2025.10.0 (#389)

Fix error with storing forecast (there was no timezone-correction)
Fixed several other errors

Author: corneel27

dao/CHANGELOG.md

@@ -1,6 +1,35 @@
 # Changelog 刀 DAO
 # Day Ahead Optimizer
 
+# 2025.10.0
+This version is similar with 2025.10.0.rc1 in the testing-channel
+- Fix error with storing forecast (there was no timezone-correction)
+- Maximised the calculationtime to 20 sec, the accuracy to 0.005 euro (whichever comes first)
+- Fixed format-error when loglevel=debug
+- Set max of y-as for soc to 102%: when soc=100% the line is visible
+- Extra information in the logging (level info) for control of the calculation of the profit
+- Fixed error calculation/presentation "time needed" charging an ev
+- Fixed error with postgresql da-database
+- Added logging of free/used memory in logging on level "debug"
+
+Know issue:
+- The forecast-calculation in the reports is nog good working
+
+
+**Interval**<br>
+This release supports two intervals: **1hour** and **15min**
+You make your choice in the settings, put in your settings somewhere in the root the setting:<br>
+"interval": "1hour" or "interval": "15min", 1hour will be default when you make no intervalsetting
+
+**source prices**<br>
+At the moment only Nordpool, Tibber and Entsoe (the last one could not be tested) supports 15min prices.<br>
+If you don't want trouble: make nordpool your price-source.
+
+**boiler**<br>
+The boiler part is completely redesigned and (in my opinion) become better.
+The (thermal and financial) value of the boiler at the beginning and the end of the calculationperiod si taken into account. 
+For every interval the electricity consumption for reheating is calculated and thee best fitting will be choosen in the algortime.
+
 # 2025.9.0
 Fix error api prognose pv_dc
 

dao/config.yaml

@@ -1,6 +1,6 @@
 ---
 name: 刀 Day Ahead Optimizer
-version: 2025.9.0
+version: 2025.10.0
 slug: day_ahead_opt
 description: Home Assistant Community Add-ons for day ahead optimizations
 url: https://github.com/corneel27/day-ahead

dao/prog/da_base.py

@@ -3,6 +3,7 @@
 import os
 import fnmatch
 import time
+import pytz
 from requests import get
 import json
 import hassapi as hass
@@ -113,11 +114,13 @@ def __init__(self, file_name: str = None):
         self.config.set("latitude", resp_dict["latitude"])
         self.config.set("longitude", resp_dict["longitude"])
         self.config.set("time_zone", resp_dict["time_zone"])
+        self.time_zone = resp_dict["time_zone"]
         self.config.set("country", resp_dict["country"])
         self.db_da = self.config.get_db_da()
         self.db_ha = self.config.get_db_ha()
         self.meteo = Meteo(self.config, self.db_da)
         self.solar = self.config.get(["solar"])
+        self.interval = self.config.get(["interval"], None, "1hour").lower()
 
         self.prices = DaPrices(self.config, self.db_da)
         self.prices_options = self.config.get(["prices"])
@@ -231,7 +234,7 @@ def generate_tasks():
             "calc_optimum": {
                 "name": "Optimaliseringsberekening zonder debug",
                 "cmd": ["python3", "../prog/day_ahead.py", "calc"],
-                "object": "DaCalc",
+                "object": "DaBase",
                 "function": "calc_optimum",
                 "file_name": "calc",
             },
@@ -341,8 +344,11 @@ def save_df(self, tablename: str, tijd: list, df: pd.DataFrame):
         df_db = pd.DataFrame(columns=["time", "code", "value"])
         df = df.reset_index(drop=True)
         columns = df.columns.values.tolist()[1:]
+        tz = pytz.timezone(self.time_zone)
         for index in range(len(tijd)):
-            utc = int(tijd[index].timestamp())
+            dt = pd.to_datetime(tijd[index])
+            dt = tz.localize(dt)
+            utc = int(dt.timestamp())
             for c in columns:
                 db_row = [str(utc), c, float(df.loc[index, c])]
                 df_db.loc[df_db.shape[0]] = db_row
@@ -513,7 +519,7 @@ def calc_optimum_met_debug(self):
         # dacalc = DaCalc("../data/tst_options/options_mirabis.json")
         dacalc.debug = True
         dacalc.calc_optimum()
-        # dacalc.calc_optimum(_start_dt=datetime.datetime(2025, 9, 16, 10, minute=45), _start_soc=98)
+        # dacalc.calc_optimum(_start_dt=datetime.datetime(2025, 9, 28, 21, minute=0), _start_soc=50)
 
