Power plant equipment state evaluation and operation mode optimization method and system

Abstract

The invention discloses a power plant equipment state evaluation and operation mode optimization method and system, and belongs to the technical field of power systems. The power plant equipment state evaluation method solves the problem that in the prior art, the power plant equipment cannot be subjected to real-time state evaluation and operation mode optimization, so that faults or performance reduction of the power plant equipment is easily caused, state data are collected and processed, an equipment state evaluation model is constructed by using a deep learning technology, new state data are predicted, and the power plant equipment state evaluation method is high in practicability. Therefore, whether the power plant equipment has a fault or performance declines is judged, state evaluation of the power plant equipment is realized, and the optimal operation mode can be found through a search algorithm by combining the prediction result with the operation data of the power plant equipment, so that powerful support is provided for intelligent management of the power plant equipment, stable operation of the power plant equipment is realized, and the reliability of the power plant equipment is improved. Meanwhile, a user interaction interface is also provided, so that an operator can intuitively monitor and manage results of equipment state evaluation and operation mode optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method and system for evaluating the status of power plant equipment and optimizing operating modes. Background Art

[0002] In the power system, the normal operation of equipment is crucial to the stability and reliability of the entire system. The main equipment of the power plant includes generators, boilers, desulfurization and dust removal, cooling towers, regulating pools, etc. These equipment work together to ensure the normal operation of the power plant. However, due to the complexity of power plant equipment and the influence of various factors, such as changes in environmental parameters such as temperature, pressure, and vibration, it is impossible to evaluate the real-time status of power plant equipment and optimize the operation mode, which easily leads to failure or performance degradation of power plant equipment; therefore, it does not meet the existing needs. For this, we propose a method and system for evaluating the status of power plant equipment and optimizing the operation mode. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for power plant equipment status assessment and operation mode optimization. By collecting and processing status data and using deep learning technology to build an equipment status assessment model, new status data is predicted to determine whether the power plant equipment has a fault or whether its performance has declined. The prediction results are combined with the operating data of the power plant equipment, and the best operating mode can be found through a search algorithm to achieve stable operation of the power plant equipment. At the same time, a user interaction interface is provided, so that the operator can intuitively monitor and manage the results of equipment status assessment and operation mode optimization, thereby solving the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solution: a method for evaluating the status of power plant equipment and optimizing the operation mode, comprising the following steps:

[0005] Collect and process status data of power plant equipment over a period of time, including temperature, pressure and vibration;

[0006] Use deep learning technology to build an equipment condition assessment model for evaluating power plant equipment;

[0007] Use the equipment status assessment model to predict new status data and predict whether the power plant equipment will fail or its performance will decline;

[0008] Based on the results of equipment status assessment and combined with the operating data of power plant equipment, the search algorithm is used to find the best operating mode for power plant equipment;

[0009] Provides a user interface that allows operators to monitor and manage the results of power plant equipment status assessment and operation mode optimization.

[0010] Furthermore, collecting and processing the status data of the power plant equipment in the past period of time includes the following steps:

[0011] Obtain the status data of power plant equipment over a period of time through historical records, including temperature, pressure and vibration;

[0012] The obtained status data is preprocessed, including removing outliers, filling missing values ​​and normalizing data, where:

[0013] Remove outliers: Use box plots or Z-score methods to identify outliers in the status data, mark all outliers and remove them from the status data;

[0014] Fill missing values: Use mean, median and mode to fill missing values ​​in status data;

[0015] Normalized data: Use the z-score standardization method to calculate the mean and standard deviation of each value of the status data, and normalize the status data to a range with a mean of 0 and a standard deviation of 1.

[0016] Furthermore, a device status assessment model for power plant equipment is constructed using deep learning technology, including the following steps:

[0017] Extracting features used to represent the status of power plant equipment from the status data, including index features and nonlinear features;

[0018] Based on the extracted features and combined with deep learning technology, an equipment status assessment model for predicting the status of power plant equipment is constructed;

[0019] Train and optimize the constructed equipment status assessment model, and conduct an assessment after the equipment status assessment model training is completed;

[0020] After the equipment status assessment model is trained, it is deployed to actual applications.

