Scenarized modeling method and system for operation and maintenance service

Abstract

The invention discloses a scene modeling method and system for operation and maintenance service, and relates to the field of operation and maintenance service, and the method comprises the steps: collecting operation and maintenance environment real-time data, carrying out the preprocessing of the operation and maintenance environment real-time data, and constructing a scene library; forming a knowledge graph based on expert experience and historical data, writing the knowledge graph into a scene library, generating a dynamic scene adaptive model, and performing dynamic scene adaptive model training; the operation and maintenance efficiency and cost are evaluated based on a dynamic scene adaptive model training result, potential problems of the model are analyzed and optimized, the model can continuously learn and improve through analysis and optimization of the potential problems of the model, the prediction accuracy and reliability are improved, the operation and maintenance cost is further reduced, and the operation and maintenance efficiency is improved. And the overall operation and maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of operation and maintenance services, and in particular to a scenario-based modeling method and system for operation and maintenance services. Background Art

[0002] With the rapid development of modern industry and information technology, operation and maintenance services are gradually shifting from traditional manual operations to intelligent and automated management. Traditional operation and maintenance methods usually rely on manual inspections and regular maintenance. Although this method can ensure the normal operation of equipment to a certain extent, it also has many limitations. With the continuous advancement of sensor technology, the Internet of Things (IoT), big data analysis, and artificial intelligence (AI), the management method of operation and maintenance services is undergoing profound changes. By collecting real-time data on equipment and the environment, combining expert experience and historical data, and building an intelligent operation and maintenance model, the efficiency and accuracy of operation and maintenance can be greatly improved, operating costs can be reduced, and the reliability and safety of equipment can be improved.

[0003] At present, many enterprises have begun to adopt intelligent operation and maintenance systems based on the Internet of Things and big data technologies. These systems usually include modules such as data acquisition, data preprocessing, fault diagnosis and prediction, and intelligent scheduling, which can realize real-time monitoring and management of equipment operation status. However, the existing intelligent operation and maintenance systems still have some shortcomings in practical applications. First, many systems lack unified standards and specifications in the data acquisition and preprocessing stages, resulting in uneven data quality and affecting the accuracy of subsequent analysis. Secondly, the existing fault diagnosis and prediction models are often based on a single data source, lack the ability to comprehensively analyze multi-source heterogeneous data, and have difficulty in accurately identifying complex fault modes. In addition, most of the existing systems are static models, lacking the ability to adapt to dynamic changes in the actual operation and maintenance environment, resulting in poor performance in complex and changeable operation and maintenance scenarios. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present invention is proposed.

[0005] Therefore, the existing technical problem to be solved by the present invention is that the existing fault diagnosis and prediction models are often based on a single data source, lack the ability to comprehensively analyze multi-source heterogeneous data, and are difficult to accurately identify complex fault models.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a scenario modeling method for operation and maintenance service, which includes collecting real-time data of operation and maintenance environment, preprocessing the real-time data of operation and maintenance environment and building a scenario library;

[0007] Form a knowledge graph based on expert experience and historical data, and write the knowledge graph into the scenario library;

[0008] Generate a dynamic scene adaptation model based on the scene library and perform dynamic scene adaptation model training;

[0009] The operation and maintenance efficiency and cost are evaluated based on the training results of the dynamic scene adaptation model, and the dynamic scene adaptation model is optimized through feedback.

[0010] As a preferred solution of the scenario-based modeling method and system for operation and maintenance services described in the present invention, the real-time data of the operation and maintenance environment collected includes equipment vibration spectrum data, thermal imaging data, gas composition analysis data, electromagnetic field strength data, process parameter correlation data, and human-computer interaction behavior data;

[0011] The preprocessing of real-time data includes cleaning the real-time data, removing noise and outliers, and standardizing and normalizing the data.

[0012] As a preferred solution of the scenario-based modeling method and system for operation and maintenance services described in the present invention, the knowledge graph formed based on expert experience and historical data includes:

[0013] Generation of knowledge nodes: Identify and extract key knowledge points related to equipment failure modes and maintenance strategies based on expert experience and historical data;

[0014] Definition of knowledge relations: According to the logical relations and relevance between knowledge nodes, the connection relations between knowledge nodes are established to form a knowledge network;

[0015] Construction of knowledge graph: Visualize knowledge nodes and their relationships to form a structured knowledge graph and store it in the scenario library.

