Multi-stage multi-source data-driven ship equipment reliability growth method and system

Abstract

The application belongs to the technical field of data processing, and specifically discloses a multi-stage multi-source data driven ship equipment reliability growth method and system. The method comprises the following steps: collecting heterogeneous data sources of each development stage of ship equipment, and performing data standardization on the heterogeneous data sources; acquiring a confidence degree comprehensive score of each heterogeneous data source, and determining a dynamic weight of the confidence degree according to the confidence degree comprehensive score; constructing a Bayesian hierarchical model, and updating a failure rate parameter of the Bayesian hierarchical model according to stages by using the dynamic weight; and generating a reliability optimization decision instruction according to a posterior mean value of a current stage of the Bayesian hierarchical model and in combination with a set failure rate confidence interval. The application can solve the problems of data fusion adaptability, evaluation timeliness and resource allocation rationality, and provide precise decision support for ship equipment life cycle reliability growth.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and more specifically, relates to a multi-stage and multi-source data-driven method and system for increasing the reliability of ship equipment. Background Art

[0002] The development cycle for shipbuilding equipment is long, involving multiple phases, including design verification, onshore bench testing, and offshore mooring and navigation testing. Traditional reliability assessment methods (such as single-stage test data modeling) struggle to integrate data from these multiple phases, resulting in delayed assessments. Data from multiple sources (such as simulation, bench testing, and offshore testing) exhibit varying confidence levels, and existing methods lack dynamic weighting mechanisms, resulting in inefficient data utilization. The reliability growth process relies on static models, which lack real-time feedback on the effectiveness of improvement measures, leading to wasted resources.

[0003] Compared with existing traditional technologies, reliability growth analysis based on the AMSAA model is only applicable to single-stage failure data; the weighted average method based on expert experience is highly subjective and lacks a probabilistic statistical basis; and the traditional Bayesian method does not consider multi-stage progression and the weighting of multi-source data. Therefore, these solutions also have certain limitations, resulting in delayed evaluation results, low data utilization efficiency, and difficulty in providing real-time feedback on the effectiveness of improvement measures, ultimately wasting resources and extending development cycles.

[0004] Therefore, how to solve the deficiencies in data fusion adaptability, evaluation timeliness and rationality of resource allocation, and provide accurate decision-making support for the reliability growth of ship equipment throughout its life cycle, is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] In response to the defects of the existing technology, the purpose of this application is to provide a multi-stage and multi-source data-driven ship equipment reliability growth method and system, aiming to solve the problems of poor accuracy and low efficiency in ship equipment reliability assessment due to the inability to dynamically integrate data, evaluation timeliness and rationality of resource allocation.

[0006] To achieve the above objectives, in a first aspect, the present application provides a multi-stage and multi-source data-driven method for increasing the reliability of ship equipment, comprising:

[0007] Collect heterogeneous data sources at various stages of ship equipment development and standardize the data;

[0008] Obtain the comprehensive confidence scores of each heterogeneous data source and determine the dynamic weight of the confidence based on the comprehensive confidence scores;

[0009] Constructing a Bayesian hierarchical model, and updating the failure rate parameters of the Bayesian hierarchical model in stages using dynamic weights;

[0010] According to the posterior mean of the current stage of the Bayesian hierarchical model, combined with the set failure rate confidence interval, a reliability optimization decision instruction is generated.

[0011] Optionally, the heterogeneous data sources include fault data, environmental data, and historical experience data;

[0012] The fault data includes the number of faults and the operating time;

[0013] The environmental data includes temperature, humidity and operating conditions; the operating conditions are quantified as severity coefficients;

[0014] The historical experience data includes a priori parameters of failures of historically similar equipment.

[0015] Optionally, the data standardization process includes:

[0016] Determine the environmental severity of the ship's typical missions;

[0017] In combination with the environmental severity and the severity coefficient corresponding to the operating conditions, the operating times of different data sources are uniformly converted into standard operating times under an equivalent standard environment.

