Short-Term Load Forecasting Method with Dynamic Response to Time-of-Use Electricity Pricing

In the context of the steady development of modern power systems, short-term load forecasting is crucial to the planning and operation of the power grid. Traditional load forecasting methods are effective for linear problems, but often do not work well when dealing with highly fluctuating or nonlinear loads. Therefore, this paper adopts a POA-TCN-BiLSTM model for short-term power load forecasting. The Pelican Optimization Algorithm (POA) adjusts the periodic components of the data to reduce noise interference. Temporal convolutional network (TCN) effectively captures the local features of load data through convolution, while bidirectional long short-term memory (BiLSTM) exploits its bidirectional structure to extract long-term dependencies in time series data. In addition, the application of demand response (DR) in power systems is also explored. By dynamically responding to time-of-use electricity prices, we optimize grid operations and reduce energy costs and carbon emissions. The results show that the accuracy of this algorithm is more outstanding compared with other models. This method adjusts the power load according to changes in electricity prices, effectively reduces electricity consumption during periods of high electricity prices and increases electricity consumption during periods of low electricity prices, ensuring economic operation and enhancing grid stability. This research provides an efficient and economical method for power system load forecasting and dispatching.