Fault Detection Method for Continuous Processes Based on Integrated Kernel Local Preserving Projection

Abstract

A continuous process fault detection method based on integrated kernel local preservation projection, which involves a continuous process fault detection method for kernel parameter selection, including using historical data under normal conditions as the training set of modeling data, and using Gaussian kernel function to model The data is transformed by kernel, and the EKLPP method is used for nonlinear continuous process modeling and fault detection. First, normalize the training data matrix. The KLPP method is used to convert it into a kernel matrix to replace the data to be detected, and the Bayesian decision is used to convert each detection result into a form of probability of failure. According to whether the integrated statistics exceed the control limit, it is judged whether the data at this time is normal. If the integration statistic exceeds the control limit, the data is faulty at that moment. If the test shows that the system fails, the staff needs to find out the situation and eliminate the dangerous situation. The invention effectively solves the problem that the applicable parameters are different for different faults.

Description

technical field [0001] The invention relates to a non-linear process fault detection method, in particular to a partial fault detection method dealing with different fault core parameter selections. Background technique [0002] As one of the important fields of industrial production, the chemical production process is increasingly showing nonlinear characteristics. The traditional global fault detection method can only retain the global information of the original data and destroy its local structure. How to effectively extract the faults in the production process? The local information of the original data to monitor the nonlinear process has become an important content of fault detection technology research. [0003] The kernel locality preserving projection method has obvious advantages in dealing with the problem of preserving the local structure of nonlinear data. However, people usually use the Gaussian kernel function when processing data with the traditional kernel...

AI Technical Summary

Problems solved by technology

However, the faults in the production process are usually not unique, and the kernel parameters selected from this are not necessarily applicable to all faults. For different faults, the applicable parameters will be significantly different.

Images