A performance evaluation method and system for railway prefabricated box beams

Abstract

The present application provides a performance evaluation method and system for railway prefabricated box girders, which relate to the field of transportation technology. The method includes: introducing intelligent process performance indicators to construct an improved quality and performance evaluation index system, and constructing an initial structural equation model corresponding to the railway prefabricated box girder; calculating the degree of fit of the initial structural equation model based on a preset degree of fit index; when the degree of fit of the initial structural equation model is greater than or equal to a first preset threshold, obtaining the production data, test data, and evaluation information of the intelligent process performance index of the railway prefabricated box girder; based on the production data, test data, and evaluation information of the intelligent process performance index of the railway prefabricated box girder, the target structural equation model obtained after adjusting the parameters of the initial structural equation model is used to evaluate the performance of the railway prefabricated box girder, and obtain the performance evaluation results of the railway prefabricated box girder. The present application improves the effectiveness of the performance evaluation results.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of transportation technology, and in particular to a performance evaluation method and system for a railway prefabricated box girder. Background Art

[0002] Because the simply supported box girder has the advantages of high rigidity, good force-bearing performance, simple structure and prefabricated construction, it has become the main structural type of railway bridges in my country's high-speed railway construction.

[0003] At present, the commonly used spans of simply supported box girders are 24m, 32m and 40m, single-track and double-track, ballasted and ballastless and other beam types. Regardless of the beam type, the prefabricated erection method is mainly used for construction. Therefore, the simply supported box girders used in railway construction and constructed by the prefabricated erection method can also be called railway prefabricated box girders.

[0004] In short, railway precast box girders are concrete beams prefabricated in factories or prefabrication yards. They use box sections and prestressed technology and have high rigidity and bending resistance.

[0005] Railway prefabricated box girders are constructed through the prefabrication erection method and are widely used in the construction of bridges for high-speed railways and ordinary railways to ensure the safety and efficiency of railway bridges.

[0006] In order to ensure that the railway prefabricated box girders have good quality, the railway industry has always managed the railway prefabricated box girders as products in the production process, formulated the "Implementation Rules for the Production License of Prestressed Concrete Railway Bridge Simply Supported Beams", and compiled special technical specifications "Prefabricated Post-Tensioned Prestressed Concrete Simply Supported Beams for High-Speed ​​Railways", as well as requiring performance evaluation of the finished products in the testing process (such as appearance, size and static load test acceptance). Through the specific operations in the above production and testing links, reliable multi-level quality assurance measures are provided for the railway prefabricated box girders throughout the entire production process, and they have played a very important role in the promotion and application of new equipment, new technologies and new processes for a long time.

[0007] In short, the railway industry has established a set of traditional quality and performance evaluation index system for railway prefabricated box girders, and based on this traditional quality and performance evaluation index system for railway prefabricated box girders, it manages the production process of railway prefabricated box girders and conducts performance testing in the post-production testing process.

[0008] However, with the widespread application of new generation information technology in the engineering field, the construction of prefabricated railway box girders has gradually developed towards informatization, digitization and intelligence, which has not only brought about efficiency improvements, but also posed new challenges to the traditional quality and performance evaluation index system of prefabricated railway box girders. Therefore, the traditional quality and performance evaluation index system of prefabricated railway box girders can no longer fully meet this new requirement.

[0009] For example, in the early stages of the application of intelligent tensioning technology, there were accidents of concrete cracking in beams due to uncontrolled tensioning force. There were also quality accident cases caused by excessive control of both tensioning force and elongation, or problems such as poor equipment durability and low reliability, which affected the smooth progress of project construction.

[0010] In order to meet these challenges, it is urgent to establish a new quality and performance evaluation index system that comprehensively covers intelligent equipment, intelligent management, and intelligent construction.

[0011] In other words, faced with the continuous improvement of the intelligent construction standards of railway prefabricated box girders, the corresponding quality and performance evaluation index system has not yet been formed. If the traditional quality and performance evaluation index system of railway prefabricated box girders is still used, it will to a certain extent restrict the large-scale promotion and application of intelligence.

[0012] Therefore, how to propose a performance evaluation method for railway prefabricated box girders based on the traditional quality and performance evaluation index system of railway prefabricated box girders is an urgent problem to be solved. Summary of the invention

[0013] The embodiments of the present application provide a performance evaluation method and system for railway prefabricated box girders, which are used to solve the problem in the prior art that the performance of railway prefabricated box girders cannot be effectively evaluated due to the insufficient adaptability of the traditional evaluation index system used in the construction environment.

[0014] In a first aspect, an embodiment of the present application provides a performance evaluation method for a railway prefabricated box girder, comprising:

[0015] An improved quality and performance evaluation index system is constructed by introducing intelligent process performance indicators, and an initial structural equation model corresponding to the railway prefabricated box girder is constructed based on the improved quality and performance evaluation index system;

[0016] The goodness of fit of the initial structural equation model was calculated based on the preset goodness of fit indices;

[0017] When the degree of fit of the initial structural equation model is greater than or equal to a first preset threshold, obtaining production data, test data, and evaluation information of intelligent process performance indicators of railway prefabricated box girders;

[0018] Based on the production data and inspection data of railway prefabricated box girders and the evaluation information of intelligent process performance indicators, the performance of railway prefabricated box girders is evaluated by using a target structural equation model to obtain a performance evaluation result of the railway prefabricated box girders; wherein the target structural equation model is obtained after adjusting the parameters of the initial structural equation model.

