Anomaly detection method of multidimensional carbon data in zero-carbon park based on GMM algorithm

T=During the operation of zero- carbon park, accurate monitoring and analysis of multi-dimensional carbon data is the key to evaluate the effect of energy saving and emission reduction, optimize the energy utilization structure and improve the green development level of the park. However, due to the complexity and diversity of energy carbon data, abnormal data cannot be found and processed in time, thus affecting the green and low-carbon development process of the park. Therefore, this paper proposes an anomaly detection method of multidimensional carbon data in zero-carbon park based on GMM algorithm. The objective function of minimum carbon emission and minimum total operating cost for abnormal detection of multidimensional carbon data in zero-carbon park is constructed. According to this objective function, the constraint conditions of exchange power and energy balance in park operation are set, and GMM algorithm is introduced to realize abnormal detection of multidimensional carbon data in zero-carbon park. The experimental results show that this method can complete the anomaly detection of multi-dimensional carbon data in zero-carbon park, and the method has less missed detection times and better detection efficiency.