[0016] As a preferred solution of the scenario-based modeling method and system for operation and maintenance services described in the present invention, the writing of the knowledge graph into the scenario library and the generation of a dynamic scenario adaptive model are expressed as:

[0017]

[0018] Among them, K represents the knowledge graph, which consists of several knowledge nodes k i D represents the dataset in the scene library, which consists of several data points d j Composition, S represents a dynamic scene set, each scene is s k , M adapt Represented as a dynamic scene adaptation model, N k is the total number of knowledge nodes; N d is the total number of data points; N s is the total number of scenes, k i (t) represents the knowledge node k at time t i The value of dj (t) represents the data point d at time t j The value of s k (t) represents the scene s at time t k The value of , F, G, H: represent the complex information processing and filtering functions for knowledge nodes, data points and scenarios respectively, β is the control parameter, regulating the characteristics of the function, and γ is the regularization parameter;

[0019] Functions F, G, and H represent the complex information processing and filtering of knowledge nodes, data points, and scenarios respectively. The specific expressions are as follows:

[0020]

[0021] Among them, k i (t) represents the knowledge node k at time t i The value of Represents knowledge node k i The value at the nth moment, represents the control parameter λ n An exponential function used to adjust k i (t) is the functional characteristic, Indicates the k i (t) 2 The integral of , a, b represent the upper and lower limits of the integral interval;

[0022]

[0023] Among them, d j (t) represents the data point d at time t j , the value of Represents data point d j The value at the mth moment, ln represents the natural logarithmic function, Represents data point d j A complex summation function of the control parameter θ p The exponential function, θ p Represents the control parameter, used to adjust d j (t) is the function characteristic, P represents the summation upper limit;

[0024]

[0025] where s k (t) represents the scene s at time t k The value of e -μt represents the exponential function of the control parameter μ, which is used to adjust s k (t) is the functional characteristic, Indicates that s is calculated on the interval [0,T] kThe weighted integral of (t), Indicates scenes k The complex summation function, Q represents the summation upper limit;

[0026]

[0027] Where R is the upper limit of the sum of regularization terms, m r are model parameters, is the square term of the model parameter, and contains the model parameter m r The exponential function of . As a preferred solution of the scenario-based modeling method and system for operation and maintenance services described in the present invention, the output value of the evaluation of operation and maintenance efficiency and cost based on the model training result is M adapt ,

[0028] When M adapt When the value is greater than the preset threshold, it indicates that the model has good adaptability to the current scenario, knowledge nodes, and scenario library data;

[0029] When M adapt When the value is less than the preset threshold, it means that the model is less adaptable to the current scenario, knowledge nodes, and scenario library data.

[0030] As a preferred solution of the scenario-based modeling method and system for operation and maintenance services described in the present invention, the formula for evaluating the operation and maintenance efficiency and cost based on the training results of the dynamic scenario adaptive model is established:

[0031]

[0032] Among them, E OM is the comprehensive evaluation result of operation and maintenance efficiency and cost, η eff Represents the operation and maintenance efficiency, which changes with time t. op represents the operation and maintenance cost, which changes with time t, T is the upper limit of the time integral, F(η eff (t)) is the operation and maintenance efficiency filtering function at time t, G(C op (t)) is the operation and maintenance cost filtering function at time t, γ is the regularization parameter, R(M) is the regularization term used to prevent overfitting, a, b are the upper and lower limits of the integral interval, N, P, R are the corresponding summation upper limits, and the operation and maintenance efficiency and cost evaluation results E OM The range of is the real number set R;

[0033] When E OM When the value is greater than the preset threshold, it indicates high O&M efficiency and low cost, indicating excellent overall performance;

[0034] When E OMWhen the value is greater than the preset threshold, it indicates low O&M efficiency and high cost, indicating poor overall performance.

[0035] As a preferred solution of the scenario-based modeling method for operation and maintenance services described in the present invention, the analysis of potential model problems and optimization includes analyzing equipment vibration spectrum data, thermal imaging data, gas composition analysis data, electromagnetic field intensity data, process parameter correlation data, and human-computer interaction behavior data based on the dynamic scenario adaptive model training results and updating the scenario library data.