[0019] Optionally, parameters of the initial structural equation model are adjusted to obtain a target structural equation model, including:

[0020] Obtain the evaluation opinions of domain experts on the performance indicators of intelligent processes;

[0021] Determine a static prior distribution of intelligent process performance indicators based on the evaluation opinion text;

[0022] Generating a non-static prior distribution of intelligent process performance indicators based on the static prior distribution;

[0023] Calculating the posterior distribution based on the non-static prior distribution and the production data and test data of the railway prefabricated box girder and the evaluation information of the intelligent process performance index;

[0024] The parameters of the initial structural equation model are updated based on the parameter estimation in the posterior distribution to obtain the structural equation model to be determined;

[0025] Calculate the goodness of fit of the structural equation model to be determined;

[0026] When the degree of fit of the structural equation model to be determined is greater than or equal to a second preset threshold, the structural equation model to be determined is determined as a target structural equation model; and the second preset threshold is greater than or equal to the first preset threshold.

[0027] Optionally, the non-static prior distribution includes a dynamic prior distribution;

[0028] The generating of the non-static prior distribution of the intelligent process performance indicator based on the static prior distribution includes:

[0029] Based on the static prior distribution, a first generation method is used to generate a dynamic prior distribution of intelligent process performance indicators; the first generation method includes a state space model construction method or a time series analysis method.

[0030] Optionally, the non-static prior distribution includes an adaptive prior distribution;

[0031] The generating of the non-static prior distribution of the intelligent process performance indicator based on the static prior distribution includes:

[0032] Based on the static prior distribution, a second generation method is used to generate an adaptive prior distribution of intelligent process performance indicators; the second generation method includes an online learning method or an adaptive algorithm.

[0033] Optionally, the target structural equation model includes a nonlinear structural equation model using a polynomial regression formula or a neural network expression.

[0034] Optionally, the target structural equation model includes a piecewise linear structural equation model.

[0035] Optionally, the intelligent process performance indicators include: intelligent production performance indicators and intelligent detection performance indicators;

[0036] Intelligent production performance indicators include at least one of the following: maturity of intelligent equipment application, quality indicator compliance or production efficiency improvement;

[0037] Intelligent detection performance indicators include at least one of the following: intelligent monitoring and detection maturity, and detection indicator compliance.

[0038] Optionally, after calculating the fit of the initial structural equation model based on a preset fit index, the method further includes:

[0039] When the fit of the initial structural equation model is less than a first preset threshold, a genetic algorithm is used to iteratively adjust the structure or parameters of the initial structural equation model based on the fit until the fit of the initial structural equation model is greater than or equal to the preset threshold.

[0040] Optionally, after the performance evaluation result of the railway prefabricated box girder is obtained by using the target structural equation model to evaluate the performance of the railway prefabricated box girder based on the production data, test data and evaluation information of the intelligent process performance index of the railway prefabricated box girder, the method further includes:

[0041] A customer satisfaction index is calculated based on the performance evaluation results of the railway prefabricated box girder.

[0042] In a second aspect, an embodiment of the present application provides a performance evaluation system for a railway prefabricated box girder, comprising:

[0043] A construction module is used to construct an improved quality and performance evaluation index system by introducing intelligent process performance indicators, and to construct an initial structural equation model corresponding to the railway prefabricated box girder based on the improved quality and performance evaluation index system;

[0044] A calculation module, used for calculating the goodness of fit of the initial structural equation model based on a preset goodness of fit index;

[0045] An acquisition module, used for acquiring production data, test data, and evaluation information of intelligent process performance indicators of railway prefabricated box girders when the fitting degree of the initial structural equation model is greater than or equal to a first preset threshold value;

[0046] The performance evaluation module is used to evaluate the performance of the railway prefabricated box girder based on the production data, test data and evaluation information of the intelligent process performance indicators of the railway prefabricated box girder using the target structural equation model to obtain the performance evaluation results of the railway prefabricated box girder; wherein the target structural equation model is obtained after adjusting the parameters of the initial structural equation model.

[0047] In an embodiment of the present application, a performance evaluation method for a railway prefabricated box girder is provided, including: constructing an improved quality and performance evaluation index system by introducing intelligent process performance indicators, and constructing an initial structural equation model corresponding to the railway prefabricated box girder based on the improved quality and performance evaluation index system; calculating the degree of fit of the initial structural equation model based on a preset degree of fit index; obtaining production data, test data, and evaluation information of the intelligent process performance index of the railway prefabricated box girder when the degree of fit of the initial structural equation model is greater than or equal to a first preset threshold; based on the production data, test data, and evaluation information of the intelligent process performance index of the railway prefabricated box girder, the target structural equation model obtained after adjusting the parameters of the initial structural equation model is used to evaluate the performance of the railway prefabricated box girder, and obtain the performance evaluation result of the railway prefabricated box girder. The embodiment of the present application constructs a new quality and performance evaluation index system suitable for intelligent equipment, intelligent management, and intelligent construction by introducing intelligent process performance indicators, thereby improving the applicability of the initial structural equation model constructed based on the new quality and performance evaluation index system in the full-process intelligent environment of intelligent equipment, intelligent management, and intelligent construction, thereby ensuring the validity of the performance evaluation result.

