Dynamic Modeling and Quantum-Enhanced Forecasting of Multi-Seasonal Energy Prices in Simulated Microgrid Environments

The study address the challenge of forecasting per unit energy prices in a microgrid environment consisting of solar and hydro power resources under multi-seasonal variations. Traditional deep learning techniques such as LSTM, GRU and ESN often struggle with non-linear dependencies and volatility in energy market. To overcome these we propose a hybrid framework incorporating Adiabatic Quantum Computing (AQC) for electricity price forecasting. The proposed AQC model encodes-32 system and market related variables into quantum states and applies adiabatic evolution to derive optimized price prediction. Simulation results using real microgrid data set-up based on HIL shows that AQC reduces forecasting error by 17.03% compared to LSTM, 14.29% to GRU and 13.88% to ESN over 24-hrs and 48-hrs horizons. The enhanced accuracy and robustness of the quantum assisted model demonstrates its potential for next generation energy market forecasting and decisions making tool. The entire framework is tested using a synthetic microgrid dataset designed to emulate real-world seasonal and operational dynamics. While this enables controlled validation of the models, the generalizability of the results to real world deployment requires further empirical evaluations on physical microgrid data set.