Spare power automatic switching pressure plate automatic switching system and switching method based on active-active architecture

Abstract

The invention discloses a backup power automatic switching pressure plate automatic switching system and switching method based on an active-active architecture, and the system comprises an active-active system which has the capability of real-time calculation. The active system and the standby system are seamlessly switched; a perfect fault detection mechanism is designed, and the communication state and the operation state of the system are monitored through heartbeat testing; data synchronization of the active-active system is a key for ensuring data consistency between the systems, measures are taken to ensure timeliness and safety of data synchronization, and data errors caused by inconsistent service data are avoided; various autonomous judgment and self-rescue functions are integrated, and fault switching, recovery and early warning can be automatically completed under abnormal conditions. The stability and safety of the system are improved, the switching efficiency of the disaster recovery system is improved, it is ensured that the main system of the OCS can be rapidly switched to the standby system under the condition that the main system is unavailable, and the influence of power grid dispatching production work is reduced to the minimum.

Description

Technical Field

[0001] The present invention relates to the technical field of power system application, and in particular to an automatic switching system and method for a standby automatic switching pressure plate based on a dual-active architecture. Background Art

[0002] In the operation of the power system, the main transformers and lines of 220kV and below voltage levels are usually operated in the radial network power supply mode. In order to ensure the reliability of power supply, a backup automatic switching device (BATS) is usually installed inside the substation. When a power grid failure occurs and the busbar is out of power, the backup automatic switching that meets the conditions will be activated to supply power to the outage busbar, which greatly reduces the probability of busbar power outage and ensures the reliability of power supply. The traditional backup automatic switching pressure plate automatic switching and withdrawal system adopts a one-way main and standby mode design. When the system fails, various key data verification work is required when switching between the main and standby to ensure the safety of power grid dispatching. This operation mode has the disadvantages of long investment time, low switching efficiency, easy aging of standby equipment, and low fault tolerance, which cannot meet the high reliability requirements of modern power systems. In addition, when the main system is out of operation and the standby system is put into operation, the calculation data and operation records cannot be automatically synchronized to the main system in reverse, and the entire system will lose its backup capability, which will affect the operation of the entire power system. If the main system is unavailable and encounters a power grid accident, it will cause significant losses to the power grid. Summary of the invention

[0003] The purpose of the present invention is to solve the above-mentioned problems and to provide a system and method for automatically launching and retiring a standby self-throwing pressure plate based on an active-active architecture. Through the design of an active-active architecture, the automatic launching and retiring of the standby self-throwing soft pressure plate can be realized, and the main and standby systems can be switched quickly. Two completely identical active systems are deployed, and operations such as data synchronization and state replication are performed to ensure that the active-active systems can be synchronized in real time and maintain consistent states and data. At the same time, each system can automatically calculate the strategy for launching and retiring the standby self-throwing soft pressure plate, but there is only one system that performs remote control operations according to the strategy. The advantages and characteristics of this method are that it can greatly improve the reliability and stability of the system, reduce the risks of manual operation, adjustment and intervention, and thus improve the availability and reliability of the system.

[0004] The technical solution adopted by the present invention to solve the technical problem is:

[0005] An automatic throw-in and throw-out system for a standby self-throwing pressure plate based on a dual-active architecture includes a real-time calculation system, a control right switching system, a fault detection system, a data synchronization system, and a data verification system.

[0006] Furthermore, the active-active system is a highly intelligent system that is deeply developed and applied based on EMS. It calculates the steps of power grid backup and automatic investment in real time as follows:

[0007] 1) Online switching strategy of automatic backup based on transformer oil temperature

[0008] In actual operation, the short-term overload capacity of the transformer is closely related to the oil temperature. The oil temperature of the main transformer is a key indicator reflecting the status of the equipment. Specifically, if the oil temperature of the associated transformer exceeds the preset value, the system will automatically exit the standby mode to avoid overheating risks; if the oil temperature is within a safe range, the system may consider switching on based on other conditions;

[0009] 2) N-1 verification

[0010] The N-1 fault analysis of the automatic backup pressure plate is a subset of the traditional N-1 calculation. According to the characteristics of the regional power grid, a new processing method is adopted. This method is based on the base state topology, establishes the automatic backup action model by searching the power supply key path, and uses the load transfer method to greatly shorten the processing time.

