Market abuse evaluation of generators with time-varying pricing

In China's electricity spot market, the time variation of quotation is an important challenge in market regulation, which directly affects the accurate assessment of market abuse. To address this issue, this paper introduces a novel evaluation approach utilizing a cost-sensitive support vector machine (CSVM), considerably enhancing the detection accuracy of violations in datasets characterized by imbalance. Experiments show that this method achieves high recall rate and low consumption in both synthetic and real data, and provides an effective tool for real-time monitoring of the power market.