An intelligent anti-misoperation checking method and system for operation tickets

Abstract

This invention discloses an intelligent anti-error verification method and system for operation tickets, relating to the field of operation ticket verification technology. An intelligent anti-error verification system for operation tickets includes an operation ticket information extraction module, an equipment matching and disambiguation module, a coupled risk analysis module, and a comprehensive risk verification module. This invention, by coupling health status and operational intensity through Copula function-based modeling, can capture the nonlinear fault risk amplification effect of equipment under high-intensity operation. This upgrades the traditional static mode of isolated equipment status assessment to a dynamic coupled risk assessment, making the calculated equipment fault risk probability more closely resemble the objective law in actual engineering experience that the older the equipment and the higher the operational intensity, the exponentially increasing the equipment fault probability. This improves the safety verification effect of this intelligent anti-error verification method and system for operation tickets.

Description

Technical Field

[0001] This invention relates to the field of operation ticket verification technology, specifically to an intelligent anti-error verification method and system for operation tickets. Background Technology

[0002] Operation tickets are written instructions in power systems used to standardize the operation procedures of electrical equipment. They clearly define the operation sequence, steps, and safety measures to ensure that operators follow the predetermined procedures. Verification of operation tickets to prevent errors is extremely necessary because misoperation (such as incorrect sequence, omitted steps, and logical conflicts) can easily lead to equipment damage, power outages, and even personal injury or death. Verification can automatically identify risks such as logical errors, insufficient safety distances, and conflicting equipment states in operation tickets, enabling mandatory pre-operation interlocking and early warning. This is a key technical safeguard for ensuring the safe and stable operation of the power grid and preventing human error.

[0003] Existing intelligent error prevention verification methods and systems for operation tickets typically only verify the text on the operation ticket, the operation logic and sequence of the equipment, the operator's qualifications, and the equipment's state before and after the operation. However, the correct operation logic and sequence of the equipment does not guarantee that the equipment will not be damaged due to its lifespan or the intensity of the operation. Similarly, even if the operator's qualifications meet the requirements, it is difficult to guarantee that the operator will not make mistakes due to differences in personal ability, physical fatigue, or the complexity of the operation. Therefore, existing intelligent error prevention verification methods and systems for operation tickets lack coupled assessment of implicit risks such as the intensity of operation and the health of the equipment, resulting in poor verification effect for operation safety.

[0004] Based on the above, this invention proposes an intelligent anti-error verification method and system for operation tickets with high security requirements. Summary of the Invention

[0005] To overcome the shortcomings of existing intelligent error prevention verification methods and systems for operation tickets, which typically only verify the text on the operation ticket, the operation logic and sequence of the equipment, the operator's qualifications, and the state of the equipment before and after the operation, this invention proposes an intelligent error prevention verification method and system for operation tickets with high security requirements. However, the correct operation logic and sequence of the equipment does not guarantee that the equipment will not be damaged due to its lifespan and the intensity of operation. Similarly, even if the operator's qualifications meet the requirements, it is difficult to guarantee that the operator will not make mistakes due to differences in personal ability, physical fatigue, and the complexity of the operation. Therefore, existing intelligent error prevention verification methods and systems for operation tickets lack coupled assessment of implicit risks such as the intensity of operation and the health of the equipment, resulting in poor verification effect for operation safety.

[0006] A method for intelligent error prevention and verification of operation tickets includes the following steps: Obtain the operation ticket to be verified against false errors, extract information from the text on the operation ticket using a pre-trained BERT model, obtain the operation task information corresponding to the operation ticket, perform statistical quantitative analysis and evaluation on the operation task information, and obtain the operation intensity data of each operation device and the operation complexity data of the operation task. A device name database is constructed, and multi-level matching and disambiguation of various operation device names in the operation task information are performed based on hash index, graph neural network and human-machine collaboration to obtain the matching results of operation device names; The system acquires the current status parameters of various operating equipment and quantifies the health status of the operating equipment. Based on the Copula function, it assesses the coupled equipment failure risk corresponding to the health status quantification result and the operation intensity, and calculates the equipment failure risk probability of the operation task according to probability. The system acquires the current physiological data of the operator and quantifies the operator's fatigue status. Based on historical operation data, it performs logistic regression fitting on the coupled misoperation risk corresponding to the fatigue status quantification result and the operation complexity, and obtains the personnel misoperation risk probability of the operation task. The failure risk of the operation task corresponding to the operation ticket is comprehensively assessed based on the probability of equipment failure and the probability of human error. The failure risk of the operation ticket is then verified based on the comprehensive assessment results to obtain the risk verification result of the operation ticket.