[0018] Optionally, the method for obtaining the comprehensive confidence score includes:

[0019] Obtain a sample size score based on the actual trial sample size and the minimum sample size requirement;

[0020] Obtain environmental consistency scores based on environmental severity and the severity coefficient corresponding to the operating conditions;

[0021] Obtain the difference data between the failure rate of the current data source and other data sources in the stage, and obtain a data consistency score based on the difference data and the maximum allowable difference;

[0022] A comprehensive confidence score is obtained based on the sample size score, environmental consistency score, and data consistency score.

[0023] Optionally, the method for obtaining the dynamic weight of the confidence level includes:

[0024] Get the cumulative score of the confidence comprehensive score of all data sources;

[0025] The dynamic weight of the confidence is obtained based on the confidence score of the current data source and the cumulative score.

[0026] Optionally, the parameter updating process of the Bayesian hierarchical model includes:

[0027] determining that the number of failures satisfies a Poisson distribution, and that a prior distribution of a failure rate is a gamma distribution; a prior parameter of the failure rate is a posterior parameter of a previous stage, and the posterior parameter of the previous stage is determined according to historical experience data;

[0028] obtaining a cumulative weighted failure number and a cumulative weighted standard operating time of the current stage; the cumulative weighted failure number is determined based on the posterior parameter of the previous stage, a dynamic weight, and the number of failures, and the cumulative weighted standard operating time is determined based on the posterior parameter of the previous stage, the dynamic weight, and a standard operating time;

[0029] using the cumulative weighted failure number and the cumulative weighted standard operating time as a posterior distribution of the current stage, and using the posterior distribution of the current stage as a prior distribution of a next stage to realize parameter updating of the failure rate of the model.

[0030] Optionally, the method for generating the reliability optimization decision instruction comprises:

[0031] obtaining a posterior mean of the Bayesian hierarchical model, and setting a confidence interval according to the posterior mean;

[0032] if the posterior mean exceeds a preset target value or an upper limit of the confidence interval reaches a safety threshold, generating an improvement instruction to adjust the failure rate;

[0033] if a width of the confidence interval exceeds an allowable value, increasing a sample size of a next stage;

[0034] if the posterior mean does not exceed the preset target value and the width of the confidence interval does not exceed the allowable value, determining that the reliability meets a standard.

[0035] In a second aspect, the application further provides a multi-stage multi-source data driven reliability growth system of a ship equipment, comprising:

[0036] a collection module configured to collect heterogeneous data sources of each development stage of the ship equipment, and to perform data standardization on the heterogeneous data sources;

[0037] a confidence module configured to obtain a confidence comprehensive score of each heterogeneous data source, and to determine a dynamic weight of the confidence according to the confidence comprehensive score;

[0038] a parameter updating module configured to construct a Bayesian hierarchical model, and to perform parameter updating of a failure rate of the Bayesian hierarchical model according to the dynamic weight and stages;

[0039] a reliability optimization module configured to generate a reliability optimization decision instruction according to a posterior mean of the Bayesian hierarchical model of a current stage and a set confidence interval of the failure rate.

[0040] In a third aspect, the present application provides an electronic device, comprising: at least one memory configured to store a program; and at least one processor configured to execute the program stored in the memory, wherein the processor is configured to execute the method described in the first aspect or any possible implementation manner of the first aspect when the program stored in the memory is executed.

[0041] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is run on a processor, the processor executes the method described in the first aspect or any possible implementation manner of the first aspect.

[0042] In a fifth aspect, the present application provides a computer program product, which, when run on a processor, causes the processor to execute the method described in the first aspect or any possible implementation manner of the first aspect.

[0043] It can be understood that the beneficial effects of the above-mentioned second aspect to fifth aspect can be referred to the related description in the first aspect, which will not be repeated here.

