A fast recognition method based on dynamic time window

Abstract

The present invention discloses a rapid identification method based on a dynamic time window, and relates to the field of power grid operation and maintenance. The present invention includes the following steps: dynamic switching strategy generation, the master station automatically generates an initial switching instruction sending sequence for the identified object according to the rapid identification time limit and the number of objects to be identified; dynamic time matching, matching calculation of the reported data, if the match is successful within the set switching time interval, the next device is sent immediately, the time interval is dynamically planned, and no longer waits for the fixed time interval to be sent; for the device that failed to be identified, the master station performs multiple rounds of re-addition, and if it still fails or exceeds the preset time, it enters the topology identification failure list. The present invention greatly reduces the time consumption caused by the timing process by streamlining the device timing process, using the instruction issuance time in the switching strategy to replace the instruction sending time; on the basis of ensuring the identification of the main process, the communication, program and other implementation processes are further optimized in time.

Description

Technical Field

[0001] The present invention relates to the field of power grid operation and maintenance, and in particular to a fast identification method based on a dynamic time window. Background Art

[0002] In recent years, with the development of research and application of technologies such as HPLC (power line high-speed carrier communication), the low-voltage side has gradually acquired technical means to identify the relationship between power supply and consumption. However, when conducting topological identification of adjacent substations on site, due to the problem of carrier signal crosstalk between substations, the application effect of topological identification is still far from expectations. Therefore, in the pilot study of substation topology identification carried out by Zhejiang Company in 2021, the use of characteristic current topological identification technology was newly proposed to effectively solve the crosstalk problem. However, when this technology is applied in production, the conventional topological identification mode used in the pilot has problems such as a single identification mode and poor adaptability when facing different needs such as timeliness and accuracy on site. It greatly affects the quality and efficiency of engineering applications, and is a realistic problem that needs to be solved urgently for the mass promotion and application of characteristic current topological identification technology.

[0003] Different from the pilot scenario, in the actual production application process, whenever the end of the month is approaching and the line loss index of the substation needs to be verified or when a customer complaint needs to be responded to quickly, etc., which has high timeliness requirements, the characteristic current topology recognition needs to output the recognition results in the fastest time. The conventional topology recognition mode of the pilot basically belongs to the state of "linear accumulation based on the number to be recognized" in terms of recognition speed and time. For example, the three low-voltage substations (a total of 150 households) under the same 10kV line are executed using the conventional characteristic current topology recognition mode of the original pilot model. It takes more than 700 minutes, more than 12 hours. When the recognition request is received in the afternoon of a weekday, it may take a day to output the recognition result, which cannot effectively support the problem solving. Therefore, how to better meet the timeliness requirements of the topology recognition function in actual production activities at the algorithm level, ensure that in fast scenarios, the system side reasonably allocates recognition tasks, and quickly realizes batch topology recognition of multiple substations is the main problem that needs to be solved. Summary of the invention

[0004] The technical problem to be solved and the technical task proposed by the present invention are to improve and perfect the existing technical solutions and provide a fast identification method based on a dynamic time window to optimize the time of the core link of topology identification. To this end, the present invention adopts the following technical solutions.

[0005] A fast recognition method based on a dynamic time window comprises the following steps:

[0006] 1) Identification device function verification

[0007] Identify the device, remove the identification sequence and provide feedback for the device that does not have the characteristic current receiving and sending function;

[0008] 2) Dynamic switching strategy generation

[0009] The master station automatically generates an initial switching instruction sending sequence for the identified object according to the fast identification time limit and the number of objects to be identified; at the same time, according to the real-time identification situation in step 6), the switching strategy is dynamically updated;

[0010] 3) Timing of characteristic current receiving equipment

[0011] The master station performs clock confirmation operation on the characteristic current receiving device. The master station reads the current clock of the receiving device, sets the clock for the device whose time difference exceeds the set time length, and reads back for confirmation to ensure that the clocks of all receiving devices are consistent before the characteristic current identification is performed; if the device fails to synchronize the time for multiple times, the corresponding device will be included in the failure list, and the type is synchronization failure;

[0012] The terminal clock is used as the only standard clock source for this identification;