[0007] As a preferred aspect of the invention, the specific steps of extracting information from the text on the operation ticket using a pre-trained BERT model to obtain the operation task information corresponding to the operation ticket, and performing statistical quantitative analysis and evaluation of the operation task information to obtain the operation intensity data of each operation device and the operation complexity data of the operation task are as follows: Input the operation ticket to be prevented from being falsely verified into the pre-trained BERT model to obtain the operation task information corresponding to the operation ticket. The operation task information includes operator information, operation condition information, operation sequence information, operation equipment name information, and corresponding operation action information. Based on the name information of the operating equipment and the corresponding operation action information, the operation frequency of each operating equipment is obtained, and the ratio of the operation frequency to the preset daily or hourly operation limit of each operating equipment is calculated to obtain the operation intensity data of each operating equipment. Based on the operation sequence information, operation device name information, and corresponding operation action information, obtain the total number of operation steps, the number of parallel operation steps, and the number of conditional branch steps for this operation task. Then, perform a normalized weighted sum on the total number of operation steps, the number of parallel operation steps, and the number of conditional branch steps to obtain the operation complexity data of the operation task.

[0008] As a preferred aspect of the invention, the specific steps for constructing a device name database and performing multi-level matching and disambiguation on various operation device names in the operation task information based on hash indexes, graph neural networks, and human-machine collaboration to obtain the matching results of the operation device names are as follows: Construct a device name database, in which the corresponding name of each operating device includes a standard name and an alias. Construct a device name hash table based on the device name database, where the key is the standard name and alias of the operating device, and the value is the unique ID of the device. The operation device name information is matched based on the device name hash table. When the edit distance between the operation device name in the operation device name information and the hash table key is less than a preset threshold and the semantic vector similarity is greater than a preset threshold, the match is successful. For operation device names that are not directly matched, all successfully identified operation device names, operation actions, and operation sequences in the operation task information are first input into a graph attention network. The nodes of the graph attention network are candidate operation devices, and the edges are the power grid topology connections between operation devices. Then, the graph attention network is used to perform topology context disambiguation and matching on the operation device names that are not directly matched, so that operation device names that meet the preset matching rules are indirectly matched. For operation device names that are not indirectly matched, manual confirmation is forcibly triggered to obtain the matching result of the operation device name.

[0009] As a preferred aspect of the invention, the specific steps of obtaining the current state parameters of each operating device and quantifying the health status of the operating device, evaluating the coupled device failure risk corresponding to the health status quantification result and the operating intensity based on the Copula function, and calculating the device failure risk probability of the operating task according to probability are as follows: Based on the matching results of the names of the operating devices, the current status parameters of each operating device are obtained, including the temperature, vibration amplitude and vibration frequency of the operating device. The health status of the operating device is quantitatively evaluated according to the deviation of the current status parameters of the operating device from the preset health status parameters, and the health status index of each operating device is obtained. For each piece of equipment, normal operation records and operational failure records are selected from historical operation data and formed into normal subsets and failure subsets, respectively. Log-normal distributions are fitted to the health status index data of the equipment corresponding to the normal subsets and failure subsets, while Beta distributions are fitted to the operational intensity data of the equipment. After obtaining the distribution parameters through maximum likelihood estimation, the cumulative distribution functions of health status and operational intensity corresponding to the normal subsets and failure subsets are obtained. Appropriate Copula functions are selected based on the distribution characteristics of the normal subsets and failure subsets, and function fitting is performed using historical operation data to obtain the Copula fitting functions corresponding to the normal subsets and failure subsets, respectively. The health status index and operational intensity of the equipment are substituted into the cumulative distribution functions of health status, operational intensity, and Copula fitting functions corresponding to the normal subsets and failure subsets, respectively, and the current joint density of normal and joint density of failure of the equipment are calculated. The normal operation probability and operational failure probability of the equipment are obtained from historical operation data, and the failure risk probability of the equipment under this health status index and operational intensity condition is calculated based on Bayes' theorem. The failure risk probability of each piece of operating equipment is analyzed using probabilistics, and the failure risk probability of the equipment for the operating task is obtained through calculation. The specific calculation formula is as follows: ,

[0010] in This indicates the probability of equipment failure during the operation task. Indicates the first operation task The probability of failure of the operating equipment. This indicates the total number of devices operated in the task.