[0011] 3) Bad data processing

[0012] Since the pressure plate activation and deactivation strategy is based on the main transformer oil temperature calibration and real-time N-1 calibration, the following bad data situations need to be considered:

[0013] a) If the oil temperature of the relevant transformer is abnormal, and the standby automatic start setting takes the oil temperature factor into consideration, the current operating state of the standby automatic start pressure plate shall be maintained unchanged;

[0014] b) In the state estimation calculation, if bad data is detected in the 220kV main transformer related to the backup automatic switch, the 220kV main transformer will not be checked;

[0015] c) In the state estimation calculation, if bad data is detected in the 110kV main transformer or 110kV line, the current operating state of the backup automatic voltage board is maintained unchanged;

[0016] 4) Handling abnormal situations

[0017] During the execution of the pressure plate launch and withdrawal of the standby automatic start-up, the following abnormal situations need to be considered: 1) SCADA telemetry data is not refreshed; 2) state estimation does not converge; 3) the maximum number of remote control times per hour for the standby automatic start-up is limited; 4) the maximum number of remote control times per day for the standby automatic start-up is limited.

[0018] Furthermore, in step 1), the platen insertion and withdrawal strategy taking into account the oil temperature and the initial load rate is as follows:

[0019] 1) 220kV main transformer oil temperature and initial load rate detection;

[0020] a) For a 220kV main transformer, if the initial load rate exceeds 0.9, the corresponding standby automatic switch should be locked;

[0021] b) For a 220kV main transformer, if the initial load rate exceeds 0.85 and the oil temperature exceeds 85℃ or the winding temperature exceeds 105℃, the corresponding standby automatic switch should also be locked;

[0022] 2) 110kV main transformer oil temperature and initial load rate detection

[0023] a) For a 110kV main transformer, if the initial load rate exceeds 0.85, the corresponding backup automatic switch should be locked.

[0024] b) For a 110kV main transformer, if the initial load rate exceeds 0.85, and at the same time the oil temperature exceeds 85°C (adjustable) or the winding temperature exceeds 105°C, the corresponding standby automatic switching should also be locked.

[0025] Further, in step 2), the fault analysis and processing method is as follows:

[0026] 1) Determine the backup automatic switch caused by N-1 fault based on power supply path search

[0027] When the backup automatic transfer system meets the charging conditions, if the working bus loses power but the backup power supply is energized, the backup automatic transfer system will start;

[0028] 2) Detailed processing based on power supply path search

[0029] When searching for power supply paths, we start with the 220kV bus and generator nodes as the root nodes. The specific logic is as follows: simulate disconnecting the 220kV line switch to establish the relationship between the node and the branch; use depth-first downward search to record the low-voltage bus power supply path encountered; establish a mapping relationship between the equipment and the backup automatic re-start bus; identify loop information, and loop equipment failure will not cause the backup automatic re-start action; if the equipment has a mapping relationship with both the working bus and the backup bus, it is regarded as a homologous device, and its failure will not trigger the backup automatic re-start; record the number of power supply points related to the backup automatic re-start, and determine the key power supply path equipment, whose failure will trigger the backup automatic re-start action;

[0030] 3) Processing of standby automatic investment and withdrawal strategies based on multiple objectives

[0031] When a specific N-1 fault will trigger N backup automatic switching operations, if all the backup automatic switching operations lead to overload of power grid components, and when less than all the backup automatic switching operations occur, no other power grid components are overloaded, it is obviously not practical to fully block all the backup automatic switching operations in this case. In fact, only some of the backup automatic switching operations need to be blocked. Therefore, a priority is set for each backup automatic switching operation.

[0032] V pri =10000×V run +1000×V pid -int(V bus )+4000-Vvip ×3000

[0033] Where: V pri is the calculated priority of standby automatic switching, the smaller the value, the higher the priority; V run It is the function start and end mark, the local exit value is 2, the automatic control value is 1, and the other values ​​are 0; V pid It is the value of the automatic pressure plate factor. If there is a pressure plate ID, the value is 1, otherwise it is 0; V bus The active power of the automatic operation busbar; V vip To prepare for the self-investment of important user factors, the value is 1 when there is an important user, otherwise it is 0.