[0013] 4) Switching command is issued

[0014] The master station automatically generates the equipment switching sequence based on the switching time interval and the initial file, and transparently transmits the switching instructions to the equipment in sequence when the master station clock reaches the switching time point;

[0015] 5) Identify event reporting

[0016] The master station monitors the characteristic current receiving events of the terminals and circuit breakers in real time during switching;

[0017] 6) Dynamic time matching

[0018] The reported data is matched and calculated. If the match is successful within the set switching time interval, the next device is sent immediately, and the time interval is dynamically planned, and no longer waits for the fixed time interval to be sent. For devices that fail to be identified, the master station performs multiple rounds of re-addition. If it still fails or exceeds the preset time, it will enter the topology identification failure list.

[0019] This technical solution aims at the demand for rapid identification scenarios of characteristic current topology identification technology in actual production applications, and establishes a rapid identification method based on a dynamic time window algorithm. On the one hand, this method greatly reduces the time consumption caused by the synchronization process by streamlining the equipment timing process and replacing the instruction sending time with the instruction issuing time in the switching strategy; on the other hand, it further optimizes the communication, program and other implementation processes on the basis of ensuring the main identification process, targeting the core links of topology identification, including: switching strategy generation, instruction issuance, result call, and result calculation. The execution of topology identification is uniformly controlled by the master station. The rapid identification method can freely select the substation topology identification time limit and dynamically plan the time interval according to the reception situation.

[0020] Relying on the characteristic current switching sequence on the master station side and the successful reception event of the characteristic current identification device, rapid topology identification is performed. The instruction issuance time in the switching strategy is used to replace the issuance time, and the input information of the identification result is synchronously modified to realize a topology identification process that matches the requirements of rapid identification scenarios. At the same time, according to the execution of each switching identification, the switching strategy is dynamically and real-time updated to optimize the time of the core link of topology identification.

[0021] As a preferred technical means: in step 3), the receiving device includes a terminal and a circuit breaker, the master station reads the current clock of the receiving device, sets the clock for devices with a difference of more than 10 seconds and reads back for confirmation, to ensure that the clocks of all receiving devices are consistent before the characteristic current identification is executed; if the device fails to synchronize the time for 3 times, the corresponding device will be included in the failure list.

[0022] As a preferred technical means: in step 4), the transparent transmission is sent by the main site name. When multiple meters share one collector on site, only one meter is taken when the switching sequence is generated, and the ownership of the remaining meters remains consistent with it.

[0023] As a preferred technical means: in step 5), the information of the characteristic current receiving event includes phase, identification success time and receiving device identification information.

[0024] As a preferred technical means: in step 6), if there is no current receiving event within the device sending sequence, it is considered to be a device that failed to be identified. For the device that failed to be identified, the master station performs 3 rounds of re-investment. If it still fails or exceeds the preset time, it will enter the topology identification failure list and record the status as failure, and the reason is reception failure.

[0025] As a preferred technical means: it also includes a comprehensive verification step, which is located before outputting the station area topology result; the comprehensive verification includes the following steps:

[0026] a) Regional terminal time synchronization

[0027] The master station automatically searches the line to which the target device to be identified belongs in the original file, calls and tests the clocks of all public transformer terminals under the line, and performs accurate time synchronization if the set time is exceeded;

[0028] b) Switching command is issued

[0029] The master station automatically generates the equipment switching time sequence based on the set switching time interval and the initial file, and transparently transmits the switching instructions to the equipment in sequence when the master station clock reaches the switching time point;

[0030] c) Identify event reception

[0031] During the execution of the switching command, the master station receives the successful identification report event of the terminal in the area; for the equipment that fails to be identified, multiple rounds of re-switching are carried out, and if it still fails, it will enter the household change identification result list;

[0032] d) Single-point attribution analysis

[0033] Confirm the device area ownership according to the matching information, and return information including the area number;

[0034] e) Comparison of topology recognition results

[0035] Obtain the topology identification results of the target substation, and obtain the topology identification results of two different technical solutions based on planned power outages and HPLC; compare them with the result information of successful characteristic current identification. If the results are consistent, it means that the credibility of the results is further improved, and enter the next step of application simulation verification; if they are inconsistent, start the secondary authentication of the currently used characteristic current identification method again, and the authentication result is marked as a research-level identification result;