[0011] As a preferred aspect of the invention, the specific steps for acquiring the operator's current physiological data and quantifying the operator's fatigue state, and performing logistic regression fitting on the coupled error risk corresponding to the fatigue state quantification result and the operation complexity based on historical operation data to obtain the probability of personnel error risk for the operation task are as follows: Acquire various physiological data of the operator, including the operator's heart rate variability, skin conductance response and electromyography signal. Quantitatively assess the operator's fatigue state based on the deviation of the operator's current physiological data from preset normal physiological data, and obtain the operator's fatigue state index. Logistic regression was performed on the operator's historical operation data to obtain the operator's error probability function. The specific formula of the function is as follows:

[0012] in This represents the probability of operator error. This indicates the operator's fatigue level index. This indicates the operational complexity of the task. , and Let be the fitting coefficients, and This indicates the coupling strength between the operator's fatigue index and the operational complexity of the task. By substituting the operator's current fatigue index and the operational complexity of the task into the error probability function, the probability of human error risk for the task can be obtained.

[0013] As a preferred aspect of the invention, the specific steps for comprehensively assessing the failure risk of the operation task corresponding to the operation ticket based on the probability of equipment failure risk and the probability of human error operation risk, and verifying the failure risk of the operation ticket based on the comprehensive assessment result to obtain the risk verification result of the operation ticket are as follows: A probabilistic analysis of the probability of equipment failure and the probability of human error are performed, and the probability of failure of the operation task corresponding to the operation ticket is calculated. The specific calculation formula is as follows:

[0014] in Indicates the probability of failure of the operation task; The operation ticket is checked for failure risk based on the failure risk probability of the operation task and the preset risk verification rules. When the failure risk probability is lower than the preset threshold, the failure risk verification result of the operation ticket is passed. When the failure risk probability is equal to or higher than the preset threshold, the failure risk verification result of the operation ticket is failed. The operation ticket is classified into failure risk levels based on the difference between the failure risk probability and the preset threshold, and a corresponding suggested handling plan is generated.

[0015] An intelligent error prevention and verification system for operation tickets includes: The operation ticket information extraction module is used to obtain the operation ticket to be verified against false errors. It extracts information from the text on the operation ticket using a pre-trained BERT model to obtain the operation task information corresponding to the operation ticket. It performs statistical quantitative analysis and evaluation on the operation task information to obtain the operation intensity data of various operation equipment and the operation complexity data of the operation task. The device matching and disambiguation module is used to build a device name database and perform multi-level matching and disambiguation on the names of various operation devices in the operation task information based on hash index, graph neural network and human-machine collaboration to obtain the matching results of operation device names; The coupled risk analysis module includes an equipment failure risk analysis unit and a personnel misoperation risk analysis unit. The equipment failure risk analysis unit is used to acquire the current status parameters of various operating equipment and quantify the health status of the operating equipment. Based on the Copula function, it evaluates the coupled equipment failure risk corresponding to the health status quantification result and the operation intensity, and calculates the equipment failure risk probability of the operation task according to probability. The personnel misoperation risk analysis unit is used to acquire the current physiological data of the operator and quantify the operator's fatigue status. Based on historical operation data, it performs logistic regression fitting on the coupled misoperation risk corresponding to the fatigue status quantification result and the operation complexity to obtain the personnel misoperation risk probability of the operation task. The comprehensive risk verification module is used to comprehensively assess the failure risk of the operation task corresponding to the operation ticket based on the probability of equipment failure risk and the probability of human error operation risk. Based on the comprehensive assessment results, the operation ticket is verified for failure risk, and the risk verification result of the operation ticket is obtained.

[0016] The present invention has the following advantages: 1. This invention uses Copula functions to couple health status and operational intensity modeling, which can capture the nonlinear failure risk amplification effect of equipment under high-intensity operation. This upgrades the traditional static mode of isolated equipment status assessment to dynamic coupled risk assessment, making the equipment failure risk probability calculation results closer to the objective law in actual engineering experience that the older the equipment and the higher the operational intensity, the exponential increase in the equipment failure probability. This improves the safety verification effect of this intelligent anti-misoperation verification method and system for operation tickets.

[0017] 2. This invention quantifies fatigue status by collecting real-time physiological data such as operator heart rate variability and skin conductance response, and obtains the coupling strength coefficient between fatigue and operational complexity by performing logistic regression fitting based on historical operation data. This not only breaks through the traditional static management model that only checks operator qualification certificates, but also dynamically senses changes in the operator's physiological cognitive load during task execution. As a result, the calculation results of the probability of personnel misoperation risk are closer to the objective law in actual engineering experience that the more fatigued the operator and the higher the operation complexity, the explosive increase in the probability of personnel misoperation risk. This improves the security verification effect of this intelligent anti-misoperation verification method and system for operation tickets.