[0034] Furthermore, the control right switching system realizes seamless switching of control rights between the main control system and the backup control system;

[0035] The system is equipped with two control right marking points, each of which has two status values, representing control right and no control right respectively, and the status of these two marking points always remain mutually exclusive. The main dispatching system is responsible for real-time monitoring of the status of the main dispatching control right marking point, while the backup dispatching system is responsible for real-time detection of the status of the backup dispatching control right marking point. The switching operation of the system is essentially completed by exchanging the status values ​​of the two control right marking points, thereby realizing seamless handover of control rights;

[0036] The calculation process of the system can be divided into two major links: strategy calculation and strategy execution. The strategy calculation link is responsible for calculating the start and stop strategies of each standby automatic start-up according to the current operating conditions of the power grid, while the strategy execution link uses the results of the strategy calculation to remotely control the start and stop of the standby automatic start-up soft pressure plate. In the main and standby dispatching systems, only the strategy execution link will detect their respective control right marking points. When the marking point shows "with control right", the remote control will be executed, otherwise it will not be executed.

[0037] Further, the fault detection system is a well-designed fault detection mechanism;

[0038] The main dispatching system and the backup dispatching system use a heartbeat detection mechanism. These two systems periodically send heartbeat messages to the message queue server to detect the running status of the server.

[0039] If the main or backup system cannot receive the heartbeat information sent by itself or the other party within the set time window, the system will immediately determine it as a communication failure and take corresponding troubleshooting measures;

[0040] When the system receives its own and the other party's heartbeat information normally again, it means that the communication link has been restored and data synchronization can continue.

[0041] Furthermore, the steps of implementing data backup and synchronization by the data synchronization system are as follows;

[0042] 1) Authentication before synchronization

[0043] The control right of the system is determined and obtained through the SCADA detection point. For this purpose, the platform provides two detection points: the main control right detection point (marked as: A1) and the backup control right detection point (marked as: B1). These two detection points are mutually exclusive in the same system, which means that at the same time, only one detection point can indicate the control right. At the same time, these two detection points correspond to each other in two different systems and have the same function and meaning.

[0044] 2) Data backup

[0045] Each strategy calculation uses the current operation mode and status to calculate the investment and withdrawal status of the standby automatic investment;

[0046] 3) Data synchronization.

[0047] Furthermore, the data synchronization steps are as follows:

[0048] 1) Data classification

[0049] It is mainly divided into the following categories:

[0050] a System interface operation data: After the user performs system parameter maintenance through the human-machine interface, the relevant maintenance information will be synchronized to the other party's system to ensure the consistency of the system parameters of both parties;

[0051] b. Operation data of the standby automatic start interface: If the user modifies the standby automatic start status through the human-machine interface, the corresponding control information must be synchronized to the other party's system in real time to maintain the synchronous update of the standby automatic start status;

[0052] c System calculation data: The two systems perform strategy calculations independently, but the calculation interval is set to 10 minutes. To avoid large deviations in strategy due to differences in calculation time points, the system with control will send the calculation time to the other system after completing the calculation. After receiving it, the other system will check it against its own time and directly correct the local timer if it exceeds the allowable error range;

[0053] d. All system data: The ABC computing system has a small amount of data, which only contains one entry to record various computing information. Currently, every 6 hours, the system with control rights will synchronize all data to the system without control rights;

[0054] e All data of backup self-investment: The amount of data of ABC backup self-investment is moderate, with about 2,000 records. Every 6 hours, the system with control rights will synchronize all the calculation data of backup self-investment to the system without control rights;

[0055] f Heartbeat data: This data is sent periodically to detect whether the connection with the server is disconnected. When the connection is disconnected and reconnected successfully, it will trigger the synchronization of all data in different systems and all data in the backup system.

[0056] 2) Communication method

[0057] In the active-active system, we use JMS message queues to implement the mechanism of message transmission and data exchange. Specifically, we set up message queues between the primary system and the backup system, and use the API provided by JMS to realize message production and consumption, thereby achieving two-way message transmission and data exchange;

[0058] 3) Synchronization direction

[0059] Except for heartbeat, other data synchronization is always synchronized from the authorized party to the unauthorized party.

[0060] Furthermore, the data verification system integrates a variety of autonomous judgment and self-help functions including the following parts:

[0061] 1) Data consistency check

[0062] The consistency check of active-active data synchronization requires an independent module to perform. This module is responsible for checking three types of data: system global parameters, standby automatic start-up control parameters, and standby automatic start-up strategies. The system will periodically export these three types of data in the form of files and send them to the comparison server for comparison. The comparison server reads the content in the file and makes a detailed comparison of the data of the two systems. The comparison results are displayed in a graphical interface, including the number of data comparisons, the consistency ratio, and the specific difference details. At the same time, according to different data types, a corresponding consistency ratio threshold will be set. If the threshold is exceeded, the switching of the active and standby systems is not allowed.

