RCM model optimization method and related device

Abstract

The invention belongs to the technical field of RCM model optimization, and discloses an RCM model optimization method and a related device. The RCM model optimization method comprises the steps of collecting and preprocessing operation data, performing dimension reduction processing on the preprocessed operation data through a principal component analysis method, extracting key feature data of the operation data after dimension reduction processing, and constructing a hybrid optimization algorithm based on fusion of a genetic algorithm and a particle swarm optimization algorithm. Inputting the key feature data into a hybrid optimization algorithm, carrying out a plurality of times of iterative optimization on parameters of the RCM model through the hybrid optimization algorithm, calculating an objective function value of each time of iteration, and when the objective function value meets a preset convergence condition or reaches a preset maximum number of iterations, stopping iteration and outputting the RCM model after iterative optimization; according to the method, the data quality can be improved, the data dimension can be reduced, the calculation complexity can be reduced, important information can be reserved, the optimization efficiency can be improved, and the performance of the RCM model can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of RCM model optimization, and particularly relates to an RCM model optimization method and related device. BACKGROUND

[0002] The RCM (Reliability Centered Maintenance) model originated in the American aviation industry in the late 1960s, and was first proposed by Nolan and Ship of American Airlines. The model aims to determine the preventive maintenance needs of equipment through systematic analysis and optimize the maintenance system. The core of the model is to comprehensively evaluate the functions, failure modes, failure causes and consequences of equipment, and then select the most appropriate maintenance strategy.

[0003] However, with the development of technology, the existing RCM model has gradually revealed problems. On the data level, the data quality is uneven, containing a large amount of noise, outliers and duplicate data, which affects the accuracy of model analysis. At the same time, the data dimension is too high, resulting in high computational complexity and low optimization efficiency.

[0004] Therefore, it is urgent to optimize the RCM model and improve the above-mentioned deficiencies of the RCM model. SUMMARY

[0005] The present application aims to provide an RCM model optimization method and related device to overcome the problems in the prior art. The present application can improve data quality, reduce data dimension, reduce computational complexity, while retaining important information, which helps to improve optimization efficiency and improve the performance of the RCM model.

[0006] To achieve the above-mentioned purposes, the technical solutions adopted by the present application are as follows: In a first aspect, the present application provides an RCM model optimization method, comprising the following steps: Step one, collecting the operation data of the RCM model and pre-processing the operation data; Step two, performing dimensionality reduction processing on the pre-processed operation data by principal component analysis method, and extracting key feature data of the operation data after dimensionality reduction processing; Step three, constructing a hybrid optimization algorithm based on the fusion of genetic algorithm and particle swarm algorithm, setting the crossover probability and mutation probability of the genetic algorithm, and setting the inertia weight and learning factor of the particle swarm algorithm; Step four, inputting the key feature data into the hybrid optimization algorithm, and performing several times of iterative optimization on the parameters of the RCM model by the hybrid optimization algorithm, and calculating the objective function value of each iteration; Step five, when the objective function value meets the preset convergence condition or reaches the preset maximum number of iterations, stopping iteration and outputting the RCM model after iterative optimization; Furthermore, the pre-processing of the operation data specifically includes: Identify and remove outliers in the running data using a machine learning algorithm to obtain outlier-free running data. Then, identify and remove duplicate values ​​in the outlier-free running data using hash comparison to obtain preprocessed running data. Furthermore, the principal component analysis method is used to perform dimensionality reduction processing on the pre-processed operating data to extract key feature data of the operating data after dimensionality reduction processing, specifically including: The pre-processed running data is standardized, a covariance matrix is ​​constructed for the standardized running data, and the covariance matrix is ​​decomposed into eigenvalues ​​to obtain eigenvalues ​​and eigenvectors corresponding to the eigenvalues. The eigenvectors are sorted according to the size of the eigenvalues, and the front vectors whose contribution rate is greater than or equal to the set threshold are extracted. k feature vectors, projecting the normalized running data onto k In the subspace composed of eigenvectors, the standardized original data matrix is ​​obtained, and the reduced-dimensional data matrix is ​​obtained by matrix multiplication of the standardized original data matrix, and several columns of the reduced-dimensional data matrix are used as key feature data; Furthermore, the normalized original data matrix is ​​multiplied by matrix to obtain a reduced-dimensional data matrix, and the formula specifically includes: Y=XV k ; Where Y represents the data matrix after dimensionality reduction; X represents the original data matrix after standardization; k represents the number of eigenvectors; Furthermore, the contribution rate is set at a threshold of 95%; Furthermore, the crossover probability ranges from 0.6 to 0.9, and the mutation probability ranges from 0.01 to 0.05; The inertia weight ranges from 0.4 to 0.9, and the learning factor ranges from 1.5 to 2.5; Furthermore, the formula of the objective function value specifically includes: ; α+β= 1; Where, F represents the objective function value; θ represents the objective function; α represents the first weight coefficient; β represents the second weight coefficient; C total It represents the total maintenance cost per unit time; λ sys Indicates the failure rate.