[0021] Furthermore, extracting features for representing the state of power plant equipment from the state data includes the following steps:

[0022] After processing the state data, the indicator features and nonlinear features are extracted from the state data, where:

[0023] The indicator characteristics are temperature, pressure and vibration, which reflect the physical status of the power plant equipment during operation;

[0024] The nonlinear characteristics are power spectrum density and power factor, which are used to predict the operating status of power plant equipment;

[0025] After feature extraction is completed, the state data is divided into training set and test set;

[0026] The state data is split into 70% training set and 30% test set, where:

[0027] Using a part of the training set to train the equipment state assessment model, and using another part of the training set to optimize the equipment state assessment model;

[0028] After the training is completed, the test set is used to evaluate the equipment status assessment model, and the performance indicators of the equipment status assessment model are calculated, including precision, recall, and F1 score.

[0029] Furthermore, the new status data monitored in real time is predicted by the equipment status assessment model, including the following steps:

[0030] Through various sensors installed on power plant equipment, the status of power plant equipment is monitored in real time, and new status data detected in real time is collected;

[0031] The collected new status data is processed by removing outliers, filling missing values ​​and normalizing, and the processed new status data is converted into the same format as the input data format of the equipment status assessment model;

[0032] Before making a prediction, the new state data is enhanced by using time shift or frequency change methods;

[0033] Use equipment condition assessment models to make predictions about new condition data, including determining whether power plant equipment has failed or whether power plant equipment performance has degraded;

[0034] The prediction structure of the equipment status assessment model is output, and based on the prediction results, an early warning notification is immediately sent to the user interface when power plant equipment fails or performance deteriorates.

[0035] Furthermore, the search algorithm is used to find the best operation mode of the power plant equipment, including the following steps:

[0036] According to the prediction results of the equipment status assessment model, the status information of the current power plant equipment is obtained;

[0037] At the same time, the operating data of power plant equipment is collected, including grid load conditions, historical operating records, environmental parameters and equipment maintenance records;

[0038] By using a search method and according to the prediction results and operation data of the power plant equipment, the best operation mode of the power plant equipment is found, wherein the search method adopts a genetic algorithm.

[0039] Furthermore, a user interaction interface is provided, including the following functions:

[0040] Data visualization function: used to display the results of power plant equipment status assessment and operation mode optimization in the form of charts;

[0041] Permission management function: used to grant permissions to operators. Authorized operators are allowed to operate the results of power plant equipment status assessment and operation mode optimization;

[0042] Log recording function: used to record every operation of the operator.

[0043] A power plant equipment status evaluation and operation mode optimization system is used to implement a power plant equipment status evaluation and operation mode optimization method, including:

[0044] Data collection module for:

[0045] Collect the status data of power plant equipment in the past period of time and pre-process the collected status data, where the status data includes temperature, pressure and vibration;

[0046] Device evaluation modules for:

[0047] Use deep learning technology to build an equipment status assessment model for power plant equipment;

[0048] Use the equipment status assessment model to predict new status data and determine whether the power plant equipment has failed or its performance has declined;

[0049] Based on the prediction results, when power plant equipment fails or performance deteriorates, an early warning notification is immediately sent to the result display module;

[0050] Equipment optimization module, used for:

[0051] Combine the prediction results of the equipment status assessment model with the operating data of the power plant equipment, and find the best operating mode of the power plant equipment through the search algorithm;

[0052] The result display module is used to:

[0053] A user interaction interface is provided to display the results of the status evaluation and operation mode optimization of power plant equipment and early warning notifications, while allowing operators to monitor and manage the results of the status evaluation and operation mode optimization of power plant equipment.

