt.close(fig1)
+            plt.close(fig2)
+
+    @staticmethod
+    def create_borehole_temp_plot(
+        segindex: list[int],
+        times: list,
+        params_ds: xr.Dataset | list[xr.Dataset],
+        results_ds: xr.Dataset | list[xr.Dataset],
+        singlerun: SingleRun,
+        out_path: Path,
+    ):
+        """
+        Only compatible with (multiple) single simulations.
+        Generates and saves depth-profiles at different timesteps in the simulation results, for fluid temperatures,
+        borehole wall temperature, subsurface temperature, subsurface heat flow and subsurface bulk thermal conductivity.
+
+        Generate borehole cross-section temperature plots at different timesteps.
+
+        Parameters
+        ----------
+        segindex : List[int]
+            Segment indices to plot.
+        times : List[float]
+            Timesteps to plot.
+        params_ds : Union[xr.Dataset, List[xr.Dataset]]
+            Dataset(s) containing simulation input parameters.
+        results_ds : Union[xr.Dataset, List[xr.Dataset]]
+            Dataset(s) containing simulation results.
+        singlerun : SingleRun
+            SingleRun object with simulation input parameters.
+        out_path : Path
+            Directory to save plots.
+
+        Raises
+        ------
+        ValueError
+            If any required parameter is missing from a DataFrame in `results_ds` or `params_ds`.
+
+        Returns
+        -------
+        None (the plots are saved directly to the specified output path).
+        """
+        method = "nearest"
+        if not isinstance(params_ds, list):
+            params_ds = [params_ds]
+            results_ds = [results_ds]
+
+        for time in times:
+            for param, results in zip(params_ds, results_ds):
+                pipe_temp = results["T_f"].sel(time=time, method=method)
+                borehole_wall_temp = results["T_b"].sel(time=time, method=method)
+
+                rb_scale = 1.2
+
+                for iseg in segindex:
+                    T_f = pipe_temp[iseg, :]
+                    T_b = borehole_wall_temp[iseg]
+
+                    singlerun.bh_design.m_flow = singlerun.sim_params.m_flow[0]
+                    singlerun.bh_design.customPipe = (
+                        singlerun.bh_design.get_custom_pipe()
+                    )
+                    custom_pipe = singlerun.bh_design.customPipe
+                    custom_pipe.update_thermal_resistances()
+
+                    R = custom_pipe.R[iseg]
+                    Q = np.linalg.solve(R, T_f - T_b)
+                    Q = np.asarray(Q).flatten()
+                    N_xy = 200
+                    r_b = custom_pipe.b.r_b
+                    x = np.linspace(-rb_scale * r_b, rb_scale * r_b, num=N_xy)
+                    y = np.linspace(-rb_scale * r_b, rb_scale * r_b, num=N_xy)
+                    X, Y = np.meshgrid(x, y)
+
+                    # Grid points to evaluate temperatures
+                    position_pipes = custom_pipe.pos
+                    pipe_r_out = custom_pipe.r_out
+                    R_fp = custom_pipe.R_f + custom_pipe.R_p[iseg]
+                    k_g = custom_pipe.k_g[iseg]
+                    k_s = custom_pipe.k_s
+                    T_b = np.asarray(T_b)
+                    (T_f, temperature, it, eps_max) = gt.pipes.multipole(
+                        position_pipes,
+                        pipe_r_out,
+                        r_b,
+                        k_s,
+                        k_g,
+                        R_fp,
+                        T_b,
+                        Q,
+                        J=3,
+                        x_T=X.flatten(),
+                        y_T=Y.flatten(),
+                    )
+                    distance = np.sqrt(X.flatten() ** 2 + Y.flatten() ** 2)
+                    temperature[distance > r_b] = np.nan
+
+                    # create figs
+                    fig, ax = plt.subplots()
+                    ax.set_xlabel("x (m)")
+                    ax.set_ylabel("y (m)")
+                    # Axis limits
+                    plt.axis(
+                        [
+                            -rb_scale * r_b,
+                            rb_scale * r_b,
+                            -rb_scale * r_b,
+                            rb_scale * r_b,
+                        ]
+                    )
+                    plt.gca().set_aspect("equal", adjustable="box")
+                    gt.utilities._format_axes(ax)
+
+                    levels = np.linspace(
+                        np.nanmin(temperature), np.nanmax(temperature), 10
+                    )
+                    cs = plt.contourf(
+                        X,
+                        Y,
+                        temperature.reshape((N_xy, N_xy)),
+                        levels=levels,
+                        cmap="viridis",
+                    )
+                    cbar = fig.colorbar(cs)
+
+                    # Borehole wall outline
+                    borewall = plt.Circle(
+                        (0.0, 0.0),
+                        radius=r_b,
+                        fill=False,
+                        linestyle="--",
+                        linewidth=2.0,
+                    )
+                    ax.add_patch(borewall)
+
+                    # Pipe outlines
+                    for pos, r_out_n in zip(position_pipes, pipe_r_out):
+                        pipe = plt.Circle(
+                            pos,
+                            radius=r_out_n,
+                            fill=False,
+                            linestyle="-",
+                            linewidth=4.0,
+                        )
+                        ax.add_patch(pipe)
+
+                    # Adjust to plot window
+                    plt.tight_layout()
+
+                    # Save fig
+                    filename = f"{out_path}_borehole_temp_{time}_seg{iseg}.png"
+                    fig.savefig(filename)
+                    plt.close(fig)
+
+    @staticmethod
+    def create_barplot(
+        results_df: pd.DataFrame | list[pd.DataFrame],
+        params_df: pd.DataFrame | list[pd.DataFrame],
+        y_variable: str,
+        outpath: Path,
+    ) -> None:
+        """
+        Only compatible with results and inputs of stochastic (MC) simulations.
