CN1996191A - Industrial process nonlinear fault diagnosis system and method based on fisher - Google Patents

Abstract

A nonlinear failure diagnostic system based fisher industrial process comprises industrial process object connected on site intelligent meter, DCS system, upper position machine, with the said DCS system made of data interface, control station, data base, the intelligent meter DCS system and the upper position machine connected sequentially, with the upper position machine composed of standardized handling module, fisher judging analysis module and diagnostic module. It also puts forward a failure diagnostic method. It provides a good failure diagnostic effect with extensive application based on fisher industrial process nonlinear failure diagnostic system and method.

Description

A kind of industrial process nonlinear fault diagnosis system and method based on fisher

(1) technical field

The present invention relates to the industrial process fault diagnosis field, especially, relate to a kind of industrial process nonlinear fault diagnosis system and method based on fisher.

(2) background technology

Because product quality, economic benefit, safety and environmental protection requirement, it is very complicated that industrial process and relevant control system become, and in order to guarantee the normal operation of industrial system, Fault Diagnosis is being played the part of very important role with detection in industrial process.In recent years, statistical study is applied to process monitoring and fault diagnosis has obtained extensive studies.

Utilize industrial measured data, adopt the method for statistics to carry out fault diagnosis, avoided complicated Analysis on Mechanism, it is convenient relatively to find the solution.But present most of method for diagnosing faults all have certain requirement to the distribution or the covariance distribution of variable, and such as requiring variable to satisfy Gaussian distribution etc., and the data in the industrial processes often probably do not satisfy these requirements.Therefore, often can not get good fault diagnosis effect.

(3) summary of the invention

In order to overcome poor for applicability, the deficiency that diagnosis effect is relatively poor of existing fault diagnosis system, the invention provides a kind of applied widely, the industrial process nonlinear fault diagnosis system and method based on fisher that can access good fault diagnosis effect.

The technical solution adopted for the present invention to solve the technical problems is:

A kind of industrial process nonlinear fault diagnosis system based on fisher comprises the field intelligent instrument, DCS system and the host computer that are connected with industrial process object, and described DCS system is made of data-interface, control station, database; Intelligence instrument, DCS system, host computer link to each other successively, and described host computer comprises:

The standardization module is used for data are carried out standardization, and the average of each variable is 0, and variance is 1, obtains input matrix X, adopts following process to finish:

1) computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i,$

2) calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}}),$

3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x},$

Wherein, TX is a training sample, and N is a number of training;

Fisher discriminatory analysis module is used for sample is carried out analyzing and diagnosing, determines sorter model, adopts following process:

1) scatter matrix B between interior scatter matrix W of the group of calculation training sample and group;

2) calculate | the generalized character root λ of B-λ W|=0 and proper vector 1;

3) get proper vector corresponding to non-0 characteristic root, following transform:

$\left\{\begin{array}{c}{Y}_1=l_{11}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{12}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{1m}{X}_m^{\&prime;}\\ .\\ .\\ .\\ Y_L=l_{L1}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{L2}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{\mathrm{Lm}}{X}_m^{\&prime;}\end{array}\right.$

Wherein L is the number of non-0 characteristic root, X _i' be the data after the standardization;

L index when 4) calculating the normal and fault of system respectively by above transform is the center of gravity of all kinds of states, gets top n by the parameter setting;

5) sample to be tested obtains L index through same conversion, gets top n, calculates it and the distance of all kinds of state centers of gravity respectively, and it is classified as a class apart from minimum;

Signal acquisition module is used to set the time slot of each sampling, the signal of collection site intelligence instrument;

The diagnostic data determination module, the data that are used for gathering are sent to the DCS real-time data base, from the real-time data base of DCS database, obtain up-to-date variable data as diagnostic data VX at each timing cycle;

Fault diagnosis module is used for data to be tested VX the time is obtained with training And σ _x ²Carry out standardization, and with the input of the data after the standardization as fisher discriminatory analysis module, the discriminant function that will input substitution training obtains, the computational discrimination functional value is differentiated the state of described industrial process.

As preferred a kind of scheme: described host computer also comprises: the discrimination model update module, be used for regularly adding the normal point of process status to training set VX, output to the standardization module, and the sorter model in the fisher discriminatory analysis module of renewal host computer.

