Digital management method and system for whole process of relay protection debugging

Abstract

The invention discloses a relay protection debugging full-process digital management method and system, and relates to the technical field of relay protection intelligent operation and maintenance, and the method comprises the steps: obtaining the real-time operation data of relay protection equipment, and carrying out the feature extraction based on the real-time data; establishing a performance score judgment mechanism, calculating a performance score based on the data after feature extraction, and judging whether the score is lower than a preset performance threshold; generating a maintenance decision vector through a Bayesian network algorithm, executing a maintenance action in the maintenance decision vector, and obtaining data after maintenance; and recalculating the performance score according to the maintained data, and verifying the maintenance effect. Multi-dimensional characteristic indexes such as the harmonic distortion rate and the output fluctuation intensity are introduced, the performance scoring model is established by fusing the entropy weight method, and the health state of the relay protection device is objectively reflected; and dynamically calculating benefit weights of different schemes in combination with the Bayesian network and historical maintenance data to realize optimal recommendation of the maintenance schemes.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent operation and maintenance of relay protection, and in particular to a full-process digital management method and system for relay protection debugging. Background Art

[0002] Relay protection is a key component of power systems, ensuring safe equipment operation and preventing faults from escalating. The correct configuration, precise commissioning, and timely maintenance of protection devices are directly related to the stability and power supply security of the power system. With the increasing complexity of power system structures, the massive integration of distributed energy resources, and the continuous advancement of smart grid construction, the commissioning and maintenance of relay protection systems face the following significant challenges: On the one hand, traditional relay protection debugging methods rely heavily on manual experience and offline test data, lack the ability to dynamically perceive the real-time operating status of the equipment, and find it difficult to comprehensively evaluate the accuracy and matching of protection actions; on the other hand, the maintenance decision-making process is usually static, failing to fully integrate the current status of the equipment, historical operating characteristics, and maintenance effect feedback, resulting in strategic lags or resource waste in some maintenance behaviors. Summary of the Invention

[0003] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

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

[0005] To solve the above technical problems, the present invention provides the following technical solution: a digital management method for the entire process of relay protection commissioning, characterized in that the method comprises the following steps: Step 1: Acquire real-time operating data of the relay protection device and perform feature extraction based on the real-time data; Step 2: Establish a performance scoring mechanism to calculate the performance score based on the feature-extracted data and determine whether the score is lower than the preset performance threshold; Step 3: If the score is lower than the preset performance threshold, a maintenance decision vector is generated using a Bayesian network algorithm based on the score and the historical maintenance database. The maintenance action in the maintenance decision vector is executed, and post-maintenance data is obtained. Step 4: Recalculate the performance score based on the data after maintenance and verify the maintenance effect.

[0006] As a preferred solution of the digital management method for the whole process of relay protection debugging described in the present invention, the real-time operation data includes: voltage / current waveform data collected from smart meters , action record data collected from relay protection equipment, including action time and the number of errors E, as well as the output curve data collected from the distributed energy controller ; right Perform feature extraction to obtain the harmonic distortion rate H; right Perform feature extraction to obtain the fluctuation intensity coefficient .

[0007] As a preferred solution of the digital management method for the entire process of relay protection commissioning described in the present invention, the calculation process of the performance score is as follows: S101: Use entropy weight method to calculate indicators: action time , error number E, harmonic distortion rate H and fluctuation intensity coefficient The information entropy of , and assign weights to the four indicators respectively: 、 、 、 ; S102: Linearly combine the above indicators according to their weights as negative indicators of device performance , and its calculation formula is: ; S103: Based on the scoring model with a full score of 100, the negative scores are deducted to obtain the final performance score PS: , where a higher value indicates better device performance.

[0008] As a preferred solution of the digital management method for the entire process of relay protection debugging described in the present invention, the specific method of maintaining the decision vector is: S201: Determine the cost set corresponding to the remaining life or service life L of the current equipment and the historical maintenance plan ; S202: Construct a Bayesian conditional probability network: The goal is to calculate the success probability of each maintenance plan under the current PS situation: .