[0048] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 A flow chart of a method for evaluating the performance of a railway prefabricated box girder provided in an embodiment of the present application;

[0051] Figure 2 It is a schematic diagram of the framework structure of the quality and performance evaluation index system of the traditional railway prefabricated box girder;

[0052] Figure 3 A structural schematic diagram corresponding to an initial structural equation model provided in an embodiment of the present application;

[0053] Figure 4 A path diagram corresponding to a preset structural equation model provided in an embodiment of the present application;

[0054] Figure 5 A schematic diagram of the structure of a performance evaluation system for a railway prefabricated box girder provided in an embodiment of the present application;

[0055] Figure 6 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0057] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as 11, 12, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to different types.

[0058] To solve this technical problem, the embodiment of the present application proposes a performance evaluation method for railway prefabricated box girders. By introducing intelligent process performance indicators, a new quality and performance evaluation index system suitable for intelligent equipment, intelligent management, and intelligent construction is constructed. The applicability of the structural equation model constructed based on the new quality and performance evaluation index system in the full-process intelligent environment of intelligent equipment, intelligent management, and intelligent construction is improved, thereby ensuring the effectiveness of the performance evaluation results.

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0060] Figure 1 A flow chart of a performance evaluation method for a railway prefabricated box girder provided in an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0061] S11. An improved quality and performance evaluation index system is constructed by introducing intelligent process performance indicators, and an initial structural equation model corresponding to railway prefabricated box girders is constructed based on the improved quality and performance evaluation index system.

[0062] In this embodiment, the improved quality and performance evaluation index system is also called a new quality and performance evaluation index system. The improved quality and performance evaluation index system introduces intelligent process performance indicators based on the traditional quality and performance evaluation index system of railway prefabricated box girders. Figure 2 As shown in the figure, the quality and performance evaluation index system of traditional railway prefabricated box girders includes quality and performance evaluation indicators in seven processes and seven tests, among which the seven processes include: steel bar cutting and forming, concrete production, transportation and pouring vibration, concrete curing and temperature control, prestressed strand threading and tensioning, channel grouting and control, beam end anchor hole roughening and sealing, and beam end anti-corrosion coating spraying and control. The seven tests include: raw material performance inspection, concrete mixture inspection, beam body concrete specimen inspection, beam body finished product appearance quality inspection, beam body finished product size measurement, beam body linear control, and beam body static load test evaluation. It should be understood that the first three of the seven tests can be intelligently inspected during the production process, and the last four can be intelligently inspected after the production is completed.

[0063] The intelligent process performance indicators include: intelligent production performance indicators corresponding to the seven processes and intelligent detection performance indicators corresponding to the seven inspections; intelligent production performance indicators include at least one of the following: maturity of intelligent equipment application, compliance with quality indicators or improvement in production efficiency. In addition, this embodiment does not specifically limit the content of intelligent production performance indicators. Exemplarily, intelligent production performance indicators may also include at least one of the production intelligence level, quality effect after implementation, and efficiency improvement. Intelligent detection performance indicators include at least one of the following: maturity of intelligent monitoring and detection, and compliance with detection indicators. In addition, this embodiment does not specifically limit the content of intelligent detection performance indicators. Exemplarily, intelligent detection performance indicators may also include at least one of the monitoring and detection intelligence level and quality effect after implementation.

[0064] In order to comprehensively evaluate the performance of railway prefabricated box girders, the embodiment of the present application can construct an initial structural equation model including a structural model and a measurement model. Specifically, the embodiment can determine hidden variables and explicit variables based on the improved quality and performance evaluation index system, and then construct a structural model and a measurement model.

[0065] Among them, the latent variables include: latent endogenous variables η and latent exogenous variables ξ, and the explicit variables include: explicit endogenous variables Y and explicit exogenous variables X.

[0066] The above four variables are explained as follows: (1) The latent endogenous variable η is determined by other variables in the model, that is, the latent endogenous variable η is the dependent variable in the structural model. In this embodiment, the latent endogenous variable η represents seven processes and seven tests, for example, Figure 2 As shown, there are 14 in total.

[0067] (2) The implicit exogenous variable ξ is an independent variable in the structural model and is not affected by other variables in the structural model. In this embodiment, the implicit exogenous variable ξ may include the following five items: intelligent equipment application maturity, quality index compliance, production efficiency improvement, intelligent monitoring and detection maturity, and detection index compliance.

[0068] (3) Explicit endogenous variables Y are variables that can be directly observed. They are the external manifestations of implicit endogenous variables. In this embodiment, explicit endogenous variables Y represent the quality and performance evaluation indicators in seven processes and seven tests. For example, Figure 2 As shown, 30 in total.

[0069] (4) Explicit exogenous variables X are variables that can be directly observed, and they are the external manifestations of implicit exogenous variables. In this embodiment, explicit exogenous variables X can include the following five items: usage of intelligent equipment, quality control indicators, production efficiency indicators, intelligent monitoring and detection indicators, and detection indicators.

[0070] Accordingly, this embodiment may be provided with three matrix equations to express the measurement model and the structural model:

[0071] ;

[0072] ;

[0073] ;

[0074] Among them, formula (1) is the expression of the structural model, which is used to reflect the correlation between the implicit exogenous variable ξ and the implicit endogenous variable η in the initial structural equation model. B represents the effect coefficient matrix between the implicit endogenous variables η, Γ represents the effect coefficient matrix of the implicit exogenous variable ξ on the implicit endogenous variable η, and any element in B and Γ can be called a path coefficient; ς represents the vector composed of residual terms, which represents random errors or unobserved influencing factors.