     def calc_optimum(self):
         from day_ahead import DaCalc

dao/prog/da_report.py

@@ -1967,9 +1967,6 @@ def get_grid_data(
                 df_prices = pd.read_sql_query(query, connection)
             logging.debug(f"Prijzen \n{df_prices.to_string()}\n")
             """
-            df_prices = self.get_price_data(
-                last_moment, tot
-            )  # +datetime.timedelta(hours=1))
 
             df_ha = pd.DataFrame()
             if source == "all" or source == "ha":
@@ -2008,6 +2005,7 @@ def get_grid_data(
                     df_ha["datasoort"] = "recorded"
                 else:
                     last_moment = vanaf
+                df_prices = self.get_price_data(vanaf, last_moment, interval="1hour")
 
             if source == "all" or source == "da":
                 if last_moment < tot:
@@ -2051,10 +2049,15 @@ def get_grid_data(
                     df_prog.index = pd.to_datetime(df_prog["tijd"])
                     df_prog["datasoort"] = "expected"
                     if len(df_prog) > 0:
+                        df_prog_prices = self.get_price_data(last_moment, tot, interval=self.interval)
                         if len(df_ha) == 0:
                             df_ha = df_prog
+                            df_prices = df_prog_prices
                         else:
                             df_ha = pd.concat([df_ha, df_prog])
+                            df_prices.index = pd.to_datetime(df_prices["time"])
+                            df_prog_prices.index = pd.to_datetime(df_prog_prices["time"])
+                            df_prices = pd.concat([df_prices, df_prog_prices])
 
             df_prices.rename(columns={"time": "tijd"}, inplace=True)
             df_prices.index = pd.to_datetime(df_prices["tijd"])

dao/prog/day_ahead.py

@@ -28,7 +28,6 @@ def __init__(self, file_name=None):
         super().__init__(file_name=file_name)
         if self.config is None:
             return
-        self.interval = self.config.get(["interval"], None, "1hour").lower()
         self.interval_s = 3600 if self.interval == "1hour" else 900
         self.interval_name = "uur" if self.interval == "1hour" else "kwartier"
         self.steps_day = 24 if self.interval == "1hour" else 96
@@ -122,8 +121,8 @@ def calc_optimum(
             return
         if u_data <= 8 or u_prices <= 8:
             logging.warning(
-                f"Er ontbreken voor een aantal uur gegevens "
-                f"(meteo en/of dynamische prijzen)\n"
+                f"Er ontbreken voor een aantal uur meteogegevens "
+                f"(en/of dynamische prijzen)\n"
                 f"controleer of alle gegevens zijn opgehaald"
             )
             if self.notification_entity is not None:
@@ -234,8 +233,6 @@ def calc_optimum(
             uur.append(hour)
             tijd.append(dtime)
             gr = row.glob_rad
-            if self.interval == "15min":
-                gr *= 4
             global_rad.append(gr)
             pv_total = 0
             if first_interval:
@@ -248,10 +245,8 @@ def calc_optimum(
                 interval_fraction.append(1)
             for s in range(solar_num):
                 prod = self.calc_prod_solar(
-                    self.solar[s], row.time, gr, interval_fraction[-1]
+                    self.solar[s], row.time, gr, hour_fraction[-1]
                 )
-                if self.interval == "15min":
-                    prod /= 4
                 solar_prod[s].append(prod)
                 pv_total += prod
             pv_org_ac.append(pv_total)
@@ -268,7 +263,7 @@ def calc_optimum(
                         self.battery_options[b]["solar"][s],
                         row.time,
                         gr,
-                        interval_fraction[-1],
+                        hour_fraction[-1],
                     )
                     pv_total += prod
             pv_org_dc.append(pv_total)
@@ -492,8 +487,6 @@ def calc_optimum(
                         global_rad[u],
                         hour_fraction[u],
                     )
-                    if self.interval =="15min":
-                        prod_dc /= 4
                     eff = 1
                     for ds in range(DS[b]):
                         if discharge_stages[b][ds]["power"] / 1000 > prod_dc:
@@ -1820,8 +1813,8 @@ def calc_optimum(
                 max_heat_power = stages[-1]["max_power"] * stages[-1]["cop"] / 1000
 