As preferred another kind of scheme: described host computer also comprises: display module as a result, be used for fault diagnosis result is passed to the DCS system, and at the control station procedure for displaying state of DCS, and by DCS system and fieldbus process status information is delivered to operator station and shows.

A kind of described method for diagnosing faults of industrial process nonlinear fault diagnosis system method based on fisher may further comprise the steps:

(1), determine the key variables that fault diagnosis is used, from the historical data base of DCS database respectively during the normal and fault of acquisition system the data of described variable as training sample TX;

(2), in the fisher of host computer discriminatory analysis module, parameters such as discriminant function number N are set, and set the sampling period among the DCS;

(3), training sample TX in host computer, data are carried out standardization, make that the average of each variable is 0, variance is 1, obtains input matrix X, adopts following process to finish:

3.1) computation of mean values: $\stackrel{\&OverBar;}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{TX}}_i,$

3.2) the calculating variance: ${\mathrm{\&sigma;}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}({\mathrm{TX}}_i-\stackrel{\&OverBar;}{\mathrm{TX}}),$

3.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\&OverBar;}{\mathrm{TX}}}{{\mathrm{\&sigma;}}_x},$

Wherein, N is a number of training.

(4), again data are carried out the fisher discriminatory analysis, concrete step is:

4.1) scatter matrix B between scatter matrix W and group in the group of calculation training sample;

4.2) calculate | the generalized character root λ of B-λ W|=0 and proper vector 1;

4.3) get proper vector corresponding to non-0 characteristic root, get following transform:

$\left\{\begin{array}{c}{Y}_1=l_{11}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{12}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{1m}{X}_m^{\&prime;}\\ .\\ .\\ .\\ Y_L=l_{L1}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{L2}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{\mathrm{Lm}}{X}_m^{\&prime;}\end{array}\right.$

Wherein L is the number of non-0 characteristic root, X _i' be the data after the standardization;

4.4) L index when calculating the normal and fault of system respectively by above transform, be the center of gravity of all kinds of states, get top n by the parameter setting;

4.5) sample to be tested obtains L index through same conversion, gets top n, calculates it and the distance of all kinds of state centers of gravity respectively, and it is classified as a class apart from minimum;

(5), the data of gathering are sent in the DCS real-time data base, from the real-time data base of DCS database, obtain up-to-date variable data at each timing cycle as diagnostic data VX; Data to be tested VX the time is obtained with training

And σ _x ²Carry out standardization, and with the input of the data after the standardization as fisher discriminatory analysis module, the discriminant function that will input substitution training obtains, the computational discrimination functional value is differentiated the state of described industrial process.

As preferred a kind of scheme: described method for diagnosing faults also comprises: (6), regularly the normal point of process status is added among the training set VX, repeat the training process of (3), so that the sorter model in the fisher discriminatory analysis module of the host computer that upgrades in time.

As preferred another scheme: in described (5), the computational discrimination functional value, and on the man-machine interface of host computer the state of procedure for displaying, host computer is passed to the DCS system with fault diagnosis result, and, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows at the control station procedure for displaying state of DCS.

Variable is distributed the Fisher criterion and covariance distribution does not require, applied widely, therefore extensively applies to the every field of discriminatory analysis.The top priority of fault diagnosis is that the sample from industrial process is divided into normal and fault two classes, can be summed up as the discriminatory analysis problem, so the present invention introduces the fault diagnosis that the fisher criterion is carried out process, can be widely used in various industrial processs.

Beneficial effect of the present invention mainly shows: 1, applied widely, can access good fault diagnosis effect; 2, can be widely used in various industrial processs.

(4) description of drawings

Fig. 1 is the hardware structure diagram of fault diagnosis system proposed by the invention.

Fig. 2 is a fault diagnosis system functional block diagram proposed by the invention.

Fig. 3 is the theory diagram of host computer of the present invention.

(5) embodiment

Below in conjunction with accompanying drawing the present invention is further described.

