An intelligent evaluation system and method for power production PMS business work tickets

Abstract

The present invention relates to the field of intelligent assessment technology, and specifically to an intelligent assessment system and method for work tickets for power production PMS business. First, the work ticket and the typical ticket are input into the BERT pre-trained language model, and the word embedding results of the work ticket and the typical ticket are output. A co-occurrence matrix is ​​constructed based on the word embedding results of the work ticket and the typical ticket, and the similarity between the work ticket and the typical ticket is calculated based on the co-occurrence matrix; the work ticket and the work ticket content indicators are analyzed to obtain the safety index of the work ticket; based on the similarity and the safety index of the work ticket, it is judged whether the simulation training of the trainees is qualified. The intelligent assessment system for work tickets for power production PMS business proposed by the present invention can intelligently and objectively evaluate the trainees' ticketing skills, improve the training effect, reduce the error rate in actual operation, and provide a strong guarantee for the safe operation of the power system.

Description

Technical Field

[0001] The present invention relates to the field of intelligent assessment technology, and in particular to an intelligent assessment system and method for power production PMS business work tickets. Background Art

[0002] The power production management system PMS (Power Management System) is a comprehensive software system used to manage and optimize power production, distribution and use. PMS mainly has the following functions: 1. Real-time monitoring of the operating status of the power system; 2. Optimization of power production and distribution; 3. Improvement of the safety and stability of the power system; 4. Reduction of operating costs. Among them, optimizing power production and distribution is the core function of PMS. Through intelligent scheduling and load forecasting, PMS can adjust the power generation plan in time when power demand fluctuates to ensure efficient and stable power supply.

[0003] In the power industry, the work ticket system is a safeguard for power safety, an important measure to ensure reliable and safe maintenance and construction of power equipment, and an important line of defense to protect the personal safety of power workers and the property safety of power equipment. With the continuous expansion and complexity of the power grid, the requirements for the professional skills and operational norms of power workers are becoming higher and higher.

[0004] Traditional training methods, such as completing training through accumulation of practical operation experience, are not only costly and time-consuming, but also pose potential safety risks. Therefore, simulation training systems have emerged to simulate the working conditions and operating environment of actual systems, thus avoiding the dangers that may be caused by the actual operation of the system and reducing the high cost.

[0005] However, most of the existing work ticket simulation training systems are still at the basic simulation operation level, lacking the real-time evaluation and supervision function of the trainees' operation process, and unable to conduct a systematic and objective evaluation of the trainees' ticketing skills. This makes it difficult to quantify the training effect, and trainees are still prone to making mistakes in actual operations, bringing potential risks to the safe operation of the power system.

[0006] In order to solve the above problems, an intelligent evaluation system and method for work tickets based on the applicable power production PMS business came into being. Summary of the invention

[0007] The purpose of the present invention is to provide an intelligent assessment system and method for power production PMS business work tickets, so as to solve the technical problem that the existing solutions cannot systematically and objectively evaluate the trainees' invoicing skills.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] On the one hand, the present invention proposes an intelligent evaluation system for work tickets applicable to power production PMS business, the system comprising a work ticket acquisition module, a similarity calculation module, a safety index calculation module and an intelligent evaluation module;

[0010] The work ticket acquisition module is used to obtain the work tickets prepared by trainees during the simulation training process;

[0011] A similarity calculation module is used to input the work ticket and the typical ticket into the BERT pre-trained language model, output the word embedding results of the work ticket and the typical ticket, construct a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, and calculate the similarity between the work ticket and the typical ticket based on the co-occurrence matrix;

[0012] A safety index calculation module is used to parse the work ticket and the work ticket content indicators to obtain the safety index of the work ticket;

[0013] The intelligent assessment module is used to judge whether the trainees’ simulation training is qualified based on the similarity and safety index of the work ticket.

[0014] Furthermore, the working ticket and the typical ticket are input into the BERT pre-trained language model, and the word embedding results of the working ticket and the typical ticket are output. The specific process includes the following:

[0015] Step 1: Text preprocessing of work tickets: traverse the text information of the work ticket and remove abnormal text information of the work ticket based on the Ransac algorithm;

[0016] Step 2: Take the preprocessed work ticket text as input and give the vector representation of each word in the text information. Each word is represented by a mixture of word vector, segment vector and position vector. The vector representation result is used as the input of the BERT pre-trained language model.

