A Quantum Controlled-NOT Neural Network-Based Load Forecast and Management Model for Smart Grid

The uncertainty and unpredictable power generation and usage of resources at demand as well as supply side causes the mismatch problem. Therefore, the power demand and supply management of regional smart meters have become an essential and challenging research issue in the current era. The existing solutions have inadequately explored the relationship between forecasted load and power demand supply management of smart meters, which imposes a huge impact on the performance of power services to the regional customers. In this context, a novel quantum machine learning-driven power distribution and supply management (QPD-SM) model is proposed, which estimates the power demand of regional smart meters proactively by utilizing the load forecasting values and managing power supply efficiently. The model deploys a newly developed “quantum controlled-NOT gate induced feed-forward neural network” (QCNN) which is optimized using the quantum adaptive differential evolutionary (QuADE) learning algorithm to forecast regional load with precise accuracy. Further, a demand–supply management approach is designed to analyze and estimate the over-/under-supply of power units relative to different regions and govern the power generation and supply management accordingly. Experimental results from the Irish Commission for Energy Regulation and the Australian Smart Grid Smart City show the improvement of root-mean-squared error up to 77.02% and 78.92%, respectively, by the proposed model as compared to the existing state-of-the-art works.