[0044] Overall, compared with the prior art, the above technical solutions conceived by the present application have the following beneficial effects:

[0045] (1) The present application integrates different stages of heterogeneous data sources at the data fusion level, and through the construction of a confidence scoring system, dynamically adjusts the weight of each data source, ensures that the system adopts high-confidence real-time data first, avoids evaluation deviation caused by static fusion rules, and improves evaluation accuracy; In terms of evaluation timeliness, the phased parameter updating mechanism of the Bayesian hierarchical model can quickly convert new data into posterior failure rate estimates, improving evaluation efficiency; In terms of resource allocation rationality, the optimized instructions generated by combining the posterior mean and the pre-set confidence interval can accurately locate high-failure-risk links, thereby improving support efficiency and reducing life cycle cost.

[0046] (2) The present application integrates heterogeneous data sources of each development stage of ship equipment, and realizes standardized processing, solving the problem of non-uniform data format and difficulty in integration in traditional methods. The standardized data can be uniformly analyzed, and the confidence dynamic weight mechanism further optimizes the real-time performance of data fusion, solving the problem that the existing technology analyzes single-stage failure data and cannot dynamically fuse data, so that the reliability growth process can generate decision basis based on the latest and most reliable data, improving accuracy.

[0047] (3) This application iteratively updates the failure rate parameters by constructing a Bayesian hierarchical model. New data at each stage are used to modify the prior distribution through Bayesian inference to form a posterior estimate. This dynamic update mechanism overcomes the limitations of traditional fixed-parameter models, thereby more accurately reflecting the actual reliability status.

[0048] (4) This application automatically triggers decision instructions based on the Bayesian posterior mean and a preset confidence interval of the failure rate (such as a 95% confidence level). This data-driven decision logic reduces reliance on subjective experience. Especially when the confidence interval exceeds the threshold, it can quickly locate weak links and prioritize optimization to avoid wasting resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is one of the flow charts of the multi-stage multi-source data driven ship equipment reliability growth method provided in the embodiment of the present application;

[0050] Figure 2 This is the second flow chart of the multi-stage multi-source data driven ship equipment reliability growth method provided in the embodiment of the present application;

[0051] Figure 3 This is the third flow chart of the multi-stage multi-source data driven ship equipment reliability growth method provided in the embodiment of the present application;

[0052] Figure 4 Schematic diagram of the structure of a multi-stage and multi-source data-driven ship equipment reliability growth system provided by an embodiment of the present application;

[0053] Figure 5 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0055] The term "and / or" as used herein describes an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. The symbol " / " as used herein indicates that the related objects are in an "or" relationship, for example, A / B means either A or B.

[0056] The terms "first" and "second" in this specification and claims are used to distinguish different objects rather than to describe a specific order of objects. For example, "first response message" and "second response message" are used to distinguish different response messages rather than to describe a specific order of response messages.

[0057] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0058] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more, for example, multiple processing units means two or more processing units, etc.; multiple elements means two or more elements, etc.

[0059] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0060] Reference Figure 1 The present application provides a multi-stage and multi-source data-driven method for ship equipment reliability growth, including:

[0061] S101. Collect heterogeneous data sources at various stages of ship equipment development and standardize the data from the heterogeneous data sources;

[0062] S102. Obtaining a comprehensive confidence score for each heterogeneous data source and determining a dynamic confidence weight based on the comprehensive confidence score;

[0063] S103. Constructing a Bayesian hierarchical model and updating the failure rate parameters of the Bayesian hierarchical model by stage using dynamic weights;

[0064] S104. Generate a reliability optimization decision instruction based on the posterior mean of the current stage of the Bayesian hierarchical model and the set failure rate confidence interval.

[0065] Specifically, data collection and integration are performed through step S101, and multi-source heterogeneous data of each development stage (design verification, bench test, mooring and navigation test) are collected throughout the life cycle of the ship equipment.

[0066] Furthermore, the heterogeneous data sources include fault data, environmental data, and historical experience data;

[0067] The fault data includes the number of faults and the operating time;

[0068] The environmental data includes temperature, humidity and operating conditions; the operating conditions are quantified as severity coefficients;

[0069] The historical experience data includes a priori parameters of failures of historically similar equipment.