[0007] In a second aspect, the present application provides an RCM model optimization system, comprising: a data acquisition and preprocessing module configured to acquire operation data of the RCM model and to preprocess the operation data; a dimension reduction processing and key feature data extraction module configured to perform dimension reduction processing on the preprocessed operation data by a principal component analysis method and to extract key feature data of the operation data after the dimension reduction processing; a hybrid optimization algorithm construction and setting module configured to construct a hybrid optimization algorithm based on a fusion of a genetic algorithm and a particle swarm algorithm, and to set a crossover probability and a mutation probability of the genetic algorithm and an inertia weight and a learning factor of the particle swarm algorithm; an iterative optimization module configured to input the key feature data into the hybrid optimization algorithm, to perform several times of iterative optimization on parameters of the RCM model by the hybrid optimization algorithm, and to calculate a target function value of each iteration; an RCM model output module configured to stop the iteration and to output the RCM model after the iterative optimization when the target function value meets a preset convergence condition or a preset maximum number of iterations is reached.

[0008] In a third aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0009] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program implements the steps of the above method when executed by a processor.

[0010] The above technical solution has the following advantages or beneficial effects: In a first aspect, the present application provides an RCM model optimization method. By collecting and preprocessing operation data, noise and outliers can be removed, data quality can be improved, and a reliable foundation can be provided for subsequent analysis. By principal component analysis dimension reduction processing, key feature data can be extracted, data dimension can be reduced, calculation complexity can be reduced, important information can be retained, and optimization efficiency can be improved. By constructing a hybrid optimization algorithm based on the fusion of genetic algorithm and particle swarm algorithm, the advantages of the two algorithms are combined. The global search ability of genetic algorithm is strong, and the convergence speed of particle swarm algorithm is fast. By reasonably setting the crossover probability, mutation probability, inertia weight and learning factor, the algorithm is more adaptable and efficient. By inputting the key feature data into the hybrid optimization algorithm for iterative optimization, the RCM model parameters can be adjusted more accurately, and the model performance can be improved. When the target function value meets the convergence condition or the maximum iteration number is reached, the iteration is stopped, and the optimized model is output, ensuring the effectiveness of the optimization process and the reliability of the results, and providing a higher quality RCM model for practical application.

[0011] Further, by using machine learning algorithms to identify and remove outliers, abnormal values in the operation data can be effectively removed, avoiding interference of these abnormal data on subsequent analysis and optimization process, ensuring that the data truly reflects the model operation, and improving data accuracy. On this basis, by using hash comparison to identify and remove duplicate values, data is further purified, data redundancy is reduced, subsequent calculation complexity is reduced, and data processing efficiency is improved. After these two preprocessing steps, high-quality preprocessed operation data is obtained, providing a more reliable data basis for subsequent principal component analysis dimension reduction, hybrid optimization algorithm iterative optimization and other steps, which helps to optimize the performance of the RCM model.

