[BUG]: class_ems nd_array not JSON serializable

Big Bugfix - not sure if everything works

src/akkudoktoreos/class_ems.py

@@ -58,15 +58,15 @@ class EnergieManagementSystem:
         total_hours = ende - start_stunde
 
         # Pre-allocate arrays for the results, optimized for speed
-        last_wh_pro_stunde = np.zeros(total_hours)
-        netzeinspeisung_wh_pro_stunde = np.zeros(total_hours)
-        netzbezug_wh_pro_stunde = np.zeros(total_hours)
-        kosten_euro_pro_stunde = np.zeros(total_hours)
-        einnahmen_euro_pro_stunde = np.zeros(total_hours)
-        akku_soc_pro_stunde = np.zeros(total_hours)
-        eauto_soc_pro_stunde = np.zeros(total_hours)
-        verluste_wh_pro_stunde = np.zeros(total_hours)
-        haushaltsgeraet_wh_pro_stunde = np.zeros(total_hours)
+        last_wh_pro_stunde = np.full((total_hours), np.nan)
+        netzeinspeisung_wh_pro_stunde = np.full((total_hours), np.nan)
+        netzbezug_wh_pro_stunde = np.full((total_hours), np.nan)
+        kosten_euro_pro_stunde = np.full((total_hours), np.nan)
+        einnahmen_euro_pro_stunde = np.full((total_hours), np.nan)
+        akku_soc_pro_stunde = np.full((total_hours), np.nan)
+        eauto_soc_pro_stunde = np.full((total_hours), np.nan)
+        verluste_wh_pro_stunde = np.full((total_hours), np.nan)
+        haushaltsgeraet_wh_pro_stunde = np.full((total_hours), np.nan)
 
         # Set initial state
         akku_soc_pro_stunde[0] = self.akku.ladezustand_in_prozent()
@@ -78,7 +78,7 @@ class EnergieManagementSystem:
 
             # Accumulate loads and PV generation
             verbrauch = self.gesamtlast[stunde]
-
+            verluste_wh_pro_stunde[stunde_since_now] = 0.0
             if self.haushaltsgeraet:
                 ha_load = self.haushaltsgeraet.get_last_fuer_stunde(stunde)
                 verbrauch += ha_load
@@ -117,7 +117,7 @@ class EnergieManagementSystem:
             akku_soc_pro_stunde[stunde_since_now] = self.akku.ladezustand_in_prozent()
 
         # Total cost and return
-        gesamtkosten_euro = np.sum(kosten_euro_pro_stunde) - np.sum(
+        gesamtkosten_euro = np.nansum(kosten_euro_pro_stunde) - np.nansum(
             einnahmen_euro_pro_stunde
         )
 
@@ -131,35 +131,10 @@ class EnergieManagementSystem:
             "Einnahmen_Euro_pro_Stunde": einnahmen_euro_pro_stunde,
             "Gesamtbilanz_Euro": gesamtkosten_euro,
             "E-Auto_SoC_pro_Stunde": eauto_soc_pro_stunde,
-            "Gesamteinnahmen_Euro": np.sum(einnahmen_euro_pro_stunde),
-            "Gesamtkosten_Euro": np.sum(kosten_euro_pro_stunde),
+            "Gesamteinnahmen_Euro": np.nansum(einnahmen_euro_pro_stunde),
+            "Gesamtkosten_Euro": np.nansum(kosten_euro_pro_stunde),
             "Verluste_Pro_Stunde": verluste_wh_pro_stunde,
-            "Gesamt_Verluste": np.sum(verluste_wh_pro_stunde),
+            "Gesamt_Verluste": np.nansum(verluste_wh_pro_stunde),
             "Haushaltsgeraet_wh_pro_stunde": haushaltsgeraet_wh_pro_stunde,
         }
-
-        # List output keys where the first element needs to be changed to None
-        keys_to_modify = [
-            "Last_Wh_pro_Stunde",
-            "Netzeinspeisung_Wh_pro_Stunde",
-            "akku_soc_pro_stunde",
-            "Netzbezug_Wh_pro_Stunde",
-            "Kosten_Euro_pro_Stunde",
-            "Einnahmen_Euro_pro_Stunde",
-            "E-Auto_SoC_pro_Stunde",
-            "Verluste_Pro_Stunde",
-            "Haushaltsgeraet_wh_pro_stunde"
-        ]
-
-        # Loop through each key in the list
-        for key in keys_to_modify:
-            # Convert the NumPy array to a list
-            element_list = out[key].tolist()
-
-            # Change the first value to None
-            element_list[0] = None
-
-            # Assign the modified list back to the dictionary
-            out[key] = element_list
-
         return out

src/akkudoktoreos/class_haushaltsgeraet.py

@@ -16,7 +16,6 @@ class Haushaltsgeraet:
         :param start_hour: The hour at which the device should start.
         """
         self.reset()
-
         # Check if the duration of use is within the available time frame
         if start_hour + self.dauer_h > self.hours:
             raise ValueError("The duration of use exceeds the available time frame.")