[0036] f) Line loss calculation simulation verification

[0037] The application scenario of the substation topology identification results is simulated, and a line loss calculation model based on the new substation user relationship is established according to the identification result information. The substation line loss is calculated based on the power data of the power consumption information collection system, and compared with the historical line loss data of the substation. If there is a positive improvement, it is considered that the line loss trial calculation simulation verification has passed, and the identification result has a high credibility. The identification result is recorded as an application-level identification result. If there is no positive improvement, a single-point verification is performed again for the questionable equipment in the substation. If there is no change in the verification result, it is converted to a research-level identification result. If there is a change, the line loss trial calculation simulation verification is performed again. If there is no improvement, it is also recorded as a research-level identification result.

[0038] Comprehensive verification solves the uncertainty problem of topology identification results caused by various on-site problems such as command downlink delay. On the basis of characteristic current topology identification, it comprehensively considers three aspects, namely, integrity optimization of identification results, verification of similar methods, and scenario simulation, to form a comprehensive verification method for characteristic current topology identification suitable for production applications, which effectively improves the accuracy of characteristic current topology identification.

[0039] Comprehensive verification is not only applicable to the result verification scenario of characteristic current topology identification, but also can be easily adjusted to other scenarios of substation topology identification based on other technologies in the past or in the future. It can solve the problems of single identification mode and poor adaptability of conventional topology identification mode and improve the quality and efficiency of engineering application.

[0040] Beneficial effects:

[0041] On the one hand, this method greatly reduces the time consumption caused by the synchronization process by streamlining the equipment synchronization process and replacing the instruction sending time with the instruction issuing time in the switching strategy; on the other hand, for the core link of topology identification, the communication, program and other implementation processes are further optimized on the basis of ensuring the main identification process. The execution of topology identification is uniformly controlled by the master station. The fast identification method can freely select the substation topology identification time limit and dynamically plan the time interval according to the reception situation.

[0042] This technical solution better meets the needs of rapid identification, realizes the rapid output of the results of the substation topology identification based on the characteristic current, can make quick responses to indicator assessments or customer complaints, improve customer service levels, and increase customer satisfaction. The application of characteristic current topology identification technology that adapts to the actual needs of on-site scenarios accelerates the practical production application of characteristic current topology identification technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of the present invention.

[0044] Figure 2 It is a schematic diagram of the dynamic process execution timing of the present invention.

[0045] Figure 3 It is a comprehensive verification flow chart of the present invention. DETAILED DESCRIPTION

[0046] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0047] This embodiment 1:

[0048] This embodiment is a fast identification method based on a dynamic time window algorithm.

[0049] The rapid identification method mainly relies on the characteristic current switching sequence on the master side and the successful reception event of the characteristic current identification device for rapid topology identification. In this mode, the instruction issuance time in the switching strategy is used instead of the issuance time, and the input information of the identification result algorithm is synchronously modified to realize the topology identification process that matches the needs of the rapid identification scenario; at the same time, according to the execution of each switching identification, the switching strategy is dynamically updated in real time to optimize the core link time of topology identification. The main process links include identification device function verification, switching strategy generation, switching instruction issuance, identification time reporting, sending and receiving time matching, and equipment ownership identification. The specific steps include:

[0050] 1. Identify device function verification

[0051] The prerequisite for the master station to execute the topology identification function is that the device has the characteristic current receiving and sending function. For unsupported devices, the identification sequence will be removed and feedback will be given. This link is the pre-identification process and generally does not take up the execution time. The master station may manually or automatically filter the identification objects according to custom conditions.

[0052] 2. Dynamic switching strategy generation

[0053] The master station automatically generates the initial switching command sending sequence for the identification object according to the fast identification time limit and the number of objects to be identified; at the same time, according to the real-time identification situation in the sixth stage, the switching strategy is dynamically updated; the master station configures the appropriate fast identification time according to the user's identification time limit requirements. The system can prompt when the number of objects to be identified is too large or the time limit is too short. During this process, the system background will arrange the identification objects in rounds according to the principle that the substations / users under the same 10kV line cannot enable tasks at the same time. Objects in the same round can be identified in parallel, and objects in different rounds are identified serially.