[0018] 3. This invention constructs a device name database and builds a three-level matching and disambiguation architecture based on hash index, graph neural network, and human-machine collaboration. This not only enables fast and accurate matching of common device names, but also performs topological context disambiguation and intelligent inference on device names that use device aliases, abbreviations, or have input errors. This significantly improves the accuracy of device name matching, completely solves the problem of missed detection caused by device aliases and ambiguous descriptions that traditional systems cannot handle, and enhances the verification effect of this intelligent anti-error verification method and system for operation tickets. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating an intelligent error prevention and verification method for operation tickets used in an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of an intelligent anti-error verification system for operation tickets used in an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this invention.

[0022] Example 1: A method for intelligent error prevention and verification of operation tickets, such as... Figure 1 As shown, it includes the following steps: Obtain the operation ticket to be verified against false errors, extract information from the text on the operation ticket using a pre-trained BERT model, obtain the operation task information corresponding to the operation ticket, perform statistical quantitative analysis and evaluation on the operation task information, and obtain the operation intensity data of each operation device and the operation complexity data of the operation task. A device name database is constructed, and multi-level matching and disambiguation of various operation device names in the operation task information are performed based on hash index, graph neural network and human-machine collaboration to obtain the matching results of operation device names; The system acquires the current status parameters of various operating equipment and quantifies the health status of the operating equipment. Based on the Copula function, it assesses the coupled equipment failure risk corresponding to the health status quantification result and the operation intensity, and calculates the equipment failure risk probability of the operation task according to probability. The system acquires the current physiological data of the operator and quantifies the operator's fatigue status. Based on historical operation data, it performs logistic regression fitting on the coupled misoperation risk corresponding to the fatigue status quantification result and the operation complexity, and obtains the personnel misoperation risk probability of the operation task. The failure risk of the operation task corresponding to the operation ticket is comprehensively assessed based on the probability of equipment failure and the probability of human error. The failure risk of the operation ticket is then verified based on the comprehensive assessment results to obtain the risk verification result of the operation ticket.

[0023] The specific steps for extracting information from the text on the operation ticket using a pre-trained BERT model to obtain the corresponding operation task information, and then performing statistical quantitative analysis and evaluation of the operation task information to obtain the operation intensity data of each operation device and the operation complexity data of the operation task are as follows: Input the operation ticket to be prevented from being falsely verified into the pre-trained BERT model to obtain the operation task information corresponding to the operation ticket. The operation task information includes operator information, operation condition information, operation sequence information, operation equipment name information, and corresponding operation action information. Based on the name information of the operating equipment and the corresponding operation action information, the operation frequency of each operating equipment is obtained, and the ratio of the operation frequency to the preset daily or hourly operation limit of each operating equipment is calculated to obtain the operation intensity data of each operating equipment. Based on the operation sequence information, operation device name information, and corresponding operation action information, obtain the total number of operation steps, the number of parallel operation steps, and the number of conditional branch steps for this operation task. Then, perform a normalized weighted sum on the total number of operation steps, the number of parallel operation steps, and the number of conditional branch steps to obtain the operation complexity data of the operation task.

[0024] The specific steps for constructing the device name database and performing multi-level matching and disambiguation of various operation device names in the operation task information based on hash indexes, graph neural networks, and human-machine collaboration to obtain the matching results of operation device names are as follows: Construct a device name database, in which the corresponding name of each operating device includes a standard name and an alias. Construct a device name hash table based on the device name database, where the key is the standard name and alias of the operating device, and the value is the unique ID of the device. The operation device name information is matched based on the device name hash table. When the edit distance between the operation device name in the operation device name information and the hash table key is less than a preset threshold and the semantic vector similarity is greater than a preset threshold, the match is successful. For operation device names that are not directly matched, all successfully identified operation device names, operation actions, and operation sequences in the operation task information are first input into a graph attention network. The nodes of the graph attention network are candidate operation devices, and the edges are the power grid topology connections between operation devices. Then, the graph attention network is used to perform topology context disambiguation and matching on the operation device names that are not directly matched, so that operation device names that meet the preset matching rules are indirectly matched. For operation device names that are not indirectly matched, manual confirmation is forcibly triggered to obtain the matching result of the operation device name.

