MATLAB Examples

A Benchmark Case for Statistic Process Monitoring - Cranfield Multiphase Flow Facility

This package contains data sets collected in the Cranfield Multiphase Flow Facility aiming to serve as a benchmark case for statistic process monitoring. Details of the benchmark case are presented in [1]. Some examples of Canonical Variate Analysis (CVA) on these data sets are presented as follows.

Contents

Reference

[1] C. Ruiz-Cárcel, Y. Cao, D. Mba, L.Lao and R. T. Samuel, Statistical Process Monitoring of a Multiphase Flow Facility, Control Engineering Practice , V. 42, PP. 74–88, 2015, http://www.sciencedirect.com/science/article/pii/S0967066115000866#

Section 1 Parameters

vIndex = 1:23; % measurement index
alpha = 0.99;  % confidence level
n = 25;        % retained state dimension
p = 15;        % length of past observation
f = 15;        % length of future observation

Section 2 Training with normal data sets 2 and 3

This block loads the available training data sets and selects the two data sets used to train the algorithm. The measurements included in both data sets are selected according to vIndex

load Training
X1 = T2(:,vIndex);
X2 = T3(:,vIndex);

Section 3 Fault detection for Faulty Case 3

This block loads the data set used for the first monitoring example, including the same measurements as the training data sets (vIndex)

load FaultyCase3
XT = Set3_1(:,vIndex);

Section 4 Fault detection result Figure 7(a) and (b) in the paper

cvatutor(alpha, n,p,f,X1,X2,XT);

Section 5 Fault detection for Faulty Case 5

This block loads the data set used for the second monitoring example, including the same measurements as the training data sets (vIndex)

load FaultyCase5
XT2 = Set5_2(:,vIndex);

Section 6 Fault detection result Figure 9(a) and (b) in the paper

cvatutor(alpha, n,p,f,X1,X2,XT2);