Simplification of raw data set during the fault detection process

A novel simplification of database technique is proposed to explicitly account for compromise of low cost and high detection performance used to fault identification in the practical industrial processes. Based on the principle of Mahalanobis distance, the samples with the similar characteristics are replaced by the mean of them, so that the number of raw data set is reduced easily. Moreover, the supper ball domains of mean and variance of samples are presented, which not only retain the statistical properties of raw data set but also avoid the reduction of data unlimitedly. Finally, numerical examples and simulations are given to illustrate the effectiveness of the proposed method.