[0025] For example, a typical case of using graph attention networks to perform topological context disambiguation and matching on operation device names that did not directly match is as follows: When the operation ticket states "open the main transformer side disconnect switch" without specifying the side, and the preceding step is "check that the 101 switch on the 110kV side of the #1 main transformer is open", first, based on the keyword "main transformer", all main transformer nodes in the substation (such as #1 main transformer, #2 main transformer) are filtered. Then, the topological distance from each main transformer to the equipment specified in the operation ticket (such as "101 circuit breaker") is calculated. It is found that there is a direct path between the #1 main transformer and the 101 bay, while the #2 main transformer has no direct connection. At this time, the topological connection strength is automatically learned through GAT, so that the #1 main transformer receives a higher attention weight. Finally, the probability distribution is output as P(#1 main transformer side disconnect switch) = 0.82, P(#2 main transformer side disconnect switch) = 0.13. When the highest probability > 0.75 and the leading second-best probability > 0.2, it is determined that the #1 main transformer side disconnect switch is open.

[0026] The above steps, by constructing a device name database and building a three-level matching and disambiguation architecture of hash index, graph neural network and human-machine collaboration, can not only achieve fast and accurate matching of common device names, but also perform topological context disambiguation and intelligent inference for device names that reference device aliases, abbreviations or have input errors. This significantly improves the accuracy of device name matching, completely solves the problem of missed detection caused by device aliases and ambiguous descriptions that traditional systems cannot handle, and improves the verification effect of this intelligent anti-misoperation verification method and system for operation tickets.

[0027] The specific steps for obtaining the current status parameters of each operating device and quantifying the health status of the operating device, evaluating the coupled device failure risk corresponding to the health status quantification result and operation intensity based on the Copula function, and calculating the device failure risk probability of the operation task according to probability are as follows: Based on the matching results of the names of the operating devices, the current status parameters of each operating device are obtained, including the temperature, vibration amplitude and vibration frequency of the operating device. The health status of the operating device is quantitatively evaluated according to the deviation of the current status parameters of the operating device from the preset health status parameters, and the health status index of each operating device is obtained. For each piece of equipment, normal operation records and operational failure records are selected from historical operation data and formed into normal subsets and failure subsets, respectively. Log-normal distributions are fitted to the health status index data of the equipment corresponding to the normal subsets and failure subsets, while Beta distributions are fitted to the operational intensity data of the equipment. After obtaining the distribution parameters through maximum likelihood estimation, the cumulative distribution functions of health status and operational intensity corresponding to the normal subsets and failure subsets are obtained. Appropriate Copula functions are selected based on the distribution characteristics of the normal subsets and failure subsets, and function fitting is performed using historical operation data to obtain the Copula fitting functions corresponding to the normal subsets and failure subsets, respectively. The health status index and operational intensity of the equipment are substituted into the cumulative distribution functions of health status, operational intensity, and Copula fitting functions corresponding to the normal subsets and failure subsets, respectively, and the current joint density of normal and joint density of failure of the equipment are calculated. The normal operation probability and operational failure probability of the equipment are obtained from historical operation data, and the failure risk probability of the equipment under this health status index and operational intensity condition is calculated based on Bayes' theorem. The failure risk probability of each piece of operating equipment is analyzed using probabilistics, and the failure risk probability of the equipment for the operating task is obtained through calculation. The specific calculation formula is as follows:

[0028] in This indicates the probability of equipment failure during the operation task. Indicates the first operation task The probability of failure of the operating equipment. This indicates the total number of devices operated in the task.

[0029] The above steps, by coupling the health status and operational intensity modeling based on the Copula function, can capture the nonlinear failure risk amplification effect of equipment under high-intensity operation. This upgrades the traditional static mode of isolated equipment status assessment to dynamic coupled risk assessment, making the equipment failure risk probability calculation results closer to the objective law in actual engineering experience that the older the equipment and the higher the operational intensity, the exponential increase in the equipment failure probability. This improves the safety verification effect of this intelligent anti-misoperation verification method and system for operation tickets.

[0030] The specific steps for obtaining the operator's current physiological data and quantifying the operator's fatigue state, and then performing logistic regression fitting on the coupled error risk corresponding to the fatigue state quantification result and the operation complexity based on historical operation data to obtain the probability of personnel error risk for the operation task are as follows: Acquire various physiological data of the operator, including the operator's heart rate variability, skin conductance response and electromyography signal. Quantitatively assess the operator's fatigue state based on the deviation of the operator's current physiological data from preset normal physiological data, and obtain the operator's fatigue state index. Logistic regression was performed on the operator's historical operation data to obtain the operator's error probability function. The specific formula of the function is as follows:

[0031] in This represents the probability of operator error. This indicates the operator's fatigue level index. This indicates the operational complexity of the task. , and Let be the fitting coefficients, and This indicates the coupling strength between the operator's fatigue index and the operational complexity of the task. By substituting the operator's current fatigue index and the operational complexity of the task into the error probability function, the probability of human error risk for the task can be obtained.