Embodiment 1

With reference to Fig. 1, Fig. 2, Fig. 3, a kind of industrial process nonlinear fault diagnosis system based on fisher, comprise the field intelligent instrument 2, DCS system and the host computer 6 that are connected with industrial process object 1, described DCS system is made of data-interface 3, control station 4, database 5; Intelligence instrument 2, DCS system, host computer 6 link to each other successively by fieldbus, and described host computer 6 comprises:

Standardization module 7 is used for data are carried out standardization, makes that the average of each variable is 0, variance is 1, and obtains input matrix X, adopts following process to finish:

1) computation of mean values: $\stackrel{\&OverBar;}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{TX}}_i,$

2) calculate variance: ${\mathrm{\&sigma;}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}({\mathrm{TX}}_i-\stackrel{\&OverBar;}{\mathrm{TX}}),$

3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\&OverBar;}{\mathrm{TX}}}{{\mathrm{\&sigma;}}_x},$

Wherein, TX is a training sample, and N is a number of training;

Fisher discriminatory analysis module 8 is used for sample is carried out analyzing and diagnosing, determines sorter model, adopts following process:

1) scatter matrix B between interior scatter matrix W of the group of calculation training sample and group;

2) calculate | the generalized character root λ of B-λ W|=0 and proper vector l;

3) get proper vector corresponding to non-0 characteristic root, following transform:

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="Y"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">Y</figure-callout>}}_1=l_{11}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{12}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{1m}{X}_m^{\&prime;}\\ .\\ .\\ .\\ Y_L=l_{L1}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{L2}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{\mathrm{Lm}}{X}_m^{\&prime;}\end{array}\right.$

Wherein L is the number of non-0 characteristic root, X _i' be the data after the standardization;

L index when 4) calculating the normal and fault of system respectively by above transform is the center of gravity of all kinds of states, gets top n by the parameter setting;

5) sample to be tested obtains L index through same conversion, gets top n, calculates it and the distance of all kinds of state centers of gravity respectively, and it is classified as a class apart from minimum;

Signal acquisition module 9 is used to set the time slot of each sampling, the signal of collection site intelligence instrument;

Diagnostic data determination module 10, the data that are used for gathering are sent to the DCS real-time data base, from the real-time data base of DCS database, obtain up-to-date variable data as diagnostic data VX at each timing cycle;

Fault diagnosis module 11 is used for data to be tested VX the time is obtained with training And σ _x ²Carry out standardization, and with the input of the data after the standardization as fisher discriminatory analysis module, the discriminant function that will input substitution training obtains, the computational discrimination functional value is differentiated the state of described industrial process.

Described host computer also comprises: discrimination model update module 12, be used for regularly adding the normal point of process status to training set VX, and output to the standardization module, and the sorter model in the fisher discriminatory analysis module of renewal host computer.

Described host computer also comprises: display module 13 as a result, are used for fault diagnosis result is passed to the DCS system, and at the control station procedure for displaying state of DCS, and by DCS system and fieldbus process status information are delivered to operator station and show.

The hardware structure diagram of the industrial process fault diagnosis system of present embodiment as shown in Figure 1, described fault diagnosis system core is made of the host computer 6 that comprises standardization module 7 and 8 liang of big functional modules of fisher discriminatory analysis module and man-machine interface, comprise in addition: field intelligent instrument 2, DCS system and fieldbus.Described DCS system is made of data-interface 3, control station 4, database 5; Industrial process object 1, intelligence instrument 2, DCS system, host computer 6 link to each other successively by fieldbus, realize uploading and assigning of information flow.Fault diagnosis system is moved on host computer 6, can carry out message exchange with first floor system easily, in time the answering system fault.

The functional block diagram of the fault diagnosis system of present embodiment mainly comprises standardization module 7, fisher discriminatory analysis module 8 etc. as shown in Figure 2.