[0012] Further, by standardizing the data, the influence of different dimensions and orders of magnitude is eliminated, making each data equally important in analysis, improving the accuracy and reliability of subsequent analysis; by constructing a covariance matrix and performing eigenvalue decomposition, the internal structure and correlation between data can be clearly revealed; by sorting the eigenvectors according to the eigenvalue size and extracting the first k characteristic vectors with a contribution rate greater than a set threshold, the main features of the data are captured and redundant information is removed, while retaining key data information, significantly reducing data dimension, reducing calculation and storage requirements, and improving the efficiency of subsequent model optimization; by projecting the standardized data into a subspace composed of k the first characteristic vectors, a dimension-reduced data matrix is obtained, and key feature data is extracted from it. The key feature data set embodies the main features of the original data and can be more effectively used in subsequent hybrid optimization algorithm iterative optimization, which helps the optimization algorithm to more accurately adjust the RCM model parameters and improve the model performance, providing a higher quality RCM model for practical application.

[0013] Further, the reduced dimension data matrix retains the main features of the original data, so that the hybrid optimization algorithm based on the data can more accurately capture the data regularity, thereby more effectively adjusting the RCM model parameters, helping to improve the performance and accuracy of the model, so that the optimized model can better meet the needs in practical applications.

[0014] Further, the key feature data extracted under the threshold can make the hybrid optimization algorithm more accurately adjust the RCM model parameters and improve the model performance. The threshold is reasonable and easy to operate, providing a reliable data basis for model optimization, which helps to obtain a better RCM model in practical applications.

[0015] Further, in the genetic algorithm, the crossover probability is 0.6-0.9, which can make the individuals have a higher probability of gene exchange, effectively promote population diversity, avoid the algorithm falling into local optimum too early, and also not cause the excellent genes to be destroyed due to too high crossover probability; the mutation probability is 0.01-0.05, which can introduce new genes to a certain extent and enhance the global search ability of the algorithm, and also will not destroy the stability of the population due to too frequent mutation; in the particle swarm algorithm, the inertia weight is 0.4-0.9, which can balance the global search and local search ability of the particle, so that the particle can flexibly adjust the search direction and speed at different stages; the learning factor is 1.5-2.5, which can guide the particle to learn from the individual optimal and global optimal positions, and accelerate the convergence speed.

[0016] In the second aspect, the application provides an RCM model optimization system. The data acquisition and preprocessing module ensures that the system can obtain accurate and high-quality operation data, improves the data reliability by collecting data and removing outliers and duplicate values, lays a solid foundation for subsequent analysis, and avoids interference of abnormal and redundant data on the optimization process; the dimension reduction processing and key feature data extraction module uses principal component analysis method to effectively reduce the data dimension, remove redundant information, and extract key feature data, which not only reduces the calculation complexity, but also enables the optimization algorithm to focus on important features and improve optimization efficiency and accuracy; the hybrid optimization algorithm construction and setting module combines genetic algorithm and particle swarm algorithm, reasonably sets key parameters, fully utilizes the advantages of the two algorithms, enhances the global search and local search ability of the algorithm, and makes the optimization process more efficient and accurate; the iterative optimization module iteratively optimizes the RCM model parameters multiple times and calculates the objective function value, which can gradually approach the optimal model, and the RCM model output module stops iteration and outputs the result in time when the conditions are met, ensuring the effectiveness of the optimization process and the reliability of the result, providing a RCM model with better performance for users, and helping to improve the decision-making quality and efficiency in practical applications.

[0017] In a third aspect, the present application provides a computer device capable of efficiently implementing the steps of the method of the present application by executing a specific computer program through a processor, which can accurately perform numerical calculation and logical judgment when performing data processing tasks, thus avoiding errors caused by human factors; meanwhile, since the computer program has high stability and reliability, the accuracy and consistency of the data processing results can be ensured.

[0018] In a fourth aspect, the present application provides a computer readable storage medium, in which the steps of the method of the present application are programmed as a computer program and stored on the computer readable storage medium, so that users can easily load these programs onto any compatible computer device and execute them without the need to rewrite or convert the codes, thus greatly improving the convenience and flexibility of program execution. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 A flowchart of the RCM model optimization method of the present application; Figure 2 A structural diagram of the computer device of the present application. DETAILED DESCRIPTION

[0020] The present application will be further described in detail below with reference to specific examples, which are intended to explain but not limit the present application. In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts should fall within the scope of the present application. It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to include only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0021] Embodiment: The present application is applicable to but not limited to the thermal control of power plants, power stations, boiler systems, steam turbine systems and water turbine systems.

