Maintenance decision-making method and system for ammonia engine fuel supply system based on D-S evidence theory

Abstract

The invention provides an ammonia engine fuel supply system maintenance decision-making method and system based on a D-S evidence theory, and the method comprises the steps: building a maintenance suggestion framework according to the maintenance complexity of equipment in an ammonia engine fuel supply system; the maintenance suggestion framework comprises single maintenance suggestions and composite maintenance suggestions; establishing a decision evidence set according to the fault hazard degree, the system state prediction value and the operation duration of the equipment; establishing a maintenance decision basic probability distribution matrix according to the maintenance suggestion framework and the decision evidence set; performing decision-level fusion on the normalized maintenance decision basic probability distribution matrix by adopting a D-S evidence theory to obtain a comprehensive probability value of the maintenance suggestion of the ammonia engine fuel supply system; and selecting the comprehensive probability value based on a preset decision rule to obtain a maintenance decision result of the ammonia engine fuel supply system equipment. The maintenance decision-making method provided by the invention can provide comprehensive and effective maintenance suggestions for the ammonia engine fuel supply system.

Description

Technical Field

[0001] The present invention belongs to the technical field of maintenance decision-making for power machinery equipment, and in particular relates to a maintenance decision-making method and system for an ammonia engine fuel supply system based on DS evidence theory. Background Art

[0002] The use of ammonia engines can significantly reduce greenhouse gas emissions and have high power output, playing an important role in promoting carbon neutrality. Therefore, research on them is of strategic significance in achieving environmental benefits. Among them, the ammonia engine fuel supply system is a key system for providing fuel to the ammonia engine. Its reliable operation is the basic prerequisite for the normal operation of the ammonia engine. Therefore, it is of great significance to carry out timely and accurate maintenance activities on the ammonia engine fuel supply system.

[0003] Maintenance decision-making refers to providing scientific maintenance suggestions based on monitoring data and status prediction information during the maintenance activities of a system or equipment. Reasonable maintenance decision-making methods can effectively guide the maintenance work of the system or equipment and guide the selection of the best maintenance time. It is an important auxiliary means to improve its operating safety and stability. In practical applications, maintenance decisions directly affect the effectiveness of maintenance activities of the ammonia engine fuel supply system and are related to its working status. In addition, in the maintenance decision-making process, it is necessary to comprehensively refer to a variety of information, and the uncertainty factors in the information should also be considered when making decisions. The above situations are the key to maintenance decisions. Therefore, studying a comprehensive and effective maintenance decision-making method to provide accurate auxiliary maintenance suggestions to the ammonia engine fuel supply system is crucial to reducing the failure rate of the ammonia engine fuel supply system and improving the safety and economy of ammonia engine operation.

[0004] In the prior art, a maintenance decision design method for an aircraft engine based on failure probability prediction includes the following steps: obtaining historical operation data of an aircraft engine collected by multiple sensors; extracting and screening degradation features based on the historical operation data of an aircraft engine collected by multiple sensors; predicting the failure probability of different time windows in the future through a single-step iteration method based on the screened degradation features; and evaluating the maintenance cost online based on the failure probability prediction results to determine the best time for predictive maintenance. The maintenance decision for the aircraft engine can be realized, which can maximize the system life and reduce the maintenance cost. "The disadvantage is that this method only formulates the maintenance strategy of the aircraft engine based on the failure probability, and cannot fully reflect its actual operating status, thereby oversimplifying the maintenance decision process and making it difficult to make targeted maintenance decisions for the aircraft engine, which will ultimately affect the accuracy and objectivity of the decision results. When the structure of the maintenance object is complex, a single decision evidence is difficult to fully characterize its state, which will increase the uncertainty of the decision process and ultimately affect the scientific nature of the decision results. Therefore, it is urgent to study a maintenance decision method that can combine multiple decision evidences and fully consider the uncertainty therein, so as to provide reliable maintenance suggestions for the ammonia engine fuel supply system. Summary of the invention

