Machine learning model training method and system for power load identification

The embodiment of the invention provides a machine learning model training method and system for power load identification. Specifically, actually measured electrical parameter data is taken as a basis; basic electrical parameter data is unified in format, trained and input into a neural network model for continuous optimization; parameters of the model are continuously adjusted by verifying a data set so as to select an optimal model; meanwhile, the performance of the model is evaluated by utilizing the test data set; the optimal effect is achieved; and the model can be further applied to a power load identification system. According to the method, the model can be trained according to the input sampling data, so that the electric equipment in use can be identified according to waveform sampling data of specific voltage, current and active power; therefore, manual parameter adjustment and feature extraction are not needed, and the feature parameters required for identifying the powerload can be autonomously