Industrial process fault diagnosis method based on multiple classifiers and D-S evidence fusion

Abstract

The invention discloses an industrial process fault diagnosis method based on multiple classifiers and D-S evidence fusion. The method comprises the steps that firstly, independent repeated sampling is conducted according to fault data in the industrial process; secondly, the multiple classifiers are applied to new training data, respective off-line modeling models are obtained, and meanwhile the properties of all the classifiers are represented in the form of a fusion matrix; thirdly, different types of elementary probability valuation functions are calculated according to the D-S evidence theory, decisions of the multiple classifiers are selectively integrated and synthesized according to the similarity index, a combined elementary probability valuation function is obtained, and a final classified diagnosis result is obtained by means of comparison. Compared with other methods in the prior art, the industrial process fault diagnosis method can greatly improve the diagnosis effect of the industrial process, shorten delayed diagnosis time and increase the diagnosis accuracy rate, improves the monitoring performance to a great extent, enhances the comprehension ability and operation confidence of process operators in the process, and is more beneficial to automatic implementation of the industrial process.

Description

technical field [0001] The invention belongs to the field of industrial process control, in particular to an industrial process fault diagnosis method based on multi-classifier and Dempster-Shafer (D-S) evidence fusion. Background technique [0002] In recent years, the monitoring of industrial production process has been paid more and more attention by the industry and academia. On the one hand, the actual industrial process is complex, has many operating variables, and has nonlinear, non-Gaussian, and dynamic stages. Under a single assumption, the monitoring effect of a certain method has great limitations. On the other hand, if the process is not well monitored and possible faults are diagnosed, operational accidents may occur, which may affect the quality of the product in the slightest, and cause loss of life and property in the severest case. Therefore, finding a better process monitoring method and making timely and correct forecasts has become one of the research ho...

AI Technical Summary

Problems solved by technology

Traditional monitoring methods assume that the process operates under a single condition, which can no longer meet the monitoring requirements of actual industrial processes

Even if the different operating conditions of the process are modeled separately, satisfactory monitoring results cannot be achieved

Because when monitoring new process data, it is necessary to combine process knowledge to judge the working conditions of the data and select the corresponding monitoring model, which greatly enhances the dependence of monitoring methods on process knowledge, which is not conducive to the automation of industrial processes implement

Images