[0005] In order to solve the problem of providing maintenance suggestions for an ammonia engine fuel supply system under multiple decision-making evidences, the present invention provides a maintenance decision method and system for an ammonia engine fuel supply system based on DS evidence theory. The method adopts the equipment failure hazard degree, system state prediction value and operating time of the ammonia engine fuel supply system as a set of decision-making evidences, and sets up composite maintenance suggestions to fully consider the uncertainty contained in the decision-making evidences, so as to provide auxiliary decision support for maintenance personnel.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A maintenance decision method for an ammonia engine fuel supply system based on DS evidence theory includes the following steps:

[0008] According to the maintenance complexity of the equipment in the ammonia engine fuel supply system, a maintenance suggestion framework is established; wherein the maintenance suggestion framework includes single maintenance suggestions and compound maintenance suggestions;

[0009] Establish a decision-making evidence set based on the failure hazard level of the equipment in the ammonia engine fuel supply system, the system state prediction value and the operating time;

[0010] Establishing a maintenance decision basic probability distribution matrix according to the maintenance suggestion framework and the decision evidence set;

[0011] The maintenance decision basic probability distribution matrix is ​​normalized, and the DS evidence theory is used to perform decision-level fusion on the normalized maintenance decision basic probability distribution matrix to obtain a comprehensive probability value of maintenance suggestions for the ammonia engine fuel supply system;

[0012] The comprehensive probability value is selected based on a preset decision rule to obtain a maintenance decision result of the ammonia engine fuel supply system equipment.

[0013] Preferably, the method for establishing the basic probability distribution matrix of maintenance decision-making includes:

[0014] Obtaining a basic probability distribution of a single maintenance suggestion corresponding to each decision evidence according to a membership function between the single maintenance suggestion and the decision evidence set;

[0015] Based on the basic probability distribution of a single maintenance suggestion, the basic probability distribution of a composite maintenance suggestion corresponding to each decision evidence is obtained;

[0016] The maintenance decision basic probability distribution matrix is ​​obtained based on the basic probability distribution of the single maintenance suggestion and the basic probability distribution of the composite maintenance suggestion.

[0017] Preferably, the method for obtaining the comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system comprises:

[0018] According to the basic probability distribution of maintenance suggestions corresponding to different decision evidences, the conflict factor used to measure the degree of contradiction between decision evidences is obtained;

[0019] Based on the conflict factors and the basic probability distribution of maintenance suggestions corresponding to different decision evidences, a comprehensive probability value of maintenance suggestions for the ammonia engine fuel supply system is obtained.

[0020] Preferably, the preset decision rules include:

[0021] If the maximum comprehensive probability value corresponds to a composite maintenance recommendation, the maintenance decision on the ammonia engine fuel supply system is postponed;

[0022] If the maximum comprehensive probability value corresponds to a single maintenance suggestion, the single maintenance suggestion is used as the final decision result of the ammonia engine fuel supply system.

[0023] The present invention also provides an ammonia engine fuel supply system maintenance decision system based on DS evidence theory, which is used to implement a maintenance decision method, including:

[0024] A maintenance suggestion framework building module is used to establish a maintenance suggestion framework according to the maintenance complexity of the equipment in the ammonia engine fuel supply system; wherein the maintenance suggestion framework includes a single maintenance suggestion and a composite maintenance suggestion;

[0025] A decision evidence set building module is used to build a decision evidence set according to the failure hazard degree of the equipment in the ammonia engine fuel supply system, the system state prediction value and the operation time;

[0026] A basic probability allocation module, used to establish a basic probability allocation matrix for maintenance decision according to the maintenance suggestion framework and the decision evidence set;

[0027] A comprehensive probability value calculation module is used to normalize the maintenance decision basic probability distribution matrix, and use DS evidence theory to perform decision-level fusion on the normalized maintenance decision basic probability distribution matrix to obtain a comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system;

[0028] The maintenance decision result acquisition module is used to select the comprehensive probability value based on a preset decision rule to obtain the maintenance decision result of the ammonia engine fuel supply system equipment.