[0022] Referring to Figure 1 The application provides an RCM model optimization method, comprising the following steps: Step one, collecting the operation data of the RCM model, and pre-processing the operation data; Specifically, the pre-processing of the operation data comprises: identifying and removing outliers in the operation data through a machine learning algorithm to obtain operation data with abnormal values removed, and then identifying and removing duplicate values in the operation data with abnormal values removed through hash comparison to obtain pre-processed operation data; Preferably, the machine learning algorithm can be a supervised learning algorithm, an unsupervised learning algorithm, a semi-supervised learning algorithm, a reinforcement learning algorithm, or other machine learning algorithms; the supervised learning algorithm can be linear regression, logistic regression, decision tree, random forest, support vector machine or neural network, etc.; the unsupervised learning algorithm can be clustering algorithm, dimensionality reduction algorithm or association rule learning algorithm, etc.; the reinforcement learning algorithm can be Q-learning, policy gradient method or deep reinforcement learning, etc. Step two, performing dimensionality reduction processing on the pre-processed operation data through a principal component analysis method, and extracting key feature data of the operation data after the dimensionality reduction processing; Specifically, the pre-processed operation data is subjected to standardization processing, a covariance matrix is constructed for the operation data after the standardization processing, the covariance matrix is subjected to eigenvalue decomposition to obtain eigenvalues and eigenvectors corresponding to the eigenvalues, the eigenvectors are sorted according to the sizes of the eigenvalues, and the first k eigenvectors with a contribution rate greater than or equal to a set threshold value are extracted, the operation data after the standardization processing is projected into a subspace composed of the k eigenvectors to obtain a standardized original data matrix, and the standardized original data matrix is subjected to matrix multiplication to obtain a dimensionality-reduced data matrix, and several columns of the dimensionality-reduced data matrix are taken as the key feature data; Specifically, the standardized original data matrix is subjected to matrix multiplication to obtain a dimensionality-reduced data matrix, and the formula is specifically as follows: Y=XV k ; In the formula, Y represents the dimensionality-reduced data matrix; X represents the standardized original data matrix; k represents the number of eigenvectors; Specifically, the set threshold value of the contribution rate is 90-98%; Preferably, the set threshold value of the contribution rate can be 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or any value within 90-98%; Step three, a hybrid optimization algorithm based on the fusion of genetic algorithm and particle swarm algorithm is constructed, and the crossover probability and mutation probability of the genetic algorithm and the inertia weight and learning factor of the particle swarm algorithm are set; Specifically, the value range of the crossover probability is 0.6-0.9, the value range of the mutation probability is 0.01-0.05, the value range of the inertia weight is 0.4-0.9, and the value range of the learning factor is 1.5-2.5. Preferably, the crossover probability can be 0.6, 0.7, 0.8, 0.9, or any value within 0.6-0.9. Preferably, the mutation probability can be 0.01, 0.002, 0.003, 0.004, 0.005, or any value within 0.01-0.05. Preferably, the inertia weight can be 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, or any value within 0.4-0.9. Preferably, the learning factor can be 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, or any value within 1.5-2.5. Step four, input the key feature data into the hybrid optimization algorithm, and perform several times of iteration optimization on the parameters of the RCM model through the hybrid optimization algorithm to calculate the objective function value of each iteration. Specifically, the formula of the objective function value specifically includes: ; α+β= 1; In the formula, F denotes the objective function value; θ denotes the objective function; α denotes the first weight coefficient; β denotes the second weight coefficient; C total denotes the total maintenance cost per unit time; λ sys denotes the failure rate; Step five, when the objective function value meets the preset convergence condition or reaches the preset maximum number of iterations, stop iteration and output the RCM model after iteration optimization.