[0075] Formula (2) and (3) together constitute the expression of the measurement model. Formula (2) and (3) respectively reflect the relationship between the implicit endogenous variable η and the explicit endogenous variable Y, and the relationship between the implicit exogenous variable ξ and the explicit exogenous variable X. and They represent the regression coefficients (or factor loading matrices) of the implicit endogenous variable η and the implicit exogenous variable ξ, respectively, and ε and σ represent the measurement errors of the explicit endogenous variable Y and the explicit exogenous variable X, respectively.

[0076] based on Figure 2 Given the 30 indicators and the expression of the initial structural equation model, the embodiment of the present application can design the explicit endogenous variable Y. The 30 indicators involved in the explicit endogenous variable Y are shown in Table 1:

[0077] Table 1 Textual description of each explicit endogenous variable Y

[0078] ;

[0079] Based on the design of the explicit endogenous variable Y, this embodiment can obtain the following Figure 3 The initial structural equation model is shown.

[0080] Optionally, in the performance evaluation process of railway prefabricated box girders, in order to make the analysis more specific, this embodiment can replace the explicit endogenous variable Y described in text with a quantified value. Exemplarily, the embodiment of the present application uses a Likert scale four-point table for questionnaire survey, i.e., a 1-4 point system, and refines the 14 first-level indicators (i.e., seven processes and seven tests) in the initial structural equation model into 30 second-level indicators (i.e., Y1~Y30), and uses a Likert scale four-point table rating to assign values ​​to each indicator.

[0081] based on Figure 3 And formula (2), the partial equations of the measurement model can be specifically: ;

[0082] in, All of them are implicit endogenous variables.

[0083] Based on the above formula (1), the equation of the structural model can be specifically: ;

[0084] Among them, each β represents the effect coefficient between two implicit endogenous variables.

[0085] Optionally, this embodiment may use a four-point Likert scale to grade and quantify the scores of five implicit exogenous variables ξ (i.e., intelligent equipment application maturity, quality index compliance, production efficiency improvement, intelligent monitoring and detection maturity, and detection index compliance). For example, the grading and quantitative scoring information of the five explicit exogenous variables X are shown in Tables 2 to 6 respectively:

[0086] Table 2 Classification and quantitative scoring information of intelligent equipment application maturity

[0087]

[0088] Table 3 Quality indicator compliance classification and quantitative scoring information

[0089]

[0090] Table 4 Classification and quantitative scoring information of production efficiency improvement

[0091]

[0092] Table 5 Classification and quantitative scoring information of intelligent monitoring and detection maturity

[0093]

[0094] Table 6 Grading and quantitative scoring information of test index compliance

[0095]

[0096] Among them, the main control items (A-level indicators) in Table 3 and Table 6 refer to key indicators that have a decisive influence on the performance evaluation of railway prefabricated box girders. The key indicator can refer to at least one of the primary indicators and / or secondary indicators. Key indicators usually involve the core functions and safety performance of railway prefabricated box girders. If the key indicators are not up to standard, the quality of railway prefabricated box girders may be seriously degraded, or even unusable or have safety hazards. Therefore, the main control items occupy an extremely important position in the performance evaluation of railway prefabricated box girders. The remaining indicators are other indicators except the main control items.

[0097] For the above indicators, this embodiment can obtain data from the following two aspects.

[0098] (1) Some public documents have confirmed the idea of ​​information-based and intelligent construction. Through the use of intelligent equipment in each construction process, combined with the use of the Internet of Things and big data information technology, the construction content such as the location of intelligent equipment used, the number of processes, the processing size, the quality of process completion, etc., is automatically uploaded to the platform and recorded, and the data can then be obtained from the platform;

[0099] (2) Obtain relevant data from documents on information storage, real-time monitoring, quality and safety management, etc. provided by the railway information management platform to ensure the accuracy and breadth of information sources.

[0100] This embodiment uses statistical analysis software to perform statistical analysis based on the historical data obtained by the above two methods, and then uses expert validity as the test indicator of content validity, using the Likert five-point scale shown in Table 7:

[0101] Table 7 Likert five-point scale

[0102]

[0103] Based on the structural equation model, the historical data were processed using the structural equation model processing software to obtain a standardized path diagram with path coefficients (such as Figure 4 Path diagram corresponding to the initial structural equation model shown).

[0104] S12. Calculate the goodness of fit of the initial structural equation model based on the preset goodness of fit index.

[0105] The preset goodness of fit indicators include at least one of the following: Minimum Chi-Square Statistic (CMIN), Degrees of Freedom (DF), P value (P), Root Mean Square Residual (RMR), Goodness-of-Fit Index (GFI), Adjusted Goodness-of-Fit Index (AGFI), Normed Fit Index (NFI), Relative Fit Index (RFI), Comparative Fit Index (CFI), and Root Mean Square Error of Approximation (RMSEA).

[0106] Exemplarily, the preset fit indicators used in the embodiments of the present application when calculating the fit of the initial structural equation model include CMIN, DF, P, RMR, GFI, AGFI, NFI, RFI, CFI and RMSEA.

[0107] S13. When the degree of fit of the initial structural equation model is greater than or equal to a first preset threshold, obtaining production data, test data, and evaluation information of intelligent process performance indicators of the railway prefabricated box girder.