                 # een of meer intervallen minder als boiler via wp gaat
-                if boiler_heated_by_heatpump and boiler_start_index < U:
-                    boiler_int = est_needed_intv[U - 1]
+                if boiler_heated_by_heatpump and boiler_start < U:
+                    boiler_int = est_needed_intv[boiler_start] + 1
                 else:
                     boiler_int = 0
                 max_heat_prod = sum(
@@ -2727,17 +2720,16 @@ def calc_optimum(
             + boiler_storage
         )
 
-        logging.info("Calculation profit after optimize in €")
-        logging.info(f"Cost before optimize               {old_cost_da: 7.2f}")
-        logging.info(f"Cost consumption   {cost_consumption: 7.2f}")
-        logging.info(f"Profit production  {profit_production: 7.2f}")
-        logging.info(f"Cycle cost         {total_cycle_cost: 7.2f}")
-        logging.info(f"Battery storage    {battery_storage: 7.2f}")
-        logging.info(f"Boiler storage     {boiler_storage: 7.2f}")
-        logging.info(f"Total              {total_cost: 7.2f}")
-        logging.info(f"Cost after optimize               {cost.x: 7.2f}")
-        logging.info(
-            f"Profit:                             {old_cost_da - cost.x: 7.2f}"
+        logging.info("/nCalculation profit after optimize in €\n"
+            f"Cost before optimize            {old_cost_da: 7.2f}\n"
+            f"Cost consumption   {cost_consumption: 7.2f}\n"
+            f"Profit production  {profit_production: 7.2f}\n"
+            f"Cycle cost         {total_cycle_cost: 7.2f}\n"
+            f"Battery storage    {battery_storage: 7.2f}\n"
+            f"Boiler storage     {boiler_storage: 7.2f}\n"
+            f"Total              {total_cost: 7.2f}\n"
+            f"Cost after optimize            {cost.x: 7.2f}\n"
+            f"Profit:                        {old_cost_da - cost.x: 7.2f}"
         )
 
         # doorzetten van alle settings naar HA

dao/prog/db_manager.py

@@ -426,7 +426,7 @@ def get_prognose_data(self, start, end=None, interval="1hour"):
                 fld_df = self.get_prognose_field(field, start, end, interval)
                 # fld_df.index = pd.to_datetime(fld_df["tijd"])
                 # fld_df = interpolate(fld_df, field, 15, (field == "gr"))
-                fld_df = interpolate(fld_df, field, (field == "gr"))
+                fld_df = interpolate(fld_df, field, False)
                 if result_df is None:
                     result_df = fld_df
                 else:

release-testing/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog 刀 DAO
 # Day Ahead Optimizer
+# 2025.10.0.rc1
+This version is similar with 2025.10.0 in the stable-channel
+- Fix error with storing forecast (there was no timezone-correction)
+
+Know issue:
+- The forecast in the reports still nog good working
+
 # 2025.9.1.rc7
 - Fix error (Signal SIGSEGV caught) in rc6 on x86 (arch=amd64) machines
 

release-testing/config.yaml

@@ -1,6 +1,6 @@
 ---
 name: 刀 Day Ahead Optimizer (TESTING)
-version: 2025.9.1.rc7
+version: 2025.10.0.rc1
 stage: experimental
 slug: day_ahead_opt-testing
 description: Beta version of DAO. Use only for testing!