[0029] Preferably, the basic probability allocation module includes:

[0030] A single maintenance suggestion basic probability allocation unit, used to obtain a basic probability allocation of a single maintenance suggestion corresponding to each decision evidence according to a membership function between the single maintenance suggestion and the decision evidence set;

[0031] A composite maintenance suggestion basic probability allocation unit is used to obtain a basic probability allocation of a composite maintenance suggestion corresponding to each decision evidence based on the basic probability allocation of a single maintenance suggestion;

[0032] The basic probability distribution matrix acquisition unit is used to obtain the maintenance decision basic probability distribution matrix based on the basic probability distribution of the single maintenance suggestion and the basic probability distribution of the compound maintenance suggestion.

[0033] Preferably, the comprehensive probability value calculation module includes:

[0034] A conflict factor calculation unit is used to obtain a conflict factor for measuring the degree of contradiction between decision evidences according to the basic probability distribution of maintenance suggestions corresponding to different decision evidences;

[0035] The comprehensive probability value calculation unit is used to obtain the comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system based on the conflict factor and the basic probability distribution of the maintenance suggestions corresponding to different decision evidences.

[0036] Preferably, the preset decision rules in the maintenance decision result acquisition module include:

[0037] If the maximum comprehensive probability value corresponds to a composite maintenance recommendation, the maintenance decision on the ammonia engine fuel supply system is postponed;

[0038] If the maximum comprehensive probability value corresponds to a single maintenance suggestion, the single maintenance suggestion is used as the final decision result of the ammonia engine fuel supply system.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The maintenance decision method for an ammonia engine fuel supply system based on DS evidence theory proposed in the present invention combines three types of decision evidence, namely, the degree of hazard of equipment failure, the predicted value of the state, and the operating time, to provide reliable maintenance suggestions for the ammonia engine fuel supply system, thereby avoiding the problem of inaccurate decision results and lack of objectivity when a single decision evidence is used for maintenance decisions. In addition, the present invention uses the membership function method to set the basic probability distribution for the decision evidence to make the distribution result more objective, and reflects the uncertainty in the decision-making process of the ammonia engine fuel supply system by setting a single maintenance suggestion and a composite maintenance suggestion, thereby comprehensively improving the accuracy and pertinence of the maintenance decision results. The maintenance decision results of the present invention can assist in guiding maintenance personnel to carry out maintenance activities on the ammonia engine fuel supply system and ensure the safe operation of the ammonia engine fuel supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 It is a flow chart of a maintenance decision method of an ammonia engine fuel supply system based on DS evidence theory according to an embodiment of the present invention;

[0043] Figure 2 This is a maintenance decision execution flow chart of an embodiment of the present invention;

[0044] Figure 3 Schematic diagram of membership function according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] Embodiment 1

[0048] like Figure 1 , Figure 2 As shown, a maintenance decision method for an ammonia engine fuel supply system based on DS evidence theory includes the following steps:

[0049] S1: According to the maintenance complexity of the equipment in the ammonia engine fuel supply system, a maintenance recommendation framework A = {a1, a2, ..., a j The maintenance suggestion framework A of the ammonia engine fuel supply system includes j-1 single maintenance suggestions (a1, a2, …, a j-1 ) and single composite repair recommendations a j The composite maintenance recommendation is composed of all the single maintenance recommendations and represents the uncertainty in making decisions about the ammonia engine fuel supply system.

[0050] S2: According to the failure hazard level m1 of the equipment in the ammonia engine fuel supply system, the system state prediction value m2 and the operation time m3, a decision evidence set M = {m1,m2,…,m i}, used to guide maintenance decisions for ammonia engine fuel supply system equipment.

[0051] S3: Establish the basic probability distribution matrix of maintenance decision based on the maintenance recommendation framework and decision evidence set;

[0052] A further implementation method is that the method for establishing a basic probability distribution matrix for maintenance decision-making includes:

[0053] S31: According to the membership function between the single maintenance suggestion and the decision evidence set, the basic probability distribution of the single maintenance suggestion corresponding to each decision evidence is obtained; specifically, the basic probability distribution m of the single maintenance suggestion i (a j-1 ) By establishing m i and a j-1 This function quantifies the maintenance suggestion a j-1 and decision evidence m i The degree of correlation between them reflects the support of the i-th decision evidence for the j-1-th maintenance suggestion. The specific form is:

[0054] m i (a j-1 )=f(m i ,a j-1 ),

[0055] Among them, f is the membership function.

