Abnormity monitoring method for non-stationary nonlinear industrial process

Abstract

The invention discloses an anomaly monitoring method for a non-stationary nonlinear industrial process, and belongs to the field of industrial process anomaly monitoring. On the basis of probabilistic stationary subspace analysis, a kernel probability stationary subspace analysis method is provided by using a kernel skill, and non-stationary and nonlinear characteristics of a complex industrial process can be processed at the same time. The method comprises the following steps: firstly, mapping nonlinear data to a high-dimensional feature space, and establishing a linear model in the high-dimensional feature space; secondly, estimating parameters of the model by using an expectation maximization algorithm; by introducing a kernel skill, realizing a parameter learning process by using a kernel function without knowing an explicit expression of nonlinear mapping. Based on a kernel probability stationary subspace model, two detection indexes are proposed for monitoring a non-stationary nonlinear industrial process. Compared with an original probabilistic stationary subspace analysis method, the non-linear relation in the measured variables can be effectively extracted, and therefore the method is more suitable for monitoring the non-stationary industrial process with the non-linear characteristic at the same time.

Description

technical field [0001] The invention belongs to the field of industrial process abnormality monitoring, and in particular relates to an abnormality monitoring method for non-stationary nonlinear industrial processes. Background technique [0002] With the increasing complexity of modern industrial processes and the increasing demands on system safety and production efficiency, process monitoring technology is crucial to ensure the normal and efficient operation of industrial systems. Process monitoring methods can generally be classified into model-based, knowledge-based and data-driven methods. Compared with the other two methods, the data-driven monitoring method does not require physical models and expert knowledge, so it has received more and more attention in recent years. With the development of sensor and measurement technology, a large amount of industrial process operation data can be recorded in the distributed control system (DCS), which also promotes the progres...

AI Technical Summary

Problems solved by technology

Among them, the neural network has strong nonlinear representation and learning capabilities, but its training process often consumes a lot of time, and the structure design and parameter adjustment largely depend on experience and skills.

Images