[0070] The details are as follows:

[0071] a. Failure data: number of failures , running time (Unit: hours);

[0072] b. Environmental data: temperature, humidity, operating conditions (quantified as severity coefficient );

[0073] c. Expert experience data: a priori parameters of failures of historically similar equipment 、 .

[0074] Then, the collected data is subjected to data standardization, and the data standardization process includes:

[0075] Determine the environmental severity of the ship's typical missions;

[0076] In combination with the environmental severity and the severity coefficient corresponding to the operating conditions, the operating times of different data sources are uniformly converted into standard operating times under an equivalent standard environment.

[0077] Specifically, the time normalization method is used to normalize the test time of different data sources. Converted to the running time under equivalent standard environment:

[0078]

[0079] in, The severity of the typical mission environment of the ship (such as =0.8).

[0080] A confidence evaluation system is constructed through step S102. A scoring system is constructed based on the integrity, timeliness, and stability of the data source, etc.

[0081] Methods for obtaining comprehensive confidence scores include:

[0082] Obtain a sample size score based on the actual trial sample size and the minimum sample size requirement;

[0083] Obtain environmental consistency scores based on environmental severity and the severity coefficient corresponding to the operating conditions;

[0084] Obtain the difference data between the failure rate of the current data source and other data sources in the stage, and obtain a data consistency score based on the difference data and the maximum allowable difference;

[0085] A comprehensive confidence score is obtained based on the sample size score, environmental consistency score, and data consistency score.

[0086] The details are as follows:

[0087] For each data source Calculating a composite confidence score , which contains the following dimensions:

[0088] Sample size rating: / Minimum sample size requirement (take 1 if more than 1)

[0089] Environmental consistency score: ;

[0090] Data consistency score:

[0091] Overall confidence score:

[0092] The weight calculation process is as follows:

[0093] Get the cumulative score of the confidence comprehensive score of all data sources;

[0094] The dynamic weight of the confidence is obtained based on the confidence score of the current data source and the cumulative score.

[0095] Specifically, the weight is calculated based on the confidence indicators of the data source (such as sample size, environmental severity, and data consistency). , the specific calculation formula is:

[0096]

[0097] Furthermore, a Bayesian hierarchical model is constructed using step S103, and the failure rate parameters of the Bayesian hierarchical model are updated in stages using dynamic weights.

[0098] The number of failures is assumed to follow a Poisson distribution. This is because the Poisson distribution is suitable for describing sparse random events that occur per unit time, such as equipment failures. The parameter of the Poisson distribution is the failure rate, which reflects the average number of failures per unit time. Furthermore, the prior distribution for the failure rate is the gamma distribution.

[0099] The number of failures is distributed according to the Poisson distribution , failure rate The prior distribution of (Gamma) Distribution:

[0100]

[0101] in is the posterior parameter of the previous stage, which is determined based on historical experience data.

[0102] The parameter updating process of the Bayesian hierarchical model includes:

[0103] Determining that the number of failures satisfies a Poisson distribution and that the prior distribution of the failure rate is a gamma distribution; the prior parameters of the failure rate are the posterior parameters of the previous stage, which are determined based on historical experience data;

[0104] Obtaining the cumulative weighted number of failures and the cumulative weighted standard operating time for the current phase; the cumulative weighted number of failures is determined based on the a posteriori parameters, dynamic weights, and number of failures in the previous phase; and the cumulative weighted standard operating time is determined based on the a posteriori parameters, dynamic weights, and standard operating time in the previous phase;

[0105] The cumulative weighted number of failures and the cumulative weighted standard running time are used as the posterior distribution of the current stage;

[0106] The posterior distribution of the current stage is used as the prior distribution of the next stage to update the failure rate parameters of the model.

[0107] During the dynamic model update process, the cumulative weighted failure count and the cumulative weighted standard operating time for the current phase must first be calculated. The cumulative weighted failure count refers to the number of failures recorded during the current phase. Weighting can more effectively reflect the importance of failures or the conditions under which they occur (for example, failures under different operating environments or load conditions).