[0056] S32: Based on the basic probability distribution of a single maintenance suggestion, the basic probability distribution of the composite maintenance suggestion corresponding to each decision evidence is obtained. The calculation formula is as follows:

[0057] m i (a j )=1-max(m i (a j-1 )),

[0058] Among them, m i (a j ) represents the basic probability distribution of each decision evidence corresponding to the composite maintenance recommendation, m i (a j-1 ) represents the basic probability distribution of a single maintenance suggestion corresponding to each decision evidence.

[0059] S33: Based on the basic probability distribution of the single maintenance suggestion and the basic probability distribution of the compound maintenance suggestion, a maintenance decision basic probability distribution matrix is ​​obtained.

[0060] Specifically, according to S31, maintenance suggestion a is obtained. j and decision evidence m i The relationship between them is obtained, and the basic probability distribution m of the maintenance suggestion corresponding to each decision evidence is obtained according to S32 and S33 i (a j ), establish the basic probability distribution matrix P = (m i (a j )) i×j , the specific form is:

[0061]

[0062] S4: The maintenance decision basic probability distribution matrix is ​​normalized, and the DS evidence theory is used to perform decision-level fusion on the normalized maintenance decision basic probability distribution matrix to obtain the comprehensive probability value of the maintenance recommendation for the ammonia engine fuel supply system.

[0063] In this embodiment, the basic probability distribution matrix P is normalized to ensure that m under each decision evidence is i (a j ) is 1. The basic probability distribution normalization formula is:

[0064]

[0065] Among them, m i (a j )' represents the normalized result, and n is the number of maintenance recommendations. For the sake of simplicity, the basic probability distribution in the following content defaults to the normalized result and is uniformly represented as mi (a j ) in the form of.

[0066] In a further embodiment, the method for obtaining a comprehensive probability value of a maintenance suggestion for an ammonia engine fuel supply system comprises:

[0067] According to the basic probability distribution of maintenance suggestions corresponding to different decision evidences, the conflict factor between decision evidences is obtained; specifically, in DS evidence theory, the conflict factor k is used to measure the degree of contradiction between decision evidences. This parameter can affect the final combination result, and its calculation formula is:

[0068]

[0069] Among them, m z (a zj ) and m q (a qj ) represent the basic probability distribution of maintenance suggestions corresponding to two different decision evidences when making maintenance decisions on the ammonia engine fuel supply system, a zj and a qj The decision evidence is m z 、m q Maintenance recommendations when needed.

[0070] Based on the conflict factors and the basic probability distribution of maintenance suggestions corresponding to different decision evidences, the comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system is obtained. Specifically, the basic probability distribution combination formula of two different decision evidences is as follows:

[0071]

[0072] in, Representative decision evidence m z and m q Repair suggestions after assembly j The comprehensive probability value of .

[0073] S5: Select the comprehensive probability value based on the preset decision rule to obtain the maintenance decision result of the ammonia engine fuel supply system equipment.

[0074] A further implementation method is that the preset decision rule includes:

[0075] If the maximum comprehensive probability value corresponds to a composite maintenance recommendation, the maintenance decision on the ammonia engine fuel supply system is postponed;

[0076] If the maximum comprehensive probability value corresponds to a single maintenance suggestion, the single maintenance suggestion is used as the final decision result of the ammonia engine fuel supply system.

