Research on the online monitoring of metal oxide arrester based on genetic algorithm

To tackle the aging problem of metal oxide arresters (MOA) in online monitoring, a novel genetic algorithm (GA)-based method is proposed for monitoring MOA degradation. Using the operating voltage and measured leakage current of MOA, the GA's optimization capability is employed to determine the parameters k and c in the MOA equivalent model, which vary with aging. This approach ultimately enables the monitoring of MOA degradation. Additionally, simulations using Matlab are conducted to analyze the effects of voltage harmonics, frequency fluctuations, and voltage fluctuations in the power grid on the algorithm. The study demonstrates that the new algorithm can adapt the calculated leakage current to match the measured leakage current of MOA (with a standard deviation of 1.498%) and accurately calculate the parameters k and c . Furthermore, Matlab simulation results show that the maximum errors of k and c are only 0.08% and 0.05% respectively, even under voltage harmonics, voltage fluctuations, and frequency fluctuations. This demonstrates the strong anti-interference robustness of the proposed algorithm. By relying on only two simple parameters, the method also simplifies the monitoring process, improving both computational efficiency and practical applicability for online MOA aging assessment.