[0009] S203: Cost of introducing the solution Construct the cost-performance objective function with the remaining life or service life L of the current equipment: ; S204: Find the maximum objective function value among all candidate solutions {Solution 1, ..., Solution n}: ; Among them, the output This is the maintenance instruction set within this scheduling cycle.

[0010] As a preferred solution of the digital management method for the entire process of relay protection commissioning described in the present invention, in step 4, the specific method of verifying the maintenance effect is to quantify the cost-effectiveness improvement effect of the maintenance action by using the performance change under unit cost, and the calculation formula is:

[0011] in, represents the performance score recalculated after maintenance, Indicates the system's preset minimum maintenance cost; like > , Indicates a significant improvement in the threshold, or a recalculated fluctuation intensity coefficient , Represents the fluctuation tolerance threshold, then the entropy weight method weights and Bayesian network parameters are updated.

[0012] As a preferred solution of the digital management method for the whole process of relay protection commissioning described in the present invention, the update logic of the entropy weight method weight is: calculate the harmonic distortion rate H and the fluctuation intensity coefficient characteristic deviation rate; After obtaining the feature deviation rate, the system will adjust the weight of each feature in the entropy weight method accordingly: the adjustment method is: ; The update logic of the Yes network parameters is: calculate the current maintenance effect coefficient The historical average effect coefficient The difference between them is used to construct an exponential decay function to correct the conditional probability of each maintenance scheme in the Bayesian network; it is expressed as: , represents the attenuation factor.

[0013] As a preferred solution of the digital management method for the entire process of relay protection commissioning described in the present invention, M is matched with the historical decision database for similarity; if the matching degree exceeds a threshold, an associated maintenance item is added to M.

[0014] A management system applied to the above-mentioned digital management method for the entire process of relay protection commissioning includes: a data acquisition and integration module for real-time acquisition and aggregation of various operating data of the relay protection system; The feature extraction and performance evaluation module processes the collected data, extracts key feature indicators, and evaluates performance scores. The intelligent diagnosis and maintenance decision module generates the optimal maintenance plan based on the Bayesian network when the performance score does not meet the standard. The maintenance execution and status monitoring module executes maintenance actions, records post-maintenance status data, and monitors changes in key features; as well as the closed-loop feedback and model self-optimization module, compares the effects before and after maintenance, automatically corrects the entropy weights and Bayesian parameters, and achieves adaptive optimization.

[0015] The present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned method for digital management of the entire process of relay protection debugging when executing the computer program.

[0016] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for digital management of the entire process of relay protection debugging are realized.

[0017] Beneficial effects of the present invention: 1. This invention introduces multi-dimensional characteristic indicators such as harmonic distortion rate and output fluctuation intensity, integrates the entropy weight method to establish a performance scoring model, and objectively reflects the health status of the relay protection device. It also establishes an intelligent generation mechanism for maintenance strategies, combines the Bayesian network with historical maintenance data, and dynamically calculates the benefit weights of different plans to achieve optimal recommendation of maintenance plans.

[0018] 2. The present invention uses feedback indicators such as the maintenance effect coefficient and feature deviation rate to make real-time corrections to the scoring weights and Bayesian network conditional probabilities, thereby realizing an intelligent closed-loop debugging process. Through the similarity matching mechanism between the decision vector and the historical maintenance database, highly correlated maintenance items are automatically added to improve the integrity of the decision and the ability to transfer experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them: Figure 1 This is a schematic diagram of the overall structure of a digital management method and system for the entire process of relay protection debugging proposed by the present invention. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0023] Reference Figure 1 , as an embodiment of the present invention, provides a digital management method and system for the entire process of relay protection commissioning, the method comprising the following steps: Step 1: Obtain real-time operating data of the relay protection equipment and perform feature extraction based on the real-time data. The real-time operating data includes: voltage / current waveform data collected from smart meters , action record data collected from relay protection equipment, including action time (This data reflects the delay of relay protection response) and the number of errors E (this data is used to reflect abnormal behaviors such as misoperation and refusal to operate), as well as the output curve data collected from the distributed energy controller ; right Perform feature extraction to obtain the harmonic distortion rate H (reflecting the degree of distortion of the voltage / current waveform); right Perform feature extraction to obtain the fluctuation intensity coefficient (Reflects the degree of dynamic disturbance of the power supply system).