[0077] On the other hand, the present invention also provides an ammonia engine fuel supply system maintenance decision system based on DS evidence theory, which is used to implement a maintenance decision method, including:

[0078] A maintenance suggestion framework building module is used to establish a maintenance suggestion framework according to the maintenance complexity of the equipment in the ammonia engine fuel supply system; wherein the maintenance suggestion framework includes a single maintenance suggestion and a composite maintenance suggestion;

[0079] A decision evidence set building module is used to build a decision evidence set according to the failure hazard degree of the equipment in the ammonia engine fuel supply system, the system state prediction value and the operation time;

[0080] The basic probability allocation module is used to establish the basic probability allocation matrix of maintenance decision according to the maintenance suggestion framework and decision evidence set;

[0081] The comprehensive probability value calculation module is used to normalize the maintenance decision basic probability distribution matrix, and use the DS evidence theory to perform decision-level fusion on the normalized maintenance decision basic probability distribution matrix to obtain the comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system;

[0082] The maintenance decision result acquisition module is used to select the comprehensive probability value based on the preset decision rules to obtain the maintenance decision result of the ammonia engine fuel supply system equipment.

[0083] A further implementation is that the basic probability allocation module comprises:

[0084] A single maintenance suggestion basic probability allocation unit is used to obtain a basic probability allocation of a single maintenance suggestion corresponding to each decision evidence according to a membership function between the single maintenance suggestion and the decision evidence set;

[0085] A composite maintenance suggestion basic probability allocation unit is used to obtain a basic probability allocation of a composite maintenance suggestion corresponding to each decision evidence based on the basic probability allocation of a single maintenance suggestion;

[0086] The basic probability distribution matrix acquisition unit is used to obtain the basic probability distribution matrix of maintenance decision based on the basic probability distribution of the single maintenance suggestion and the basic probability distribution of the compound maintenance suggestion.

[0087] A further implementation method is that the comprehensive probability value calculation module includes:

[0088] A conflict factor calculation unit is used to obtain a conflict factor for measuring the degree of contradiction between decision evidences according to the basic probability distribution of maintenance suggestions corresponding to different decision evidences;

[0089] The comprehensive probability value calculation unit is used to obtain the comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system based on the conflict factor and the basic probability distribution of the maintenance suggestion corresponding to different decision evidences.

[0090] A further implementation method is that the preset decision rules in the maintenance decision result acquisition module include:

[0091] If the maximum comprehensive probability value corresponds to a composite maintenance recommendation, the maintenance decision on the ammonia engine fuel supply system is postponed;

[0092] If the maximum comprehensive probability value corresponds to a single maintenance suggestion, the single maintenance suggestion is used as the final decision result of the ammonia engine fuel supply system.

[0093] Embodiment 2

[0094] This embodiment discloses a maintenance decision method for a high-pressure fuel pump of an ammonia engine fuel supply system based on DS evidence theory. Figure 2 As shown, the specific implementation steps are as follows:

[0095] Step 1: According to the maintenance complexity of the high-pressure fuel pump in the ammonia engine fuel supply system, a maintenance suggestion framework A = {a1, a2, a3, a4} is established. The framework includes single maintenance suggestions a1, a2, a3 for the high-pressure fuel pump, and a composite maintenance suggestion a4 = (a1, a2, a3). Among them, a1, a2 and a3 represent strengthening monitoring, cleaning the pump body and replacing the pump body, respectively, and a4 represents that a1, a2 or a3 can be used as maintenance suggestions, that is, the decision result made on the high-pressure fuel pump based on the current decision evidence has a large uncertainty.

[0096] Step 2: Define the decision evidence set M = {m1, m2, m3}, where m1, m2, m3 are the predicted value of the ammonia engine fuel supply system state, the failure hazard level of the high-pressure fuel pump m2, and the operating time of the high-pressure fuel pump m3, respectively.

[0097] Step 3: Follow the repair suggestions in steps 1 and 2 j and decision evidence m j The relationship between them is used to set the basic probability distribution m for the maintenance suggestion corresponding to each decision evidence of the high-pressure fuel pump. i (a j ), the specific steps are:

[0098] (1) First, establish m i The basic probability distribution m of a single maintenance suggestion by the membership function between a1, a2, and a3 i (a p ) is set. This function quantifies the single maintenance recommendation (a1, a2, a3) of the high-pressure fuel pump and the decision evidence mi The degree of association between them, the membership function established is as follows:

[0099]