src/akkudoktoreos/class_optimize.py

@@ -15,7 +15,7 @@ from akkudoktoreos.visualize import visualisiere_ergebnisse
 class optimization_problem:
     def __init__(
         self,
-        prediction_hours: int = 24,
+        prediction_hours: int = 48,
         strafe: float = 10,
         optimization_hours: int = 24,
         verbose: bool = False,
@@ -138,7 +138,7 @@ class optimization_problem:
         """
         try:
             o = self.evaluate_inner(individual, ems, start_hour)
-        except Exception:
+        except Exception as e:
             return (100000.0,)  # Return a high penalty in case of an exception
 
         gesamtbilanz = o["Gesamtbilanz_Euro"] * (-1.0 if worst_case else 1.0)
@@ -326,6 +326,35 @@ class optimization_problem:
             extra_data=extra_data,
         )
 
+        # List output keys where the first element needs to be changed to None
+        keys_to_modify = [
+            "Last_Wh_pro_Stunde",
+            "Netzeinspeisung_Wh_pro_Stunde",
+            "akku_soc_pro_stunde",
+            "Netzbezug_Wh_pro_Stunde",
+            "Kosten_Euro_pro_Stunde",
+            "Einnahmen_Euro_pro_Stunde",
+            "E-Auto_SoC_pro_Stunde",
+            "Verluste_Pro_Stunde",
+            "Haushaltsgeraet_wh_pro_stunde",
+        ]
+
+        # Loop through each key in the list
+        for key in keys_to_modify:
+            # Convert the NumPy array to a list
+            element_list = o[key].tolist()
+
+            # Change the first value to None
+            element_list[0] = None
+            # Change the NaN to None (JSON)
+            element_list = [
+                None if isinstance(x, (int, float)) and np.isnan(x) else x
+                for x in element_list
+            ]
+
+            # Assign the modified list back to the dictionary
+            o[key] = element_list
+
         # Return final results as a dictionary
         return {
             "discharge_hours_bin": discharge_hours_bin,

src/akkudoktoreosserver/flask_server.py

@@ -3,14 +3,14 @@
 import os
 from datetime import datetime
 from typing import Any, TypeGuard
-import json
+
 import matplotlib
 
 # Sets the Matplotlib backend to 'Agg' for rendering plots in environments without a display
 matplotlib.use("Agg")
 
 import pandas as pd
-from flask import Flask, jsonify, redirect, request, send_from_directory, url_for, Response
+from flask import Flask, jsonify, redirect, request, send_from_directory, url_for
 
 from akkudoktoreos.class_load import LoadForecast
 from akkudoktoreos.class_load_container import Gesamtlast
@@ -211,6 +211,12 @@ def flask_pvprognose():
 
 @app.route("/optimize", methods=["POST"])
 def flask_optimize():
+    with open(
+        "C:\\Users\\drbac\\OneDrive\\Dokumente\\PythonPojects\\EOS\\debug_output.txt",
+        "a",
+    ) as f:
+        f.write("Test\n")
+
     if request.method == "POST":
         from datetime import datetime
 
@@ -243,18 +249,17 @@ def flask_optimize():
                 {"error": f"Missing parameter: {', '.join(missing_params)}"}
             ), 400  # Return error for missing parameters
 
-        # Perform optimization simulation
-        result = opt_class.optimierung_ems(
-            parameter=parameter, start_hour=datetime.now().hour
-        )
-
         # Optional min SoC PV Battery
         if "min_soc_prozent" not in parameter:
             parameter["min_soc_prozent"] = None
 
+        # Perform optimization simulation
+        result = opt_class.optimierung_ems(
+            parameter=parameter, start_hour=datetime.now().hour
+        )
+        print(result)
         # convert to JSON (None accepted by dumps)
-        json_data = json.dumps(result)
-        return Response(json_data, mimetype='application/json')
+        return jsonify(result)
 
 
 @app.route("/visualisierungsergebnisse.pdf")