A method for identifying deaerator faults based on correlation degree

Abstract

The application discloses a kind of deaerator fault identification methods based on correlation degree, the fault identification framework of the application determines deaerator, the basic probability assignment of fault is output;According to the basic probability assignment of fault, the average basic probability assignment of fault is calculated;The mutual correlation degree of each sensor basic probability assignment is calculated;The average correlation degree between each sensor basic probability assignment and average basic probability assignment is calculated;The absolute correlation degree of each sensor basic probability assignment is calculated;According to absolute correlation degree, the weight of each sensor basic probability assignment is calculated;According to weight, the weighted basic probability assignment of fault is calculated;The weighted basic probability assignment is fused using D-S fusion rule, and the final fusion result is output;The final fusion result is judged according to decision rule, and the diagnostic result is output.The application can make full use of known information, comprehensively represent the correlation between fault basic probability assignment, and obtain accurate fault identification result.

Description

Technical Field

[0001] This invention belongs to the field of deaerator fault identification, specifically relating to a deaerator fault identification method based on correlation. Background Technology

[0002] Deaerators, as devices that remove dissolved oxygen from boiler feedwater, protect the boiler from oxygen corrosion. Their main function is to remove oxygen and other gases from the feedwater, ensuring its quality. Simultaneously, the deaerator itself acts as a mixing heater in the feedwater reheating system, heating the feedwater and increasing its temperature. As a crucial piece of equipment in industrial production, a deaerator malfunction can result in significant economic losses. Therefore, appropriate fault identification methods should be employed to promptly identify and resolve faults in the deaerator, restoring production as soon as possible. Summary of the Invention

[0003] The purpose of this invention is to provide a deaerator fault identification method based on correlation degree to solve the problem of deaerator fault identification.

[0004] A deaerator fault identification method based on correlation includes the following steps:

[0005] S1, Determine the fault identification framework for the deaerator G = {G1, G2, ..., G...} n}, G i This represents the i-th fault in the fault identification framework G, where i = 1, 2, ..., n, and n is the number of faults. h sets of sensors are used to identify the characteristic parameters of the fault, and the basic probability value m of the fault is output. j (G i ), where j represents the j-th sensor group, j = 1, 2, ..., h;

[0006] S2, Calculate the average basic probability value of the fault based on the basic probability assignment.

[0007]

[0008] Where, m j (G i ) represents the basic probability assignment of the ith fault of the j-th sensor, where h is the number of sensors;

[0009] S3, Calculate the correlation between the basic probability assignments of each sensor.

[0010]

[0011] Where, m j (G i ), m p (G i) represent the basic probability assignments for the ith fault of the j-th and p-th sensors, respectively;

[0012] S4, Calculate the average correlation between the base probability assignment and the average base probability assignment for each sensor.

[0013]

[0014] Where, m j (G i ), These represent the basic probability assignment and the average basic probability assignment for the i-th fault of the j-th sensor, respectively;

[0015] S5, calculate the absolute correlation of the basic probability assignments for each sensor.

[0016] R j =X j +A j (4)

[0017] Among them, X j A j These represent the cross-correlation degree and average correlation degree of the basic probability assignment of the j-th sensor, respectively;

[0018] S6, Calculate the weight of the basic probability assignment for each sensor based on the absolute correlation degree.

[0019]

[0020] Among them, R j This represents the absolute correlation between the base probability assignments of the j-th sensor and the base probability assignments.

[0021] S7, calculate and assign the weighted basic probability of the fault based on the weights.

[0022]

[0023] Among them, w1, w2, ..., w h Let m1(G) represent the weights assigned to the basic probabilities of the 1st, 2nd, ..., hth sensors, respectively. i ), m2(G i ), m h (G i ), i = 1, 2, ..., n, representing the basic probability assignment of the i-th fault of the 1st, 2nd, ..., h-th sensors respectively;

[0024] S8. Using the DS fusion rule, the weighted basic probability m'(Y) is assigned and fused h-1 times, where h is the number of sensors. The final fusion result is output, and the calculation formula is as follows:

[0025]

[0026] Where k is the conflict coefficient of the DS fusion rule;

[0027] S9 uses the decision rule that assigns the fault corresponding to the highest basic probability value as the current fault of the deaerator, judges the final fusion result, and outputs the diagnostic result.

[0028] The beneficial effects of this invention are as follows:

[0029] 1. This invention utilizes the basic probability assignment of faults to calculate the average basic probability assignment, cross-correlation degree, and average correlation degree, which can make full use of known information.

[0030] 2. This invention obtains the absolute correlation degree by calculating the mutual correlation degree and the average correlation degree, which can comprehensively represent the correlation relationship between the basic probability assignments of faults.

[0031] 3. This invention is simple to operate and easy to implement. Attached Figure Description

[0032] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0033] The present invention will now be further described with reference to the accompanying drawings.

[0034] like Figure 1 As shown, a deaerator fault identification method based on correlation includes the following steps:

[0035] S1, Determine the fault identification framework for the deaerator G = {G1, G2, ..., G...} n}, G i This represents the i-th fault in the fault identification framework G, where i = 1, 2, ..., n, and n is the number of faults. h sets of sensors are used to identify the characteristic parameters of the fault, and the basic probability value m of the fault is output. j (G i ), where j represents the j-th sensor group, j = 1, 2, ..., h;

[0036] Three sets of sensors were used to monitor the deaerator that malfunctioned, and the basic probability values ​​for each malfunction are shown in Table 1.