[0108] It should be understood that the preset goodness of fit index can be one or more, so the goodness of fit of the initial structural equation model is also one or more. When there are multiple preset goodness of fit indexes, in this step, the goodness of fit of the initial structural equation model is greater than or equal to the first preset threshold value, which means that for each preset goodness of fit index, the goodness of fit of the initial structural equation model must be greater than or equal to the first preset threshold value corresponding to the index. The first preset threshold values ​​corresponding to different goodness of fit indexes may be the same or different, and this embodiment does not specifically limit this.

[0109] S14. Based on the production data, inspection data of the railway prefabricated box girder, and the evaluation information of the intelligent process performance indicators, the target structural equation model is used to evaluate the performance of the railway prefabricated box girder, and the performance evaluation results of the railway prefabricated box girder are obtained; wherein, the target structural equation model is obtained after adjusting the parameters of the initial structural equation model.

[0110] Optionally, the target structural equation model has the same paths as the initial structural equation model, but some path coefficients of the target structural equation model may be different from those of the initial structural equation model.

[0111] The embodiment of the present application introduces intelligent process performance indicators to construct a new quality and performance evaluation index system suitable for intelligent equipment, intelligent management, and intelligent construction, thereby improving the applicability of the initial structural equation model constructed based on the new quality and performance evaluation index system in the full-process intelligent environment of intelligent equipment, intelligent management, and intelligent construction, thereby ensuring the validity of the performance evaluation results.

[0112] In some possible embodiments, after S12, calculating the fit of the initial structural equation model based on a preset fit index, the method further includes:

[0113] Step 15: When the fit of the initial structural equation model is less than the first preset threshold, a genetic algorithm is used to iteratively adjust the structure or parameters of the initial structural equation model based on the fit until the fit of the initial structural equation model is greater than or equal to the first preset threshold.

[0114] In the process of adjustment, the structure or parameters of the initial structural equation model can also be iteratively adjusted in combination with the path coefficient and the goodness of fit. Among them, iteratively adjusting the structure or parameters of the initial structural equation model refers to iteratively adjusting the path or path coefficient, regression coefficient and other parameters of the initial structural equation model.

[0115] The embodiment of the present application can improve the fit of the initial structural equation model. Specifically, by adjusting the path coefficient, the relationship between variables can be more accurately reflected, so that the initial structural equation model can better fit the data. The optimization of the path coefficient helps to reduce the residual of the initial structural equation model and improve the overall fit of the initial structural equation model. According to the feedback of the fit index (such as CMIN, RMSEA, CFI, etc.), the unreasonable path or the missing path in the initial structural equation model can be identified and adjusted accordingly. For example, some paths can be added or deleted to improve the fit of the initial structural equation model.

[0116] In some possible embodiments, after S14, based on the production data, test data, and evaluation information of the intelligent process performance index of the railway prefabricated box girder, the target structural equation model is used to perform performance evaluation on the railway prefabricated box girder to obtain the performance evaluation result of the railway prefabricated box girder, the method further includes:

[0117] Step 16: Calculate the Customer Satisfaction Index (CSI) based on the performance evaluation results of the railway prefabricated box girder.

[0118] Exemplarily, the formula used to calculate the satisfaction index is: ;in, represents the customer satisfaction index, is the error term The expected or average value of represents the minimum value of the error term, ; represents the maximum value of the error term, ,in, is the weight of each explicit exogenous variable, is the measured score of each explicit exogenous variable. In some embodiments, n=5.

[0119] The customer satisfaction index reflects the customer's satisfaction with railway prefabricated box girders. Calculating the customer satisfaction index can prompt enterprises to make continuous improvements in product quality and performance. Based on the customer satisfaction index, enterprises can allocate resources more reasonably and give priority to solving key issues that affect customer satisfaction.

[0120] The production and testing links of railway prefabricated box girders may encounter various dynamic changes, such as seasonal effects, aging of intelligent equipment, process improvements, etc. The traditional static prior distribution may not be able to fully capture these dynamic changes, resulting in inaccurate model evaluation. To solve this problem, the embodiment of the present application constructs a target structural equation model based on a dynamic prior distribution or an adaptive prior distribution. In some optional embodiments, the parameters of the initial structural equation model are adjusted to obtain the target structural equation model, including:

[0121] Step 141: Obtain the evaluation opinion text of the domain experts on the intelligent process performance indicators.

[0122] Exemplarily, when the intelligent process performance indicators include the maturity of intelligent equipment application, quality indicator compliance, production efficiency improvement, intelligent monitoring and testing maturity, and testing indicator compliance, the evaluation opinion text of the intelligent process performance indicators includes: experts believe that the maturity of intelligent equipment application is between 0.7 and 0.9, historical data show that the mean quality indicator compliance is 0.85, and the standard deviation is 0.05; experts believe that the production efficiency improvement is between 10% and 20%; historical data show that the mean intelligent monitoring and testing maturity is 0.8, and the standard deviation is 0.1; experts believe that the testing indicator compliance is between 0.85 and 0.95.

[0123] Step 142: Determine the static prior distribution of the intelligent process performance indicator based on the evaluation opinion text.