[0054] 3. Timing of characteristic current receiving equipment (simplified steps)

[0055] The master station performs clock confirmation operations on the characteristic current receiving device (terminal, circuit breaker). First, the master station reads the current clock of the receiving device, sets the clock for the device with a difference of more than 10 seconds and reads back for confirmation to ensure that the clocks of all receiving devices are consistent before the characteristic current identification is performed. If the device fails to synchronize the time for 3 times, the corresponding device will be included in the failure list, and the type is synchronization failure.

[0056] This process is the pre-stage of the core identification process, which aims to ensure the clock uniformity of terminals, meters and other devices involved in topology identification. It generally includes clock matching, setting, and abnormal handling of hundreds of objects, which takes a long time. Under the fast identification method, the terminal clock is used as the only standard clock source for this identification, and the switching time in the switching strategy replaces the instruction sending time required for subsequent algorithm calculations, which simplifies this link and greatly reduces the execution time of topology identification.

[0057] 4. Switching command is issued

[0058] The master station automatically generates the equipment switching sequence based on the switching time interval and the initial file, and transmits the switching instructions to the equipment in sequence when the master station clock reaches the switching time point, and the name of the master station is transmitted. When multiple meters share one collector on site, only one meter is taken when the switching sequence is generated, and the ownership relationship of the other meters is consistent with it.

[0059] 5. Identify incident reports

[0060] The master station monitors the characteristic current receiving events of the terminal and the circuit breaker in real time during switching. The event information includes phase, successful identification time, and receiving device identification.

[0061] 6. Dynamic time matching

[0062] The reported data is matched and calculated. If the match is successful within the set switching time interval, the next device is sent immediately, and the time interval is dynamically planned, and no longer waiting for the fixed time interval to be sent. The device without current receiving events within the device sending sequence is a device that failed to be identified. For the device that failed to be identified, 3 rounds of re-addition are performed. If it still fails or exceeds the preset time, it will enter the topology (user change) identification failure list and record the status as failure, and the reason is reception failure. According to the matching results of the sending and receiving time, the background automatically performs topology analysis according to the topology identification algorithm, and based on the reporting time of the identification result, it dynamically iterates the new switching sequence with the objects that are not currently identified.

[0063] 7. Form a sending and receiving matching record

[0064] Record and display the results of substation topology identification.

[0065] Aiming at the demand for rapid identification scenarios of characteristic current topology identification technology in actual production applications, a rapid identification method based on a dynamic time window algorithm is established. On the one hand, this mode greatly reduces the time consumption caused by the synchronization process by streamlining the equipment timing process and replacing the instruction sending time with the instruction issuing time in the switching strategy; on the other hand, for the core links of topology identification (switching strategy generation, instruction issuing, result call, and result calculation), the communication, program and other implementation processes are further optimized on the basis of ensuring the main identification algorithm. The execution of topology identification is uniformly controlled by the master station. The rapid identification method can freely select the substation topology identification time limit and dynamically plan the time interval according to the receiving situation.

[0066] The above-mentioned rapid identification method based on the dynamic time window algorithm will better meet the needs of rapid identification, realize the rapid output of the substation topology identification results based on the characteristic current, and make quick responses to indicator assessments or customer complaints, thereby improving customer service levels and customer satisfaction. Based on the rapid identification method of substation topology, three substations (a total of 150 users) under the same 10kV line were selected for a pilot project, and the identification results were obtained within 3 hours, realizing the application of characteristic current topology identification technology that adapts to the actual needs of the on-site scene, and accelerating the practical production application of characteristic current topology identification technology.

[0067] Embodiment 2:

[0068] In order to improve the accuracy of topology recognition, a comprehensive verification step can be added.