Described method for diagnosing faults is realized according to following steps:

1, determine the key variables that fault diagnosis is used, from the historical data base of DCS database 5 respectively during the normal and fault of acquisition system the data of these variablees as training sample TX;

2, in the fisher of host computer 6 discriminatory analysis module 8, parameters such as discriminant function number N are set, and set the sampling period among the DCS;

3, training sample TX passes through functional modules such as standardization 7, fisher discriminatory analysis 8 successively in host computer 6, adopts following steps to finish the training of diagnostic system:

1) in the standardization functional module 7 of host computer 6, data are carried out standardization, make that the average of each variable is 0, variance is 1, obtains input matrix X.Adopt following process to finish:

1. computation of mean values: $\stackrel{\&OverBar;}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{TX}}_i,$

2. calculate variance: ${\mathrm{\&sigma;}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}({\mathrm{TX}}_i-\stackrel{\&OverBar;}{\mathrm{TX}}),$

3. standardization: $X=\frac{\mathrm{TX}-\stackrel{\&OverBar;}{\mathrm{TX}}}{{\mathrm{\&sigma;}}_x},$

4. wherein N is a number of training.

The standardization that the standardization functional module 7 of host computer 6 is carried out can eliminate each variable because the influence that the dimension difference causes.

2) scatter matrix B between interior scatter matrix W of the group of calculation training sample and group;

3) calculate | the generalized character root λ of B-λ W|=0 and proper vector l;

4) get proper vector corresponding to non-0 characteristic root, following transform:

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="Y"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">Y</figure-callout>}}_1=l_{11}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{12}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{1m}{X}_m^{\&prime;}\\ .\\ .\\ .\\ Y_L=l_{L1}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{L2}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{\mathrm{Lm}}{X}_m^{\&prime;}\end{array}\right.$

Wherein L is the number of non-0 characteristic root, X _i' be the data after the standardization;

L index when 5) calculating the normal and fault of system respectively by above transform as the center of gravity of all kinds of states, got top n by the parameter setting;

4, system begins to put into operation:

1) uses timer, set the time interval of each sampling;

2) field intelligent instrument 2 testing process data and being sent in the real-time data base of DCS database 5;

3) host computer 6 from the real-time number pick storehouse of DCS database 5, obtains up-to-date variable data, as diagnostic data VX at each timing cycle;

4) data to be tested VX in the standardization functional module 7 of host computer 6, obtains during with training And σ _x ²Carry out standardization, and with the input of the data after the standardization as fisher discriminatory analysis module 8;

5) the fisher discriminatory analysis module 8 in the host computer 6, the discriminant function that will input substitution training obtains, the computational discrimination functional value is differentiated and the state of procedure for displaying on the man-machine interface of host computer 6;

6) host computer 6 is passed to DCS with fault diagnosis result, and at the control station 4 procedure for displaying states of DCS, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows, makes the execute-in-place worker in time to tackle.

5, discrimination model upgrades

In system put into operation process, regularly the point that the process status coin is normal added among the training set VX, and the training process of repeating step 3 is so that the sorter model in the fisher discriminatory analysis module 8 of the host computer 6 that upgrades in time keeps discrimination model to have effect preferably.

Embodiment 2

With reference to Fig. 1, Fig. 2, Fig. 3, a kind of described method for diagnosing faults of industrial process nonlinear fault diagnosis system method based on fisher may further comprise the steps:

(1), determine the key variables that fault diagnosis is used, from the historical data base of DCS database 5 respectively during the normal and fault of acquisition system the data of described variable as training sample TX;

(2), in the fisher of host computer 6 discriminatory analysis module 8, parameters such as discriminant function number N are set, and set the sampling period among the DCS;

(3), training sample TX in host computer, carry out standardization in 7 pairs of data of standardized module, make that the average of each variable is 0, variance is 1, obtains input matrix X, adopts following process to finish:

3.1) computation of mean values: $\stackrel{\&OverBar;}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{TX}}_i,$

3.2) the calculating variance: ${\mathrm{\&sigma;}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}({\mathrm{TX}}_i-\stackrel{\&OverBar;}{\mathrm{TX}}),$

3.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\&OverBar;}{\mathrm{TX}}}{{\mathrm{\&sigma;}}_x},$

Wherein, N is a number of training.