[0124] Specifically, the evaluation opinions are used as prior knowledge, and a static prior distribution is generated based on the prior knowledge. For example, for the maturity of intelligent equipment application, the embodiment of the present application sets the prior distribution P(θ 智能装备运用成熟度 For example, a uniform distribution U(0.7,0.9) is selected. For the quality index compliance, the embodiment of the present application sets the prior distribution P(θ 质量指标符合度 ). For example, a normal distribution N(0.85,0.052) is selected. Regarding the improvement of production efficiency, the embodiment of the present application sets the prior distribution P(θ 生产效率提升度 ). For example, a uniform distribution U(0.1,0.2) is selected. For intelligent monitoring and detection maturity, the embodiment of the present application sets the prior distribution P(θ 智能监测及检测成熟度 For example, a normal distribution N(0.8,0.12) is selected. For the detection index compliance, the present application embodiment sets the prior distribution P(θ 检测指标符合度 ). For example, choose a uniform distribution U(0.85,0.95).

[0125] Step 143: Generate a non-static prior distribution of the intelligent process performance indicator based on the static prior distribution.

[0126] As an optional implementation method, the non-static prior distribution includes a dynamic prior distribution; step 143, generating a non-static prior distribution of intelligent process performance indicators based on the static prior distribution, includes: based on the static prior distribution, using a first generation method to generate a dynamic prior distribution of intelligent process performance indicators; the first generation method includes a state space model (State Space Model, SSM) construction method or a time series analysis method. Among them, the time series analysis method is an autoregressive integrated moving average model (Autoregressive Integrated Moving Average, ARIMA), or a long short-term memory network (Long Short-Term Memory, LSTM).

[0127] In other words, this embodiment can use the state space model to describe the dynamic changes of seasonal effects, intelligent equipment aging, process improvements, etc., and dynamically update the static prior distribution through methods such as Kalman Filter or Extended Kalman Filter. Alternatively, this embodiment can use time series analysis methods to model data in the form of time series and dynamically adjust the parameters of the static prior distribution.

[0128] Dynamic prior distribution can establish a prior distribution that can change over time based on time series data. This method can better reflect the uncertainty that changes over time and improve the adaptability and accuracy of the target structural equation model.

[0129] As another optional implementation method, the non-static prior distribution includes an adaptive prior distribution; step 143, generating a non-static prior distribution of intelligent process performance indicators based on the static prior distribution, includes: based on the static prior distribution, using a second generation method to generate an adaptive prior distribution of intelligent process performance indicators; the second generation method includes an online learning method or an adaptive algorithm.

[0130] In other words, this embodiment uses an online learning algorithm (such as online gradient descent, online Bayesian update, etc.) to achieve dynamic update of real-time data such as production data, test data, and evaluation information of intelligent process performance indicators of railway prefabricated box girders. Alternatively, this embodiment uses an adaptive algorithm (such as adaptive Kalman filtering, adaptive Bayesian update, etc.) to automatically adjust the static prior distribution according to real-time data to obtain an adaptive prior distribution.

[0131] The adaptive prior distribution can automatically adjust the static prior distribution based on the latest collected production data, inspection data of railway prefabricated box girders, and evaluation information of intelligent process performance indicators to better adapt to the actual situation.

[0132] Step 144: Calculate the posterior distribution based on the non-static prior distribution and the production data and test data of the railway prefabricated box girder and the evaluation information of the intelligent process performance index.

[0133] Step 145: Update the parameters of the initial structural equation model based on the parameter estimation in the posterior distribution to obtain the structural equation model to be determined.

[0134] Step 146: Calculate the goodness of fit of the structural equation model to be determined.

[0135] In step 146, this embodiment calculates the fitness of the structural equation model to be determined based on the preset fitness index. The description of the preset fitness index refers to the description in S12, which will not be repeated here.

[0136] In step 146, the preset goodness of fit index may be one or more, so the goodness of fit of the structural equation model to be determined is also one or more. When there are multiple preset goodness of fit indexes, in this step, the goodness of fit of the structural equation model to be determined is greater than or equal to the first preset threshold, which means that for each preset goodness of fit index, the goodness of fit of the structural equation model to be determined must be greater than or equal to the second preset threshold corresponding to the index. The second preset thresholds corresponding to different goodness of fit indexes may be the same or different, and this embodiment does not specifically limit this.

[0137] Step 147: When the degree of fit of the structural equation model to be determined is greater than or equal to a second preset threshold, the structural equation model to be determined is determined as the target structural equation model; and the second preset threshold is greater than or equal to the first preset threshold.

[0138] In this embodiment, the above process is a technical solution that combines Bayesian statistical methods and structural equation model optimization, which is used to optimize the quality control and performance evaluation in the production process of railway prefabricated box girders, and can improve the accuracy and reliability of the target structural equation model.

[0139] It should be noted that there may be a nonlinear relationship between implicit endogenous variables and implicit exogenous variables. For example, in the performance prediction of railway prefabricated box girders, a certain implicit endogenous variable (such as steel bar cutting and forming) may be affected by complex nonlinear factors such as the maturity of intelligent equipment application and the compliance of quality indicators. For example, the maturity of intelligent equipment application may have different effects on implicit endogenous variables at different levels, and the compliance of quality indicators may also have different effects on implicit endogenous variables at different levels.

[0140] In order to capture these relationships more accurately, the embodiment of the present application can establish a nonlinear structural equation model or a piecewise linear model. Among them, the nonlinear structural equation model allows the relationship between implicit endogenous variables and implicit exogenous variables to be nonlinear. This embodiment can be implemented by a variety of nonlinear regression methods, such as polynomial regression, neural network, etc. Therefore, in some optional embodiments, the target structural equation model includes a nonlinear structural equation model using a polynomial regression formula or a neural network expression.

