The invention discloses a power equipment fault detection method and device, computer equipment and a storage medium. The method comprises the steps of obtaining scanning image data of a power grid customization list; performing information extraction and identification on the scanned image data by using a pre-established information extraction neural network to obtain rated operation parameters of the to-be-tested power equipment; obtaining actual operation parameters of the to-be-tested power equipment; comparing the rated operation parameters with the actual operation parameters to obtain acomparison result; and judging whether the to-be-tested power equipment has a fault or not based on the comparison result. The scanning image data of the custom list is subjected to information recognition and extraction through the neural network, automatic analysis of the custom list is achieved, then the custom list is compared with the actual operation parameters to judge whether the power equipment breaks down or not, and the fault detection efficiency of the power equipment is improved.