[0024] What needs to be said specifically is that the feature extraction calculation method of the harmonic distortion rate H is: Perform wavelet transform to separate high-frequency components and fundamental components, calculate the ratio of high-frequency component energy to fundamental wave energy, and calculate the average value within the time window to obtain H; and the fluctuation intensity coefficient The feature extraction calculation method is as follows: Calculate the first-order derivative, integrate the square of the derivative in the time interval [0, T], and take the square root of the integral result to obtain the fluctuation intensity coefficient .

[0025] Step 2: Establish a performance scoring mechanism to calculate the performance score based on the feature-extracted data and determine whether the score is lower than the preset performance threshold. The performance score calculation process is as follows: S101: Use entropy weight method to calculate indicators: action time , error number E, harmonic distortion rate H and fluctuation intensity coefficient The information entropy of , and assign weights to the four indicators respectively: 、 、 、 ; The logic is as follows: First, construct the indicator matrix and ,E,H, Perform normalization processing, and then calculate the information entropy of each indicator separately , used to measure its discreteness, the formula is: ; in, The normalization constant representing entropy is defined as ,make sure ∈[0,1], n represents the total number of data samples, is the normalized value of the j-th sample on the i-th index.

[0026] Then, weights are assigned inversely based on information entropy (the smaller the entropy value and the greater the difference, the higher the weight): ; (satisfying ∑ =1) This method avoids the subjectivity of manual experience in weight setting, and the weight distribution is objective and dynamically adaptable.

[0027] S102: Linearly combine the above indicators according to their weights as negative indicators of device performance , and its calculation formula is: ; S103: Based on the scoring model with a full score of 100, the negative scores are deducted to obtain the final performance score PS: , where a higher value indicates better device performance.

[0028] In summary, a clear chain is formed through "indicator collection → entropy weight allocation → penalty item calculation → final score".

[0029] In step 3, if the score is lower than the preset performance threshold, a maintenance decision vector is generated using the Bayesian network algorithm based on the score and the historical maintenance database. The maintenance action in the maintenance decision vector is executed, and post-maintenance data is obtained. Since PS < (preset performance threshold), the system determines that the equipment operating status is abnormal. Therefore, different maintenance strategies need to be analyzed and evaluated in combination with historical data. The specific method of the maintenance decision vector is as follows: S201: Determine the cost set corresponding to the remaining life or service life L of the current equipment and the historical maintenance plan ; S202: Construct a Bayesian conditional probability network: The goal is to calculate the success probability of each maintenance plan under the current PS situation: This conditional probability is trained through historical maintenance data (e.g., "Maintenance plan A is adopted under a certain PS value, and the success rate is 90%)," forming the node dependency structure of the Bayesian network. S203: If only probability is used Choosing the best will ignore the cost factor. If we only consider the cost , then an inefficient solution may be chosen, thus introducing the solution cost Construct the cost-performance objective function with the remaining life or service life L of the current equipment: ;in It indicates the cost-effectiveness of maintenance cost to life, that is, the value of equipment life that can be brought by unit cost.

[0030] S204: Find the maximum objective function value among all candidate solutions {Solution 1, ..., Solution n}: .