[0108] The cumulative weighted standard operating time reflects the effective time the equipment was operating normally during the assessment period. Standardization eliminates the impact of abnormal operating time due to maintenance, inspections, etc. The effective operating time during this period is calculated using a weighted method to reflect actual usage under different conditions.

[0109] The posterior distribution for the current phase is constructed using the cumulative weighted number of failures and the cumulative weighted standard runtime. Using Bayes' theorem, the posterior distribution for the current phase is calculated by combining the prior distribution (the gamma distribution) with the likelihood function (the Poisson distribution constructed from the failure data for the current phase).

[0110] After completing the posterior distribution update for the current phase, it is used as the prior distribution for the next phase. By leveraging Bayesian reasoning for continuous iterative updates, we ensure that model parameters are always based on the latest data and results in future analyses, thereby improving the accuracy and reliability of failure rate predictions. This iterative process not only increases the model's flexibility in responding to changing circumstances but also makes reliability analysis results more robust and optimizes maintenance strategies.

[0111] Specifically, the parameter update rules are as follows:

[0112] In the stage In the fusion of weighted fault count and test time:

[0113]

[0114]

[0115] in, Reflects the cumulative weighted number of failures, Reflects the cumulative equivalent test time, i.e. the cumulative weighted standard running time.

[0116] The process of passing across stages, stages The posterior distribution of As a stage Prior distribution of , realizing historical information inheritance.

[0117] Reference Figure 2 , Figure 2 It is a diagram of multi-stage parameter transfer, showing 、 Cross-stage updates.

[0118] Step S104: Generate a reliability optimization decision instruction based on the posterior mean of the current stage of the Bayesian hierarchical model and the set failure rate confidence interval.

[0119] The method for generating reliability optimization decision instructions specifically includes:

[0120] Obtaining a posterior mean of the Bayesian hierarchical model, and setting a confidence interval based on the posterior mean;

[0121] If the posterior mean exceeds a preset target value, or the upper limit of the confidence interval reaches a safety threshold, an improvement instruction is generated to adjust the failure rate;

[0122] If the confidence interval width exceeds the allowed value, the sample size for the next stage of testing will be increased;

[0123] If the posterior mean does not exceed the preset target value and the confidence interval width does not exceed the allowable value, it is determined that the reliability meets the standard.

[0124] Specifically, the calculation formula of the posterior mean in this embodiment is as follows:

[0125]

[0126] Let the 95% confidence interval be:

[0127]

[0128] Closed-loop decision making:

[0129] a. Design improvement trigger: If Or if the upper limit of the confidence interval exceeds the threshold, an improvement instruction is generated (such as optimizing the design of a subsystem);

[0130] b. Test resource allocation: If the confidence interval width , increase the sample size for the next stage of testing;

[0131] c. Termination of the development phase: If and , approved to enter the equipment production stage.

[0132] in, is the preset target value, are allowed values, which are all data values ​​set in advance.

[0133] Reference Figure 3 , Figure 3 It is a complete flow chart of an embodiment of the present application, including four stages: data collection → weight allocation → Bayesian update → decision output.

[0134] The solution of this application is described in detail below with reference to specific embodiments:

[0135] The development of a certain type of ship power system consists of three stages: design verification (stage 1), bench test (stage 2), offshore mooring test (stage 3), and target failure rate ;

[0136] Stage 0: Initial Expert Priors

[0137] Prior parameters: Based on historical similar equipment data, set , , corresponding to the initial failure rate .

[0138] Phase 1: Design Verification (Simulation and Expert Data Fusion)

[0139] (1) Data source:

[0140] a. Simulation data:

[0141] Number of failures , test time hours, environmental severity ;

[0142] Sample size score (The minimum sample size requirement for Phase 1 is 200 hours);

[0143] Data consistency scoring (The difference from expert estimates is within the allowable range).

[0144] b. Experts’ estimates:

[0145] Number of failures , test time hours, environmental severity ;

[0146] Sample size score , data consistency score .

