The invention discloses a
sewage treatment fault diagnosis method based on a weighted
extreme learning machine integrated
algorithm, and the method comprises the steps: employing an integrated
algorithm Adaboost as the overall
algorithm frame of classification learning; initializing the method through employing an improved
sample weight value; employing a weighted
extreme learning machine as a base classifier, carrying out the iterative updating of the characteristics of the
sample weight value through an integrated algorithm,
processing the imbalance data, and combining with the nonlinear mapping of a kernel function to improve the linearly separable degree of data. On the basis of the integrated algorithm, the method employs a weighted
extreme learning machine as the base classifier, can achieve the classification of imbalance data of a plurality of classes, improves the classification performance of imbalance data, and effectively improves the fault diagnosis accuracy in a
sewage treatment process.