[0023] In an embodiment of the present application, an RCM model optimization method is provided, comprising the following steps: Step one, collect the equipment operation data of the past year from the RCM model, including equipment state, maintenance record, fault history, etc., use random forest algorithm to detect anomalies in the operation data, identify and remove outliers in the data, get the operation data without outliers, detect repeated values in the operation data without outliers through hash comparison technology, and remove repeated records to get preprocessed operation data; Step two, standardize the preprocessed operation data to make the data have zero mean and unit variance, construct a covariance matrix for the standardized operation data to analyze the correlation between the data, perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues and eigenvectors corresponding to the eigenvalues, sort the eigenvectors according to the size of the eigenvalues, select the first k characteristic vectors with a contribution rate greater than or equal to 95%, project the standardized original data matrix into the subspace composed of k characteristic vectors, and obtain the dimensionality reduced data matrix through matrix multiplication, and use several columns of the dimensionality reduced data matrix as key feature data for subsequent model optimization; Step three, construct a hybrid optimization algorithm based on the fusion of genetic algorithm and particle swarm algorithm, set the crossover probability of genetic algorithm to 0.8 and the mutation probability to 0.03, and set the inertia weight of particle swarm algorithm to 0.7 and the learning factor to 2.0; Step four, input the extracted key feature data into the hybrid optimization algorithm, and perform several iterations of optimization on the parameters of the RCM model through the hybrid optimization algorithm, and calculate the objective function value after each iteration; Step five, when the change rate of the objective function value is less than 1% for 5 consecutive iterations, it is considered that the objective function value meets the preset convergence condition, or when the preset maximum number of iterations is 100, the iteration is stopped, and the RCM model after iteration optimization is output.

[0024] In an embodiment of the present application, an RCM model optimization system is provided, comprising: a data acquisition and preprocessing module for acquiring operation data of an RCM model and preprocessing the operation data; a dimensionality reduction processing and key feature data extraction module for performing dimensionality reduction processing on the preprocessed operation data through principal component analysis method and extracting key feature data of the dimensionally reduced operation data; a hybrid optimization algorithm construction and setting module for constructing a hybrid optimization algorithm based on the fusion of genetic algorithm and particle swarm algorithm, setting the crossover probability and mutation probability of genetic algorithm, and setting the inertia weight and learning factor of particle swarm algorithm; An iterative optimization module is configured to input the key feature data into a hybrid optimization algorithm, and perform several times of iterative optimization on parameters of the RCM model by the hybrid optimization algorithm to calculate a target function value of each iteration; An RCM model output module after iterative optimization is configured to stop the iteration and output the RCM model after iterative optimization when the target function value meets a preset convergence condition or reaches a preset maximum number of iterations.

[0025] Referring to Figure 2 In an embodiment of the present application, a computer device is provided, which comprises a processor and a memory, the memory is configured to store a computer program, the computer program comprises program instructions, and the processor is configured to execute the program instructions stored in the computer storage medium; the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, and are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement a corresponding method flow or a corresponding function; the processor in the embodiment of the present application can be used for the operation of the RCM model optimization method.

[0026] In an embodiment of the present application, a computer readable storage medium, specifically a Memory, is provided. The computer readable storage medium is a memory device in a computer device, and is configured to store programs and data. It should be understood that the computer readable storage medium can include an internal storage medium of the computer device, and can also include an external storage medium supported by the computer device. The computer readable storage medium provides a storage space, and the storage space stores an operating system of the terminal. In addition, one or more instructions stored in the storage space are adapted to be loaded and executed by a processor, and the instructions can be one or more computer programs (including program codes). It should be understood that the computer readable storage medium can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory. The processor can load and execute the one or more instructions stored in the computer readable storage medium, so as to implement the corresponding steps of the RCM model optimization method in the embodiment.

[0027] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. In addition, the present application can adopt a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0028] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device implemented in the flowcharts and / or block diagrams. Figure 1 The function of one or more flows and / or blocks Figure 1 The function of one or more flows and / or blocks

[0029] These computer program instructions can also be stored in a computer readable storage medium capable of directing the computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable storage medium produce a manufactured product including instruction devices, which implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The function of one or more flows and / or blocksFigure 1 the function specified in the one or more blocks.

[0030] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, so that the instructions executed on the computer or other programmable data processing devices provide processes for implementing the flows Figure 1 the flows or the plurality of flows and / or blocks Figure 1 the steps of the function specified in the one or more blocks.