[0036] The scenario-based modeling system for operation and maintenance services is characterized by: comprising a data collection module, a scenario recognition module, and a model application module;

[0037] As a preferred solution of the scenario-based modeling system for operation and maintenance services described in the present invention, the data collection module is used to collect real-time data of the operation and maintenance environment, perform preprocessing, and store the data in the system to provide basic data for subsequent scenario recognition and model application;

[0038] The scene recognition module identifies and defines different operation and maintenance scenarios based on the collected data and knowledge graph, builds a scene library, and generates a dynamic scene adaptive model;

[0039] The model application module applies a dynamic scenario adaptive model to make operation and maintenance decisions and optimization, thereby improving operation and maintenance efficiency and reducing operation and maintenance costs.

[0040] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the scenario-based modeling method for operation and maintenance services described in any one of claims 1 to 7 when executing the computer program.

[0041] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a scenario-based modeling method for operation and maintenance services.

[0042] The beneficial effects of the present invention are: expert knowledge and historical data are converted into structured knowledge graphs to provide guidance for the model. The model's intelligence and self-adaptation capabilities are enhanced, so that the system can automatically adjust the operation and maintenance strategy according to actual conditions, improving the flexibility and response speed of the system, thereby more accurately predicting and handling equipment failures and improving operation and maintenance efficiency;

[0043] By evaluating the operation and maintenance efficiency and cost based on the dynamic scenario adaptive model training results, analyzing the potential problems of the model and optimizing it, the model can continuously learn and improve, improve the accuracy and reliability of the prediction, thereby further reducing the operation and maintenance costs and improving the overall operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 It is a flow chart of the present invention. Figure 2 This is a system module diagram of the present invention. DETAILED DESCRIPTION

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings.

[0047] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0048] Example 1

[0049] Reference Figure 1 , Figure 2 , which is the first embodiment of the present invention, and provides a scenario-based modeling method for operation and maintenance services, which includes the following steps:

[0050] S1: Collect real-time data of the operation and maintenance environment, pre-process the real-time data of the operation and maintenance environment and build a scenario library;

[0051] First, real-time data on the operation and maintenance environment is collected. These data include equipment vibration spectrum data, thermal imaging data, gas composition analysis data, electromagnetic field strength data, process parameter correlation data, and human-computer interaction behavior data. Through the real-time collection of these data, real-time information on the equipment's operating status can be provided, which helps to promptly detect and prevent potential faults and improve operation and maintenance efficiency.

[0052] Next, the collected real-time data is preprocessed. Data preprocessing includes cleaning data, removing noise and outliers, and standardizing and normalizing the data. Through these steps, the quality and consistency of the data can be ensured, providing a reliable data foundation for subsequent feature extraction and model training. A high-quality data foundation is the key to building and optimizing models, and can significantly enhance the stability and accuracy of the model.

[0053] S2: Form a knowledge graph based on expert experience and historical data, write the knowledge graph into the scene library, generate a dynamic scene adaptation model, and train the dynamic scene adaptation model;

[0054] After data preprocessing, a knowledge graph is established based on expert experience and historical data. A knowledge graph is a graphical structure used to represent domain knowledge, including nodes and edges. Nodes represent concepts or entities, and edges represent the relationship between concepts. By combining expert experience and historical data, a knowledge graph that reflects the operation and maintenance environment and equipment status is generated, which facilitates the storage, query and use of information. The establishment of a knowledge graph provides a structured knowledge representation for generating dynamic scene adaptive models, supporting the processing and decision-making of complex information.

[0055] Then, a dynamic scenario adaptive model is generated and trained. Based on the knowledge graph and real-time data, the model can dynamically adjust and optimize the operation and maintenance strategy through continuous training and updating to adapt to different operation and maintenance scenarios. The generation and training of the dynamic scenario adaptive model is the core step to improve operation and maintenance efficiency and reduce costs. Through model training, operation and maintenance decisions are continuously optimized, the flexibility and adaptability of the model are improved, and the overall performance of the system is ultimately improved.