[0069] The comprehensive verification of this embodiment mainly solves the uncertainty problem of topology recognition results caused by various field problems such as instruction downlink delay. On the basis of characteristic current topology recognition, it comprehensively considers three aspects, namely, integrity optimization of recognition results, verification of similar methods, and scenario simulation, to form a comprehensive verification method for characteristic current topology recognition suitable for production applications. The main process is as follows: Figure 2 As shown, the comprehensive verification includes the following steps:

[0070] A. Regional Terminal Time Synchronization

[0071] The master station automatically searches for the 10kV line to which the target device to be identified belongs in the original file, calls for testing of all public transformer terminal clocks under the line, and performs accurate time synchronization in over 10 seconds. Users can select the area by themselves through the front desk, especially for cross-regional problems caused by manual adjustment of files, which can expand the search scope.

[0072] B. Switching command is issued

[0073] The master station automatically generates the equipment switching time sequence based on the switching time interval set by the user and the initial file, and transmits the switching instructions to the equipment in sequence when the master station clock reaches the switching time point, and the name of the master station is transmitted; when multiple meters share one collector on site, only one meter is taken when the switching sequence is generated, and the ownership relationship of the other meters is consistent with it;

[0074] C. Identify event reception

[0075] During the execution of the switching command, the master station receives the successful identification report event of the terminal in the area. When there is no current reception event in the device sending sequence, it is considered to be a device that failed to be identified. The master station performs 3 rounds of re-switching. If it still fails, it will enter the household change identification result list and record the status as failed, and the reason is reception failure.

[0076] D. Single point attribution analysis

[0077] Confirm the device region ownership based on the matching information and return the region number and other information.

[0078] E. Comparison of topology recognition results

[0079] Obtain the topological identification results of the target substation. Taking the currently used acquisition system as an example, the topological identification results of two different technical solutions, namely, planned power outage and HPLC, can be obtained. Compare with the result information of successful characteristic current identification. If the results are consistent, it indicates that the credibility of the results is further improved, and the next step is to apply simulation verification. If they are inconsistent, the secondary authentication of the currently used characteristic current identification method is started again, and the authentication result is marked as a research-level identification result.

[0080] F. Line loss calculation simulation verification

[0081] The application scenario of the substation topology identification results is simulated, and a line loss calculation model based on the new substation user relationship is established according to the identification result information. The substation line loss is calculated based on the power data of the power consumption information collection system, and compared with the historical line loss data of the substation. If there is a positive improvement, it is considered that the line loss trial simulation verification has passed, and the recognition result has a high credibility. The recognition result is recorded as an application-level recognition result. If there is no positive improvement, a single-point verification is performed again for the questionable equipment in the substation. If there is no change in the verification result, it is converted to a research-level recognition result. If there is a change, the line loss trial simulation verification is performed again. If there is no improvement, it is also recorded as a research-level recognition result.

[0082] Comprehensive verification is not only applicable to the result verification scenario of characteristic current topology identification, but also can be simply adjusted to other scenarios of substation topology identification based on other past or future technologies. Taking the currently promoted substation topology identification based on characteristic current as an example, before the final result of the main process is output, an external link is used to perform comprehensive verification on the existing results, so as to optimize the identification results without affecting the main process of topology identification.

[0083] Among them, the application-level identification result is: the topology identification result credibility level, the application level is the highest credibility, and the result can be directly used to guide on-site production applications.

[0084] The research-level identification result is: the topology identification result credibility level. The research level is the next level of credibility, but higher than the initial result credibility. The result can be used in research scenarios and can also be used to guide on-site production applications when combined with on-site confirmation.

[0085] The above-mentioned rapid identification method based on a dynamic time window is a specific embodiment of the present invention, which has embodied the substantial characteristics and progress of the present invention. According to actual use needs and under the guidance of the present invention, equivalent modifications in shape, structure, etc. can be made to the method, which are all within the protection scope of this scheme.