[0031] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A RCM model optimization method, characterized in that: The following steps are involved: Collect the operating data of the RCM model and pre-process the operating data; The pre-processed operating data is subjected to dimensionality reduction processing by principal component analysis method, and key feature data of the operating data after dimensionality reduction processing is extracted; Construct a hybrid optimization algorithm based on the fusion of genetic algorithm and particle swarm algorithm, set the crossover probability and mutation probability of genetic algorithm, and the inertia weight and learning factor of particle swarm algorithm; Input the key feature data into the hybrid optimization algorithm, perform several iterative optimizations on the parameters of the RCM model through the hybrid optimization algorithm, and calculate the objective function value of each iteration; When the objective function value meets the preset convergence condition or reaches the preset maximum number of iterations, the iteration is stopped and the iteratively optimized RCM model is output.

2. The RCM model optimization method according to claim 1, characterized in that: The preprocessing of the operation data specifically includes: The outliers in the running data are identified and removed by a machine learning algorithm to obtain the running data with outliers removed. Then, the duplicate values ​​in the running data with outliers removed are identified and removed by hash comparison to obtain the preprocessed running data.

3. The RCM model optimization method according to claim 1, characterized in that: The principal component analysis method is used to perform dimensionality reduction processing on the pre-processed operation data, and key feature data of the operation data after dimensionality reduction processing is extracted, specifically including: The pre-processed running data is standardized, a covariance matrix is ​​constructed for the standardized running data, and the covariance matrix is ​​decomposed into eigenvalues ​​to obtain eigenvalues ​​and eigenvectors corresponding to the eigenvalues. The eigenvectors are sorted according to the size of the eigenvalues, and the front vectors whose contribution rate is greater than or equal to the set threshold are extracted. k feature vectors, projecting the normalized running data onto k In the subspace composed of eigenvectors, the standardized original data matrix is ​​obtained, and the reduced-dimensional data matrix is ​​obtained by matrix multiplication of the standardized original data matrix, and several columns of the reduced-dimensional data matrix are used as key feature data.

4. The RCM model optimization method according to claim 3, characterized in that: The normalized original data matrix is ​​obtained by matrix multiplication to obtain the reduced-dimensional data matrix, and the formula specifically includes: Y=XV k ; Where Y represents the data matrix after dimensionality reduction; X represents the original data matrix after standardization; k represents the number of eigenvectors.

5. The RCM model optimization method according to claim 3, characterized in that: The contribution rate is set at a threshold of 95%.

6. The RCM model optimization method according to claim 1, characterized in that: The crossover probability ranges from 0.6 to 0.9, and the mutation probability ranges from 0.01 to 0.05; The inertia weight ranges from 0.4 to 0.9, and the learning factor ranges from 1.5 to 2.

5.

7. The RCM model optimization method according to claim 1, characterized in that: The formula of the objective function value specifically includes: ； α+β= 1； Where, F represents the objective function value; θ represents the objective function; α represents the first weight coefficient; β represents the second weight coefficient; C total It represents the total maintenance cost per unit time; λ sys Indicates the failure rate.

8. An RCM model optimization system, characterized in that: An RCM model optimization method according to any one of claims 1 to 7, comprising: Data acquisition and preprocessing module, used to collect the operating data of the RCM model and preprocess the operating data; Dimensionality reduction processing and key feature data extraction module, used to perform dimensionality reduction processing on the pre-processed operating data through the principal component analysis method, and extract the key feature data of the operating data after dimensionality reduction processing; The hybrid optimization algorithm construction and setting module is used to build a hybrid optimization algorithm based on the fusion of genetic algorithm and particle swarm algorithm, set the crossover probability and mutation probability of the genetic algorithm, and the inertia weight and learning factor of the particle swarm algorithm; Iterative optimization module, which is used to input key feature data into the hybrid optimization algorithm, perform several iterative optimizations on the parameters of the RCM model through the hybrid optimization algorithm, and calculate the objective function value of each iteration; The iteratively optimized RCM model output module is used to stop the iteration and output the iteratively optimized RCM model when the objective function value meets the preset convergence condition or reaches the preset maximum number of iterations.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images