[0147] (2) Weight calculation:

[0148]

[0149]

[0150]

[0151]

[0152] (3) Bayesian updating:

[0153]

[0154]

[0155]

[0156] Decision: The failure rate rises to 0.013, triggering a design review to optimize high-failure components.

[0157] Phase 2: Bench test (dominated by measured data)

[0158] (1) Data source:

[0159] a. Bench test:

[0160] Number of failures , test time hours, environmental severity ;

[0161] Sample size score (The minimum sample size requirement for Phase 2 is 500 hours);

[0162] Data consistency scoring (The difference from expert estimates is within the allowable range).

[0163] b. Similar equipment data:

[0164] Number of failures , test time hours, environmental severity ;

[0165] Sample size score , data consistency score .

[0166] (2) Weight calculation:

[0167]

[0168]

[0169]

[0170]

[0171] (3) Bayesian updating:

[0172]

[0173]

[0174] .

[0175] (4) Decision-making:

[0176] The failure rate decreased but still did not meet the standard, so the bench test sample size was increased to 800 hours.

[0177] Phase 3: Sea trials (real environment verification):

[0178] (1) Data source:

[0179] a. Mooring test:

[0180] Number of failures , test time hours, environmental severity ;

[0181] Sample size score (The minimum sample size requirement for Phase 3 is 800 hours);

[0182] Data consistency scoring (The difference from expert estimates is within the allowable range).

[0183] b. Sea trial data:

[0184] Number of failures , Test time Hours, Environment severity ;

[0185] Sample size score , Data consistency score .

[0186] (2) Weight calculation:

[0187]

[0188]

[0189]

[0190]

[0191] (3) Bayesian update:

[0192]

[0193]

[0194]

[0195] (4) Decision:

[0196] Failure rate 0.006 is lower than target value 0.008, and confidence interval width (0.01), approved to enter the equipment production stage.

[0197] With reference to Figure 4 The application also provides a multi-stage multi-source data driven ship equipment reliability growth system, comprising:

[0198] A collection module 410 is configured to collect heterogeneous data sources of each development stage of ship equipment, and perform data standardization on the heterogeneous data sources.

[0199] A confidence degree module 420 is configured to obtain a confidence degree comprehensive score of each heterogeneous data source, and determine a dynamic weight of the confidence degree according to the confidence degree comprehensive score.

[0200] A parameter updating module 430 is configured to construct a Bayesian hierarchical model, and perform failure rate parameter updating on the Bayesian hierarchical model according to stages by using the dynamic weight.

[0201] The reliability optimization module 440 is used to generate a reliability optimization decision instruction based on the posterior mean of the current stage of the Bayesian hierarchical model in combination with a set failure rate confidence interval.

[0202] Optionally, the heterogeneous data sources include fault data, environmental data, and historical experience data;

[0203] The fault data includes the number of faults and the operating time;

[0204] The environmental data includes temperature, humidity and operating conditions; the operating conditions are quantified as severity coefficients;

[0205] The historical experience data includes a priori parameters of failures of historically similar equipment.

[0206] Optionally, the data standardization process includes:

[0207] Determine the environmental severity of the ship's typical missions;

[0208] In combination with the environmental severity and the severity coefficient corresponding to the operating conditions, the operating times of different data sources are uniformly converted into standard operating times under an equivalent standard environment.

[0209] Optionally, the method for obtaining the comprehensive confidence score includes:

[0210] Obtain a sample size score based on the actual trial sample size and the minimum sample size requirement;

[0211] Obtain environmental consistency scores based on environmental severity and the severity coefficient corresponding to the operating conditions;

[0212] Obtain the difference data between the failure rate of the current data source and other data sources in the stage, and obtain a data consistency score based on the difference data and the maximum allowable difference;

[0213] A comprehensive confidence score is obtained based on the sample size score, environmental consistency score, and data consistency score.

[0214] Optionally, the method for obtaining the dynamic weight of the confidence level includes:

[0215] Get the cumulative score of the confidence comprehensive score of all data sources;

[0216] The dynamic weight of the confidence is obtained based on the confidence score of the current data source and the cumulative score.