[0054] Furthermore, the data collection module includes:

[0055] Data acquisition module, used to:

[0056] Obtain the status data of power plant equipment in the past period of time through historical record data;

[0057] Data processing module for:

[0058] The acquired status data is preprocessed, including removing outliers, filling missing values, and normalizing data.

[0059] Furthermore, the equipment evaluation module includes:

[0060] Feature extraction module for:

[0061] Extract features from the state data, and extract features used to represent the state of the power plant equipment from the state data, including index features and nonlinear features;

[0062] After feature extraction is completed, the state data is divided into training set and test set;

[0063] Model building modules for:

[0064] Based on the extracted features and deep learning technology, an equipment status assessment model is constructed;

[0065] Model training module, used to:

[0066] Use the training set to train and optimize the equipment status assessment model;

[0067] Use the test set to evaluate the equipment status assessment model;

[0068] Model deployment module, used to:

[0069] After the equipment status assessment model is trained, it is deployed to actual applications;

[0070] Use the equipment status assessment model to predict new status data and determine whether the power plant equipment has failed or performance has degraded;

[0071] Abnormal warning module, used for:

[0072] Based on the prediction results of the equipment status assessment model, if the power plant equipment fails or its performance deteriorates, an early warning notification will be immediately sent to the result display module.

[0073] Compared with the prior art, the present invention has the following beneficial effects:

[0074] The present invention collects and processes power plant equipment status data over a period of time to provide data support for subsequent model construction, and then uses deep learning technology to construct an equipment status assessment model for evaluating power plant equipment. The model can accurately predict the newly collected status data and determine whether the equipment has failed or its performance has declined, thereby realizing the status assessment of the power plant equipment. Based on the prediction of the equipment status assessment model, the prediction result is combined with the operating data of the power plant equipment. The search algorithm can be used to find the best operating mode of the power plant equipment, thereby providing strong support for the intelligent management of the power plant equipment. The provided user interaction interface can facilitate the operator to intuitively monitor and manage the results of equipment status assessment and operation mode optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A flow chart of the method for evaluating the state of power plant equipment and optimizing the operation mode of the present invention;

[0076] Figure 2 It is a structural schematic diagram of the power plant equipment status evaluation and operation mode optimization system of the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] In order to solve the problem that the existing technology cannot evaluate the real-time status of power plant equipment and optimize the operation mode, which may easily lead to failure or performance degradation of power plant equipment, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0079] A method for evaluating the state of power plant equipment and optimizing the operation mode comprises the following steps:

[0080] Collect and process status data of power plant equipment over a period of time, including temperature, pressure and vibration;

[0081] Use deep learning technology to build an equipment condition assessment model for evaluating power plant equipment;

[0082] Use the equipment status assessment model to predict new status data and predict whether the power plant equipment will fail or its performance will decline;

[0083] Based on the results of equipment status assessment and combined with the operating data of power plant equipment, the search algorithm is used to find the best operating mode for power plant equipment;

[0084] Provides a user interface that allows operators to monitor and manage the results of power plant equipment status assessment and operation mode optimization.

[0085] The technical effect of the above content is: by collecting and processing the status data of power plant equipment in the past period of time, data support can be provided for subsequent model construction. Based on the collected status data and using deep learning technology (such as neural networks, convolutional neural networks and recurrent neural networks, etc.), an equipment status evaluation model for evaluating power plant equipment can be constructed. The new status data is predicted by the equipment status evaluation model to predict whether the power plant equipment has failed or whether its performance has declined. The prediction results can provide a basis for the maintenance plan, preventive maintenance or equipment replacement of power plant equipment. Based on the prediction results of the equipment status evaluation and combined with the operating data of the power plant equipment, a search algorithm is used to find the best operating mode of the power plant equipment, so as to find the optimal operating mode that can maximize the operating efficiency of the power plant equipment or extend the life of the equipment under the current environment. At the same time, a user interaction interface is also provided to allow operators to monitor and manage the results of the status evaluation and operation mode optimization of the power plant equipment, so that operators can view the operating status and operating parameters of the equipment at any time.

