Build the application

## Build
Monitor real-time interval telemetry across customer branches to detect bypass leakage and low delta-T syndrome. Calculate parasitic pumping energy waste and trigger automated hydraulic inspection tickets.

## User
District cooling customer services engineer

## Workflow
1. Ingest 15-minute interval energy meter telemetry including branch flow rates and return temperatures.
2. Identify sustained thermal deviations where delta-T drops significantly below network baselines.
3. Calculate parasitic pumping power overhead caused by excessive chilled water flow.
4. Generate diagnostic tickets targeting specific customer secondary pumps or two-way bypass valves.

## Interface
Real-time telemetry scatter plot showing customer delta-T versus flow, highlighting hydraulic outliers and wasted pumping kilowatt-hours.

## Data
Generate synthetic 15-minute interval telemetry records for multiple commercial branches over 30 days, simulating normal load cycles and bypass leakage anomalies.
Starter examples below are synthetic. They are not downloaded records.

## Source
[Water Network Tool for Resilience](https://raw.githubusercontent.com/USEPA/WNTR/main/README.md)
Applies hydraulic network pressure-dependent flow calculations modeled via WNTR simulation concepts to analyze continuous flow metrics.

## Tools
pydantic-ai, pandas, numpy.
Guarantees strict type casting and validation on continuous telemetry streams and complex thermodynamic calculations.
Local or hosted LLM with strong numerical reasoning for drafting technical diagnostic tickets based on time-series anomalies.

## 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
- Identifies periods where branch delta-T falls below 8.0 degrees Celsius for more than 4 consecutive hours.
- Computes excess flow and estimates parasitic pumping energy using hydraulic resistance equations.
- Generates an actionable field diagnostic ticket for suspected bypass valve failure.
- Preserves normal tracking for compliant buildings experiencing expected partial-load thermal drops.

## Starter Examples
```json
[
  {
    "input": {
      "branch_id": "Branch-A",
      "baseline_delta_t_c": 9,
      "current_delta_t_c": 4.5,
      "flow_rate_l_s": 150,
      "consecutive_hours_below_threshold": 5
    },
    "expected_output": {
      "anomaly_detected": true,
      "excess_flow_l_s": 75,
      "estimated_parasitic_kwh_per_hour": 55.2,
      "ticket_action": "dispatch_inspection"
    }
  },
  {
    "input": {
      "branch_id": "Branch-B",
      "baseline_delta_t_c": 9,
      "current_delta_t_c": 8.5,
      "flow_rate_l_s": 120,
      "consecutive_hours_below_threshold": 0
    },
    "expected_output": {
      "anomaly_detected": false,
      "excess_flow_l_s": 0,
      "estimated_parasitic_kwh_per_hour": 0,
      "ticket_action": "monitor"
    }
  }
]
```

## Safety
Produces advisory technical tickets. All maintenance dispatches must undergo engineering review before execution.
Use mock external writes. Require human review before any real deployment.