[0063] 2) Comprehensive parameter synchronization after communication is restored

[0064] When the heartbeat detection communication is restored, the system will perform comprehensive parameter synchronization, including global parameters and standby automatic control parameters;

[0065] 3) Automatic restart strategy for abnormal process exit

[0066] When it is detected that a process exits abnormally, the system will immediately try to automatically restart the process. If the automatic restart fails, the system will continue to try, but the number of attempts is limited. If multiple consecutive attempts fail to successfully restart the process, it is considered that the problem may be permanent and have a fatal impact on system operation.

[0067] A method for automatically throwing and retracting a standby automatic pressure plate based on an active-active architecture, using the above-mentioned automatic throwing and retracting system for a standby automatic pressure plate, comprises the following steps;

[0068] aIn a dual-active system, the status of the other party and its own system is periodically monitored through heartbeat testing;

[0069] b. If the system loses connection or fails, the control right switching process needs to be started. The control right is switched according to the preset strategy and mechanism. The system control right can also be switched manually when the system is normal. Regardless of the switching method, only one system has control right.

[0070] c In the active-active system, both have the function of periodically collecting data and calculating the activation and deactivation strategies;

[0071] d. Use a master-slave mechanism, where the one with control is the master and the other is the backup;

[0072] The integrity, correctness and consistency of the synchronized data and the respective calculation result data are checked periodically, so that problems can be discovered early and corrected and updated in a timely manner.

[0073] The beneficial effects of the present invention are:

[0074] 1. High reliability. The active-active system adopts dual backup and real-time synchronization. In the event of a system failure, it can automatically switch to the backup system immediately to ensure high reliability of the system.

[0075] 2. High availability. The active-active system is highly flexible and scalable. It can dynamically perform distributed deployment and fault recovery operations on data, thereby ensuring high availability of the system.

[0076] 3. Real-time synchronization: The active-active system uses real-time data synchronization to reduce the risk of data synchronization delay and data inconsistency, and ensure the real-time and consistency of data.

[0077] 4. Improve user experience. The active-active system can provide more stable and reliable services and often has a shorter response time, thereby improving user experience and satisfaction. DETAILED DESCRIPTION

[0078] The invention discloses an automatic throw-in and throw-out system for a standby self-throwing pressure plate based on a dual-active architecture, comprising a dual-active system for real-time calculation, a control right switching system, a fault detection system, a data synchronization system, and a data verification system.

[0079] The active-active system is a high-availability system architecture whose core idea is to achieve real-time data synchronization and active-active operation of services between two or more data centers. This architecture can significantly improve the system's disaster recovery capabilities and business continuity, ensuring that when the primary data center fails, the backup data center can seamlessly take over the business, ensuring service continuity and stability.

[0080] The active-active system is a highly intelligent system that is deeply developed and applied based on EMS (energy management system). It not only makes full use of the real-time collected SCADA information, but also comprehensively and comprehensively considers multiple key factors such as the initial load rate of the main transformer, the N-1 verification results of the power grid equipment, the priority setting of the backup automatic action, and possible data abnormalities. The real-time calculation of the power grid backup automatic input and output steps is as follows:

[0081] 1) Online switching strategy of automatic backup based on transformer oil temperature

[0082] In actual operation, the short-term overload capacity of the transformer is closely related to the oil temperature. The main transformer oil temperature is a key indicator reflecting the equipment status. Therefore, in the backup automatic switch-in and switch-out strategy, we can innovatively introduce the main transformer oil temperature as a control variable. Specifically, if the oil temperature of the associated transformer exceeds the preset value, the system will automatically exit the backup automatic switch-in to avoid the risk of overheating; if the oil temperature is within a safe range, the system may consider switching on the backup automatic switch based on other conditions.