[0017] Step three: Use the BERT pre-trained language model to extract word vectors: convert text information into word vectors, send the word vectors to the pre-trained network model, and obtain the output of the specified feature layer. The output is the word embedding result of the text.

[0018] Furthermore, the process of eliminating abnormal text information of work tickets based on the Ransac algorithm specifically includes the following steps:

[0019] Step 1, set the number of iterations k of the Ransac algorithm;

[0020] Step 2: randomly select n work ticket text information as a data set, and use the least squares method to fit the preset straight line model based on the data set. The preset straight line model is ax+by+c=0, and the coordinates corresponding to each work ticket text information are , The value range is ;

[0021] Step 3, based on And the preset straight line model is used to determine whether the work ticket text information is an internal point:

[0022] ;

[0023] when , then the work ticket text information is judged to be an internal point. , then it is determined that the work ticket text information is not an internal point, where a, b and c are constants that are not 0;

[0024] Step 4: add 1 to the number of iterations. When the number of iterations is less than k, return to step 2 and calculate the next model. Each iteration corresponds to a preset straight line model. Finally, the model with the most points in the bureau is obtained according to the total number of points in the bureau of each model. The straight line corresponding to the model with the most points in the bureau is recorded as the normal text information straight line of the work ticket.

[0025] Step 5, based on the normal text information straight line of the work ticket, determine whether the work ticket contains abnormal text information: determine whether the vertical distance between the coordinates corresponding to each work ticket text information and the normal text information straight line of the work ticket exceeds the preset distance value. If so, the work ticket text information is abnormal text information and it will be eliminated. If not, the work ticket text information is not abnormal text information and it will be retained.

[0026] Furthermore, a co-occurrence matrix is ​​constructed based on the word embedding results of the work ticket and the word embedding results of the typical ticket. The similarity between the work ticket and the typical ticket is calculated based on the co-occurrence matrix, which specifically includes the following process:

[0027] Construct a co-occurrence matrix based on the word embedding results of work tickets and typical tickets :

[0028] ;

[0029] in, The word embedding vector representing the work ticket, express , The word embedding vector representing a typical ticket, express , express The transpose of

[0030] Based on the co-occurrence matrix Calculate the co-occurrence representation index :

[0031] ;

[0032] in, express The number of column vectors, express The number of column vectors, e is 2.72;

[0033] Co-occurrence-based representation index Calculate the similarity between the work ticket and the typical ticket :

[0034] .

[0035] Furthermore, based on the work ticket and the work ticket content indicators, the safety index of the work ticket is obtained, which specifically includes the following processes:

[0036] Establish a hierarchical model with a hierarchical structure, where the levels include the goal level, the criterion level and the measure level;

[0037] Constructing a judgment matrix: Determining the number of relevant influencing factors at each level , construct a set of risk factors for work tickets , ,in, is the correct subset of all substation names in the work ticket, is the correct subset of the names of the distribution stations in the work order, is the correctness rate subset of the dispatching power of the substation in the work ticket, is the dispatch power accuracy subset of the distribution station in the work ticket, from the set Take out two subsets in the same level for comparison, assign the corresponding importance according to the preset ratio, combine the importance of each level to form a judgment matrix;

[0038] Calculate the maximum eigenvalue of the judgment matrix :

[0039] ;

[0040] in, is the matrix obtained by normalizing each column vector of the judgment matrix. The value of is 1,2,...,m, For the matrix The elements of are added row by row to obtain the vector and then normalized into a matrix. For the matrix Add the elements of column by column to obtain the vector and then normalize the matrix;

[0041] Consistency test of judgment matrix:

[0042] When the judgment matrix has only one non-zero eigenvalue, it means that the matrix is ​​completely consistent. If the matrix V has more than one eigenvalue, the consistency index can be used :

[0043] ;

[0044] in, represents the order of the judgment matrix, The smaller the value of, the better the consistency, and vice versa, the greater the degree of deviation;

[0045] Will The value is multiplied by the corresponding weight to obtain the safety index of the work ticket.

[0046] Furthermore, judging whether the trainee's simulation training is qualified based on the similarity and the safety index of the work ticket specifically includes the following process:

[0047] The similarity and the safety index of the work ticket are summed to obtain the sum value;

[0048] The simulation training qualification threshold of the trainee is loaded, and it is judged whether the sum value exceeds the simulation training qualification threshold. If so, it is judged that the simulation training of the trainee is qualified; if not, it is judged that the simulation training of the trainee is unqualified.