[0100] Among them, c determines m i (a p ) reaches the peak value m i The size of m i (a p ) width, k represents m i (a p ), with x as the center point. The membership function established according to the above formula is as follows: Figure 3 As shown;

[0101] (2) Based on the basic probability distribution of the single maintenance suggestion obtained, m i (a p ) Obtain the basic probability distribution m of the composite maintenance recommendation for the high-pressure fuel pump i (a4), the calculation formula is:

[0102] m i (a4)=1-max(m i (a p ))p=1,2,3

[0103] Among them, m i (a p ) represents the basic probability distribution of each decision evidence corresponding to the maintenance suggestion a1, a2 or a3, m i (a4) represents the basic probability distribution of each decision evidence corresponding to the a4 maintenance suggestion;

[0104] (3) According to the results obtained in step (2) and step (3), the basic probability distribution matrix P of the high-pressure fuel pump maintenance decision is established. i (a j )) 3×4 , the specific form is:

[0105]

[0106] Step 4: Normalize the basic probability distribution matrix P in step 3 to ensure that m i (a j ) is 1, and the normalization formula is:

[0107]

[0108] Among them, m i (a j)' represents the normalized result. For the sake of simplicity, the basic probability distribution in the following content defaults to the normalized result and is uniformly represented as m i (a j ) in the form of.

[0109] Step 5: Combine the P in step 4 using the DS evidence theory to achieve decision-level fusion, and finally obtain the comprehensive probability value m(a j ), the specific steps are:

[0110] (1) First, the conflict factor k between decision evidences is calculated. This parameter is directly related to the final maintenance recommendation for the high-pressure fuel pump. Its calculation formula is:

[0111]

[0112] Among them, m z (a zj ) and m q (a qj ) are the basic probability distributions of two different decision evidences, a zj and a qj The decision evidence m is z and m q Maintenance suggestions below;

[0113] (2) The basic probability distribution of the high-pressure fuel pump decision evidence is combined, and the combination formula is as follows:

[0114]

[0115] in, Representative decision evidence m z and m q Repair suggestions after assembly j Since there are three decision evidences, it is necessary to first combine m1(a j ) and m2(a j ) and then the result is combined with m3(a j ) combined with the comprehensive probability distribution m(a j ), the specific formula is:

[0116]

[0117] Step 6: Select the comprehensive probability value of the decision evidence through the decision rule. If the maintenance suggestion a4 has the largest comprehensive probability value, postpone the decision on the high-pressure fuel pump. If the maintenance suggestion corresponding to the largest comprehensive probability value is a1, a2 or a3, then it is output as the final maintenance suggestion for the high-pressure fuel pump.

[0118] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A maintenance decision method for an ammonia engine fuel supply system based on DS evidence theory, characterized in that: The following steps are involved: According to the maintenance complexity of the equipment in the ammonia engine fuel supply system, a maintenance suggestion framework is established; wherein the maintenance suggestion framework includes single maintenance suggestions and compound maintenance suggestions; Establish a decision-making evidence set based on the failure hazard level of the equipment in the ammonia engine fuel supply system, the system state prediction value and the operating time; Establishing a maintenance decision basic probability distribution matrix according to the maintenance suggestion framework and the decision evidence set; The maintenance decision basic probability distribution matrix is ​​normalized, and the DS evidence theory is used to perform decision-level fusion on the normalized maintenance decision basic probability distribution matrix to obtain a comprehensive probability value of maintenance suggestions for the ammonia engine fuel supply system; The comprehensive probability value is selected based on a preset decision rule to obtain a maintenance decision result of the ammonia engine fuel supply system equipment.

2. The maintenance decision method for ammonia engine fuel supply system based on DS evidence theory according to claim 1 is characterized in that: The methods for establishing the basic probability distribution matrix for maintenance decision-making include: Obtaining a basic probability distribution of a single maintenance suggestion corresponding to each decision evidence according to a membership function between the single maintenance suggestion and the decision evidence set; Based on the basic probability distribution of a single maintenance suggestion, the basic probability distribution of a composite maintenance suggestion corresponding to each decision evidence is obtained; The maintenance decision basic probability distribution matrix is ​​obtained based on the basic probability distribution of the single maintenance suggestion and the basic probability distribution of the composite maintenance suggestion.