[0083] 2) N-1 verification

[0084] In the processing of the automatic backup pressure plate, we only conduct a detailed analysis of the N-1 faults that can trigger the automatic backup, and no in-depth analysis is required for others. Therefore, the N-1 fault analysis of the automatic backup pressure plate is a subset of the traditional N-1 calculation. Since the remote control of the automatic backup pressure plate needs to be operated online, the calculation speed of the N-1 fault is extremely high. When simulating the N-1 fault that may cause the automatic backup, the topology calculation and flow analysis need to be re-performed, which consumes a lot of computing time. In order to reduce the computing resource usage, we adopted a new processing method based on the characteristics of the regional power grid. This method is based on the base state topology, establishes the automatic backup action model by searching the key power supply path, and uses the load transfer method to greatly shorten the processing time;

[0085] 3) Bad data processing

[0086] Since the pressure plate activation and deactivation strategy is based on the main transformer oil temperature calibration and real-time N-1 calibration, the following bad data situations need to be considered:

[0087] a) If the oil temperature of the relevant transformer is abnormal, and the standby automatic start setting takes the oil temperature factor into consideration, the current operating state of the standby automatic start pressure plate shall be maintained unchanged;

[0088] b) In the state estimation calculation, if bad data is detected in the 220kV main transformer related to the backup automatic switch, the 220kV main transformer will not be checked;

[0089] c) In the state estimation calculation, if bad data is detected in the 110kV main transformer or 110kV line, the current operating state of the backup automatic voltage board is maintained unchanged;

[0090] 4) Handling abnormal situations

[0091] During the execution of the pressure plate launch and withdrawal of the standby automatic start-up, the following abnormal situations need to be considered: 1) SCADA telemetry data is not refreshed; 2) state estimation does not converge; 3) the maximum number of remote control times per hour for the standby automatic start-up is limited; 4) the maximum number of remote control times per day for the standby automatic start-up is limited.

[0092] In step 1), the platen insertion and withdrawal strategy taking into account the oil temperature and initial load rate is as follows:

[0093] 1) 220kV main transformer oil temperature and initial load rate detection;

[0094] a) For a 220kV main transformer, if the initial load rate exceeds 0.9 (adjustable), the corresponding standby automatic switch should be locked;

[0095] b) For a 220kV main transformer, if the initial load rate exceeds 0.85 (adjustable), and at the same time the oil temperature exceeds 85℃ (adjustable) or the winding temperature exceeds 105℃ (adjustable), the corresponding standby automatic switch should also be locked;

[0096] The locked standby automatic switching devices refer specifically to those automatic switching devices that may transfer the load of other 220kV network segments to this segment after operation, but the automatic switching devices that will not increase the load of this transformer after automatic switching do not need to be locked;

[0097] 2) 110kV main transformer oil temperature and initial load rate detection

[0098] a) For a 110kV main transformer, if the initial load factor exceeds 0.85 (adjustable), the corresponding standby automatic switching function should be locked.

[0099] b) For a 110kV main transformer, if the initial load rate exceeds 0.85 (adjustable), and at the same time the oil temperature exceeds 85℃ (adjustable) or the winding temperature exceeds 105℃ (adjustable), the corresponding standby automatic switching should also be locked.

[0100] The locked backup automatic transfer object specifically refers to the 10kV automatic transfer device connected to this 110kV transformer, which is intended to prevent the load transfer after the backup automatic transfer action from increasing the burden on the already high-load main transformer.

[0101] In step 2), the fault analysis and processing method is as follows:

[0102] 1) Determine the backup automatic switch caused by N-1 fault based on power supply path search

[0103] When the backup automatic transfer system meets the charging conditions, if the working bus loses power and the backup power supply is energized, the backup automatic transfer will start. Therefore, the backup automatic transfer action can be transformed into a problem of searching for the power supply path of the working power supply and the backup power supply. In the regional power grid, we regard the 220kV main transformer and the modeling generator as power supply points, search from these points downward to the power supply bus with the lowest voltage, and record the power supply path that meets the conditions;

[0104] 2) Detailed processing based on power supply path search

[0105] For regional power grids, standby automatic re-start is mainly used for 110kV, 35kV and 10kV voltage levels, while the 220kV bus usually operates as a ring network. Its line failure generally does not cause standby automatic re-start or change the number of system nodes, and can be handled by specific methods. When searching for power supply paths, we start with the 220kV bus and generator nodes as the root nodes. The specific logic is as follows: simulate disconnecting the 220kV line switch to establish the relationship between nodes and branches; use depth-first downward search to record the low-voltage bus power supply paths encountered; establish a mapping relationship between equipment and standby automatic re-start bus; identify loop information, and loop equipment failure will not cause standby automatic re-start; if the equipment has a mapping relationship with both the working bus and the standby bus, it is regarded as a homologous device, and its failure will not trigger the standby automatic re-start; record the number of power supply points related to the standby automatic re-start, and determine the key power supply path equipment, whose failure will trigger the standby automatic re-start;

[0106] 3) Processing of standby automatic investment and withdrawal strategies based on multiple objectives

[0107] When a specific N-1 fault will trigger N backup automatic switching operations, if all the backup automatic switching operations lead to overload of power grid components, and when less than all the backup automatic switching operations occur, no other power grid components are overloaded, it is obviously not practical to fully block all the backup automatic switching operations in this case. In fact, only some of the backup automatic switching operations need to be blocked. Therefore, a priority is set for each backup automatic switching operation.