[0049] Furthermore, the system's database adopts the MySql database management system.

[0050] Furthermore, judging whether the trainees have passed the simulation training also includes signature verification and time verification of the work ticket.

[0051] On the other hand, the present invention also proposes an intelligent evaluation method for a work ticket of a power production PMS business, the method comprising the following steps:

[0052] Obtain the work tickets prepared by trainees during the simulation training process;

[0053] Input the work ticket and typical ticket into the BERT pre-trained language model, output the word embedding results of the work ticket and the typical ticket, build a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, and calculate the similarity between the work ticket and the typical ticket based on the co-occurrence matrix;

[0054] Based on the work ticket and the work ticket content indicators, the safety index of the work ticket is obtained;

[0055] The similarity and safety index of the work ticket are used to determine whether the trainees have passed the simulation training.

[0056] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0057] The present invention inputs the work ticket and the typical ticket into the BERT pre-trained language model, outputs the word embedding results of the work ticket and the typical ticket, constructs a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, calculates the similarity between the work ticket and the typical ticket based on the co-occurrence matrix; performs analysis based on the work ticket and the work ticket content index to obtain the safety index of the work ticket; and judges whether the simulation training of the trainee is qualified based on the similarity and the safety index of the work ticket. The intelligent and objective evaluation of the trainee's ticketing skills improves the training effect, reduces the error rate in actual operation, and provides a strong guarantee for the safe operation of the power system.

[0058] Furthermore, the present invention can also provide a scientific and objective basis for evaluating the training effect for the training management department of the power enterprise, which is helpful to optimize the training program and improve the training quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0060] Figure 1 This is a system block diagram of an intelligent evaluation system for power production PMS business work tickets applicable to an embodiment of the present invention;

[0061] Figure 2 This is a workflow diagram of an intelligent evaluation system for power production PMS business work tickets according to an embodiment of the present invention;

[0062] Figure 3 This is another workflow diagram of an intelligent evaluation system for power production PMS business work tickets applicable to an embodiment of the present invention;

[0063] Figure 4 The present invention is a flowchart of an intelligent evaluation method for a power production PMS business work ticket applicable to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] 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.

[0065] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0066] This embodiment provides an intelligent evaluation system for power production PMS business work tickets. Figure 1 is a system block diagram of an intelligent evaluation system for power production PMS business work tickets according to an embodiment of the present invention. Figure 1 As shown, the system includes a work ticket acquisition module, a similarity calculation module, a safety index calculation module and an intelligent evaluation module;

[0067] The work ticket acquisition module is used to obtain the work tickets prepared by trainees during the simulation training process;

[0068] A similarity calculation module is used to input the work ticket and the typical ticket into the BERT pre-trained language model, output the word embedding results of the work ticket and the typical ticket, construct a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, and calculate the similarity between the work ticket and the typical ticket based on the co-occurrence matrix;

[0069] A safety index calculation module is used to parse the work ticket and the work ticket content indicators to obtain the safety index of the work ticket;

[0070] The intelligent assessment module is used to judge whether the trainees’ simulation training is qualified based on the similarity and safety index of the work ticket.

[0071] The present invention inputs the work ticket and the typical ticket into the BERT pre-trained language model, outputs the word embedding results of the work ticket and the typical ticket, constructs a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, calculates the similarity between the work ticket and the typical ticket based on the co-occurrence matrix; performs analysis based on the work ticket and the work ticket content index to obtain the safety index of the work ticket; and judges whether the simulation training of the trainee is qualified based on the similarity and the safety index of the work ticket. The intelligent and objective evaluation of the trainee's ticketing skills improves the training effect, reduces the error rate in actual operation, and provides a strong guarantee for the safe operation of the power system.

[0072] Furthermore, the present invention can also provide a scientific and objective basis for evaluating the training effect for the training management department of the power enterprise, which is helpful to optimize the training program and improve the training quality.

[0073] It is worth mentioning that the work ticket acquisition module, similarity calculation module, safety index calculation module and intelligent evaluation module can all achieve pairwise communication.

