Load data derived from the Internet of Things (IoT) often exhibits nonlinear characteristics, making it difficult for traditional statistical and machine learning methods to effectively capture the underlying data relationships. With the rapid advancement of artificial intelligence, deep learning has emerged as a powerful tool for data mining. In this context, we propose a hybrid model for short-term load forecasting that combines a Temporal Convolutional Network with TimesNet. The process initiates with a TCN employed to capture features from the initial input sequence. Following this, TimesNet converted 1D time series into a 2D tensor, which captures correlations across different scales of time sequential data within original input load data. Finally, the extracted features are fed into a fully connected layer to produce the load predictions. To validate the model’s effectiveness, we conducted experiments using a Denmark load dataset. The results demonstrate that the proposed TCN-TimesNet model excels in short-term power load forecasting.