[0108] V pri =10000×V run +1000×V pid -int(V bus )+4000-V vip ×3000

[0109] Where: V pri is the calculated priority of standby automatic switching, the smaller the value, the higher the priority; V run It is the function start and end mark, the local exit value is 2, the automatic control value is 1, and the other values ​​are 0; V pid It is the value of the automatic pressure plate factor. If there is a pressure plate ID, the value is 1, otherwise it is 0; V bus The active power of the automatic operation busbar; V vipTo prepare for the self-investment of important user factors, the value is 1 when there is an important user, otherwise it is 0.

[0110] The control right switching system realizes seamless switching of control rights between the main control system and the backup control system;

[0111] 1) Switching instructions

[0112] The system is equipped with two control right marking points, each of which has two status values, representing control right and no control right respectively, and the status of these two marking points always remain mutually exclusive. The main dispatching system is responsible for real-time monitoring of the status of the main dispatching control right marking point, while the backup dispatching system is responsible for real-time detection of the status of the backup dispatching control right marking point. The switching operation of the system is essentially completed by exchanging the status values ​​of the two control right marking points, thereby realizing seamless handover of control rights;

[0113] The calculation process of the system can be divided into two major links: strategy calculation and strategy execution. The strategy calculation link is responsible for calculating the launch and withdrawal strategies of each standby automatic switch according to the current operating conditions of the power grid, while the strategy execution link uses the results of the strategy calculation to remotely launch and withdraw the standby automatic switch soft pressure plate. In the main and standby dispatching systems, only the strategy execution link will detect their respective control rights mark points. When the mark point shows "with control rights", remote control will be executed, otherwise it will not be executed. The switching operation of the system is essentially achieved by exchanging the status values ​​of the two control rights mark points to ensure a smooth transition of control rights.

[0114] 2) Switching effect

[0115] Seamless transfer of control rights: By exchanging the status values ​​of the control rights mark points between the main control and the backup control, the system can achieve a smooth transition of control rights, ensuring that only one system has control over the backup automatic investment at any time, thereby avoiding conflicts and confusion in control rights.

[0116] Strategy execution continuity: During the switching process, the strategy execution link will decide whether to execute the remote control operation based on the status of the control right mark point. This ensures that the standby automatic investment and withdrawal strategies can be continuously executed before and after the switching, without interruption or delay due to the system switching.

[0117] The fault detection system is a well-designed fault detection mechanism;

[0118] The data synchronization between the main dispatching system and the backup dispatching system mainly relies on the message queue server, which plays a key role in data transmission and buffering. However, once the message queue server fails, the entire data synchronization link will be seriously affected and the system will not be able to complete the data synchronization work normally. Therefore, ensuring the stable operation of the message queue server becomes the core task of data synchronization fault detection.

[0119] To achieve this goal, the main dispatching system and the backup dispatching system adopt a heartbeat detection mechanism. Specifically, the two systems periodically send heartbeat messages to the message queue server to detect the running status of the server. This heartbeat message is a lightweight communication signal and will not interfere with the normal business of the system.

[0120] If the main or backup system cannot receive the heartbeat information sent by itself or the other party within the set time window, the system will immediately determine it as a communication failure and take corresponding troubleshooting measures. This timely fault detection mechanism helps to quickly locate the problem and reduce data synchronization delays and losses;

[0121] When the system receives its own and the other party's heartbeat information normally again, it means that the communication link has been restored and data synchronization can continue.

