Photovoltaic array rapid fault diagnosis method and system based on IDBO-LightGBM hybrid model

The invention belongs to the technical field of photovoltaic array fault diagnosis, and particularly relates to a photovoltaic array rapid fault diagnosis method and system based on an IDBO-Light GBM hybrid model. The method comprises the following steps: S1, inputting photovoltaic array sample data, and performing standardized conversion; s2, initializing parameters of a Light GBM model and parameters of an IDBO algorithm, and setting the number of dung beetle populations, the upper and lower limits of independent variables, the maximum number of iterations and the dimension size; s3, initializing a dung beetle population; s4, calculating the fitness value of the new position of the dung beetle according to the position updating formula of each dung beetle population; s5, judging whether the number of iterations reaches the maximum or not, if yes, ending iteration, comparing the optimal fitness values stored in all iterations, and outputting optimal parameters; if not, returning to the step S3; and S6, const