[0217] Optionally, the parameter updating process of the Bayesian hierarchical model includes:

[0218] Determining that the number of failures satisfies a Poisson distribution and that the prior distribution of the failure rate is a gamma distribution; the prior parameters of the failure rate are the posterior parameters of the previous stage, which are determined based on historical experience data;

[0219] The cumulative weighted number of failures and the cumulative weighted standard running time of the current stage are obtained; the cumulative weighted number of failures is determined based on the posterior parameters, dynamic weights and number of failures of the previous stage, and the cumulative weighted standard running time is determined based on the posterior parameters, dynamic weights and standard running time of the previous stage; the cumulative weighted number of failures and the cumulative weighted standard running time are used as the posterior distribution of the current stage, and the posterior distribution of the current stage is used as the prior distribution of the next stage to realize the update of the failure rate parameters of the model.

[0220] Optionally, the method for generating the reliability optimization decision instruction includes:

[0221] Obtaining a posterior mean of the Bayesian hierarchical model, and setting a confidence interval based on the posterior mean;

[0222] If the posterior mean exceeds a preset target value, or the upper limit of the confidence interval reaches a safety threshold, an improvement instruction is generated to adjust the failure rate;

[0223] If the confidence interval width exceeds the allowed value, the sample size for the next stage of testing will be increased;

[0224] If the posterior mean does not exceed the preset target value and the confidence interval width does not exceed the allowable value, it is determined that the reliability meets the standard.

[0225] It should be understood that the above-mentioned device is used to execute the method in the above-mentioned embodiment. The implementation principle and technical effect of the corresponding program module in the device are similar to those described in the above-mentioned method. The working process of the device can refer to the corresponding process in the above-mentioned method and will not be repeated here.

[0226] Reference Figure 5 Based on the methods in the above embodiments, an embodiment of the present application provides an electronic device, which may include: a processor (Processor) 510, a communication interface (Communications Interface) 520, a memory (Memory) 530, and a communication bus 540. The processor 510, the communication interface 520, and the memory 530 communicate with each other via the communication bus 540. The processor 510 may call logic instructions in the memory 530 to execute the methods in the above embodiments.

[0227] In addition, the logic instructions in the aforementioned memory 530 can be implemented in the form of a software functional unit and, when sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.

[0228] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0229] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0230] It is understood that the processor in the embodiments of the present application may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0231] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and storage medium can be located in an ASIC.

[0232] The above embodiments can be implemented in whole or in part using software, hardware, firmware, or any combination thereof. When implemented using software, they can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions. When loaded and executed on a computer, the computer program instructions fully or partially produce the processes or functions described in the embodiments of this application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disk, hard disk, tape), optical media (e.g., DVD), or semiconductor media (e.g., solid-state drive (SSD)).

[0233] It will be understood that the various numerical numbers involved in the embodiments of the present application are merely distinctions for the convenience of description and are not intended to limit the scope of the embodiments of the present application.

[0234] It is easy for those skilled in the art to understand that the above is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A multi-stage and multi-source data-driven ship equipment reliability growth method, characterized by: include: Collect heterogeneous data sources at various stages of ship equipment development and standardize the data; Obtain the comprehensive confidence scores of each heterogeneous data source and determine the dynamic weight of the confidence based on the comprehensive confidence scores; Constructing a Bayesian hierarchical model, and updating the failure rate parameters of the Bayesian hierarchical model in stages using dynamic weights; Generate reliability optimization decision instructions based on the posterior mean of the current stage of the Bayesian hierarchical model and the set failure rate confidence interval; The method for obtaining the confidence comprehensive score includes: Obtain a sample size score based on the actual trial sample size and the minimum sample size requirement; Obtain environmental consistency scores based on environmental severity and the severity coefficient corresponding to the operating conditions; Obtain the difference data between the failure rate of the current data source and other data sources in the stage, and obtain a data consistency score based on the difference data and the maximum allowable difference; A comprehensive confidence score is obtained based on the sample size score, environmental consistency score, and data consistency score.