[0056] Writing the knowledge graph into the scene library and generating a dynamic scene adaptation model is expressed as:

[0057]

[0058] Among them, K represents the knowledge graph, which consists of several knowledge nodes k i D represents the dataset in the scene library, which consists of several data points d j Composition, S represents a dynamic scene set, each scene is s k , M adapt Represented as a dynamic scene adaptation model, N k is the total number of knowledge nodes; N d is the total number of data points; N s is the total number of scenes, k i (t) represents the knowledge node k at time t i The value of d j (t) represents the data point d at time t j The value of s k (t) represents the scene s at time t k The value of , F, G, H: represent the complex information processing and filtering functions for knowledge nodes, data points and scenarios respectively, β is the control parameter, regulating the characteristics of the function, and γ is the regularization parameter;

[0059] Functions F, G, and H represent the complex information processing and filtering of knowledge nodes, data points, and scenarios respectively. The specific expressions are as follows:

[0060]

[0061] Among them, k i (t) represents the knowledge node k at time t i The value of Represents knowledge node k i The value at the nth moment, represents the control parameter λ n An exponential function used to adjust k i (t) is the functional characteristic, Indicates the k i (t) 2 The integral of , a, b represent the upper and lower limits of the integral interval;

[0062]

[0063] Among them, d j (t) represents the data point d at time t j , the value of Represents data point d j The value at the mth moment, ln represents the natural logarithmic function, Represents data point d j A complex summation function of the control parameter θ p The exponential function, θ p Represents the control parameter, used to adjust d j (t) is the function characteristic, P represents the summation upper limit;

[0064]

[0065] where s k (t) represents the scene s at time t k The value of e -μt represents the exponential function of the control parameter μ, which is used to adjust s k (t) is the functional characteristic, Indicates that s is calculated on the interval [0,T] k The weighted integral of (t), Indicates scenes k The complex summation function, Q represents the summation upper limit;

[0066]

[0067] Where R is the upper limit of the sum of regularization terms, m r are model parameters, is the square term of the model parameter, and contains the model parameter m r The exponential function of .

[0068] The output value of evaluating operation and maintenance efficiency and cost based on model training results is M adapt ,

[0069] When M adapt When the value is greater than the preset threshold, it indicates that the model has good adaptability to the current scenario, knowledge nodes, and scenario library data;

[0070] When M adapt When the value is less than the preset threshold, it means that the model is poorly adaptable to the current scenario, knowledge nodes, and scenario library data.

[0071] S3: Evaluate the operation and maintenance efficiency and cost based on the dynamic scenario adaptive model training results, analyze the potential problems of the model and optimize it;

[0072] The operation and maintenance efficiency and cost are evaluated based on the dynamic scenario adaptive model training results, which can be expressed as:

[0073]

[0074] Among them, E OM is the comprehensive evaluation result of operation and maintenance efficiency and cost, η eff Represents the operation and maintenance efficiency, which changes with time t. op represents the operation and maintenance cost, which changes with time t, T is the upper limit of the time integral, F(η eff (t)) is the operation and maintenance efficiency filtering function at time t, G(C op (t)) is the operation and maintenance cost filtering function at time t, γ is the regularization parameter, R(M) is the regularization term used to prevent overfitting, a, b are the upper and lower limits of the integral interval, N, P, R are the corresponding summation upper limits, and the operation and maintenance efficiency and cost evaluation results E OM The range of is the real number set R;

[0075] When E OM When the value is greater than the preset threshold, it indicates high O&M efficiency and low cost, indicating excellent overall performance;

[0076] When E OM When the value is greater than the preset threshold, it indicates low O&M efficiency and high cost, indicating poor overall performance.

[0077] Set a threshold θ for the comprehensive evaluation result of operation and maintenance efficiency and cost (EOM), according to E OM The performance of the model is judged by the value. Assuming the threshold is set to θ = 0.75,

[0078] Case 1: EOM>0.75

[0079] Note: When the EOM value is greater than 0.75, it means that the operation and maintenance efficiency is high and the cost is low. The model has good adaptability to the current scenario, knowledge nodes, and scenario library data, and has excellent overall performance.

[0080] Feedback optimization method:

[0081] Model retention strategy: Maintain the current model parameters and strategies, and record the current parameter configuration for subsequent reference.

[0082] Data update: Continue to collect real-time data and regularly update the scenario library and knowledge graph to ensure that the model always makes decisions based on the latest data.

[0083] Preventive maintenance: On the basis of efficient operation, further preventive maintenance measures are implemented to prevent potential problems from occurring.