(4), again data are carried out the fisher discriminatory analysis, concrete step is:

4.1) scatter matrix B between scatter matrix W and group in the group of calculation training sample;

4.2) calculate | the generalized character root λ of B-λ W|=0 and proper vector l;

4.3) get proper vector corresponding to non-0 characteristic root, get following transform:

$\left\{\begin{array}{c}{\mathrm{<figure-callout\; id="1"\; label="Y"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">Y</figure-callout>}}_1=l_{11}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{12}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{1m}{X}_m^{\&prime;}\\ .\\ .\\ .\\ Y_L=l_{L1}{\mathrm{<figure-callout\; id="1"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_1^{\&prime;}+l_{L2}{\mathrm{<figure-callout\; id="2"\; label="X"\; filenames="A20061015482600141.png"\; state="\{\{state\}\}">X</figure-callout>}}_2^{\&prime;}+...+l_{\mathrm{Lm}}{X}_m^{\&prime;}\end{array}\right.$

Wherein L is the number of non-0 characteristic root, X _i' be the data after the standardization;

4.4) L index when calculating the normal and fault of system respectively by above transform, be the center of gravity of all kinds of states, get top n by the parameter setting;

4.5) sample to be tested obtains L index through same conversion, gets top n, calculates it and the distance of all kinds of state centers of gravity respectively, and it is classified as a class apart from minimum;

(5), the data of gathering are sent in the DCS real-time data base, from the real-time data base of DCS database, obtain up-to-date variable data at each timing cycle as diagnostic data VX; Data to be tested VX the time is obtained with training

And σ _x ²Carry out standardization, and with the input of the data after the standardization as fisher discriminatory analysis module, the discriminant function that will input substitution training obtains, the computational discrimination functional value is differentiated the state of described industrial process.

Described method for diagnosing faults also comprises: (6), regularly the normal point of process status is added among the training set VX, repeat the training process of (3), so that the sorter model in the fisher discriminatory analysis module 8 of the host computer 6 that upgrades in time.

In described (5), the computational discrimination functional value, and on the man-machine interface of host computer 6 state of procedure for displaying, host computer 6 is passed to the DCS system with fault diagnosis result, and, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows at the control station procedure for displaying state of DCS.

Claims (6)

1, a kind of industrial process nonlinear fault diagnosis system based on fisher comprises the field intelligent instrument, DCS system and the host computer that are connected with industrial process object, and described DCS system is made of data-interface, control station, database; Intelligence instrument, DCS system, host computer link to each other successively, and it is characterized in that: described host computer comprises:

The standardization module is used for data are carried out standardization, and the average of each variable is 0, and variance is 1, obtains input matrix X, adopts following process to finish:

1) computation of mean values: $\stackrel{\&OverBar;}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}T{X}_i,$

2) calculate variance: ${\mathrm{\&sigma;}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}(T{X}_i-\stackrel{\&OverBar;}{\mathrm{TX}}),$

3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\&OverBar;}{\mathrm{TX}}}{{\mathrm{\&sigma;}}_x},$

Wherein, TX is a training sample, and N is a number of training;

Fisher discriminatory analysis module is used for sample is carried out analyzing and diagnosing, determines sorter model, adopts following process:

1) scatter matrix B between interior scatter matrix W of the group of calculation training sample and group;

2) calculate | the generalized character root λ of B-λ W|=0 and proper vector 1;

3) get proper vector corresponding to non-0 characteristic root, following transform:

$\left\{\begin{array}{c}{Y}_1=l_{11}{X}_1^{\&prime;}+l_{12}{X}_2^{\&prime;}+...+l_{1m}{X}_m^{\&prime;}\\ .\\ .\\ .\\ Y_L=l_{L1}{X}_1^{\&prime;}+l_{L2}{X}_2^{\&prime;}+...+l_{\mathrm{Lm}}{X}_m^{\&prime;}\end{array}\right.$

Wherein L is the number of non-0 characteristic root, X _i' be the data after the standardization;

L index when 4) calculating the normal and fault of system respectively by above transform is the center of gravity of all kinds of states, gets top n by the parameter setting;

5) sample to be tested obtains L index through same conversion, gets top n, calculates it and the distance of all kinds of state centers of gravity respectively, and it is classified as a class apart from minimum;

Signal acquisition module is used to set the time slot of each sampling, the signal of collection site intelligence instrument;

The diagnostic data determination module, the data that are used for gathering are sent to the DCS real-time data base, from the real-time data base of DCS database, obtain up-to-date variable data as diagnostic data VX at each timing cycle;

Fault diagnosis module is used for data to be tested VX the time is obtained with training And σ _x ²Carry out standardization, and with the input of the data after the standardization as fisher discriminatory analysis module, the discriminant function that will input substitution training obtains, the computational discrimination functional value is differentiated the state of described industrial process.