[0074] In some embodiments, Figure 2 This is a workflow diagram of an intelligent evaluation system for power production PMS business work tickets according to an embodiment of the present invention. Figure 2 As shown, the working ticket and the typical ticket are input into the BERT pre-trained language model, and the word embedding results of the working ticket and the typical ticket are output. Specifically, the following steps are included:

[0075] Step 1: Text preprocessing of work tickets: traverse the text information of the work ticket and remove abnormal text information of the work ticket based on the Ransac algorithm;

[0076] Step 2: Take the preprocessed work ticket text as input and give the vector representation of each word in the text information. Each word is represented by a mixture of word vector, segment vector and position vector. The vector representation result is used as the input of the BERT pre-trained language model.

[0077] Step three: Use the BERT pre-trained language model to extract word vectors: convert text information into word vectors, send the word vectors to the pre-trained network model, and obtain the output of the specified feature layer. The output is the word embedding result of the text.

[0078] In some embodiments, Figure 3 Another work flow chart of the intelligent evaluation system for power production PMS business work tickets according to the embodiment of the present invention is as follows: Figure 3 As shown in the figure, the process of eliminating abnormal text information of work tickets based on the Ransac algorithm specifically includes the following steps:

[0079] Step 1, set the number of iterations k of the Ransac algorithm;

[0080] Step 2: randomly select n work ticket text information as a data set, and use the least squares method to fit the preset straight line model based on the data set. The preset straight line model is ax+by+c=0, and the coordinates corresponding to each work ticket text information are , The value range is ;

[0081] Step 3: Determine whether the work ticket text information is an internal point based on the coordinates and the preset straight line model:

[0082] ;

[0083] when , then the work ticket text information is judged to be an internal point. , then it is determined that the work ticket text information is not an internal point, where a, b and c are constants that are not 0;

[0084] Step 4: add 1 to the number of iterations. When the number of iterations is less than k, return to step 2 and calculate the next model. Each iteration corresponds to a preset straight line model. Finally, the model with the most points in the bureau is obtained according to the total number of points in the bureau of each model. The straight line corresponding to the model with the most points in the bureau is recorded as the normal text information straight line of the work ticket.

[0085] Step 5, based on the normal text information straight line of the work ticket, determine whether the work ticket contains abnormal text information: determine whether the vertical distance between the coordinates corresponding to each work ticket text information and the normal text information straight line of the work ticket exceeds the preset distance value. If so, the work ticket text information is abnormal text information and it will be eliminated. If not, the work ticket text information is not abnormal text information and it will be retained.

[0086] In some embodiments, a co-occurrence matrix is ​​constructed based on the word embedding results of the work ticket and the word embedding results of the typical ticket, and the similarity between the work ticket and the typical ticket is calculated based on the co-occurrence matrix, which specifically includes the following process:

[0087] Construct a co-occurrence matrix based on the word embedding results of work tickets and typical tickets :

[0088] ;

[0089] in, The word embedding vector representing the work ticket, express , The word embedding vector representing a typical ticket, express , express The transpose of

[0090] Based on the co-occurrence matrix Calculate the co-occurrence representation index :

[0091] ;

[0092] in, express The number of column vectors, express The number of column vectors, e is 2.72;

[0093] Co-occurrence-based representation index Calculate the similarity between the work ticket and the typical ticket :

[0094] .

[0095] In some embodiments, parsing the work ticket and the work ticket content index to obtain the safety index of the work ticket specifically includes the following process:

[0096] Establish a hierarchical model with a hierarchical structure, where the levels include the goal level, the criterion level and the measure level;

[0097] Constructing a judgment matrix: Determining the number of relevant influencing factors at each level , construct a set of risk factors for work tickets , ,in, is the correct subset of all substation names in the work ticket, is the correct subset of the names of the distribution stations in the work order, is the correctness rate subset of the dispatching power of the substation in the work ticket, is the dispatch power accuracy subset of the distribution station in the work ticket, from the set Take out two subsets in the same level for comparison, assign the corresponding importance according to the preset ratio, combine the importance of each level to form a judgment matrix;

[0098] Calculate the maximum eigenvalue of the judgment matrix :

[0099] ;

[0100] in, is the matrix obtained by normalizing each column vector of the judgment matrix. The value of is 1,2,...,m, For the matrix The elements of are added row by row to obtain the vector and then normalized into a matrix. For the matrix Add the elements of column by column to obtain the vector and then normalize the matrix;

[0101] Consistency test of judgment matrix:

[0102] When the judgment matrix has only one non-zero eigenvalue, it means that the matrix is ​​completely consistent. If the matrix V has more than one eigenvalue, the consistency index can be used :

[0103] ;

[0104] in, represents the order of the judgment matrix, The smaller the value of, the better the consistency, and vice versa, the greater the degree of deviation;

[0105] Will The value is multiplied by the corresponding weight to obtain the safety index of the work ticket.