[0031] Among them, the output It is the maintenance instruction set within this round of scheduling cycle, including specific actions and priorities. M is matched with the historical decision database for similarity; if the matching degree exceeds the threshold, the associated maintenance items are added to M. That is to say, after the system completes the generation of the preliminary maintenance decision vector M, it also performs similarity matching analysis on it with the maintenance plans in the historical maintenance database. If the matching result shows that the similarity between the current decision and a certain efficient maintenance plan in the history exceeds the set threshold, the system will automatically extract the relevant maintenance items in the historical plan that are not included in the current decision, and append them to the current maintenance vector M to generate an optimized extended maintenance vector. This approach can fully leverage historical experience, improve the integrity and adaptability of maintenance strategies, and further enhance the system's intelligent closed-loop capabilities.

[0032] Step 4: Recalculate the performance score based on the data after maintenance and verify the maintenance effect. The specific method of verifying the maintenance effect is to use the performance change under unit cost to quantify the cost-effectiveness improvement effect of the maintenance action. The calculation formula is:

[0033] in, represents the performance score recalculated after maintenance, Indicates the system's preset minimum maintenance cost; like > , Indicates the significant improvement threshold, which is a "positive trigger" indicating that the system state has improved significantly. The model can be positively strengthened based on new data or recalculated to obtain the volatility intensity coefficient. , Represents the fluctuation tolerance threshold. This is a "negative trigger," meaning that maintenance behavior has caused system fluctuations and the model needs to be corrected or downgraded. Therefore, the entropy weight method weights and Bayesian network parameters are updated.

[0034] The update logic of the entropy weight method is: calculate the harmonic distortion rate H and the fluctuation intensity coefficient characteristic deviation rate; For example, the harmonic distortion rate H' obtained by the current recollection is compared with the reference value H before maintenance to obtain the deviation rate of the harmonic characteristics. for: ; Similarly, the deviation rate of the volatility intensity coefficient is calculated as: ; The above two indicators can effectively reflect the physical impact of maintenance behavior on the operating status of the equipment.

[0035] After obtaining the feature deviation rate, the system will adjust the weight of each feature in the entropy weight method accordingly: the adjustment method is: , to enhance the sensitivity of the scoring model to state changes; the calculation formula is as follows: ; ; in, Represents the learning rate.

[0036] The update logic of the Yes network parameters is: calculate the current maintenance effect coefficient The historical average effect coefficient The difference between them is used to construct an exponential decay function to correct the conditional probability of each maintenance scheme in the Bayesian network; it is expressed as: , represents the attenuation factor.

[0037] This embodiment also provides a digital management system for the entire process of relay protection commissioning, which includes: a data acquisition and integration module for real-time acquisition and aggregation of various operating data of the relay protection system; The feature extraction and performance evaluation module processes the collected data, extracts key feature indicators and evaluates the performance score; the intelligent diagnosis and maintenance decision module generates the optimal maintenance plan based on the Bayesian network when the performance score does not meet the standard; the maintenance execution and status monitoring module executes maintenance actions, records post-maintenance status data, and monitors changes in key features; and the closed-loop feedback and model self-optimization module compares the effects before and after maintenance, automatically corrects the entropy weight and Bayesian parameters, and realizes adaptive optimization.

[0038] This embodiment also provides a computer device, which is suitable for a full-process digital management method for relay protection debugging, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a full-process digital management method for relay protection debugging as proposed in the above embodiment.

[0039] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0040] This embodiment also provides a storage medium having a computer program stored thereon. When the program is executed by a processor, it implements a method for digital management of the entire process of relay protection debugging as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0041] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A digital management method for the entire process of relay protection commissioning, characterized in that: The method comprises the following steps: Step 1: Acquire real-time operating data of the relay protection device and perform feature extraction based on the real-time data; Step 2: Establish a performance scoring mechanism to calculate the performance score based on the feature-extracted data and determine whether the score is lower than the preset performance threshold; Step 3: If the score is lower than the preset performance threshold, a maintenance decision vector is generated using a Bayesian network algorithm based on the score and the historical maintenance database. The maintenance action in the maintenance decision vector is executed, and post-maintenance data is obtained. Step 4: Recalculate the performance score based on the data after maintenance and verify the maintenance effect.