[0086] Collecting and processing the status data of power plant equipment over a period of time includes the following steps:

[0087] Obtain the status data of power plant equipment over a period of time through historical records, including temperature, pressure and vibration;

[0088] The obtained status data is preprocessed, including removing outliers, filling missing values ​​and normalizing data, where:

[0089] Remove outliers: Use box plots or Z-score methods to identify outliers in the status data, mark all outliers and remove them from the status data;

[0090] Fill missing values: Use mean, median and mode to fill missing values ​​in status data;

[0091] Normalized data: Use the z-score standardization method to calculate the mean and standard deviation of each value of the status data, and normalize the status data to a range with a mean of 0 and a standard deviation of 1.

[0092] The technical effect of the above content is: by analyzing historical records to obtain the status data of power plant equipment over a period of time, and in order to ensure the availability and reliability of the data, the collected status data needs to be preprocessed. By removing outliers and filling missing values, the interference of outliers and missing values ​​on the status data can be reduced, thereby improving the quality of subsequent analysis and modeling. By normalizing the data, the subsequent model can be more easily converged and the potential relationship between the data can be discovered. The preprocessing process can not only eliminate outliers and missing values ​​in the original data, but also create a clean and stable status data, thereby providing a basis for subsequent model training.

[0093] Using deep learning technology to build an equipment condition assessment model for power plant equipment includes the following steps:

[0094] Extracting features used to represent the status of power plant equipment from the status data, including index features and nonlinear features;

[0095] Based on the extracted features and combined with deep learning technology, an equipment status assessment model for predicting the status of power plant equipment is constructed;

[0096] Train and optimize the constructed equipment status assessment model, and conduct an assessment after the equipment status assessment model training is completed;

[0097] After the equipment status assessment model is trained, it is deployed to actual applications.

[0098] Extracting features used to represent the status of power plant equipment from the status data includes the following steps:

[0099] After processing the state data, the indicator features and nonlinear features are extracted from the state data, where:

[0100] The indicator characteristics are temperature, pressure and vibration, which reflect the physical status of the power plant equipment during operation;

[0101] The nonlinear characteristics are power spectrum density and power factor, which are used to predict the operating status of power plant equipment;

[0102] After feature extraction is completed, the state data is divided into training set and test set;

[0103] The status data is divided into 70% training set and 30% test set. The division enables the equipment status assessment model to be exposed to a variety of different data distributions during training, so that it can be better generalized to unseen data.

[0104] Using a part of the training set to train the equipment state assessment model, and using another part of the training set to optimize the equipment state assessment model until the equipment state assessment model achieves satisfactory performance;

[0105] After the training is completed, the equipment state assessment model is evaluated using the test set, and the performance indicators of the equipment state assessment model are calculated, including precision, recall, and F1 score, to understand the generalization ability and accuracy of the equipment state assessment model.

[0106] The technical effect of the above content is: by extracting features from the status data, it is possible to extract status features used to represent power plant equipment, including index features such as temperature, pressure and vibration, as well as some nonlinear features such as spectral density and power. The index features can directly reflect the physical state of the equipment during operation. The nonlinear features can provide richer information, thereby more accurately predicting the operating status of power plant equipment. Based on the extracted features and combined with deep learning technology, a device status assessment model for predicting the status of power plant equipment can be constructed. After the equipment status assessment model is established, it is necessary to train and optimize the equipment status assessment model so that the equipment status assessment model can better adapt to the new status data. After the equipment status assessment model is trained, the equipment status assessment model is deployed in actual applications so that the equipment status assessment model can be used to predict the status of power plant equipment, thereby realizing the status assessment of power plant equipment.

