Build the application

## Build
Coordinate utility-scale solar inverter scheduled outages against planned generation losses and spare-parts inventory. Calculate revenue loss impacts and compile verified technician maintenance dossiers before staging field repairs.

## User
Renewable plant operations planner

## Workflow
1. Ingest inverter telemetry, lead times, and solar irradiance profiles using pvlib simulation models.
2. Simulate hourly power output loss for candidate outage windows across available spare part delivery dates.
3. Calculate estimated lost generation megawatt-hours and check warehouse buffer availability for critical components.
4. Assemble a structured outage coordination manifest with selected maintenance windows for human engineer sign-off.

## Interface
Web dashboard showing inverter health, solar production curves, spare part delivery timelines, and outage approval queues.

## Data
Generate synthetic hourly irradiance time-series formatted via pvlib model conventions, paired with normalized inverter bill-of-materials and warehouse records.
Starter examples below are synthetic. They are not downloaded records.

## Source
[PVLib solar modeling](https://raw.githubusercontent.com/pvlib/pvlib-python/main/README.md)
Applies pvlib Python performance modeling concepts for simulating photovoltaic generation profiles and calculating output loss during downtime.

## Tools
pydantic-ai, pvlib, pandas.
Ensures structured validation of technical spare-part schemas and typed solar maintenance outage payloads.
Fast hosted LLM with strong JSON adherence for technical schedule generation via standard cloud API.

## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

## Acceptance
- Evaluates pvlib generation loss across at least two alternate maintenance outage time slots.
- Validates spare-part availability and lead-time constraints against warehouse inventory schema.
- Produces a verified outage schedule manifest requiring explicit human engineer approval before simulated write.

## Starter Examples
```json
[
  {
    "input": {
      "inverter_id": "INV-SOUTH-04",
      "capacity_kw": 1250,
      "required_part": "IGBT-MODULE-1200V",
      "stock_available": 2,
      "proposed_window_start": "2025-06-15T04:00:00Z",
      "duration_hours": 6,
      "projected_irradiance_wm2": [
        120,
        450,
        780,
        920,
        850,
        610
      ]
    },
    "expected_output": {
      "status": "APPROVED_FOR_REVIEW",
      "inverter_id": "INV-SOUTH-04",
      "parts_cleared": true,
      "estimated_loss_kwh": 3730,
      "recommended_window": "2025-06-15T04:00:00Z",
      "engineer_signoff_required": true
    }
  },
  {
    "input": {
      "inverter_id": "INV-NORTH-01",
      "capacity_kw": 2000,
      "required_part": "DC-BUS-CAPACITOR-PACK",
      "stock_available": 0,
      "proposed_window_start": "2025-06-16T08:00:00Z",
      "duration_hours": 4,
      "projected_irradiance_wm2": [
        900,
        950,
        980,
        940
      ]
    },
    "expected_output": {
      "status": "BLOCKED_PARTS_UNAVAILABLE",
      "inverter_id": "INV-NORTH-01",
      "parts_cleared": false,
      "estimated_loss_kwh": 7540,
      "recommended_window": null,
      "engineer_signoff_required": true
    }
  }
]
```

## Safety
Simulates all maintenance schedule changes without directly triggering physical grid interlocks or inverter breakers.
Use mock external writes. Require human review before any real deployment.