> ## Documentation Index
> Fetch the complete documentation index at: https://docs.qubit.energy/llms.txt
> Use this file to discover all available pages before exploring further.
# Layer 2 Overview
> AI-powered forecasting and prediction engine for energy systems
# Layer 2: Prediction Engine
[](https://github.com/qubit-foundation/qubit-energy-forecasting)
[](https://www.python.org/)
[](https://opensource.org/licenses/MIT)
Layer 2 provides production-ready machine learning models for energy forecasting, demand prediction, and renewable generation analysis. Building on standardized data from Layer 1, this layer delivers sub-second predictions for energy system optimization.
## Available Features
Weather-based solar PV generation predictions with confidence intervals
Electricity demand prediction with behavioral and seasonal patterns
Automated feature extraction for energy time series data
Sub-second inference with batch and real-time prediction modes
## Quick Start
### Installation
```bash theme={null}
pip install qubit-energy-forecasting
# With specific ML backends
pip install qubit-energy-forecasting[deep] # TensorFlow
pip install qubit-energy-forecasting[prophet] # Facebook Prophet
pip install qubit-energy-forecasting[all] # All models
```
### Generate Your First Forecast
```python theme={null}
from qubit.forecasting import SolarForecaster
import pandas as pd
# Initialize forecaster
forecaster = SolarForecaster(
system_capacity_kw=1000,
panel_type="polycrystalline"
)
# Train on historical data
forecaster.fit(X_train, y_train)
# Generate 24-hour forecast
forecast = forecaster.predict(
weather_data,
horizon="24h",
confidence_levels=[0.80, 0.95]
)
print(f"Peak generation: {forecast.peak_value:.2f} kW")
print(f"Total energy: {forecast.total:.2f} kWh")
```
## Architecture
```mermaid theme={null}
graph LR
A[Layer 1 Data] --> B[Feature Engineering]
B --> C[ML Models]
C --> D[Predictions]
D --> E[Layer 3 Optimization]
subgraph "ML Pipeline"
B
C
end
subgraph "Models"
F[Solar Forecaster]
G[Demand Forecaster]
H[Wind Forecaster]
I[Price Forecaster]
end
C --> F
C --> G
C --> H
C --> I
```
## Technology Stack
* **Python 3.9+** - Primary language
* **scikit-learn** - Classical ML algorithms
* **TensorFlow/Keras** - Deep learning models
* **XGBoost/LightGBM** - Gradient boosting
* **Facebook Prophet** - Time series forecasting
* **pandas/numpy** - Data manipulation
* **Qubit Schemas** - Standardized data formats
* **Feature Store** - Automated feature engineering
* **Validation** - Schema and range validation
* **FastAPI** - REST API serving
* **Docker** - Containerization
* **Kubernetes** - Orchestration
* **Prometheus** - Monitoring and metrics
## Model Performance
Performance metrics from production deployments across 100+ energy assets.
### Solar Generation Forecasting
| Horizon | MAPE | RMSE | Peak Accuracy |
| -------- | ----- | ------- | ------------- |
| 1 hour | 8.5% | 12.3 kW | 94.2% |
| 24 hours | 15.2% | 28.1 kW | 87.6% |
| 7 days | 22.8% | 41.7 kW | 78.3% |
### Demand Forecasting
| Customer Type | MAPE | RMSE | Load Factor |
| ------------- | ----- | ------- | ----------- |
| Residential | 12.1% | 5.8 kW | 0.68 |
| Commercial | 8.9% | 15.2 kW | 0.74 |
| Industrial | 6.3% | 42.1 kW | 0.82 |
## Real-World Applications
* **15-minute** load forecasting for grid balancing
* **Day-ahead** renewable integration planning
* **Week-ahead** maintenance scheduling optimization
* **Seasonal** capacity planning and resource allocation
* **Intraday** generation forecasting for trading
* **Weather-dependent** O\&M scheduling
* **Performance** monitoring and anomaly detection
* **Financial** revenue and P\&L forecasting
* **Peer-to-peer** trading optimization
* **Storage** dispatch scheduling
* **EV charging** load coordination
* **Demand response** event planning
## Integration with Qubit Stack
Layer 2 seamlessly integrates with other Qubit Foundation components:
Consumes standardized TimeSeries from **Layer 1** (Schemas, Connectors, Adapters)
Automatically extracts time, weather, calendar, and lag features
Trains specialized models for each energy domain and asset type
Outputs forecasts in TimeSeries format for **Layer 3** optimization
## Available Models
### Time Series Models
Classical autoregressive models for stable patterns
Handles seasonality and holidays automatically
Deep learning for complex temporal dependencies
Gradient boosting for feature-rich predictions
Ensemble methods with uncertainty quantification
Combines multiple models for best accuracy
### Domain-Specific Forecasters
```python theme={null}
from qubit.forecasting.solar import SolarForecaster
forecaster = SolarForecaster(
system_capacity_kw=5000,
panel_type="monocrystalline",
tilt_angle=35,
azimuth=180
)
# Integrates weather data automatically
forecast = forecaster.predict(
historical_data,
weather_forecast=weather_df,
horizon="24h"
)
```
```python theme={null}
from qubit.forecasting.demand import DemandForecaster
forecaster = DemandForecaster(
customer_type="commercial",
include_weather=True,
include_calendar=True
)
forecast = forecaster.predict(
load_history,
temperature_forecast=temp_df,
horizon="24h"
)
```
## Deployment Options
Single-node deployment for development and small installations
Production deployment with auto-scaling and high availability
Event-driven predictions for variable workloads
Local inference for latency-sensitive applications
### Example Deployment
```yaml theme={null}
apiVersion: apps/v1
kind: Deployment
metadata:
name: energy-forecaster
spec:
replicas: 3
template:
spec:
containers:
- name: forecaster
image: qubit/energy-forecasting:latest
resources:
requests:
cpu: 1000m
memory: 2Gi
env:
- name: MODEL_CACHE_SIZE
value: "10"
```
## Next Steps
Install and run your first forecast in 5 minutes
Deep dive into available prediction models
Explore source code and contribute to development
Connect Layer 2 with your Layer 1 data pipeline
***
*Layer 2 Prediction Engine is production-ready and powering forecasts across renewable energy, utilities, and smart grid applications worldwide.*