[0107] The new status data monitored in real time is predicted by the equipment status assessment model, including the following steps:

[0108] Through various sensors installed on power plant equipment, the status of power plant equipment is monitored in real time, and new status data detected in real time is collected;

[0109] The collected new status data is processed by removing outliers, filling missing values ​​and normalizing, and the processed new status data is converted into the same format as the input data format of the equipment status assessment model;

[0110] Before making a prediction, the new state data is enhanced by using time shift or frequency change methods;

[0111] Use equipment condition assessment models to make predictions about new condition data, including determining whether power plant equipment has failed or whether power plant equipment performance has degraded;

[0112] The prediction structure of the equipment status assessment model is output, and based on the prediction results, an early warning notification is immediately sent to the user interface when power plant equipment fails or performance deteriorates.

[0113] The technical effect of the above content is: new status data is collected through various sensors, and the new status data is used to predict the status of the current power plant equipment. Before making a prediction, the new status data is enhanced using time shift or frequency change methods to improve the robustness of the model. The new status data is predicted through the equipment status evaluation model to predict whether the power plant equipment has failed or its performance has declined, thereby realizing the status evaluation of the power plant equipment. Based on the prediction results, an early warning notification can be issued to the user interface so that timely measures can be taken to avoid equipment failure or reduce equipment performance.

[0114] The search algorithm is used to find the best operation mode of the power plant equipment, including the following steps:

[0115] According to the prediction results of the equipment status assessment model, the status information of the current power plant equipment is obtained;

[0116] At the same time, the operating data of power plant equipment is collected, including grid load conditions, historical operating records, environmental parameters and equipment maintenance records;

[0117] Through the search method, and according to the prediction results and operation data of the power plant equipment, the best operation mode of the power plant equipment is found, wherein the search method adopts a genetic algorithm (the search algorithm can be a genetic algorithm, an ant colony algorithm, a simulated annealing algorithm, etc., or a combination of multiple algorithms), specifically:

[0118] Initialize the population: Generate the initial population, where each individual represents a possible operation mode. The attributes of the individual include the grid load of the power plant equipment, historical operation records, environmental parameters and equipment maintenance records;

[0119] Design fitness function: Define the fitness function to measure the adaptability of each individual. The function reflects the status of power plant equipment, environmental conditions and energy consumption. Individuals with strong adaptability are more likely to be selected for the next generation of reproduction.

[0120] Crossover operation: Calculate the similarity between two different individuals as the basis for gene recombination. This similarity can be measured using methods such as cosine similarity.

[0121] Mutation operation: set a mutation probability for each individual, and perform mutation when the fitness of the individual is lower than a certain threshold;

[0122] Selection operation: According to certain rules, such as roulette selection method, maximum and minimum ant colony selection method, etc., a part of individuals are selected from the parent population as the parents of the next generation population;

[0123] Iterative operation: Repeat the above three steps until the termination condition is met, for example, the maximum number of iterations is reached or a satisfactory solution is found;

[0124] Output result: Find the optimal solution that meets the termination conditions as the best operating mode of the power plant equipment.

[0125] The technical effect of the above content is: according to the prediction results of the equipment status assessment model, the status information of the current power plant equipment is obtained, and then the operating data of the power plant equipment is collected, so as to better understand the operating status of the equipment in order to find the best operating mode, and to find the best operating mode of the power plant equipment through search methods (such as genetic algorithms). During the search process, the prediction results of the equipment status assessment model and the operating data are combined as the input of the search algorithm, so that the search algorithm can automatically find the best operating mode based on the understanding of the equipment status, thereby optimizing the operation of the power plant equipment.

[0126] Provides a user interaction interface, including the following functions:

[0127] Data visualization function: used to display the results of power plant equipment status assessment and operation mode optimization in the form of charts;

[0128] Permission management function: used to grant permissions to operators. Authorized operators are allowed to operate the results of power plant equipment status assessment and operation mode optimization;

[0129] Log recording function: used to record every operation of the operator.