2. A digital management method for the entire process of relay protection commissioning according to claim 1, characterized in that: The real-time operation data includes: voltage / current waveform data collected from smart meters , action record data collected from relay protection equipment, including action time and the number of errors E, as well as the output curve data collected from the distributed energy controller ; right Perform feature extraction to obtain the harmonic distortion rate H; right Perform feature extraction to obtain the fluctuation intensity coefficient .

3. A digital management method for the entire process of relay protection commissioning according to claim 2, characterized in that: The calculation process of the performance score is: S101: Use entropy weight method to calculate indicators: action time , error number E, harmonic distortion rate H and fluctuation intensity coefficient The information entropy of , and assign weights to the four indicators respectively: 、 、 、 ; S102: Linearly combine the above indicators according to their weights as negative indicators of device performance , and its calculation formula is: ; S103: Based on the scoring model with a full score of 100, the negative scores are deducted to obtain the final performance score PS: , where a higher value indicates better device performance.

4. A digital management method for the entire process of relay protection commissioning according to claim 3, characterized in that: The specific method of maintaining the decision vector is: S201: Determine the cost set corresponding to the remaining life or service life L of the current equipment and the historical maintenance plan ; S202: Construct a Bayesian conditional probability network: The goal is to calculate the success probability of each maintenance plan under the current PS situation: ; S203: Cost of introducing the solution Construct the cost-performance objective function with the remaining life or service life L of the current equipment: ; S204: Find the maximum objective function value among all candidate solutions {Solution 1, ..., Solution n}: ； Among them, the output This is the maintenance instruction set within this scheduling cycle.

5. A digital management method for the entire process of relay protection commissioning according to claim 4, characterized in that: In step 4, the specific method of verifying the maintenance effect is to quantify the cost-effectiveness improvement effect of the maintenance action by using the performance change under unit cost. The calculation formula is: ；； in, represents the performance score recalculated after maintenance, Indicates the system's preset minimum maintenance cost; like > , Indicates a significant improvement in the threshold, or a recalculated fluctuation intensity coefficient , Represents the fluctuation tolerance threshold, then the entropy weight method weights and Bayesian network parameters are updated.

6. A digital management method for the entire process of relay protection commissioning according to claim 5, characterized in that: The update logic of the entropy weight method is: calculate the harmonic distortion rate H and the fluctuation intensity coefficient characteristic deviation rate; After obtaining the feature deviation rate, the system will adjust the weight of each feature in the entropy weight method accordingly: the adjustment method is: ; The update logic of the Yes network parameters is: calculate the current maintenance effect coefficient The historical average effect coefficient The difference between them is used to construct an exponential decay function to correct the conditional probability of each maintenance scheme in the Bayesian network; it is expressed as: , Represents the attenuation factor.

7. A digital management method for the entire process of relay protection commissioning according to claim 6, characterized in that: Perform similarity matching on M with the historical decision database; if the matching degree exceeds the threshold, add the associated maintenance item to M.

8. A digital management system for the entire process of relay protection commissioning, applied to the digital management method for the entire process of relay protection commissioning according to any one of claims 1 to 7, characterized in that: The system includes: a data acquisition and integration module, which is used to collect and summarize various operating data of the relay protection system in real time; Feature extraction and performance evaluation module, which processes the collected data, extracts key feature indicators and evaluates performance scores; Intelligent diagnosis and maintenance decision module, which generates the optimal maintenance plan based on Bayesian network when the performance score does not meet the standard; Maintenance execution and status monitoring module, which executes maintenance actions, records post-maintenance status data, and monitors changes in key characteristics; As well as the closed-loop feedback and model self-optimization module, it compares the effects before and after maintenance, automatically corrects the entropy weight and Bayesian parameters, and realizes adaptive optimization.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for digital management of the entire process of relay protection debugging according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for digital management of the entire process of relay protection debugging according to any one of claims 1 to 7 are implemented.

Images