[0032] The above steps quantify fatigue status by collecting real-time physiological data such as operator heart rate variability and skin conductance response, and obtain the coupling strength coefficient between fatigue and operational complexity by performing logistic regression fitting based on historical operation data. This not only breaks through the traditional static management model that only checks operator qualification certificates, but also dynamically senses the changes in physiological cognitive load of operators during task execution. As a result, the calculation results of the probability of personnel misoperation risk are closer to the objective law in actual engineering experience that the more fatigued the operator and the higher the operation complexity, the explosive increase in the probability of personnel misoperation risk. This improves the security verification effect of this intelligent anti-misoperation verification method and system for operation tickets.

[0033] The specific steps for comprehensively assessing the failure risk of the operation task corresponding to the operation ticket based on the probability of equipment failure risk and the probability of human error operation risk, and verifying the failure risk of the operation ticket based on the comprehensive assessment result, are as follows: A probabilistic analysis of the probability of equipment failure and the probability of human error are performed, and the probability of failure of the operation task corresponding to the operation ticket is calculated. The specific calculation formula is as follows:

[0034] in Indicates the probability of failure of the operation task; The operation ticket is checked for failure risk based on the failure risk probability of the operation task and the preset risk verification rules. When the failure risk probability is lower than the preset threshold, the failure risk verification result of the operation ticket is passed. When the failure risk probability is equal to or higher than the preset threshold, the failure risk verification result of the operation ticket is failed. The operation ticket is classified into failure risk levels based on the difference between the failure risk probability and the preset threshold, and a corresponding suggested handling plan is generated.

[0035] It should be noted that the comprehensive assessment method for the failure risk probability of the above-mentioned operation tasks is more suitable for indoor operation tasks and outdoor operation tasks that are less affected by environmental factors. However, for outdoor operation tasks that are greatly affected by environmental factors, the impact of environmental factors needs to be fully considered when comprehensively assessing the failure risk probability of the operation task.

[0036] Example 2: An intelligent error prevention and verification system for operation tickets, such as... Figure 2 As shown, it includes: The operation ticket information extraction module is used to obtain the operation ticket to be verified against false errors. It extracts information from the text on the operation ticket using a pre-trained BERT model to obtain the operation task information corresponding to the operation ticket. It performs statistical quantitative analysis and evaluation on the operation task information to obtain the operation intensity data of various operation equipment and the operation complexity data of the operation task. The device matching and disambiguation module is used to build a device name database and perform multi-level matching and disambiguation on the names of various operation devices in the operation task information based on hash index, graph neural network and human-machine collaboration to obtain the matching results of operation device names; The coupled risk analysis module includes an equipment failure risk analysis unit and a personnel misoperation risk analysis unit. The equipment failure risk analysis unit is used to acquire the current status parameters of various operating equipment and quantify the health status of the operating equipment. Based on the Copula function, it evaluates the coupled equipment failure risk corresponding to the health status quantification result and the operation intensity, and calculates the equipment failure risk probability of the operation task according to probability. The personnel misoperation risk analysis unit is used to acquire the current physiological data of the operator and quantify the operator's fatigue status. Based on historical operation data, it performs logistic regression fitting on the coupled misoperation risk corresponding to the fatigue status quantification result and the operation complexity to obtain the personnel misoperation risk probability of the operation task. The comprehensive risk verification module is used to comprehensively assess the failure risk of the operation task corresponding to the operation ticket based on the probability of equipment failure risk and the probability of human error operation risk. Based on the comprehensive assessment results, the operation ticket is verified for failure risk, and the risk verification result of the operation ticket is obtained.

[0037] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims. Parts not described in detail in this specification are prior art known to those skilled in the art.