[0122] The steps for the data synchronization system to achieve data backup and synchronization are as follows;

[0123] 1) Authentication before synchronization

[0124] The control right of the system is determined and obtained through the SCADA detection point. For this purpose, the platform provides two detection points: the main control right detection point (marked as: A1) and the backup control right detection point (marked as: B1). These two detection points are mutually exclusive in the same system, which means that at the same time, only one detection point can indicate the control right. At the same time, these two detection points correspond to each other in two different systems and have the same function and meaning. The combination of these two detection points provides a basis for the flexible switching of the system and the smooth transition of control right:

[0125]

[0126] As can be seen from the table above, the permissions of the two systems A / B are mutually exclusive. Because the programs are the same, there are two ways to achieve mutual exclusion. One is that the two systems A / B detect different points respectively; the other is that the two systems A / B detect the same point, but the detection values ​​are different. This has problems. If the configuration is different, there will be an overlay during the synchronous update. To prevent this problem, double verification is proposed;

[0127] The dual-verification configuration unit design binds the configuration information to the node name. When the system is running, it only reads the configuration belonging to this node, realizing the effective configuration of different requirements on different nodes for the same file.

[0128] 2) Data backup

[0129] Each strategy calculation uses the current operation mode and status to calculate the standby automatic investment and withdrawal status. Since the loss of historical status will not affect the result of strategy calculation, there is no need to back up the historical status;

[0130] 3) Data synchronization.

[0131] The steps for data synchronization are as follows:

[0132] 1) Data classification

[0133] The system involves both computing and operation functions. In order to effectively synchronize data and make rational use of resources, we need to classify the data in detail. Currently, it is mainly divided into the following categories:

[0134] a System interface operation data: After the user performs system parameter maintenance through the human-machine interface, the relevant maintenance information needs to be synchronized to the other party's system to ensure the consistency of the system parameters of both parties;

[0135] b. Operation data of the standby automatic start interface: If the user modifies the standby automatic start status through the human-machine interface, the corresponding control information must be synchronized to the other party's system in real time to maintain the synchronous update of the standby automatic start status;

[0136] c System calculation data: The two systems perform strategy calculations independently, but the calculation interval is set to 10 minutes. To avoid large deviations in strategy due to differences in calculation time points, the system with control will send the calculation time to the other system after completing the calculation. After receiving it, the other system will check it against its own time and directly correct the local timer if it exceeds the allowable error range;

[0137] d. All system data: The ABC computing system has a small amount of data, which only contains one entry to record various computing information. Considering that the network or other factors may cause synchronization differences, the system with control rights will synchronize all data to the system without control rights every 6 hours.

[0138] e All data of backup self-investment: The amount of data of ABC backup self-investment is moderate, with about 2,000 records. Also considering the possibility of synchronization differences, every 6 hours, the system with control rights will synchronize all calculation data of backup self-investment (including strategies and statistics) to the system without control rights;

[0139] f Heartbeat data: This data is sent periodically to detect whether the connection with the server is disconnected. When the connection is disconnected and reconnected successfully, it will trigger the synchronization operation of all data of different systems and all data of the backup automatic investment (from the system with control authority to the system without control authority);

[0140] 2) Communication method

[0141] In the active-active system, we use JMS message queues to implement the mechanism of message transmission and data exchange. Specifically, we set up message queues between the primary system and the backup system, and use the API provided by JMS to realize message production and consumption, thereby achieving two-way message transmission and data exchange;

[0142] In order to ensure the real-time and reliability of message transmission and data exchange between the OCS active-active system and the JMS message queue, we used the message listener provided by JMS to monitor messages in real time. In this way, when a new message appears in the message queue, the listener can respond quickly, thereby ensuring the timely response capability of the system.

[0143] In addition, when both the primary and backup systems have file data channel functions, the sending side of the JMS data channel needs to add the primary and backup tags in the JMS message. The receiving side needs to judge the running status of the channel based on the received tags to avoid responding to the messages it has sent, ensuring the stability and accuracy of the system.

[0144] 3) Synchronization direction

[0145] Except for heartbeat, other data synchronization is always synchronized from the authorized party to the unauthorized party.

[0146] The data verification system integrates a variety of autonomous judgment and self-rescue functions, including the following parts:

[0147] 1) Data consistency check

[0148] The consistency check of active-active data synchronization requires an independent module to perform. This module is responsible for checking three types of data: system global parameters, standby automatic start-up control parameters, and standby automatic start-up strategies. The system will periodically export these three types of data in the form of files and send them to the comparison server for comparison. The comparison server reads the content in the file and makes a detailed comparison of the data of the two systems. The comparison results are displayed in a graphical interface, including the number of data comparisons, the consistency ratio, and the specific difference details. At the same time, according to different data types, a corresponding consistency ratio threshold will be set. If the threshold is exceeded, the switching of the active and standby systems is not allowed.