[0130] The technical effects of the above content are: the data visualization function intuitively displays the results of power plant equipment status evaluation and operation mode optimization through various charts (such as line charts, bar charts and heat maps, etc.); the authority management function can set different authority levels for operators, and only operators with corresponding authority can operate on the results of power plant equipment status evaluation and operation mode optimization to improve the security of the system; and the log recording function can record and save every step of the operator's operation, including data input, analysis and report viewing, etc., for the convenience of later query and audit. Through the above functions, the operator can easily and comprehensively manage and control the power plant equipment status evaluation and operation mode optimization.

[0131] Specifically, this embodiment further proposes a power plant equipment status evaluation and operation mode optimization system, which is used to implement a power plant equipment status evaluation and operation mode optimization method, including:

[0132] Data collection module for:

[0133] Collect the status data of power plant equipment in the past period of time and pre-process the collected status data, where the status data includes temperature, pressure and vibration;

[0134] Device evaluation modules for:

[0135] Use deep learning technology to build an equipment status assessment model for power plant equipment;

[0136] Use the equipment status assessment model to predict new status data and determine whether the power plant equipment has failed or its performance has declined;

[0137] Based on the prediction results, when power plant equipment fails or performance deteriorates, an early warning notification is immediately sent to the result display module;

[0138] Equipment optimization module, used for:

[0139] Combine the prediction results of the equipment status assessment model with the operating data of the power plant equipment, and find the best operating mode of the power plant equipment through the search algorithm;

[0140] The result display module is used to:

[0141] A user interaction interface is provided to display the results of the status evaluation and operation mode optimization of power plant equipment and early warning notifications, while allowing operators to monitor and manage the results of the status evaluation and operation mode optimization of power plant equipment.

[0142] Data collection module, including:

[0143] Data acquisition module, used to:

[0144] Obtain the status data of power plant equipment in the past period of time through historical record data;

[0145] Data processing module for:

[0146] The acquired status data is preprocessed, including removing outliers, filling missing values, and normalizing data.

[0147] Equipment evaluation modules, including:

[0148] Feature extraction module for:

[0149] Extract features from the state data, and extract features used to represent the state of the power plant equipment from the state data, including index features and nonlinear features;

[0150] After feature extraction is completed, the state data is divided into training set and test set;

[0151] Model building modules for:

[0152] Based on the extracted features and deep learning technology, an equipment status assessment model is constructed;

[0153] Model training module, used to:

[0154] Use the training set to train and optimize the equipment status assessment model;

[0155] Use the test set to evaluate the equipment status assessment model;

[0156] Model deployment module, used to:

[0157] After the equipment status assessment model is trained, it is deployed to actual applications;

[0158] Use the equipment status assessment model to predict new status data and determine whether the power plant equipment has failed or performance has degraded;

[0159] Abnormal warning module, used for:

[0160] Based on the prediction results of the equipment status assessment model, if the power plant equipment fails or its performance deteriorates, an early warning notification will be immediately sent to the result display module.

[0161] The technical effect of the above content is: the data collection module is responsible for collecting the status data of the power plant equipment in the past period of time and preprocessing it to eliminate any abnormal values, fill in missing values ​​and normalize the data, so as to improve the quality of the status data. The equipment evaluation module uses deep learning technology to construct an equipment status evaluation model. The constructed equipment status evaluation model can predict whether the power plant equipment has a fault or whether the performance has declined through new status data, and the prediction results can provide a basis for the equipment maintenance plan, preventive maintenance, and even the equipment replacement, so as to achieve the purpose of evaluating the status of the power plant equipment. In addition, when the equipment status evaluation model predicts that the equipment has a fault or performance declines, it will immediately send an early warning notification to the result display module so that timely measures can be taken to prevent potential accidents. The equipment optimization module combines the prediction results of the equipment status evaluation model with the operating data of the power plant equipment, and finds the best operating mode of the power plant equipment through a search algorithm, so as to optimize the operation of the power plant equipment. At the same time, the result display module provides a user interaction interface, which displays the results of the status evaluation and operation mode optimization of the power plant equipment and related early warning notifications through an intuitive user interaction interface, so that the operator can better understand and control the operation of the equipment.

