Version 2026.02.2.rc1

dao/prog/da_base.py

@@ -603,7 +603,7 @@ def calc_solar_predictions(
         :param vanaf: datetime start
         :param tot: datetime tot
         :param interval: 15"min of 1 hour of None, als None wordt self.interval genomen
-        :param ml_prediction: boolean default None(= from config)
+        :param _ml_prediction: boolean default None(= from config)
         :return:
         """
         from dao.prog.solar_predictor import SolarPredictor
@@ -627,12 +627,19 @@ def calc_solar_predictions(
                 solar_prog = solar_predictor.predict_solar_device(
                     solar_option, vanaf, tot
                 )
+                if solar_prog.isnull().any().any():
+                    logging.warning(
+                        f"NaN-waarden aangetroffen in voorspelling van {solar_name}"
+                        f"Deze zijn op '0' gezet"
+                    )
+                    solar_prog.fillna(0, inplace=True)
             except FileNotFoundError as ex:
                 logging.warning(ex)
                 logging.info(
                     f"Voor {solar_option['name']} is geen model "
                     f"en dus wordt DAO-predictor gebruikt"
                 )
+
                 result = self.calc_solar_predictions(
                     solar_option, vanaf, tot, interval=interval, _ml_prediction=False
                 )

dao/prog/da_graph.py

@@ -300,8 +300,8 @@ def build(self, df, options, show=True):
                 ax.set_xlabel(haxis["title"])
             num_xas = len(df.index)
             if num_xas > 12:
-                ax.xaxis.set_major_locator(ticker.MultipleLocator(round(num_xas / 12)))
-                ax.xaxis.set_minor_locator(ticker.MultipleLocator(1))
+                ax.xaxis.set_major_locator(ticker.MultipleLocator(12))
+                ax.xaxis.set_minor_locator(ticker.MultipleLocator(2))
             if len(str(xlabels[0])) > 2:
                 ax.set_xticks(
                     ax.get_xticks(), ax.get_xticklabels(), rotation=45, ha="right"

dao/prog/da_report.py

@@ -1118,7 +1118,8 @@ def copy_col_df(
         :return: de ingevuld copy_to
         """
         dt = copy_from[col_name].dtype
-        if dt == float:
+        # Use is_numeric_dtype to catch all numeric types (float64, int64, etc.)
+        if pd.api.types.is_numeric_dtype(dt):
             copy_to[col_name] = 0.0
         else:
             copy_to[col_name] = ""

dao/prog/day_ahead.py

@@ -286,6 +286,10 @@ def calc_optimum(
 
         # nieuwe universele methode
         end = prog_data["tijd"].iloc[-1]
+        if self.interval == "1hour":
+            end += datetime.timedelta(hours=1)
+        else:
+            end += datetime.timedelta(minutes=15)
         for s in range(solar_num):
             solar_prog = self.calc_solar_predictions(
                 self.solar[s], start_interval_dt, end, self.interval
@@ -1089,7 +1093,10 @@ def calc_optimum(
             heat_rate = (
                 power_boiler * cop_boiler * self.interval_s / (spec_heat_boiler * 1000)
             )
-            max_steps = int((boiler_setpoint - boiler_ondergrens) / heat_rate) + 1
+            boiler_end_temp = max(
+                boiler_ondergrens, boiler_act_temp - U * boiler_cooling
+            )
+            max_steps = math.ceil((boiler_setpoint - boiler_end_temp) / heat_rate)
             # interval-index waarop boiler kan worden verwarmd
             if boiler_instant_start or (boiler_act_temp <= boiler_ondergrens):
                 boiler_start_index = 0
@@ -3234,10 +3241,15 @@ def calc_optimum(
                 totals = False
 
             if totals:
-                df_accu[b].at[df_accu[b].index[-1], "uur"] = "Totaal"
-                df_accu[b].at[df_accu[b].index[-1], "eff"] = "--"
-                df_accu[b].at[df_accu[b].index[-1], "o_eff"] = "--"
-                df_accu[b].at[df_accu[b].index[-1], "SoC"] = pd.NA
+                # Kolom "uur" kan string "Totaal" krijgen door eerst naar object te casten
+                df_accu[b].iloc[:, 0] = df_accu[b].iloc[:, 0].astype(object)
+                df_accu[b].iloc[:, 0] = df_accu[b].iloc[:, 0].astype(object)
+                df_accu[b].iloc[-1, 0] = "Totaal"
+                df_accu[b].iloc[-1, 2] = np.nan  # eff (ac->dc)
+                df_accu[b].iloc[-1, 6] = np.nan  # eff (dc->bat)
+                df_accu[b].iloc[-1, 8] = np.nan  # o_eff
+                df_accu[b].iloc[-1, 9] = np.nan  # SoC
+
             logging.info(
                 f"In- en uitgaande energie per {self.interval_name} batterij "
                 f"{self.battery_options[b]['name']}"
@@ -3750,7 +3762,7 @@ def calc_optimum(
                         if wf > 0:
                             sum_weight_factor += wf
                             sum_power += wf * discharge_stages[b][ds]["power"]
-                    if sum_weight_factor < 0.95:
+                    if 0.10 <= sum_weight_factor < 0.95:
                         new_state = battery_state_on_value
                         balance = False
                         netto_vermogen_bat = -round(sum_power / sum_weight_factor)

dao/prog/solar_predictor.py

@@ -1023,6 +1023,16 @@ def predict_solar_device(
         :param end: eind-tijdstip voorspelling
         :return: dataframe met berekende voorspellingen per uur
         """
+
+        def check_prediction(prediction, irradiance):
+            if irradiance <= 0.0:
+                result = 0
+            else:
+                result = prediction
+            if result < 0.0:
+                result = 0
+            return result
+
         name = self.config.get(["name"], solar_dict, "default")
         self.solar_name = name.replace(" ", "_")
         self.tilt = self.get_property_from_dict("tilt", solar_dict, 45)
@@ -1039,7 +1049,13 @@ def predict_solar_device(
         prognose = latest_dt < end
         weather_data = self.get_weatherdata(start, end, prognose=prognose)
         prediction = self.predict(weather_data)
-        prediction["prediction"] = prediction["prediction"].round(3)
+        weather_data.reset_index(inplace=True)
+        prediction["irradiance"] = weather_data["irradiance"]
+        prediction["prediction"] = prediction.apply(
+            lambda x: check_prediction(x["prediction"], x["irradiance"]), axis=1
+        )
+        prediction["prediction"].round(3)
+        prediction.drop("irradiance", axis=1, inplace=True)
         logging.info(f"ML prediction {self.solar_name}\n{prediction}")
         return prediction
 

release-testing/CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog 刀 DAO
 # Day Ahead Optimizer
+# 2026.02.2.rc1
+- Update several python modules
+- Adjusted the footer of the webpages (reported by steynovitch)
+- Removed NaN-values from ml-preditions
+- Fix error in reports (reported by @steynovich)
+- Correct totals battery summary
+- Fix error when period with minimum power is less than 1 minuut (reported by @Dogooder)
+- Fix negative or small values >0 in ml-predictions when irradiance==0 (reported by @Dogooder)
+
 # 2026.02.1.rc1
 Add missing module tzdata
 

release-testing/config.yaml

@@ -1,6 +1,6 @@
 ---
 name: 刀 Day Ahead Optimizer (TESTING)
-version: 2026.02.1.rc1
+version: 2026.02.2.rc1
 stage: experimental
 slug: day_ahead_opt-testing
 description: Beta version of DAO. Use only for testing!