Power plant NOx prediction model
================================

The ZIP contains the Python model exported from the Neural Designer project.
The two process-variable names that were interchanged in the working CSV have
been corrected against the UCI source definition without changing the input
order, model weights or numerical predictions.

Input order
-----------
1. ambient_temperature (deg C)
2. ambient_pressure (mbar)
3. ambient_humidity (%)
4. air_filter_difference_pressure (mbar)
5. gas_turbine_exhaust_pressure (mbar)
6. turbine_inlet_temperature (deg C)
7. turbine_after_temperature (deg C)
8. turbine_energy_yield (MWh)
9. compressor_discharge_pressure (approximately bar)
10. CO (mg/m3)

Output
------
NOx (mg/m3)

Usage
-----
from power_plant_nox_model import NeuralNetwork

model = NeuralNetwork()
prediction = model.calculate_outputs([
    17.713, 1013.07, 77.867, 3.926, 25.564,
    1081.428, 546.159, 133.506, 12.061, 2.372
])

This is an educational predictive model. It is not a certified emissions
monitoring system, regulatory reporting instrument or automatic control law.
