Build the application

## Build
Model district cooling plant chiller load shifting and stratified thermal storage tank charging during off-peak night hours. Reduce peak-demand electrical charges while guaranteeing daytime cooling tonnage.

## User
District Cooling Operations Lead

## Workflow
1. Ingest commercial customer cooling load forecasts and thermal energy storage (TES) state-of-charge.
2. Retrieve time-differentiated electricity tariffs and chiller coefficient-of-performance curves.
3. Formulate optimal nocturnal TES charging sequences using building energy balance principles from EnergyPlus.
4. Produce verified chiller staging plans and thermal reserve reports for operational shift sign-off.

## Interface
A real-time plant load balancing dashboard showing TES thermocline levels, chiller staging Gantt charts, and demand tariff cost deltas.

## Data
Synthesize 24-hour district cooling demand profiles (ton-hours), TES tank capacity limits, and utility tariff rate tables in JSON.
Starter examples below are synthetic. They are not downloaded records.

## Source
[EnergyPlus](https://raw.githubusercontent.com/NREL/EnergyPlus/develop/README.md)
Employs EnergyPlus chilled water loop and thermal storage tank energy balance conventions to model charging and discharging dynamics.

## Tools
dspy, pydantic, fastapi.
Compiles and optimizes structured prompt-based reasoning for complex thermodynamic scheduling decisions under tariff constraints.
Local or hosted LLM with strong math reasoning capabilities for validating thermal balance calculations.

## 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
- Maintains TES state-of-charge within safe operating bounds without overcharging.
- Demonstrates load shifting away from designated peak electrical tariff hours.
- Outputs hourly chiller stage assignments and expected peak demand cost reductions.
- Enforces mandatory engineering review check prior to transmitting setpoints to plant operators.

## Starter Examples
```json
[
  {
    "input": {
      "plant_id": "DCP-MARINA",
      "tes_capacity_ton_hours": 20000,
      "initial_tes_storage_ton_hours": 4000,
      "max_charge_rate_tons": 2500,
      "forecast_peak_cooling_demand_tons": 8000,
      "chiller_base_capacity_tons": 6000,
      "off_peak_hours": [
        0,
        1,
        2,
        3,
        4,
        5
      ]
    },
    "expected_output": {
      "recommended_night_charge_ton_hours": 15000,
      "projected_peak_coverage_deficit_tons": 0,
      "chiller_night_utilization_ratio": 0.83,
      "peak_shaving_feasible": true,
      "status": "APPROVED_FOR_REVIEW"
    }
  },
  {
    "input": {
      "plant_id": "DCP-MARINA",
      "tes_capacity_ton_hours": 10000,
      "initial_tes_storage_ton_hours": 1000,
      "max_charge_rate_tons": 1000,
      "forecast_peak_cooling_demand_tons": 9000,
      "chiller_base_capacity_tons": 6000,
      "off_peak_hours": [
        2,
        3,
        4
      ]
    },
    "expected_output": {
      "recommended_night_charge_ton_hours": 3000,
      "projected_peak_coverage_deficit_tons": 2000,
      "chiller_night_utilization_ratio": 1,
      "peak_shaving_feasible": false,
      "status": "SUPPLEMENTAL_CHILLER_REQUIRED"
    }
  }
]
```

## Safety
Generates advisory scheduling proposals; thermal setpoints require confirmation by qualified plant shift engineers.
Use mock external writes. Require human review before any real deployment.