[0162] Working principle: By collecting and processing a large amount of status data and using deep learning technology to build an equipment status assessment model, the status changes of power plant equipment can be predicted more accurately, thereby improving the accuracy of equipment status assessment. By predicting the status of power plant equipment, abnormal conditions of equipment can be discovered in a timely manner to avoid large-scale shutdowns caused by power plant equipment failures, thereby reducing repair and maintenance costs and improving the reliability of power plant equipment. The prediction results are combined with the operating data of the power plant equipment, and the best operating mode of the power plant equipment is found through the search algorithm, so as to optimize the operation of the power plant equipment and improve the safety and stability of the power plant equipment operation. Operators can monitor and manage the results of power plant equipment status assessment and operation mode optimization through the user interaction interface, provide real-time data support for operational decisions, and help operators make better decisions.

[0163] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0164] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for evaluating the state of power plant equipment and optimizing the operation mode, characterized in that: The following steps are involved: Collect and process status data of power plant equipment over a period of time, including temperature, pressure and vibration; Use deep learning technology to build an equipment condition assessment model for evaluating power plant equipment; Use the equipment status assessment model to predict new status data and predict whether the power plant equipment will fail or its performance will decline; Based on the results of equipment status assessment and combined with the operating data of power plant equipment, the search algorithm is used to find the best operating mode for power plant equipment; Provides a user interface that allows operators to monitor and manage the results of power plant equipment status assessment and operation mode optimization.

2. A method for evaluating the state of power plant equipment and optimizing the operation mode according to claim 1, characterized in that: Collecting and processing the status data of power plant equipment over a period of time includes the following steps: Obtain the status data of power plant equipment over a period of time through historical records, including temperature, pressure and vibration; The obtained status data is preprocessed, including removing outliers, filling missing values ​​and normalizing data, where: Remove outliers: Use box plots or Z-score methods to identify outliers in the status data, mark all outliers and remove them from the status data; Fill missing values: Use mean, median and mode to fill missing values ​​in status data; Normalized data: Use the z-score standardization method to calculate the mean and standard deviation of each value of the status data, and normalize the status data to a range with a mean of 0 and a standard deviation of 1.

3. A method for evaluating the state of power plant equipment and optimizing the operation mode according to claim 1, characterized in that: Using deep learning technology to build an equipment condition assessment model for power plant equipment includes the following steps: Extracting features used to represent the status of power plant equipment from the status data, including index features and nonlinear features; Based on the extracted features and combined with deep learning technology, an equipment status assessment model for predicting the status of power plant equipment is constructed; Train and optimize the constructed equipment status assessment model, and conduct an assessment after the equipment status assessment model training is completed; After the equipment status assessment model is trained, it is deployed to actual applications.

4. A method for evaluating the state of power plant equipment and optimizing the operation mode according to claim 3, characterized in that: Extracting features used to represent the status of power plant equipment from the status data includes the following steps: After processing the state data, the indicator features and nonlinear features are extracted from the state data, where: The indicator characteristics are temperature, pressure and vibration, which reflect the physical status of the power plant equipment during operation; The nonlinear characteristics are power spectrum density and power factor, which are used to predict the operating status of power plant equipment; After feature extraction is completed, the state data is divided into training set and test set; The state data is split into 70% training set and 30% test set, where: Using a part of the training set to train the equipment state assessment model, and using another part of the training set to optimize the equipment state assessment model; After the training is completed, the test set is used to evaluate the equipment status assessment model, and the performance indicators of the equipment status assessment model are calculated, including precision, recall, and F1 score.