+        Generates and saves a tornado bar plot, that shows the correlation of a specified simulation parameter
+        with other simulation input parameters and results.
+
+        This function merges simulation input parameters and results dataframes, calculates the correlation of
+        the specified simulation parameter (y_variable) with the other parameters, and visualizes the sensitivity of
+        the y_variable to changes in the other system parameters.
+
+        Parameters
+        ----------
+        results_df : Union[pd.DataFrame, List[pd.DataFrame]]
+            Simulation result dataframe(s), contain(s) simulation result parameters as columns.
+            Each DataFrame corresponds to a single simulation. It should include the columns 'samples'
+            and 'file_name' for merging purposes.
+        params_df : Union[pd.DataFrame, List[pd.DataFrame]]
+            Simulation input dataframe(s), contain(s) simulation input parameters as columns.
+            It should include the columns 'samples' and 'file_name' for merging purposes.
+            The column 'k_s' will be renamed to 'k_s_par'.
+        y_variable : str
+            Variable to analyze sensitivity for. The simulation parameter (either input or result) to be plotted
+            on the y-axis. Correlation with this parameter is visualized in the plot.
+        outpath : Path
+            Directory to save the plot.
+
+        Returns
+        -------
+        None (plots are saved directly to the specified output path)
+
+        Notes
+        -----
+        - The function selects only the simulation (input and result) parameters with numeric values and with a non-zero
+            standard deviation.
+        - If the specified y_variable does not exist in the merged DataFrame, no plot is created. No error is raised.
+        """
+
+        sns.set_theme(style="whitegrid")
+
+        # Convert single dataframe inputs to lists
+        if not isinstance(params_df, list):
+            params_df = [params_df]
+        if not isinstance(results_df, list):
+            results_df = [results_df]
+
+        # Rename columns if needed
+        params_df = [df.rename(columns={"k_s": "k_s_par"}) for df in params_df]
+
+        # Merge the dataframes on common columns
+        merged_dfs = [
+            pd.merge(df1, df2, on=["samples", "file_name"])
+            for df1, df2 in zip(params_df, results_df)
+        ]
+
+        # Concatenate merged dataframes if there are multiple
+        merged_df = pd.concat(merged_dfs)
+
+        # Group by 'file_name'
+        grouped_df = merged_df.groupby("file_name")
+
+        # Initialize the color palette for different groups
+        palette = sns.color_palette("husl", n_colors=len(grouped_df))
+
+        # Initialize the figure
+        plt.figure(figsize=(10, 6))  # Adjust the figure size if needed
+
+        # Iterate over each group and plot the data
+        for i, (group_name, group_df) in enumerate(grouped_df):
+            # Filter out columns with string values
+            numeric_columns = group_df.select_dtypes(
+                include=["float64", "int64"]
+            ).columns
+            group_numeric = group_df[numeric_columns]
+
+            # Filter out columns with zero standard deviation
+            non_constant_columns = group_numeric.columns[
+                group_numeric.std().round(7) != 0
+            ]
+            group_numeric_filtered = group_numeric[non_constant_columns]
+
+            sorted_sensitivity = pd.Series()
+            if y_variable in non_constant_columns:
+                # Calculate the correlation matrix
+                correlation_matrix = group_numeric_filtered.corr()
+                sensitivity_to_y_variable = correlation_matrix[y_variable].drop(
+                    y_variable
+                )  # Remove 'y_variable' itself from the list
+                sorted_sensitivity = pd.concat(
+                    (
+                        sorted_sensitivity.astype(sensitivity_to_y_variable.dtypes),
+                        sensitivity_to_y_variable,
+                    )
+                )
+                sorted_sensitivity = sorted_sensitivity.sort_values(
+                    ascending=False, na_position="last"
+                )
+                # sorted_sensitivity = sorted_sensitivity.reindex(sorted_sensitivity.abs().sort_values(ascending=False, na_position='last').index)
+
+                # Plot covariance matrix using a bar chart with different colors for each group
+                sns.barplot(
+                    x=sorted_sensitivity.values,
+                    y=sorted_sensitivity.index,
+                    color=palette[i],
+                    label=group_name,
+                    alpha=0.5,
+                )
+
+        # Finalize the plot
+        if y_variable in non_constant_columns:
+            plt.title(f"Correlation of {y_variable} with input parameters and results")
+            plt.xlabel("Correlation Coefficient")
+            plt.ylabel("Input and results variables")
+            plt.legend(
+                bbox_to_anchor=(0, -0.1), loc="upper left"
+            )  # Adjust legend position if needed
+
+            save_path = outpath.with_name(outpath.name + f"_sensitivity_{y_variable}")
+            plt.savefig(save_path, bbox_inches="tight")
+            plt.close()
+        else:
+            plt.close()
+
+    @staticmethod
+    def plot_temperature_field(
+        time: int,
+        temperature_result_da: xr.DataArray,
+        Tmin: float,
+        Tmax: float,
+        out_path: Path,
+    ) -> None:
+        """
+        Only compatible with results and numerical (FINVOL model) simulations.
