ems/scripts/analysis/pv_delta_profile_diagnostics.sql

-- Diagnostika: z kterých kalendářních dní (Europe/Prague) se skládá váha pro delta profil
-- (zarovnáno s ems.fn_pv_forecast_delta_profile: eff z site_pv_forecast_calibration, best s learning_eligible,
-- agregace slotů na úroveň site pro day_rank / váhy w — stejné jako slot_totals v R__078).
--
-- Uprav params (site_id, okno, half_life, threshold, top_n_days / non_top / gamma) a spusť v psql.
-- Jedna řádka = jeden kalendářní den v okně; p_top_n_days mění tier u vah (ne počet řádků).

WITH params AS (
  SELECT
    2::int AS site_id,
    (now() - interval '60 days')::timestamptz AS p_data_from,
    now()::timestamptz AS p_data_to,
    14::numeric AS half_life_days,
    150::int AS threshold_w,
    NULL::int AS p_top_n_days,
    0.02::numeric AS p_non_top_day_factor,
    1.0::numeric AS p_day_weight_gamma
),
tz AS (
  SELECT coalesce(nullif(trim(s.timezone), ''), 'Europe/Prague') AS tz_name
  FROM ems.site s
  JOIN params p ON s.id = p.site_id
),
eff AS (
  SELECT
    coalesce(cal.delta_learn_min_ts, timestamptz '2026-04-11T22:00:00Z') AS delta_learn_min_ts,
    coalesce(cal.half_life_days, p.half_life_days) AS half_life_days,
    coalesce(cal.threshold_w, p.threshold_w) AS threshold_w,
    coalesce(cal.top_n_days, p.p_top_n_days) AS top_n_days,
    coalesce(cal.non_top_day_factor, p.p_non_top_day_factor) AS non_top_day_factor,
    coalesce(cal.day_weight_gamma, p.p_day_weight_gamma) AS day_weight_gamma
  FROM params p
  JOIN ems.site s ON s.id = p.site_id
  LEFT JOIN ems.site_pv_forecast_calibration cal ON cal.site_id = s.id
),
bounds AS (
  SELECT
    greatest(p.p_data_from, p.p_data_to - interval '120 days', e.delta_learn_min_ts) AS ts_from,
    p.p_data_to AS ts_to,
    greatest(e.half_life_days, 1::numeric) AS half_life_days,
    greatest(e.threshold_w, 0::numeric) AS threshold_w
  FROM params p
  CROSS JOIN eff e
),
best AS (
  SELECT
    fa.interval_start,
    fa.pv_array_id,
    fa.forecast_power_w,
    fa.actual_power_w,
    fa.forecast_created_at,
    row_number() OVER (
      PARTITION BY fa.interval_start, fa.pv_array_id
      ORDER BY fa.forecast_created_at DESC
    ) AS rn
  FROM ems.forecast_accuracy fa
  CROSS JOIN bounds b
  JOIN params p ON fa.site_id = p.site_id
  WHERE fa.interval_start >= b.ts_from
    AND fa.interval_start < b.ts_to
    AND fa.actual_power_w IS NOT NULL
    AND fa.forecast_created_at <= fa.interval_start
    AND coalesce(fa.learning_eligible, true) IS TRUE
),
slots_array AS (
  SELECT
    b.interval_start,
    b.pv_array_id,
    b.forecast_power_w::numeric AS forecast_w,
    b.actual_power_w::numeric AS actual_w,
    (
      (extract(hour FROM (b.interval_start AT TIME ZONE tz.tz_name))::int * 60)
      + extract(minute FROM (b.interval_start AT TIME ZONE tz.tz_name))::int
    ) / 15 AS slot_of_day,
    (b.interval_start AT TIME ZONE tz.tz_name)::date AS day_local,
    extract(epoch FROM (now() - b.interval_start)) / 86400.0 AS age_days
  FROM best b
  CROSS JOIN tz
  WHERE b.rn = 1
),
slots AS (
  SELECT
    sa.interval_start,
    sum(sa.forecast_w)::numeric AS forecast_total_w,
    sum(sa.actual_w)::numeric AS actual_total_w,
    sa.slot_of_day,
    sa.day_local,
    max(sa.age_days) AS age_days
  FROM slots_array sa
  GROUP BY sa.interval_start, sa.slot_of_day, sa.day_local
),
day_energy AS (
  SELECT s.day_local, sum(s.actual_total_w)::numeric / 4000.0 AS energy_kwh
  FROM slots s
  GROUP BY s.day_local
),
ref AS (
  SELECT percentile_cont(0.5) WITHIN GROUP (ORDER BY de.energy_kwh) AS med_kwh
  FROM day_energy de
),
slot_steps AS (
  SELECT
    s.*,
    lag(s.actual_total_w) OVER (PARTITION BY s.day_local ORDER BY s.interval_start) AS prev_actual_w
  FROM slots s
  WHERE s.slot_of_day BETWEEN 20 AND 80
    AND s.actual_total_w > (SELECT threshold_w FROM bounds)
),
day_jump AS (
  SELECT
    ss.day_local,
    percentile_cont(0.5) WITHIN GROUP (ORDER BY abs(ss.actual_total_w - ss.prev_actual_w)) AS med_jump_w