[0084] Case 2: 0.50≤EOM≤0.75

[0085] Note: When the EOM value is between 0.50 and 0.75, it means that the operation and maintenance efficiency and cost are at a medium level, the model adaptability is average, and moderate optimization is required.

[0086] Feedback optimization method:

[0087] Parameter adjustment: Fine-tune the parameters of the dynamic scene adaptation model, especially the control parameter β\betaβ and the regularization parameter γ\gammaγ, to improve the adaptability of the model.

[0088] Feature selection: Further feature selection is performed on the preprocessed data to remove redundant features and select the features that have the greatest impact on operation and maintenance efficiency and cost for key analysis.

[0089] Expert participation: Incorporate expert experience, analyze the scenarios identified in the model, and add expert knowledge graph nodes to improve the decision-making quality of the model.

[0090] Case 3: EOM < 0.50

[0091] Note: When the EOM value is less than 0.50, it means that the operation and maintenance efficiency is low and the cost is high. The model has poor adaptability to the current scenario, knowledge nodes, and scenario library data, and the overall performance is poor.

[0092] Feedback optimization method:

[0093] Model retraining: Retrain the dynamic scenario adaptive model based on the latest real-time data and knowledge graph of the operation and maintenance environment, and adjust the model structure and parameters.

[0094] Anomaly Detection: Perform detailed anomaly detection on historical and current data to identify data points and scenarios that may cause the model to perform poorly.

[0095] Strategy adjustment: Based on the results of model feedback, adjust the operation and maintenance strategy and implement more proactive maintenance and optimization measures, such as increasing the maintenance frequency or replacing inefficient equipment.

[0096] Finally, based on the training results of the dynamic scenario adaptive model, the operation and maintenance efficiency and cost are evaluated. By evaluating the training results of the model, potential problems of the model can be identified and solved, and the model performance can be continuously optimized. Model evaluation and optimization ensure the effectiveness and accuracy of the model in practical applications, improve the operation and maintenance efficiency and cost-effectiveness, and promote the intelligence and efficiency of operation and maintenance business.

[0097] In summary, this scenario-based modeling method for operation and maintenance services collects and preprocesses real-time data of the operation and maintenance environment, establishes a knowledge graph, generates and trains dynamic scenario adaptive models, and finally performs model evaluation and optimization, which significantly improves operation and maintenance efficiency and reduces operation and maintenance costs, and has important practical application value.

[0098] Collect real-time data of the operation and maintenance environment, including equipment vibration spectrum data, thermal imaging data, gas composition analysis data, electromagnetic field strength data, process parameter correlation data, and human-computer interaction behavior data.

[0099] Clean the real-time data, remove noise and outliers, standardize and normalize the data, extract and select features from the preprocessed data and store them in the scene library.

[0100] Analyze potential model problems and optimize them, including analyzing equipment vibration spectrum data, thermal imaging data, gas composition analysis data, electromagnetic field intensity data, process parameter correlation data, and human-computer interaction behavior data based on the dynamic scenario adaptive model training results and updating the scenario library data.

[0101] Example 2

[0102] Reference Figure 1 , which is the second embodiment of the present invention. This embodiment is based on the previous embodiment, and the implementation process is as follows:

[0103] Construction of scene library:

[0104] Based on expert experience and historical data, a knowledge graph of operation and maintenance is formed, covering equipment failure modes, maintenance strategies and historical operation and maintenance records.

[0105] Write the knowledge graph into the scene library and generate a dynamic scene adaptation model.

[0106] Generation and training of dynamic scene adaptation models:

[0107] Use the preprocessed real-time data to train the dynamic scenario adaptive model and adjust the model parameters to optimize its adaptability to different operation and maintenance scenarios.

[0108] During the training process, supervised learning methods are used to optimize the model using fault records and maintenance effects in historical data.

[0109] Evaluation of operation and maintenance efficiency and cost:

[0110] Based on the output of the dynamic scenario adaptive model, the operation and maintenance efficiency (ηeff) and operation and maintenance cost (Cop) of each device are evaluated in real time.

[0111] Use the formula:

[0112]

[0113] The evaluation results are used to analyze potential problems of the model and continuously optimize the model to ensure its effectiveness and accuracy in practical applications.