[0106] In some embodiments, judging whether the simulation training of trainees is qualified based on the similarity and the safety index of the work ticket specifically includes the following process:

[0107] The similarity and the safety index of the work ticket are summed to obtain the sum value;

[0108] The simulation training qualification threshold of the trainee is loaded, and it is judged whether the sum value exceeds the simulation training qualification threshold. If so, it is judged that the simulation training of the trainee is qualified; if not, it is judged that the simulation training of the trainee is unqualified.

[0109] It is worth mentioning that the system's database uses the MySql database management system.

[0110] Furthermore, judging whether the trainee's simulation training is qualified also includes signature verification and time verification of the work ticket:

[0111] Verify the planned working time, the person who issued the work ticket and the time of issuance, the time of receipt of the work ticket, the signature of the operation and maintenance staff on duty, the signature of the work supervisor, the signature of the work permit issuer, the signature of the work supervisor, the permitted start time and the signatures of the work team members. If there are any missing information, the trainee's simulation training will be judged as unqualified.

[0112] In some embodiments, Figure 4 : is a work flow chart of an intelligent evaluation method for a power production PMS business work ticket according to an embodiment of the present invention, such as Figure 4 As shown, the method comprises the following steps:

[0113] Step S401: obtaining the work ticket prepared by the trainee during the simulation training process;

[0114] Step S402: input the work ticket and the typical ticket into the BERT pre-trained language model, output the word embedding results of the work ticket and the typical ticket, construct a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, and calculate the similarity between the work ticket and the typical ticket based on the co-occurrence matrix;

[0115] Step S403: Analyze the work ticket and the work ticket content index to obtain the safety index of the work ticket;

[0116] Step S404: Determine whether the trainee's simulation training is qualified based on the similarity and the safety index of the work ticket.

[0117] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0118] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0121] 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 distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent evaluation system for work tickets of power production PMS business, characterized in that: The system includes a work ticket acquisition module, a similarity calculation module, a safety index calculation module and an intelligent assessment module; The work ticket acquisition module is used to obtain the work tickets prepared by trainees during the simulation training process; The similarity calculation module is used to input the work ticket and the typical ticket into the BERT pre-trained language model, output the word embedding results of the work ticket and the typical ticket, build a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, and calculate the similarity between the work ticket and the typical ticket based on the co-occurrence matrix. The process includes: Construct a co-occurrence matrix based on the word embedding results of work tickets and typical tickets : ； in, The word embedding vector representing the work ticket, express , The word embedding vector representing a typical ticket, express , express The transpose of Based on the co-occurrence matrix Calculate the co-occurrence representation index : ； in, express The number of column vectors, express The number of column vectors, e is 2.72; Co-occurrence-based representation index Calculate the similarity between the work ticket and the typical ticket : ； The safety index calculation module is used to parse the work ticket and the work ticket content indicators to obtain the safety index of the work ticket, which specifically includes the following processes: Establish a hierarchical model with a hierarchical structure, where the levels include the goal level, the criterion level and the measure level; Constructing a judgment matrix: Determining the number of relevant influencing factors at each level , construct a set of risk factors for work tickets , ,in, is the correct subset of all substation names in the work ticket, is the correct subset of the names of the distribution stations in the work order, is the correctness rate subset of the dispatching power of the substation in the work ticket, is the dispatch power accuracy subset of the distribution station in the work ticket, from the set Take out two subsets in the same level for comparison, assign the corresponding importance according to the preset ratio, combine the importance of each level to form a judgment matrix; Calculate the maximum eigenvalue of the judgment matrix : ； in, is the matrix obtained by normalizing each column vector of the judgment matrix. The value of is 1,2,...,m, For the matrix The elements of are added row by row to obtain the vector and then normalized into a matrix. For the matrix Add the elements of column by column to obtain the vector and then normalize the matrix; Consistency test of judgment matrix: When the judgment matrix has only one non-zero eigenvalue, it means that the matrix is ​​completely consistent. If the matrix V has more than one eigenvalue, the consistency index can be used : ； in, represents the order of the judgment matrix, The smaller the value of, the better the consistency, and vice versa, the greater the degree of deviation; Will The value is multiplied by the corresponding weight to obtain the safety index of the work ticket; The intelligent assessment module is used to judge whether the trainees’ simulation training is qualified based on the similarity and safety index of the work ticket.