[0141] The polynomial regression formula is: ;

[0142] in, Indicates steel bar cutting and forming, is the intercept, is the regression coefficient, It is the maturity of intelligent equipment application. is the quality indicator compliance, is the error term.

[0143] Among them, the neural network expression is: ;

[0144] in, is the activation function of the hidden layer, which can be a ReLU activation function or other types of activation functions. is the bias term, is the weight matrix, is the vector form of implicit exogenous variables.

[0145] In this embodiment, the use of higher-order polynomial regression can more accurately capture these complex nonlinear relationships. Alternatively, in the performance prediction of railway prefabricated box girders, a hidden endogenous variable may be subject to complex interactions of multiple factors. For example, there may be complex nonlinear interaction effects between the maturity of intelligent equipment application and the compliance of quality indicators. The use of deeper neural networks can better capture these complex interactions and improve the accuracy of the performance evaluation of the target structural equation model.

[0146] It should be understood that in addition to using a nonlinear structural equation model, this embodiment can also use a piecewise linear model to more accurately capture the linear relationship at different stages and improve the applicability and accuracy of the target structural equation model. The piecewise linear model divides the range of the implicit exogenous variable into multiple intervals and uses different linear relationships in each interval to capture the nonlinear relationship. Therefore, in some optional embodiments, the target structural equation model includes a piecewise linear structural equation model.

[0147] Exemplary, steel bar cutting and forming The piecewise linear structural equation model expression is: ;in, is the slope of the different intervals, is the intercept for different intervals, It is a segmentation point, used to divide different intervals.

[0148] The piecewise linear model captures nonlinear relationships by dividing the range of the implicit exogenous variable into multiple intervals and using different linear relationships in each interval.

[0149] Therefore, this embodiment can more accurately capture the nonlinear relationship between latent variables by introducing a piecewise linear model. The target structural equation model can better reflect the complexity of the production and detection links of railway prefabricated box girders, and improve the prediction and evaluation accuracy of the target structural equation model.

[0150] Figure 5 A schematic diagram of a performance evaluation system for a railway prefabricated box girder provided in an embodiment of the present application is shown in FIG. Figure 5 As shown, the system includes:

[0151] The construction module 51 is used to construct an improved quality and performance evaluation index system by introducing intelligent process performance indicators, and to construct an initial structural equation model corresponding to the railway prefabricated box girder based on the improved quality and performance evaluation index system.

[0152] The calculation module 52 is used to calculate the fitness of the initial structural equation model based on a preset fitness index.

[0153] The acquisition module 53 is used to acquire the production data, test data and evaluation information of the intelligent process performance index of the railway prefabricated box girder when the fitting degree of the initial structural equation model is greater than or equal to the first preset threshold.

[0154] The performance evaluation module 54 is used to evaluate the performance of the railway prefabricated box girder based on the production data, test data and evaluation information of the intelligent process performance indicators of the railway prefabricated box girder using the target structural equation model to obtain the performance evaluation results of the railway prefabricated box girder; wherein the target structural equation model is obtained after adjusting the parameters of the initial structural equation model.

[0155] Figure 5 The performance evaluation system for railway prefabricated box girders can be performed Figure 1 The performance evaluation method for railway prefabricated box girders described in the illustrated embodiment, its implementation principle and technical effect are not described in detail. The specific manner in which each module and unit performs operations in the performance evaluation system for railway prefabricated box girders in the above embodiment has been described in detail in the embodiment of the method, and will not be described in detail here.

[0156] In one possible design, Figure 5 The performance evaluation system for railway prefabricated box girders of the embodiment shown can be implemented as a computing device, such as Figure 6 As shown, the computing device may include a storage component 61 and a processing component 62 .

[0157] The storage component 61 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 62 .

[0158] The processing component 62 is used to: construct an improved quality and performance evaluation index system by introducing intelligent process performance indicators, and construct an initial structural equation model corresponding to the railway prefabricated box girder based on the improved quality and performance evaluation index system; calculate the fitness of the initial structural equation model based on a preset fitness index; when the fitness of the initial structural equation model is greater than or equal to a first preset threshold, obtain the production data, inspection data, and evaluation information of the intelligent process performance indicators of the railway prefabricated box girder; based on the production data, inspection data, and evaluation information of the intelligent process performance indicators of the railway prefabricated box girder, use the target structural equation model to evaluate the performance of the railway prefabricated box girder, and obtain the performance evaluation result of the railway prefabricated box girder; wherein, the target structural equation model is obtained after adjusting the parameters of the initial structural equation model.

[0159] The processing component 62 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0160] The storage component 61 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0161] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0162] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0163] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0164] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0165] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The performance evaluation method of the railway prefabricated box girder of the illustrated embodiment.