[0114] Optimization process:

[0115] Through detailed analysis of equipment vibration spectrum data, thermal imaging data, gas composition analysis data, etc., potential abnormal patterns are discovered and the data in the scenario library is updated.

[0116] Adjust model parameters to improve the model's adaptability to new scenarios and enhance the model's prediction accuracy and decision-making effect in actual operation and maintenance.

[0117] Table 1 Experimental data comparison

[0118]

[0119] Through the above test data, it can be seen that the proposed scenario modeling method for operation and maintenance has significant advantages in operation and maintenance efficiency and cost. First of all, from the operation and maintenance efficiency (ηeff) in the table, the operation and maintenance efficiency of each device is high, especially the operation and maintenance efficiency of motor A1 and transformer A5, which reached 0.85 and 0.82 respectively, which shows that the model has good adaptability in the scenarios of these devices.

[0120] In terms of operation and maintenance cost (Cop), the method of the present invention also shows good results. The table shows that the operation and maintenance costs of pump A2 and transformer A5 are relatively low, 180 and 210 respectively. This shows that the operation and maintenance costs can be effectively reduced through the optimization of the dynamic scene adaptive model.

[0121] In contrast, the traditional operation and maintenance methods usually perform poorly in terms of operation and maintenance efficiency and cost under the same conditions. Assuming that under the same experimental conditions, the average operation and maintenance efficiency (ηeff) of the traditional method is 0.70, and the average operation and maintenance cost (Cop) is 250. Obviously, through the proposed method, the operation and maintenance efficiency is improved by about 15% on average, while the operation and maintenance cost is reduced by about 16% on average.

[0122] In addition, from the specific data, for example, after using the proposed method, the operation and maintenance efficiency of compressor A3 reached 0.75, while the operation and maintenance cost was 250. Compared with the traditional method, the operation and maintenance efficiency was increased by about 7% and the operation and maintenance cost was reduced by about 10%. This improvement is of great significance for improving the reliability of equipment operation and reducing the total operation and maintenance cost.

[0123] In summary, through the comparative analysis of table data, the proposed scenario modeling method for operation and maintenance business has shown significant advantages in operation and maintenance efficiency and cost control. This method can not only improve operation and maintenance efficiency and reduce operation and maintenance costs, but also provide more accurate decision support through dynamic analysis of real-time data and continuous optimization of models, reflecting its innovation and practicality in practical applications.

[0124] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the invention, which should be included in the scope of the claims of the present invention.

Claims

1. A scenario-based modeling method for operation and maintenance services, characterized in that: include, Collect real-time data of the operation and maintenance environment, pre-process the real-time data of the operation and maintenance environment and build a scenario library; Form a knowledge graph based on expert experience and historical data, and write the knowledge graph into the scenario library; Generate a dynamic scene adaptation model based on the scene library and perform dynamic scene adaptation model training; The operation and maintenance efficiency and cost are evaluated based on the training results of the dynamic scene adaptation model, and the dynamic scene adaptation model is optimized through feedback.

2. A scenario-based modeling method for operation and maintenance services according to claim 1, characterized in that: The real-time data collected for the operation and maintenance environment includes equipment vibration spectrum data, thermal imaging data, gas composition analysis data, electromagnetic field strength data, process parameter correlation data, and human-computer interaction behavior data; The preprocessing of real-time data includes cleaning the real-time data, removing noise and outliers, and standardizing and normalizing the data.

3. A scenario-based modeling method for operation and maintenance services as claimed in claim 2, characterized in that: The formation of a knowledge graph based on expert experience and historical data includes: Generation of knowledge nodes: Identify and extract key knowledge points related to equipment failure modes and maintenance strategies based on expert experience and historical data; Definition of knowledge relations: According to the logical relations and relevance between knowledge nodes, the connection relations between knowledge nodes are established to form a knowledge network; Construction of knowledge graph: Visualize knowledge nodes and their relationships to form a structured knowledge graph and store it in the scenario library.

