Fault Identification of Chemical Processes Based on k-NN Variable Contribution and CNN Data Reconstruction Methods

Abstract

Data-driven fault detection and identification methods are important in large-scale chemical processes. However, some traditional methods often fail to show superior performance owing to the self-limitations and the characteristics of process data, such as nonlinearity, non-Gaussian distribution, and multi-operating mode. To cope with these issues, the k-NN (k-Nearest Neighbor) fault detection method and extensions have been developed in recent years. Nevertheless, these methods are primarily used for fault detection, and few papers can be found that examine fault identification. In this paper, in order to extract effective fault information, the relationship between various faults and abnormal variables is studied, and an accurate “fault–symptom” table is presented. Then, a novel fault identification method based on k-NN variable contribution and CNN data reconstruction theories is proposed. When there is an abnormality, a variable contribution plot method based on k-NN is used to calculate the contribution index of each variable, and the feasibility of this method is verified by contribution decomposition theory, which includes a feasibility analysis of a single abnormal variable and multiple abnormal variables. Furthermore, to identify all the faulty variables, a CNN (Center-based Nearest Neighbor) data reconstruction method is proposed; the variables that have the larger contribution indices can be reconstructed using the CNN reconstruction method in turn. The proposed search strategy can guarantee that all faulty variables are found in each sample. The reliability and validity of the proposed method are verified by a numerical example and the Continuous Stirred Tank Reactor system.

Optical frequency multiplication using residual network with random forest regression

“… is the half-wave switching voltage of the two modulators, and are the modulation indexes of MZM1 and MZM2. The two MZMs are operated in a push-pull mode, which helps in achieving balanced modulation and cancellation of unwanted signals [ 44 ].…”

“…Additionally, a modified CNN model has been presented that uses a random forest (RF) classifier [ 43 ]. A data reconstruction method based on k-Nearest Neighbour (k-NN) variable contribution and convolutional neural network (CNN) is proposed to extract effective error information [ 44 ]. A network comprising of CNN layers and a multi-output regressor is proposed to estimate points [ 45 ].…”

Survey on Exact kNN Queries over High-Dimensional Data Space

“…Applications in sensor networks include intrusion detection systems [ 6 , 7 ], fault detection [ 8 ], fault identification [ 9 ], fault classification [ 10 ], fall prediction [ 11 ], indoor localisation [ 12 , 13 ], etc. The intrusion detection system [ 6 ] can distinguish between unusual and common nodes by monitoring their anomalous actions.…”

A Multi-Sensor System for Silkworm Cocoon Gender Classification via Image Processing and Support Vector Machine

“…Cai et al [16] demonstrated a prototype with an appropriate classification tool for successfully discriminating the gender of silkworm by X-ray imaging the cocoons. Shape features such as major axis, minor axis, ratio of major axis to minor axis, eccentricity, roundness, rectangularity, complexity, concave and convex characteristics of the chrysalis are extracted from X-ray images and inputted to pre-trained classifiers such as k-Nearest Neighbor (kNN) [17], Linear Discriminant Analysis (LDA) [18], Neural Networks (NN) [19] and Support Vector Machine (SVM) [20] to accurately classify the cocoons as male or female. The authors considered 1071 samples from three hybrid breeds and have reported an accuracy of 93.68% with kNN classifier.…”