The increasing penetration of renewable energy sources, particularly solar and wind power, along with the growing deployment of flexible loads and demand response programs, has introduced greater uncertainty into electricity consumption patterns and made load forecasting more challenging. In many conventional forecasting approaches, environmental factors, consumer behavior, and demand response effects are incorporated into a single integrated model. Such an approach limits the ability to distinguish the contribution of each component to the overall load pattern and may reduce forecasting accuracy.This paper proposes a hybrid framework for load forecasting that explicitly separates these components. First, to mitigate temporal shifts in load patterns caused by seasonal variations, a temporal alignment method is applied in which the sunrise and sunset times of all days of the year are mapped to a reference day. Next, the base load component is extracted and forecast using a Long Short‑Term Memory (LSTM) neural network. The effect of price signals is then modeled through an economic framework based on utility functions to estimate demand response behavior. The final load forecast is obtained by combining the predicted base load with the price-responsive component. Evaluation using real-world data shows that the proposed framework improves forecasting accuracy compared with direct total load prediction, reducing the Mean Absolute Percentage Error (MAPE) to below 7.3%. The results demonstrate the effectiveness of the proposed approach for load behavior analysis and decision support in power system operation and demand response–enabled electricity markets.