4. A scenario-based modeling method for operation and maintenance services as claimed in claim 3, characterized in that: The generation of a dynamic scene adaptive model based on a scene library is expressed as: Among them, K represents the knowledge graph, which consists of multiple knowledge nodes k i D represents the dataset in the scene library, which consists of multiple data points d j Composition, S represents a dynamic scene set, each scene is s k , M adapt Represented as a dynamic scene adaptation model, N k is the total number of knowledge nodes; N d is the total number of data points; N s is the total number of scenes, k i (t) represents the knowledge node k at time t i The value of d j (t) represents the data point d at time t j The value of s k (t) represents the scene s at time t k The value of , F, G, H: represent the complex information processing and filtering functions for knowledge nodes, data points and scenarios respectively, β is the control parameter, regulating the characteristics of the function, and γ is the regularization parameter; Functions F, G, and H represent the complex information processing and filtering of knowledge nodes, data points, and scenarios respectively. The specific expressions are as follows: Among them, k i (t) represents the knowledge node k at time t i The value of Represents knowledge node k i The value at the nth moment, represents the control parameter λ n An exponential function used to adjust k i (t) is the functional characteristic, Indicates the k i (t) 2 The integral of , a, b represent the upper and lower limits of the integral interval; Among them, d j (t) represents the data point d at time t j , the value of d jm Represents data point d j The value at the mth moment, ln represents the natural logarithmic function, Represents data point d j A complex summation function of the control parameter θ p The exponential function, θ p Represents the control parameter, used to adjust d j (t) is the function characteristic, P represents the summation upper limit; where s k (t) represents the scene s at time t k The value of e -μt represents the exponential function of the control parameter μ, which is used to adjust s k (t) is the functional characteristic, Indicates that s is calculated on the interval [0,T] k The weighted integral of (t), Indicates scenes k Complex summation function R , Q represents the upper limit of the summation; Where R is the upper limit of the sum of regularization terms, m r are model parameters, is the square term of the model parameter, is the model parameter m r The exponential function of .

5. A scenario-based modeling method for operation and maintenance services according to claim 4, characterized in that: The output value of evaluating the operation and maintenance efficiency and cost based on the model training result is M adapt , When M adapt When the value is greater than the preset threshold, it indicates that the model is adaptively adapted to the current scenario, knowledge nodes, and scenario library data; When M adapt When the value is less than the preset threshold, it means that the model is not adaptable to the current scenario, knowledge nodes, and scenario library data.

6. A scenario-based modeling method for operation and maintenance services according to claim 5, characterized in that: The operation and maintenance efficiency and cost are evaluated based on the dynamic scenario adaptive model training results, which is expressed as: Among them, E OM is the comprehensive evaluation result of operation and maintenance efficiency and cost, η eff Represents the operation and maintenance efficiency, which changes with time t. op represents the operation and maintenance cost, which changes with time t, T is the upper limit of the time integral, F(η eff (t)) is the operation and maintenance efficiency filtering function at time t, G(C op (t)) is the operation and maintenance cost filtering function at time t, γ is the regularization parameter, R(M) is the regularization term used to prevent overfitting, a, b are the upper and lower limits of the integral interval, N, P, R are the corresponding summation upper limits, and the operation and maintenance efficiency and cost evaluation results E OM The range of is the real number set R; When E OM When the value is greater than the preset threshold, it indicates high O&M efficiency and low cost, indicating excellent overall performance; When E OM When the value is greater than the preset threshold, it indicates low O&M efficiency and high cost, indicating poor overall performance.

7. A scenario-based modeling method for operation and maintenance services according to claim 6, characterized in that: The analysis of potential model problems and optimization includes analyzing equipment vibration spectrum data, thermal imaging data, gas composition analysis data, electromagnetic field strength data, process parameter correlation data, and human-computer interaction behavior data based on the dynamic scenario adaptive model training results and updating the scenario library data.

8. A scenario-based modeling system for operation and maintenance services according to any one of claims 1 to 7, characterized in that: Including data collection module, scene recognition module, and model application module; The data collection module is used to collect real-time data of the operation and maintenance environment, perform preprocessing, and store the data in the system to provide basic data for subsequent scene recognition and model application; The scene recognition module identifies and defines different operation and maintenance scenarios based on the collected data and knowledge graph, builds a scene library, and generates a dynamic scene adaptive model; The model application module applies a dynamic scenario adaptive model to make operation and maintenance decisions and optimization, thereby improving operation and maintenance efficiency and reducing operation and maintenance costs.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the scenario-based modeling method for operation and maintenance business described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the scenario-based modeling method for operation and maintenance services described in any one of claims 1 to 7 are implemented.

Images