Use dad's real P1 data; add export-rate input for terugleveringskosten

- scripts/import_dad_p1.py converts his 15-min cumulative meter dump
  (~/Downloads/P1e-2025-1-1-2026-1-1.csv) into our hourly format.
  Net of PV exports — power_w can be negative on surplus hours.
  Aligned to our 2023-09 → 2024-09 window by (month, day, hour) so
  prices and irradiance stay seasonally matched.
- data/raw/p1_hourly.csv is now dad's data; ours preserved as
  p1_hourly_michiel.csv for diff/comparison.
- LP gE upper bound was 'pv[t]'; now max(0, pv − demand). This was
  wrong before but invisible because demand was always positive; with
  dad's net data demand goes negative and the LP went infeasible.
- New 'Export rate' field in Advanced. Override saldering toggle with
  a literal €/kWh — set to −0.106 for terugleveringskosten.

scripts/import_dad_p1.py

@@ -0,0 +1,105 @@
+#!/usr/bin/env python3
+"""Convert dad's 15-min P1 export into our hourly p1 format.
+
+Source: ~/Downloads/P1e-2025-1-1-2026-1-1.csv (cumulative T1/T2 import/export
+meter readings, every 15 min, 2025 calendar year, Europe/Amsterdam local time).
+
+Output: data/raw/p1_hourly.csv (timestamp UTC, power_w mean over hour). Net
+of PV exports — `power_w` can be negative when his solar exceeded household
+demand. The simulator interprets `demand_kwh = power_w / 1000`, and a plug-in
+battery's discharge cap `max(0, demand - pv)` correctly handles negative
+demand by forbidding discharge during surplus hours.
+
+The simulator's window is 2023-09-01 → 2024-09-01 UTC. Dad's data is the
+2025 calendar year. We align by (month, day, hour) so seasonal patterns
+(winter peaks, summer surplus) line up with our prices and irradiance.
+Feb 29 in our 2024 leap year falls back to Feb 28 from dad's 2025.
+
+Run from project root:
+    uv run python scripts/import_dad_p1.py
+"""
+from __future__ import annotations
+
+import argparse
+from pathlib import Path
+
+import pandas as pd
+
+
+def main() -> None:
+    p = argparse.ArgumentParser(description=__doc__)
+    p.add_argument("--src", type=Path,
+                   default=Path.home() / "Downloads/P1e-2025-1-1-2026-1-1.csv")
+    p.add_argument("--out", type=Path, default=Path("data/raw/p1_hourly.csv"))
+    p.add_argument("--window-start", default="2023-09-01T00:00:00Z")
+    p.add_argument("--window-end",   default="2024-09-01T00:00:00Z")
+    args = p.parse_args()
+
+    print(f"Reading dad's CSV from {args.src}")
+    dad = pd.read_csv(args.src, parse_dates=["time"])
+    dad["time"] = dad["time"].dt.tz_localize(
+        "Europe/Amsterdam", ambiguous="infer", nonexistent="shift_forward"
+    )
+    dad = dad.set_index("time").sort_index()
+
+    # Cumulative net energy (signed). Imports add, exports subtract.
+    net_cum = (
+        dad["Import T1 kWh"] + dad["Import T2 kWh"]
+        - dad["Export T1 kWh"] - dad["Export T2 kWh"]
+    )
+
+    # Sample at exact hour boundaries (the meter records every 15 min, so the
+    # H:00 reading exists). Diff to get energy consumed in each hour.
+    hourly_stock = net_cum.resample("1h").first()
+    hourly_kwh = hourly_stock.diff().dropna()
+    # Now hourly_kwh.index[i] = end of the hour just consumed; shift to start.
+    hourly_kwh.index = hourly_kwh.index - pd.Timedelta(hours=1)
+    hourly_kwh = hourly_kwh.tz_convert("UTC")
+
+    annual = hourly_kwh.sum()
+    print(f"Dad's hourly net (kWh): n={len(hourly_kwh)}, "
+          f"annual net = {annual:.0f} kWh "
+          f"(min hour {hourly_kwh.min():.2f}, max {hourly_kwh.max():.2f}, "
+          f"{(hourly_kwh < 0).sum()} export hours)")
+
+    # Build (month, day, hour) → kWh lookup. dad's full year covers every key
+    # except a possible duplicate around DST (averaged).
+    lookup = (
+        hourly_kwh.groupby([
+            hourly_kwh.index.month,
+            hourly_kwh.index.day,
+            hourly_kwh.index.hour,
+        ]).mean()
+    )
+
+    # Build target timestamps for our simulator window.
+    target = pd.date_range(args.window_start, args.window_end,
+                           freq="1h", inclusive="left", tz="UTC")
+    out = []
+    missing = 0
+    for ts in target:
+        m, d, h = ts.month, ts.day, ts.hour
+        if (m, d, h) in lookup.index:
+            kwh = lookup.loc[(m, d, h)]
+        elif m == 2 and d == 29 and (2, 28, h) in lookup.index:
+            kwh = lookup.loc[(2, 28, h)]
+        else:
+            missing += 1
+            continue
+        out.append((ts, kwh * 1000.0))   # kWh per hour → mean watts
+
+    if missing:
+        print(f"WARNING: {missing} target hours had no dad-data match")
+
+    df_out = pd.DataFrame(out, columns=["timestamp", "power_w"])
+    df_out["timestamp"] = df_out["timestamp"].dt.strftime("%Y-%m-%dT%H:%M:%SZ")
+
+    args.out.parent.mkdir(parents=True, exist_ok=True)
+    df_out.to_csv(args.out, index=False)
+    print(f"Wrote {len(df_out)} hours → {args.out}")
+    print(f"  annualised mean: {df_out['power_w'].mean():.0f} W "
+          f"({df_out['power_w'].sum() / 1000:.0f} kWh net per year)")
+
+
+if __name__ == "__main__":
+    main()