[0166] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0167] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0168] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0169] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A performance evaluation method for railway prefabricated box girders, characterized in that: include: An improved quality and performance evaluation index system is constructed by introducing intelligent process performance indicators, and an initial structural equation model corresponding to the railway prefabricated box girder is constructed based on the improved quality and performance evaluation index system; The intelligent process performance indicators include: intelligent production performance indicators and intelligent detection performance indicators; intelligent production performance indicators include at least one of the following: intelligent equipment application maturity, quality indicator compliance or production efficiency improvement; intelligent detection performance indicators include at least one of the following: intelligent monitoring and detection maturity, detection indicator compliance; The goodness of fit of the initial structural equation model was calculated based on the preset goodness of fit indices; When the degree of fit of the initial structural equation model is greater than or equal to a first preset threshold, obtaining production data, test data, and evaluation information of intelligent process performance indicators of railway prefabricated box girders; Based on the production data, test data and evaluation information of intelligent process performance indicators of railway prefabricated box girders, the target structural equation model is used to evaluate the performance of railway prefabricated box girders, and the performance evaluation results of railway prefabricated box girders are obtained; wherein the target structural equation model is obtained after adjusting the parameters of the initial structural equation model; The parameters of the initial structural equation model were adjusted to obtain the target structural equation model, including: Obtain the evaluation opinions of domain experts on the performance indicators of intelligent processes; Determine a static prior distribution of intelligent process performance indicators based on the evaluation opinion text; Generating a non-static prior distribution of intelligent process performance indicators based on the static prior distribution, wherein the non-static prior distribution includes a dynamic prior distribution or an adaptive prior distribution; Calculating the posterior distribution based on the non-static prior distribution and the production data and test data of the railway prefabricated box girder and the evaluation information of the intelligent process performance index; The parameters of the initial structural equation model are updated based on the parameter estimation in the posterior distribution to obtain the structural equation model to be determined; Calculate the goodness of fit of the structural equation model to be determined; When the degree of fit of the structural equation model to be determined is greater than or equal to a second preset threshold, the structural equation model to be determined is determined as a target structural equation model; and the second preset threshold is greater than or equal to the first preset threshold.

2. The method according to claim 1, characterized in that The non-static prior distribution includes a dynamic prior distribution; The generating of the non-static prior distribution of the intelligent process performance indicator based on the static prior distribution includes: Based on the static prior distribution, a first generation method is used to generate a dynamic prior distribution of intelligent process performance indicators; the first generation method includes a state space model construction method or a time series analysis method.

3. The method according to claim 1, characterized in that The non-static prior distribution includes an adaptive prior distribution; The generating of the non-static prior distribution of the intelligent process performance indicator based on the static prior distribution includes: Based on the static prior distribution, a second generation method is used to generate an adaptive prior distribution of intelligent process performance indicators; the second generation method includes an online learning method or an adaptive algorithm.

4. The method according to claim 1, characterized in that: The target structural equation model includes a nonlinear structural equation model using a polynomial regression formula or a neural network expression.

5. The method according to claim 1, characterized in that The target structural equation model includes a piecewise linear structural equation model.

6. The method according to claim 1, characterized in that After calculating the fit of the initial structural equation model based on the preset fit index, the method further includes: When the fit of the structural equation model is less than a first preset threshold, a genetic algorithm is used to iteratively adjust the structure or parameters of the initial structural equation model based on the fit until the fit of the initial structural equation model is greater than or equal to the first preset threshold.

7. The method according to claim 1, characterized in that After the performance evaluation result of the railway prefabricated box girder is obtained by using the target structural equation model to evaluate the performance of the railway prefabricated box girder based on the production data, test data, and evaluation information of the intelligent process performance index of the railway prefabricated box girder, the method further includes: A customer satisfaction index is calculated based on the performance evaluation results of the railway prefabricated box girder.

8. A performance evaluation system for railway prefabricated box girders, characterized in that: include: A construction module is used to construct an improved quality and performance evaluation index system by introducing intelligent process performance indicators, and to construct an initial structural equation model corresponding to the railway prefabricated box girder based on the improved quality and performance evaluation index system; The intelligent process performance indicators include: intelligent production performance indicators and intelligent detection performance indicators; intelligent production performance indicators include at least one of the following: intelligent equipment application maturity, quality indicator compliance or production efficiency improvement; intelligent detection performance indicators include at least one of the following: intelligent monitoring and detection maturity, detection indicator compliance; A calculation module, used for calculating the goodness of fit of the initial structural equation model based on a preset goodness of fit index; An acquisition module, used for acquiring production data, test data, and evaluation information of intelligent process performance indicators of railway prefabricated box girders when the fitting degree of the initial structural equation model is greater than or equal to a first preset threshold value; A performance evaluation module is used to evaluate the performance of the railway prefabricated box girder based on the production data, test data, and evaluation information of the intelligent process performance indicators of the railway prefabricated box girder using a target structural equation model to obtain a performance evaluation result of the railway prefabricated box girder; wherein the target structural equation model is obtained after adjusting the parameters of the initial structural equation model; The parameters of the initial structural equation model were adjusted to obtain the target structural equation model, including: Obtain the evaluation opinions of domain experts on the performance indicators of intelligent processes; Determine a static prior distribution of intelligent process performance indicators based on the evaluation opinion text; Generating a non-static prior distribution of intelligent process performance indicators based on the static prior distribution, wherein the non-static prior distribution includes a dynamic prior distribution or an adaptive prior distribution; Calculating the posterior distribution based on the non-static prior distribution and the production data and test data of the railway prefabricated box girder and the evaluation information of the intelligent process performance index; The parameters of the initial structural equation model are updated based on the parameter estimation in the posterior distribution to obtain the structural equation model to be determined; Calculate the goodness of fit of the structural equation model to be determined; When the degree of fit of the structural equation model to be determined is greater than or equal to a second preset threshold, the structural equation model to be determined is determined as a target structural equation model; and the second preset threshold is greater than or equal to the first preset threshold.

Images