2. The intelligent evaluation system for power production PMS business work tickets according to claim 1 is characterized in that: Inputting the work ticket and typical ticket into the BERT pre-trained language model and outputting the word embedding results of the work ticket and the typical ticket specifically includes the following process: Step 1: Text preprocessing of work tickets: traverse the text information of the work ticket and remove abnormal text information of the work ticket based on the Ransac algorithm; Step 2: Take the preprocessed work ticket text as input and give the vector representation of each word in the text information. Each word is represented by a mixture of word vector, segment vector and position vector. The vector representation result is used as the input of the BERT pre-trained language model. Step three: Use the BERT pre-trained language model to extract word vectors: convert text information into word vectors, send the word vectors to the pre-trained network model, and obtain the output of the specified feature layer. The output is the word embedding result of the text.

3. The intelligent evaluation system for power production PMS business work tickets according to claim 2 is characterized in that: The specific process of eliminating abnormal text information in work tickets based on the Ransac algorithm includes the following: Step 1, set the number of iterations k of the Ransac algorithm; Step 2: randomly select n work ticket text information as a data set, and use the least squares method to fit the preset straight line model based on the data set. The preset straight line model is ax+by+c=0, and the coordinates corresponding to each work ticket text information are , The value range is ; Step 3, based on And the preset straight line model is used to determine whether the work ticket text information is an internal point: ； when , then the work ticket text information is judged to be an internal point. , then it is determined that the work ticket text information is not an internal point, where a, b and c are constants that are not 0; Step 4: add 1 to the number of iterations. When the number of iterations is less than k, return to step 2 and calculate the next model. Each iteration corresponds to a preset straight line model. Finally, the model with the most points in the bureau is obtained according to the total number of points in the bureau of each model. The straight line corresponding to the model with the most points in the bureau is recorded as the normal text information straight line of the work ticket. Step 5, based on the normal text information straight line of the work ticket, determine whether the work ticket contains abnormal text information: determine whether the vertical distance between the coordinates corresponding to each work ticket text information and the normal text information straight line of the work ticket exceeds the preset distance value. If so, the work ticket text information is abnormal text information and it will be eliminated. If not, the work ticket text information is not abnormal text information and it will be retained.

4. The intelligent evaluation system for power production PMS business work tickets according to claim 1 is characterized in that: Judging whether the trainees' simulation training is qualified based on similarity and the safety index of the work ticket specifically includes the following processes: The similarity and the safety index of the work ticket are summed to obtain the sum value; The simulation training qualification threshold of the trainee is loaded, and it is judged whether the sum value exceeds the simulation training qualification threshold. If so, it is judged that the simulation training of the trainee is qualified; if not, it is judged that the simulation training of the trainee is unqualified.

5. The intelligent evaluation system for power production PMS business work tickets according to claim 1 is characterized in that: The system's database uses the MySql database management system.

6. The intelligent evaluation system for power production PMS business work tickets according to claim 1 is characterized in that: Judging whether the trainees' simulation training is qualified also includes signature verification and time verification of the work ticket.

7. An intelligent evaluation method for power production PMS business work tickets, characterized in that: Based on the intelligent evaluation system for power production PMS business work tickets applicable to any one of claims 1 to 6, the method comprises the following steps: Obtain the work tickets prepared by trainees during the simulation training process; Input the work ticket and typical ticket into the BERT pre-trained language model, output the word embedding results of the work ticket and the typical ticket, build a co-occurrence matrix based on the word embedding results of the work ticket and the typical ticket, and calculate the similarity between the work ticket and the typical ticket based on the co-occurrence matrix; Based on the work ticket and the work ticket content indicators, the safety index of the work ticket is obtained; The similarity and safety index of the work ticket are used to determine whether the trainees have passed the simulation training.

Images