According to the current development of the electricity market, China's electricity meter equipment is in a stage of rapid growth. At the same time, with the progress and update of technology, meter equipment is also constantly upgraded. However, due to the influence of climate and other factors, the country’s power supply shortage and energy shortage are becoming increasingly prominent. In order to cope with these adverse situations and ensure the safe and economical operation of the power plant, it is necessary to strengthen the analysis and research on the relevant parameters and risks in the operation process of coal to electricity, and establish a reasonable and effective early warning mechanism to solve these problems. In this paper, based on the graph convolutional network technology, the operation analysis model of coal to electricity is designed, and the night heavy overload warning mechanism is developed. Then, this paper tests the performance of the model by simulating the operation process of the mechanism model. The test results show that the initial value of the early warning time of the mechanism optimized by the graph convolutional network technology is about 51, and shows a downward trend. This decreasing trend may represent the shortening of the early warning time of the anomaly detection system based on graph convolutional network technology, which shows the rapid early warning ability of the algorithm.