Smart power grid attack positioning method based on convolutional neural network

The invention discloses a smart power grid attack positioning method based on a convolutional neural network, and the method comprises the steps: obtaining measurement data based on a nonlinear measurement equation of a smart power grid under an AC model, transmitting the data to a bad data detection device, carrying out the data preprocessing, removing bad data, and enabling the preprocessed measurement data to serve as a training sample; then, describing a smart power grid attack positioning problem as a multi-label classification problem, and extracting internal high-dimensional features of measurement data under an alternating current model by adopting a convolutional neural network; and finally, taking the trained convolutional neural network as a multi-label classifier, and embedding the multi-label classifier into a bad data detection device to realize real-time online attack positioning. By adopting the method, attack positioning of the smart power grid can be realized, and the method has the followin