2, the industrial process nonlinear fault diagnosis system based on fisher as claimed in claim 1, it is characterized in that: described host computer also comprises:

The discrimination model update module is used for regularly adding the normal point of process status to training set VX, outputs to the standardization module, and the sorter model in the fisher discriminatory analysis module of renewal host computer.

3, the industrial process nonlinear fault diagnosis system based on fisher as claimed in claim 1 or 2, it is characterized in that: described host computer also comprises:

Display module is used for fault diagnosis result is passed to the DCS system as a result, and at the control station procedure for displaying state of DCS, and by DCS system and fieldbus process status information is delivered to operator station and shows.

4, a kind of usefulness method for diagnosing faults of realizing based on the industrial process nonlinear fault diagnosis system of fisher as claimed in claim 1, it is characterized in that: described method for diagnosing faults may further comprise the steps:

(1), determine the key variables that fault diagnosis is used, from the historical data base of DCS database respectively during the normal and fault of acquisition system the data of described variable as training sample TX;

(2), in the fisher of host computer discriminatory analysis module, parameters such as discriminant function number N are set, and set the sampling period among the DCS;

(3), training sample TX in host computer, data are carried out standardization, make that the average of each variable is 0, variance is 1, obtains input matrix X, adopts following process to finish:

3.1) computation of mean values: $\stackrel{\&OverBar;}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}T{X}_i,$

3.2) the calculating variance: ${\mathrm{\&sigma;}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}(T{X}_i-\stackrel{\&OverBar;}{\mathrm{TX}}),$

3.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\&OverBar;}{\mathrm{TX}}}{{\mathrm{\&sigma;}}_x},$

Wherein, N is a number of training.

(4), again data are carried out the fisher discriminatory analysis, concrete step is:

4.1) scatter matrix B between scatter matrix W and group in the group of calculation training sample;

4.2) calculate | the generalized character root λ of B-λ W|=0 and proper vector 1;

4.3) get proper vector corresponding to non-0 characteristic root, get following transform:

$\left\{\begin{array}{c}{Y}_1=l_{11}{X}_1^{\&prime;}+l_{12}{X}_2^{\&prime;}+...+l_{1m}{X}_m^{\&prime;}\\ .\\ .\\ .\\ Y_L=l_{L1}{X}_1^{\&prime;}+l_{L2}{X}_2^{\&prime;}+...+l_{\mathrm{Lm}}{X}_m^{\&prime;}\end{array}\right.$

Wherein L is the number of non-0 characteristic root, X _i' be the data after the standardization;

4.4) L index when calculating the normal and fault of system respectively by above transform, be the center of gravity of all kinds of states, get top n by the parameter setting;

4.5) sample to be tested obtains L index through same conversion, gets top n, calculates it and the distance of all kinds of state centers of gravity respectively, and it is classified as a class apart from minimum;

(5), the data of gathering are sent in the DCS real-time data base, from the real-time data base of DCS database, obtain up-to-date variable data at each timing cycle as diagnostic data VX; Data to be tested VX the time is obtained with training

And σ _x ²Carry out standardization, and with the input of the data after the standardization as fisher discriminatory analysis module, the discriminant function that will input substitution training obtains, the computational discrimination functional value is differentiated the state of described industrial process.

5, a kind of industrial process nonlinear fault diagnosis method as claimed in claim 4 based on fisher, it is characterized in that: described method for diagnosing faults also comprises:

(6), regularly process status is put normally and added among the training set VX, repeat the training process of (3), so that the sorter model in the fisher discriminatory analysis module of the host computer that upgrades in time.

6, as claim 4 or 5 described a kind of industrial process nonlinear fault diagnosis methods based on fisher, it is characterized in that: in described (5), the computational discrimination functional value, and on the man-machine interface of host computer the state of procedure for displaying, host computer is passed to the DCS system with fault diagnosis result, and, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows at the control station procedure for displaying state of DCS.

Images