Unlock instant, AI-driven research and patent intelligence for your innovation.
A Multi-fault Diagnosis Method of Industrial Process Based on Discriminant Analysis
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
An industrial process and discriminant analysis technology, applied in the direction of instruments, testing/monitoring control systems, control/regulation systems, etc., can solve problems that cannot be compatible with low computational complexity and high diagnostic accuracy at the same time, to ensure diagnostic accuracy, The effect of reducing computational complexity
Inactive Publication Date: 2021-06-11
HUAZHONG UNIV OF SCI & TECH
View PDF11 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0005] The present invention provides a multi-fault diagnosis method for industrial process based on discriminant analysis, which is used to solve the technical problem that the existing industrial process fault diagnosis technology cannot be compatible with low computational complexity and high diagnostic accuracy when complex multi-dimensional faults exist in fault samples
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0057] A method 100 for multi-fault diagnosis of industrial processes based on discriminant analysis, such as figure 1 shown, including:
[0058] Step 110, based on the normal sample set of the industrial process, determine the fault samples of the industrial process and establish the fault sample set;
[0059] Step 120, respectively project the normal sample set and the faulty sample set onto the same line in their space, and based on Fisher discriminant analysis, aim at the maximum inter-class scatter matrix between the projected normal sample set and the faulty sample set, and obtain the projection coefficient vector;
[0060] Step 130, based on the absolute value of each element in the projection coefficient vector, determine the reconstruction priority of each variable direction in the fault sample;
[0061] Step 140, perform multi-dimensional reconstruction on the fault samples along the first k variable directions in the reconstruction priority, so that the reconstruc...
example 1
[0130] Example 1, the specific description of the Monte Carlo numerical simulation example is as follows:
[0131]
[0132] Among them, x is the sample under normal working conditions, t 1 ,t 2 ,t 3 It is three latent variable signals with mean value 0 and standard deviation 1, 0.8 and 0.6 respectively, conforming to Gaussian distribution, noise is white noise with mean value 0 and standard deviation 0.2, noise~N(0,0.2); total 1000 normal samples to form the training set.
[0133] The fault samples to be detected are expressed in the following general form:
[0134] In Example 1, ξ=[0 1 1 0 0 0], f=1 means a step fault, and the fault starts from the 160th sample, and there are a total of 1000 fault samples to be tested.
[0135] Use the fault diagnosis method provided in this example to diagnose the fault of Example 1. The specific steps are as follows:
[0136] (1) Standardize the normal sample set matrix X, establish a principal component analysis model, select 3 p...
Embodiment 2
[0148] A storage medium, in which instructions are stored, and when a computer reads the instructions, the computer is made to execute any of the above-mentioned multi-fault diagnosis methods for industrial processes based on discriminant analysis.
[0149] The relevant technical solutions are the same as those in Embodiment 1, and will not be repeated here.
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
PUM
Login to View More
Abstract
The present invention disclosed a multi -fault diagnosis method based on the analysis of the analysis, including: normal sample set based on the industrial process to determine the faulty sample of the industrial process and establish a fault sample set; use the improved Fisher to determine the analysis of the analysis and extract the fault sample eachThe characteristic attribute of the variable; the characteristic attribute reflects the degree of the variable departing from the normal distribution. Taking the characteristics of the characteristic attribute as the reconstruction order, multi -dimensional reconstruction is performed in turn until the monitoring statistics are less than the control limit after the reconstruction.The present invention determines all the faulty variables through multi -dimensional reconstruction, and distinguishes the main faulty variables and secondary process variables through characteristic attributes; and multi -dimensional reconstruction methods are not affected by the "pollution" effect between the variables.The failure, therefore, the present invention can diagnose the failure in complex multiple faults, overcoming the "pollution" effect of traditional single -dimensional reconstruction, and at the same time, the diagnostic result can provide the fault information of the primary and secondary fault variables.
Description
technical field [0001] The invention belongs to the technical field of industrial process fault diagnosis and isolation, in particular to a multi-fault diagnosis method for industrial process based on discriminant analysis. Background technique [0002] In industrial manufacturing processes, effective process monitoring and fault diagnosis can ensure safety and improve product quality. Model-based and data-based methods have always been regarded as two schools with numerous research results. However, model-based methods often face difficulties in understanding the system mechanism in depth. With the rapid development of computer technology, more attention has been paid to data-based methods. [0003] Data-based methods include principal component analysis (PCA), partial least squares (PLS), and independent component analysis (ICA). Among them, the principal component analysis (PCA) is widely used in process monitoring. Its core idea is to decompose the high-dimensional dat...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to View More 
Login to View More