[0037] Table 1 Basic Probability Assignments

[0038] <![CDATA[m1]]> 0.8 0.1 0.1 <![CDATA[m2]]> 0.3 0.6 0.1 <![CDATA[m3]]> 0.7 0.2 0.1

[0039] S2, Calculate the average basic probability value of the fault based on the basic probability assignment.

[0040]

[0041] Where, m j (G i ) represents the basic probability assignment of the ith fault of the j-th sensor, where h is the number of sensors;

[0042]

[0043]

[0044]

[0045] S3, Calculate the correlation between the basic probability assignments of each sensor.

[0046]

[0047] Where, m j (G i ), m p (G i ) represent the basic probability assignments for the ith fault of the j-th and p-th sensors, respectively;

[0048] X1 = 1.4715,

[0049] X2 = 1.2117,

[0050] X3 = 1.6316;

[0051] S4, Calculate the average correlation between the base probability assignment and the average base probability assignment for each sensor.

[0052]

[0053] Where, m j (G i ), These represent the basic probability assignment and the average basic probability assignment for the i-th fault of the j-th sensor, respectively;

[0054] A1 = 0.8472,

[0055] A2 = 0.8120,

[0056] A3 = 0.9681;

[0057] S5, calculate the absolute correlation of the basic probability assignments for each sensor.

[0058] R j =X j +A j (4)

[0059] Among them, X j A jThese represent the cross-correlation degree and average correlation degree of the basic probability assignment of the j-th sensor, respectively;

[0060] R1 = 2.3187,

[0061] R² = 2.0237

[0062] R3 = 2.5997;

[0063] S6, Calculate the weight of the basic probability assignment for each sensor based on the absolute correlation degree.

[0064]

[0065] Among them, R j This represents the absolute correlation between the base probability assignments of the j-th sensor and the base probability assignments.

[0066] w1 = 0.3340,

[0067] w2 = 0.2915,

[0068] w3 = 0.3745;

[0069] S7, calculate and assign the weighted basic probability of the fault based on the weights.

[0070] Among them, w1, w2, ..., w h Let m1(G) represent the weights assigned to the basic probabilities of the 1st, 2nd, ..., hth sensors, respectively. i ), m2(G i ), m h (G i ), i = 1, 2, ..., n, representing the basic probability assignment of the i-th fault of the 1st, 2nd, ..., h-th sensors respectively;

[0071] m'(Y)=[0.6168 0.2832 0.100]

[0072] S8. Using the DS fusion rule, the weighted basic probability m'(Y) is assigned and fused h-1 times, where h is the number of sensors. The final fusion result is output, and the calculation formula is as follows:

[0073]

[0074] Where k is the conflict coefficient of the DS fusion rule;

[0075] According to formula (7), the final fusion result is output as shown in Table 2;

[0076] Table 2 Fusion Results

[0077] <![CDATA[m 1,2 ]]> 0.8083 0.1704 0.0213 <![CDATA[m 1,2,3 ]]> 0.9082 0.0879 0.0039

[0078] S9 uses the decision rule that assigns the fault corresponding to the highest basic probability value as the current fault of the deaerator, judges the final fusion result, and outputs the diagnostic result.

[0079] As shown in Table 2 of step S8, in the final fusion result, m 1,2,3 (G1) = 0.9082. According to the decision rule that the fault corresponding to the highest basic probability is the current fault of the deaerator, it can be concluded that the current fault of the deaerator is G1.

[0080] As can be seen from the above, the method of the present invention can effectively identify deaerator malfunctions.

[0081] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for deaerator fault identification based on correlation, characterized in that, Includes the following steps: S1, Determine the fault identification framework for the deaerator G = {G1, G2, ..., G...} n }, G i This represents the i-th fault in the fault identification framework G, where i = 1, 2, ..., n, and n is the number of faults. h sets of sensors are used to identify the characteristic parameters of the fault, and the basic probability value m of the fault is output. j (G i ), where j represents the j-th sensor group, j = 1, 2, ..., h; S2, Calculate the average basic probability value of the fault based on the basic probability assignment. Where, m j (G i ) represents the basic probability assignment of the ith fault of the j-th sensor, where h is the number of sensors; S3, Calculate the correlation between the basic probability assignments of each sensor. Where, m j (G i ), m p (G i ) represent the basic probability assignments for the ith fault of the j-th and p-th sensors, respectively; S4, Calculate the average correlation between the base probability assignment and the average base probability assignment for each sensor. Where, m j (G i ), These represent the basic probability assignment and the average basic probability assignment for the i-th fault of the j-th sensor, respectively; S5, calculate the absolute correlation of the basic probability assignments for each sensor. R j =X j +A j (4) Among them, X j A j These represent the cross-correlation degree and average correlation degree of the basic probability assignment of the j-th sensor, respectively; S6, Calculate the weight of the basic probability assignment for each sensor based on the absolute correlation. Among them, R j This represents the absolute correlation between the base probability assignments of the j-th sensor and the base probability assignments. S7, Calculate and assign the weighted basic probability of the fault based on the weights; Among them, w1, w2, ..., w h Let m1(G) represent the weights assigned to the basic probabilities of the 1st, 2nd, ..., hth sensors, respectively. i ), m2(G i ), m h (G i ), i = 1, 2, ..., n, representing the basic probability assignment of the i-th fault of the 1st, 2nd, ..., h-th sensors respectively; S8. Using the DS fusion rule, the weighted basic probability m'(Y) is assigned and fused h-1 times, where h is the number of sensors. The final fusion result is output, and the calculation formula is as follows: Where k is the conflict coefficient of the DS fusion rule; S9 uses the decision rule that assigns the fault corresponding to the highest basic probability value as the current fault of the deaerator, judges the final fusion result, and outputs the diagnostic result.

Images