  FROM slot_steps ss
  WHERE ss.prev_actual_w IS NOT NULL
  GROUP BY ss.day_local
),
day_med AS (
  SELECT
    s.day_local,
    percentile_cont(0.5) WITHIN GROUP (ORDER BY s.actual_total_w) AS p50_actual_w
  FROM slots s
  WHERE s.actual_total_w > (SELECT threshold_w FROM bounds)
  GROUP BY s.day_local
),
day_stats AS (
  SELECT
    de.day_local,
    de.energy_kwh,
    dj.med_jump_w,
    dm.p50_actual_w,
    CASE
      WHEN (SELECT med_kwh FROM ref) IS NULL OR (SELECT med_kwh FROM ref) <= 0 THEN 0.5
      ELSE greatest(
        0.0,
        least(
          1.0,
          (de.energy_kwh - (SELECT med_kwh FROM ref) * 0.55)
          / nullif((SELECT med_kwh FROM ref) * 0.35, 0)
        )
      )
    END AS w_energy,
    CASE
      WHEN dj.med_jump_w IS NULL OR dm.p50_actual_w IS NULL THEN 0.35
      ELSE greatest(
        0.0,
        least(
          1.0,
          1.0
          - (
            dj.med_jump_w
            / nullif(greatest(300.0, dm.p50_actual_w * 0.25), 0)
          )
        )
      )
    END AS w_smooth
  FROM day_energy de
  LEFT JOIN day_jump dj ON dj.day_local = de.day_local
  LEFT JOIN day_med dm ON dm.day_local = de.day_local
),
day_rank AS (
  SELECT
    ds.day_local,
    row_number() OVER (
      ORDER BY
        (coalesce(ds.w_energy, 0.35) * coalesce(ds.w_smooth, 0.35)) DESC,
        ds.day_local DESC
    ) AS rn,
    (coalesce(ds.w_energy, 0.35) * coalesce(ds.w_smooth, 0.35)) AS day_score
  FROM day_stats ds
),
filtered AS (
  SELECT
    s.day_local,
    s.slot_of_day,
    exp(-s.age_days / nullif((SELECT half_life_days FROM bounds), 0))
      * (
        CASE
          WHEN (SELECT top_n_days FROM eff) IS NULL THEN 1::numeric
          WHEN (SELECT top_n_days FROM eff) < 1 THEN 1::numeric
          WHEN dr.rn <= (SELECT top_n_days FROM eff) THEN 1::numeric
          ELSE greatest(
            0::numeric,
            least(1::numeric, coalesce((SELECT non_top_day_factor FROM eff), 0.02))
          )
        END
      )
      * (
        0.05
        + 0.95
          * power(
            greatest(
              0.0,
              least(1.0, coalesce(ds.w_energy, 0.35) * coalesce(ds.w_smooth, 0.35))
            ),
            greatest(
              0.25,
              least(coalesce((SELECT day_weight_gamma FROM eff), 1.0), 8.0)
            )
          )
      ) AS w
  FROM slots s
  CROSS JOIN bounds b
  LEFT JOIN day_stats ds ON ds.day_local = s.day_local
  LEFT JOIN day_rank dr ON dr.day_local = s.day_local
  WHERE s.slot_of_day BETWEEN 0 AND 95
    AND (s.actual_total_w > b.threshold_w OR s.forecast_total_w > b.threshold_w)
),
by_day AS (
  SELECT day_local, sum(w) AS w_sum, count(*)::bigint AS slot_rows
  FROM filtered
  GROUP BY day_local
),
tot AS (
  SELECT sum(w_sum) AS w_tot FROM by_day
)
SELECT
  dr.rn AS day_rank,
  ds.day_local,
  round(ds.energy_kwh::numeric, 2) AS energy_kwh,
  round(ds.w_energy::numeric, 3) AS w_energy,
  round(ds.w_smooth::numeric, 3) AS w_smooth,
  round(dr.day_score::numeric, 4) AS day_score,
  round(coalesce(bd.w_sum, 0)::numeric, 2) AS sum_w_in_filtered,
  coalesce(bd.slot_rows, 0::bigint) AS slot_rows,
  round(
    (100.0 * coalesce(bd.w_sum, 0) / nullif((SELECT w_tot FROM tot), 0))::numeric,
    2
  ) AS pct_of_total_weight
FROM day_stats ds
JOIN day_rank dr ON dr.day_local = ds.day_local
LEFT JOIN by_day bd ON bd.day_local = ds.day_local
ORDER BY dr.rn;

-- Shrnutí okna + výstup funkce (stejné parametry jako v params výše):
-- SELECT (ems.fn_pv_forecast_delta_profile(
--   (SELECT site_id FROM params),
--   (SELECT p_data_from FROM params),
--   (SELECT p_data_to FROM params),
--   (SELECT half_life_days FROM params),
--   (SELECT threshold_w FROM params),
--   (SELECT p_top_n_days FROM params),
--   (SELECT p_non_top_day_factor FROM params),
--   (SELECT p_day_weight_gamma FROM params)
-- ))->'data_from' AS data_from,
--      (ems.fn_pv_forecast_delta_profile(...))->'data_to' AS data_to;