+        Generates and saves a plot of the calculated 3D (z=len(1)) temperature grid around the borehole, for a specific
+        timestep.
+
+        Parameters
+        ----------
+        time : int
+            Index of the timestep to create the plot for.
+        temperature_result_da : xr.DataArray
+            Temperature DataArray, containing 3D (z=len(1)) grid of
+            calculated temperature values around the borehole.
+        Tmin : float
+            Minimum temperature in the DataArray for color scaling.
+        Tmax : float
+            Maximum temperature in the DataArray for color scaling.
+        out_path : Path
+            Directory to save the plot.
+
+        Returns
+        -------
+        None (plots are saved directly to the specified output path)
+        """
+
+        temperature_data = temperature_result_da.T.isel(time=time, z=0).transpose()
+
+        if any(i < 0 for i in temperature_result_da.x):
+            ymin = min(temperature_result_da.x)
+            ymax = max(temperature_result_da.x)
+        else:
+            ymin = max(temperature_result_da.x)
+            ymax = min(temperature_result_da.x)
+
+        # Create a figure and set size
+        fig = plt.figure(figsize=(10, 5))
+
+        # Add subplot
+        ax1 = fig.add_subplot()
+
+        # Define contour levels with even spacing every 5 degrees Celsius
+        levels = range(int(Tmin), int(Tmax) + 5, 5)
+
+        temperature_data.plot.contourf(
+            ax=ax1,
+            ylim=(ymin, ymax),
+            cmap="magma",
+            vmin=Tmin,
+            vmax=Tmax,
+            cbar_kwargs={"label": "Temperature [C]"},
+            levels=levels,
+        )
+
+        timestep = temperature_result_da.time.isel(time=time).values  # in hours
+
+        timestep_days = round(int(timestep) / 24, 0)
+
+        plt.title(
+            f"Temperature field over depth, in radial direction after {timestep_days} days"
+        )
+        plt.xlabel("Distance from well (m)")
+        plt.ylabel("Depth from top of well (m)")
+
+        plt.tight_layout()
+
+        filename = out_path.with_name(out_path.name + f"_T_field_{timestep_days}.png")
+        plt.savefig(filename)
+        plt.close()
+
+    @staticmethod
+    def create_temperature_field_movie(in_path: Path) -> None:
+        """
+        Only compatible with results and numerical (FINVOL model) simulations.
+        Generates and saves a clip of a sequence of the temperature grid plots creates using the method plot_temperature_field.
+
+        Parameters
+        ----------
+        in_path : Path
+            Directory containing temperature field images.
+
+        Returns
+        -------
+        None (plots are saved directly to the specified output path)
+        """
+        # Get list of image filenames
+        image_filenames = [
+            f.name for f in Path(in_path).iterdir() if f.suffix == ".png"
+        ]
+
+        image_filenames.sort(key=PlotResults.extract_timestep)
+
+        images = []
+        for image in image_filenames:
+            im_obj = Image.open((in_path / image), "r")
+            images.append(im_obj)
+
+        fig, ax = plt.subplots()
+
+        ims = []
+        for i in range(len(images)):
+            im = ax.imshow(images[i], animated=True)
+            if i == 0:
+                # set initial image
+                ax.imshow(images[i], animated=True)
+            ims.append([im])
+
+        ax.axis("off")
+
+        ani = animation.ArtistAnimation(fig, ims, interval=500, repeat_delay=100)
+
+        plt.tight_layout()
+        plt.close()
+
+        movie_name = "Tfield_animation.gif"
+        ani.save(in_path / movie_name)
+
+    @staticmethod
+    def extract_timestep(filename: str) -> float:
+        """
+        Extracts timestep from the filename, used for the chronological ordering of temperature grid plots in the
+        method create_temperature_field_movie.
+
+        Parameters
+        ----------
+        filename : str
+            Name of the image file.
+
+        Returns
+        -------
+        float
+            Extracted timestep or infinity if not found.
+        """
+        match = re.search(r"T_field_(\d+\.\d+)\.png", filename)
+        if match:
+            return float(match.group(1))
+        else:
+            # If no match is found, return a very large number
+            return float("inf")