Transformer state monitoring method and system

The invention provides a transformer state monitoring method and system, and the method comprises the steps: extracting an oil sample, analyzing the gas content, and respectively inputting a pre-trained generation model and a pre-trained discrimination model through employing each gas content and three gas ratios in a three-ratio method, and correcting a result output by the pre-trained generation model through an output result of the pre-trained discrimination model to obtain a fault type of the to-be-tested transformer. The method has the beneficial effects that the concept in a three-ratio method is integrated, and the output results of the three gas ratios and the gas content are used for correction, so that the model can be prevented from outputting an over-fitting output result, the prediction accuracy of the neural network model is improved, and the result is more real and credible. 本发明提供了一种变压器状态监测方法及系统，其中，方法包括：通过提取油样本，并分析处气体含量，利用各个气体含量以及三比值法中的三个气体比值，分别输入预训练的生成模型和预训练的判别模型，通过所述预训练的判别模型的输出结果对所述预训练的生成模型输出