3. The maintenance decision method for ammonia engine fuel supply system based on DS evidence theory according to claim 2 is characterized in that: The method for obtaining a comprehensive probability value for maintenance recommendations for the ammonia engine fuel supply system includes: According to the basic probability distribution of maintenance suggestions corresponding to different decision evidences, the conflict factor used to measure the degree of contradiction between decision evidences is obtained; Based on the conflict factors and the basic probability distribution of maintenance suggestions corresponding to different decision evidences, a comprehensive probability value of maintenance suggestions for the ammonia engine fuel supply system is obtained.

4. The maintenance decision method for ammonia engine fuel supply system based on DS evidence theory according to claim 2 is characterized in that: The preset decision rules include: If the maximum comprehensive probability value corresponds to a composite maintenance recommendation, the maintenance decision on the ammonia engine fuel supply system is postponed; If the maximum comprehensive probability value corresponds to a single maintenance suggestion, the single maintenance suggestion is used as the final decision result of the ammonia engine fuel supply system.

5. A maintenance decision system for an ammonia engine fuel supply system based on DS evidence theory, used to implement the maintenance decision method according to any one of claims 1 to 4, characterized in that: include: A maintenance suggestion framework building module is used to establish a maintenance suggestion framework according to the maintenance complexity of the equipment in the ammonia engine fuel supply system; wherein the maintenance suggestion framework includes a single maintenance suggestion and a composite maintenance suggestion; A decision evidence set building module is used to build a decision evidence set according to the failure hazard degree of the equipment in the ammonia engine fuel supply system, the system state prediction value and the operation time; A basic probability allocation module, used to establish a basic probability allocation matrix for maintenance decision according to the maintenance suggestion framework and the decision evidence set; A comprehensive probability value calculation module is used to normalize the maintenance decision basic probability distribution matrix, and use DS evidence theory to perform decision-level fusion on the normalized maintenance decision basic probability distribution matrix to obtain a comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system; The maintenance decision result acquisition module is used to select the comprehensive probability value based on a preset decision rule to obtain the maintenance decision result of the ammonia engine fuel supply system equipment.

6. The ammonia engine fuel supply system maintenance decision system based on DS evidence theory according to claim 5 is characterized in that: The basic probability assignment module includes: A single maintenance suggestion basic probability allocation unit, used to obtain a basic probability allocation of a single maintenance suggestion corresponding to each decision evidence according to a membership function between the single maintenance suggestion and the decision evidence set; A composite maintenance suggestion basic probability allocation unit is used to obtain a basic probability allocation of a composite maintenance suggestion corresponding to each decision evidence based on the basic probability allocation of a single maintenance suggestion; The basic probability distribution matrix acquisition unit is used to obtain the maintenance decision basic probability distribution matrix based on the basic probability distribution of the single maintenance suggestion and the basic probability distribution of the compound maintenance suggestion.

7. The ammonia engine fuel supply system maintenance decision system based on DS evidence theory according to claim 5 is characterized in that: The comprehensive probability value calculation module includes: A conflict factor calculation unit is used to obtain a conflict factor for measuring the degree of contradiction between decision evidences according to the basic probability distribution of maintenance suggestions corresponding to different decision evidences; The comprehensive probability value calculation unit is used to obtain the comprehensive probability value of the maintenance suggestion for the ammonia engine fuel supply system based on the conflict factor and the basic probability distribution of the maintenance suggestions corresponding to different decision evidences.

8. The ammonia engine fuel supply system maintenance decision system based on DS evidence theory according to claim 5 is characterized in that: The preset decision rules in the maintenance decision result acquisition module include: If the maximum comprehensive probability value corresponds to a composite maintenance recommendation, the maintenance decision on the ammonia engine fuel supply system is postponed; If the maximum comprehensive probability value corresponds to a single maintenance suggestion, the single maintenance suggestion is used as the final decision result of the ammonia engine fuel supply system.

Images