Multi-dimensional Analysis and Early Warning Model of Converter Transformer

Due to the difference of operation environment and working conditions of converter equipment, the threshold set by relevant operation and maintenance specifications has certain limitations in abnormal diagnosis. Based on the engineering application, a multi-dimensional analysis and early warning method is proposed. The multi-dimensional analysis algorithm of temperature, oil level and cooling capacity are established to evaluate the current state of converterfor the key parameters of converter transformer. On the basis of multi-dimensional analysis, the oil temperature prediction algorithm based on LSTM is proposed to realize the trend identification of the operation state of the converter. The algorithm model is deployed and applied in ±800 kV Suidong substation, and the results show that the algorithm model can effectively identify the abnormal operation state of converter transformer.