AI-Based Forecasting of Electricity Prices in Deregulated Markets

Abstract

This study, titled "AI-Based Forecasting of Electricity Prices in Deregulated Markets," evaluates the predictive accuracy, grid fuel mix dynamics, spot price volatility, and financial feasibility of artificial intelligence algorithms in wholesale power trading. Deregulated electricity markets are characterized by high price volatility, intermittent renewable generation, and extreme demand spikes. A five-year project lifecycle (2021-2025) of an AIdriven energy trading platform is evaluated using capital budgeting metrics: Net Present Value (NPV), Internal Rate of Return (IRR), Payback Period (PBP), and Benefit-Cost Ratio (BCR). Quantitative analysis indicates that renewable sources (solar and wind) account for 35% of power generation, driving spot price fluctuations. Implementing a hybrid Transformer-LSTM deep learning architecture reduces price forecasting Root Mean Square Error (RMSE) to 3.8 $/MWh compared to 18.5 $/MWh under baseline ARIMA models. Improved forecasting accuracy lowers annual market price volatility from 28.5% to 15.2%, supporting a grid peak demand of 75 GW and average spot prices of 7.4 INR/kWh by 2025. The financial model yields a positive NPV of 284.5 Crores and an IRR of 38.6%, far exceeding the 10% discount hurdle rate. The study concludes that investing in AI price forecasting software is highly viable, providing power traders, utilities, and grid operators with enhanced hedging capabilities and minimized imbalance charges.