Claims

1. A method for intelligent error prevention and verification of operation tickets, characterized in that, Includes the following steps: Obtain the operation ticket to be verified against false errors, extract information from the text on the operation ticket using a pre-trained BERT model, obtain the operation task information corresponding to the operation ticket, perform statistical quantitative analysis and evaluation on the operation task information, and obtain the operation intensity data of each operation device and the operation complexity data of the operation task. A device name database is constructed, and multi-level matching and disambiguation of various operation device names in the operation task information are performed based on hash index, graph neural network and human-machine collaboration to obtain the matching results of operation device names; The system acquires the current status parameters of various operating equipment and quantifies the health status of the operating equipment. Based on the Copula function, it assesses the coupled equipment failure risk corresponding to the health status quantification result and the operation intensity, and calculates the equipment failure risk probability of the operation task according to probability. The system acquires the current physiological data of the operator and quantifies the operator's fatigue status. Based on historical operation data, it performs logistic regression fitting on the coupled misoperation risk corresponding to the fatigue status quantification result and the operation complexity, and obtains the personnel misoperation risk probability of the operation task. The failure risk of the operation task corresponding to the operation ticket is comprehensively assessed based on the probability of equipment failure and the probability of human error. The failure risk of the operation ticket is then verified based on the comprehensive assessment results to obtain the risk verification result of the operation ticket.

2. The intelligent anti-error verification method for operation tickets according to claim 1, characterized in that, The specific steps for extracting information from the text on the operation ticket using a pre-trained BERT model to obtain the corresponding operation task information, and then performing statistical quantitative analysis and evaluation of the operation task information to obtain the operation intensity data of each operation device and the operation complexity data of the operation task are as follows: Input the operation ticket to be prevented from being falsely verified into the pre-trained BERT model to obtain the operation task information corresponding to the operation ticket. The operation task information includes operator information, operation condition information, operation sequence information, operation equipment name information, and corresponding operation action information. Based on the name information of the operating equipment and the corresponding operation action information, the operation frequency of each operating equipment is obtained, and the ratio of the operation frequency to the preset daily or hourly operation limit of each operating equipment is calculated to obtain the operation intensity data of each operating equipment. Based on the operation sequence information, operation device name information, and corresponding operation action information, obtain the total number of operation steps, the number of parallel operation steps, and the number of conditional branch steps for this operation task. Then, perform a normalized weighted sum on the total number of operation steps, the number of parallel operation steps, and the number of conditional branch steps to obtain the operation complexity data of the operation task.

3. The intelligent anti-error verification method for operation tickets according to claim 2, characterized in that, The specific steps for constructing the device name database and performing multi-level matching and disambiguation of various operation device names in the operation task information based on hash indexes, graph neural networks, and human-machine collaboration to obtain the matching results of operation device names are as follows: Construct a device name database, in which the corresponding name of each operating device includes a standard name and an alias. Construct a device name hash table based on the device name database, where the key is the standard name and alias of the operating device, and the value is the unique ID of the device. The operation device name information is matched based on the device name hash table. When the edit distance between the operation device name in the operation device name information and the hash table key is less than a preset threshold and the semantic vector similarity is greater than a preset threshold, the match is successful. For operation device names that are not directly matched, all successfully identified operation device names, operation actions, and operation sequences in the operation task information are first input into a graph attention network. The nodes of the graph attention network are candidate operation devices, and the edges are the power grid topology connections between operation devices. Then, the graph attention network is used to perform topology context disambiguation and matching on the operation device names that are not directly matched, so that operation device names that meet the preset matching rules are indirectly matched. For operation device names that are not indirectly matched, manual confirmation is forcibly triggered to obtain the matching result of the operation device name.

4. The intelligent anti-error verification method for operation tickets according to claim 3, characterized in that, The specific steps for obtaining the current status parameters of each operating device and quantifying the health status of the operating device, evaluating the coupled device failure risk corresponding to the health status quantification result and operation intensity based on the Copula function, and calculating the device failure risk probability of the operation task according to probability are as follows: Based on the matching results of the names of the operating devices, the current status parameters of each operating device are obtained, including the temperature, vibration amplitude and vibration frequency of the operating device. The health status of the operating device is quantitatively evaluated according to the deviation of the current status parameters of the operating device from the preset health status parameters, and the health status index of each operating device is obtained. For each piece of equipment, normal operation records and operational failure records are selected from historical operation data and formed into normal subsets and failure subsets, respectively. Log-normal distributions are fitted to the health status index data of the equipment corresponding to the normal subsets and failure subsets, while Beta distributions are fitted to the operational intensity data of the equipment. After obtaining the distribution parameters through maximum likelihood estimation, the cumulative distribution functions of health status and operational intensity corresponding to the normal subsets and failure subsets are obtained. Appropriate Copula functions are selected based on the distribution characteristics of the normal subsets and failure subsets, and function fitting is performed using historical operation data to obtain the Copula fitting functions corresponding to the normal subsets and failure subsets, respectively. The health status index and operational intensity of the equipment are substituted into the cumulative distribution functions of health status, operational intensity, and Copula fitting functions corresponding to the normal subsets and failure subsets, respectively, and the current joint density of normal and joint density of failure of the equipment are calculated. The normal operation probability and operational failure probability of the equipment are obtained from historical operation data, and the failure risk probability of the equipment under this health status index and operational intensity condition is calculated based on Bayes' theorem. The failure risk probability of each piece of operating equipment is analyzed using probabilistics, and the failure risk probability of the equipment for the operating task is obtained through calculation. The specific calculation formula is as follows: ， in This indicates the probability of equipment failure during the operation task. Indicates the first operation task The probability of failure of the operating equipment. This indicates the total number of devices operated in the task.