2. The multi-stage and multi-source data-driven ship equipment reliability growth method according to claim 1 is characterized in that: The heterogeneous data sources include fault data, environmental data and historical experience data; The fault data includes the number of faults and the operating time; The environmental data includes temperature, humidity and operating conditions; the operating conditions are quantified as severity coefficients; The historical experience data includes a priori parameters of failures of historically similar equipment.

3. The multi-stage and multi-source data driven ship equipment reliability growth method according to claim 2 is characterized in that: The data standardization process includes: Determine the environmental severity of the ship's typical missions; In combination with the environmental severity and the severity coefficient corresponding to the operating conditions, the operating times of different data sources are uniformly converted into standard operating times under an equivalent standard environment.

4. The multi-stage and multi-source data driven ship equipment reliability growth method according to claim 1 is characterized in that: The method for obtaining the dynamic weight of the confidence level includes: Get the cumulative score of the confidence comprehensive score of all data sources; The dynamic weight of the confidence is obtained based on the confidence score of the current data source and the cumulative score.

5. The multi-stage and multi-source data driven ship equipment reliability growth method according to claim 3 is characterized in that: The parameter updating process of the Bayesian hierarchical model includes: Determining that the number of failures satisfies a Poisson distribution and that the prior distribution of the failure rate is a gamma distribution; the prior parameters of the failure rate are the posterior parameters of the previous stage, which are determined based on historical experience data; Obtaining the cumulative weighted number of failures and the cumulative weighted standard operating time for the current phase; the cumulative weighted number of failures is determined based on the a posteriori parameters, dynamic weights, and number of failures in the previous phase; and the cumulative weighted standard operating time is determined based on the a posteriori parameters, dynamic weights, and standard operating time in the previous phase; The cumulative weighted number of failures and the cumulative weighted standard running time are used as the posterior distribution of the current stage; The posterior distribution of the current stage is used as the prior distribution of the next stage to update the failure rate parameters of the model.

6. The multi-stage and multi-source data driven ship equipment reliability growth method according to claim 1 is characterized in that: The method for generating the reliability optimization decision instruction includes: Obtaining a posterior mean of the Bayesian hierarchical model, and setting a confidence interval based on the posterior mean; If the posterior mean exceeds a preset target value, or the upper limit of the confidence interval reaches a safety threshold, an improvement instruction is generated to adjust the failure rate; If the confidence interval width exceeds the allowed value, the sample size for the next stage of testing will be increased; If the posterior mean does not exceed the preset target value and the confidence interval width does not exceed the allowable value, it is determined that the reliability meets the standard.

7. A multi-stage and multi-source data-driven ship equipment reliability growth system, characterized by: include: The acquisition module is used to collect heterogeneous data sources at various development stages of ship equipment and standardize the data of the heterogeneous data sources; The confidence module is used to obtain the comprehensive confidence scores of various heterogeneous data sources and determine the dynamic weight of the confidence based on the comprehensive confidence scores; A parameter updating module is used to construct a Bayesian hierarchical model and update the failure rate parameters of the Bayesian hierarchical model in stages using dynamic weights; The reliability optimization module is used to generate reliability optimization decision instructions based on the posterior mean of the current stage of the Bayesian hierarchical model and the set failure rate confidence interval; The method for obtaining the confidence comprehensive score includes: Obtain a sample size score based on the actual trial sample size and the minimum sample size requirement; Obtain environmental consistency scores based on environmental severity and the severity coefficient corresponding to the operating conditions; Obtain the difference data between the failure rate of the current data source and other data sources in the stage, and obtain a data consistency score based on the difference data and the maximum allowable difference; A comprehensive confidence score is obtained based on the sample size score, environmental consistency score, and data consistency score.

8. An electronic device, characterized in that: include: at least one memory for storing a computer program; At least one processor is used to execute the program stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed on a processor, the processor is caused to execute the method according to any one of claims 1 to 6.

Images