5. A method for evaluating the state of power plant equipment and optimizing the operation mode according to claim 1, characterized in that: The new status data monitored in real time is predicted by the equipment status assessment model, including the following steps: Through various sensors installed on power plant equipment, the status of power plant equipment is monitored in real time, and new status data detected in real time is collected; The collected new status data is processed by removing outliers, filling missing values ​​and normalizing, and the processed new status data is converted into the same format as the input data format of the equipment status assessment model; Before making a prediction, the new state data is enhanced by using time shift or frequency change methods; Use equipment condition assessment models to make predictions about new condition data, including determining whether power plant equipment has failed or whether power plant equipment performance has degraded; The prediction structure of the equipment status assessment model is output, and based on the prediction results, an early warning notification is immediately sent to the user interface when power plant equipment fails or performance deteriorates.

6. A method for evaluating the state of power plant equipment and optimizing the operation mode according to claim 1, characterized in that: The search algorithm is used to find the best operation mode of the power plant equipment, including the following steps: According to the prediction results of the equipment status assessment model, the status information of the current power plant equipment is obtained; At the same time, the operating data of power plant equipment is collected, including grid load conditions, historical operating records, environmental parameters and equipment maintenance records; By using a search method and according to the prediction results and operation data of the power plant equipment, the best operation mode of the power plant equipment is found, wherein the search method adopts a genetic algorithm.

7. A method for evaluating the state of power plant equipment and optimizing the operation mode according to claim 1, characterized in that: Provides a user interaction interface, including the following functions: Data visualization function: used to display the results of power plant equipment status assessment and operation mode optimization in the form of charts; Permission management function: used to grant permissions to operators. Authorized operators are allowed to operate the results of power plant equipment status assessment and operation mode optimization; Log recording function: used to record every operation of the operator.

8. A power plant equipment status evaluation and operation mode optimization system, used to implement the power plant equipment status evaluation and operation mode optimization method according to any one of claims 1 to 7, characterized in that: include: Data collection module for: Collect the status data of power plant equipment in the past period of time and pre-process the collected status data, where the status data includes temperature, pressure and vibration; Device evaluation modules for: Use deep learning technology to build an equipment status assessment model for power plant equipment; Use the equipment status assessment model to predict new status data and determine whether the power plant equipment has failed or its performance has declined; Based on the prediction results, when power plant equipment fails or performance deteriorates, an early warning notification is immediately sent to the result display module; Equipment optimization module, used for: Combine the prediction results of the equipment status assessment model with the operating data of the power plant equipment, and find the best operating mode of the power plant equipment through the search algorithm; The result display module is used to: A user interaction interface is provided to display the results of the status evaluation and operation mode optimization of power plant equipment and early warning notifications, while allowing operators to monitor and manage the results of the status evaluation and operation mode optimization of power plant equipment.

9. A power plant equipment status assessment and operation mode optimization system according to claim 8, characterized in that: The data collection module comprises: Data acquisition module, used to: Obtain the status data of power plant equipment in the past period of time through historical record data; Data processing module for: The acquired status data is preprocessed, including removing outliers, filling missing values, and normalizing data.

10. A power plant equipment status assessment and operation mode optimization system according to claim 8, characterized in that: The equipment evaluation module comprises: Feature extraction module for: Extract features from the state data, and extract features used to represent the state of the power plant equipment from the state data, including index features and nonlinear features; After feature extraction is completed, the state data is divided into training set and test set; Model building modules for: Based on the extracted features and deep learning technology, an equipment status assessment model is constructed; Model training module, used to: Use the training set to train and optimize the equipment status assessment model; Use the test set to evaluate the equipment status assessment model; Model deployment module, used to: After the equipment status assessment model is trained, it is deployed to actual applications; Use the equipment status assessment model to predict new status data and determine whether the power plant equipment has failed or performance has degraded; Abnormal warning module, used for: Based on the prediction results of the equipment status assessment model, if the power plant equipment fails or its performance deteriorates, an early warning notification will be immediately sent to the result display module.

Images