5. The intelligent anti-error verification method for operation tickets according to claim 4, characterized in that, The specific steps for obtaining the operator's current physiological data and quantifying the operator's fatigue state, and then performing logistic regression fitting on the coupled error risk corresponding to the fatigue state quantification result and the operation complexity based on historical operation data to obtain the probability of personnel error risk for the operation task are as follows: Acquire various physiological data of the operator, including the operator's heart rate variability, skin conductance response and electromyography signal. Quantitatively assess the operator's fatigue state based on the deviation of the operator's current physiological data from preset normal physiological data, and obtain the operator's fatigue state index. Logistic regression was performed on the operator's historical operation data to obtain the operator's error probability function. The specific formula of the function is as follows: ， in This represents the probability of operator error. This indicates the operator's fatigue level index. This indicates the operational complexity of the task. , and Let be the fitting coefficients, and This indicates the coupling strength between the operator's fatigue index and the operational complexity of the task. By substituting the operator's current fatigue index and the operational complexity of the task into the error probability function, the probability of human error risk for the task can be obtained.

6. The intelligent anti-error verification method for operation tickets according to claim 5, characterized in that, The specific steps for comprehensively assessing the failure risk of the operation task corresponding to the operation ticket based on the probability of equipment failure risk and the probability of human error operation risk, and verifying the failure risk of the operation ticket based on the comprehensive assessment result, are as follows: A probabilistic analysis of the probability of equipment failure and the probability of human error are performed, and the probability of failure of the operation task corresponding to the operation ticket is calculated. The specific calculation formula is as follows: ， in Indicates the probability of failure of the operation task; The operation ticket is checked for failure risk based on the failure risk probability of the operation task and the preset risk verification rules. When the failure risk probability is lower than the preset threshold, the failure risk verification result of the operation ticket is passed. When the failure risk probability is equal to or higher than the preset threshold, the failure risk verification result of the operation ticket is failed. The operation ticket is classified into failure risk levels based on the difference between the failure risk probability and the preset threshold, and a corresponding suggested handling plan is generated.

7. An intelligent error prevention and verification system for operation tickets, applied to the intelligent error prevention and verification method for operation tickets as described in any one of claims 1-6, characterized in that, Including: The operation ticket information extraction module is used to obtain the operation ticket to be verified against false errors. It extracts information from the text on the operation ticket using a pre-trained BERT model to obtain the operation task information corresponding to the operation ticket. It performs statistical quantitative analysis and evaluation on the operation task information to obtain the operation intensity data of various operation equipment and the operation complexity data of the operation task. The device matching and disambiguation module is used to build a device name database and perform multi-level matching and disambiguation on the names of various operation devices in the operation task information based on hash index, graph neural network and human-machine collaboration to obtain the matching results of operation device names; The coupled risk analysis module includes an equipment failure risk analysis unit and a personnel misoperation risk analysis unit. The equipment failure risk analysis unit is used to acquire the current status parameters of various operating equipment and quantify the health status of the operating equipment. Based on the Copula function, it evaluates the coupled equipment failure risk corresponding to the health status quantification result and the operation intensity, and calculates the equipment failure risk probability of the operation task according to probability. The personnel misoperation risk analysis unit is used to acquire the current physiological data of the operator and quantify the operator's fatigue status. Based on historical operation data, it performs logistic regression fitting on the coupled misoperation risk corresponding to the fatigue status quantification result and the operation complexity to obtain the personnel misoperation risk probability of the operation task. The comprehensive risk verification module is used to comprehensively assess the failure risk of the operation task corresponding to the operation ticket based on the probability of equipment failure risk and the probability of human error operation risk. Based on the comprehensive assessment results, the operation ticket is verified for failure risk, and the risk verification result of the operation ticket is obtained.

Images