Model, price and monitor your own solar and battery system
Start here Where to Get Your Own Half-Hourly Electricity Data (And Why It Changes Everything)
- Sizing From Your Own Consumption Data 1 supporting page7 articles 17 min
- Forecasting What Your Roof Will Actually Generate 1 supporting page6 articles 12 min
- Cost Models and AI Quote Analysis 1 supporting page5 articles 12 min
- Monitoring and Anomaly Detection 1 supporting page4 articles 13 min
- Battery and Tariff Optimisation 1 supporting page4 articles 13 min
- AI Tools, Prompts and Data Pipelines 1 supporting page5 articles 12 min
Latest writing
- Discount Rates, Energy Inflation and the Honest Payback Number Replaces simple payback with NPV and IRR, shows how sensitive the result is to assumed energy price inflation, and argues for quoting a range with stated assumptions instead of a single reassuring year count. 6 min
- Heat Pump, EV or Neither: Sizing for the Consumption You'll Have in 2030 Builds forward load scenarios by grafting modelled heat pump and EV profiles onto your existing data, arguing that sizing to today's consumption locks in an undersized array just before your demand doubles. 7 min
- Degradation, Soiling and Tilt: Building an Honest Loss Stack Assembles every derate factor from module to meter with UK-specific values and evidence for each, then shows how to apply them multiplicatively rather than the lazy flat 'minus 20%' most models use. 8 min
- Local LLMs vs Cloud for Your Household Energy Data Assesses the real trade-off for energy analysis specifically — half-hourly data is an occupancy record — comparing local model capability on tabular reasoning against cloud quality, with a hybrid split that keeps raw data at home. 6 min
- Arbitrage Maths: When Charging From the Grid Beats Storing Your Own Solar Derives the break-even condition where cheap import plus round-trip losses undercuts the opportunity cost of stored solar, and shows the counter-intuitive result that on some tariffs you should export in summer and import at night. 7 min