Claims

1. A fast recognition method based on dynamic time window, characterized in that The following steps are involved: 1) Identification device function verification Identify the device, remove the identification sequence and provide feedback for the device that does not have the characteristic current receiving and sending function; 2) Dynamic switching strategy generation The master station automatically generates an initial switching instruction sending sequence for the identified object according to the fast identification time limit and the number of objects to be identified; at the same time, according to the real-time identification situation in step 6), the switching strategy is dynamically updated; 3) Timing of characteristic current receiving equipment The master station performs clock confirmation operation on the characteristic current receiving device. The master station reads the current clock of the receiving device, sets the clock for the device whose time difference exceeds the set time, and reads back for confirmation to ensure that the clocks of all receiving devices are consistent before the characteristic current identification is performed; If the device fails to synchronize time after multiple times, the corresponding device will be included in the failure list and the type will be synchronization failure; The terminal clock is used as the only standard clock source for characteristic current identification; 4) Switching command is issued The master station automatically generates the equipment switching sequence based on the switching time interval and the initial file, and transparently transmits the switching instructions to the equipment in sequence when the master station clock reaches the switching time point; 5) Identify incident reports The master station monitors the characteristic current receiving events of the terminals and circuit breakers in real time during switching; 6) Dynamic time matching The reported data is matched and calculated. If the match is successful within the set switching time interval, the next device is sent immediately, and the time interval is dynamically planned, and no longer waits for the fixed time interval to be sent. For devices that fail to be identified, the master station performs multiple rounds of re-addition. If it still fails or exceeds the preset time, it will enter the topology identification failure list.

2. The fast recognition method based on dynamic time window according to claim 1, characterized in that: In step 3), the receiving devices include terminals and circuit breakers. The master station reads the current clock of the receiving device, sets the clock for devices with a difference of more than 10 seconds and reads back for confirmation to ensure that the clocks of all receiving devices are consistent before the characteristic current identification is performed; if the device fails to synchronize the time for 3 times, the corresponding device will be included in the failure list.

3. The fast recognition method based on dynamic time window according to claim 2, characterized in that: In step 4), the transparent transmission is sent by the main site name. When multiple meters share one collector on site, only one meter is taken when the switching sequence is generated, and the ownership of the other meters remains consistent with it.

4. The fast recognition method based on dynamic time window according to claim 3 is characterized in that: In step 5), the information of the characteristic current receiving event includes phase, identification success time and receiving device identification information.

5. The fast identification method based on dynamic time window according to claim 4 is characterized in that: In step 6), if there is no current receiving event within the device sending sequence, it is considered to be a device that failed to be identified. For the device that failed to be identified, the master station performs 3 rounds of re-investment. If it still fails or exceeds the preset time, it will enter the topology identification failure list and record the status as failure, and the reason is reception failure.

6. The fast identification method based on dynamic time window according to claim 5, characterized in that: It also includes a comprehensive verification step, which is located before outputting the result; The comprehensive verification includes the following steps: The master station automatically searches the line to which the target device to be identified belongs in the original file, calls and tests the clocks of all public transformer terminals under the line, and performs accurate time synchronization if the set time is exceeded; The master station automatically generates the equipment switching time sequence based on the set switching time interval and the initial file, and transparently transmits the switching instructions to the equipment in sequence when the master station clock reaches the switching time point; During the execution of the switching command, the master station receives the successful identification report event of the terminal in the area; for the equipment that fails to be identified, multiple rounds of re-switching are carried out, and if it still fails, it will enter the household change identification result list; Confirm the device area ownership according to the matching information, and return information including the area number; Obtain the topology identification results of the target substation, and obtain the topology identification results of two different technical solutions based on planned power outages and HPLC; compare them with the result information of successful characteristic current identification. If the results are consistent, it means that the credibility of the results is further improved, and enter the next step of application simulation verification; if they are inconsistent, start the secondary authentication of the currently used characteristic current identification method again, and the authentication result is marked as a research-level identification result; The application scenario of the substation topology identification results is simulated, and a line loss calculation model based on the new substation-user relationship is established according to the identification result information. The substation line loss is calculated based on the electricity data of the electricity consumption information collection system, and compared with the historical line loss data of the substation. If there is a positive improvement, it is considered that the line loss trial simulation verification has passed, and the identification result has a high credibility. The identification result is recorded as an application-level identification result. If there is no positive improvement, a single-point verification is performed again for the questionable equipment in the substation. If there is no change in the verification result, it is converted to a research-level identification result. If there is a change, the line loss trial simulation verification is performed again. If there is no improvement, it is also recorded as a research-level identification result.

Images