[0149] 2) Comprehensive parameter synchronization after communication is restored

[0150] When the heartbeat detection communication is restored, the system will perform comprehensive parameter synchronization, including global parameters and backup automatic control parameters. There is only one global parameter, and although the backup automatic control parameters are synchronized according to the records of the backup automatic device, considering that there are only more than 2,000 backup automatic devices in Guangzhou, the overall data volume is not large, and the total resources occupied by a single synchronization operation are only a few MB;

[0151] 3) Automatic restart strategy for abnormal process exit

[0152] The completion of a series of operations of dual-active depends on the stable operation of related processes on the active and standby systems. Once these processes exit abnormally, all operations will not be completed smoothly. To solve this problem, the system has designed an automatic pull-up mechanism. When a process exits abnormally, the system will immediately try to automatically restart the process. If the automatic pull-up fails, the system will continue to try, but the number of attempts is limited. If multiple consecutive attempts fail to successfully pull up the process, it is considered that the problem may be permanent and have a fatal impact on system operation.

[0153] A method for automatically throwing and retracting a standby automatic pressure plate based on an active-active architecture, using the above-mentioned automatic throwing and retracting system for a standby automatic pressure plate, comprises the following steps;

[0154] aIn a dual-active system, the status of the other party and its own system is periodically monitored through heartbeat testing;

[0155] b. If the system loses connection or fails, the control right switching process needs to be started. The control right is switched according to the preset strategy and mechanism. The system control right can also be switched manually when the system is normal. Regardless of the switching method, only one system has control right.

[0156] c In the active-active system, both have the function of periodically collecting data and calculating the activation and deactivation strategies;

[0157] d. The master-slave mechanism is adopted. The one with control right is the master and the other is the backup. Only the master system can perform data maintenance and issue policy control commands, and synchronize maintenance information and control information to the backup system through the message bus;

[0158] The integrity, correctness and consistency of the synchronized data and the respective calculation result data are checked periodically, so that problems can be discovered early and corrected and updated in a timely manner.

[0159] The relationship between the automatic switching system of the backup pressure plate (ABC) and the active-active system can be analyzed from multiple dimensions such as functional coordination, system architecture, data synchronization and mutual backup switching. 1. Functional coordination: ABC is mainly responsible for the calculation and execution of the state strategy of the backup pressure plate in the power system. The active-active system can immediately start the other system when any system fails by realizing real-time data synchronization and mutual backup switching of the two ABC systems, so as to achieve seamless switching and ensure that business continuity is not affected. 2. System architecture; independence and complementarity: As part of the power automation control system, the ABC system focuses on the automatic switching function of the backup pressure plate. The active-active system starts from the system architecture level, deploys two independent ABC systems, and realizes data synchronization and mutual backup switching between them to improve the redundancy and reliability of the entire system. Integration and coordination: In the active-active architecture, the two ABC systems need to be effectively integrated to form a unified and highly coordinated whole. This includes the standardization of data interfaces, the unification of communication protocols, and the precise design of mutual backup switching logic. 3. Data synchronization and mutual backup switching; Data synchronization: The active-active system needs to achieve real-time data synchronization between the two ABC systems to ensure that the two systems have completely consistent data status at any time. Mutual backup switching: When an ABC system fails, the active-active system needs to be able to trigger a mutual backup switching operation to smoothly switch the business traffic to another normal ABC system.

Claims

1. A system for automatically launching and retreating a standby pressure plate based on a dual-active architecture, characterized in that: It includes a dual-active system for real-time computing, a control switching system, a fault detection system, a data synchronization system, and a data verification system.

2. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 1 is characterized in that: The active-active system is a highly intelligent system that is deeply developed and applied based on EMS. It calculates the steps of power grid backup and automatic investment in real time as follows: 1) Online switching strategy of automatic backup based on transformer oil temperature In actual operation, the short-term overload capacity of the transformer is closely related to the oil temperature. The oil temperature of the main transformer is a key indicator reflecting the status of the equipment. Specifically, if the oil temperature of the associated transformer exceeds the preset value, the system will automatically exit the standby mode to avoid overheating risks; if the oil temperature is within a safe range, the system may consider switching on based on other conditions; 2) N-1 verification The N-1 fault analysis of the automatic backup pressure plate is a subset of the traditional N-1 calculation. According to the characteristics of the regional power grid, a new processing method is adopted. This method is based on the base state topology, establishes the automatic backup action model by searching the power supply key path, and uses the load transfer method to greatly shorten the processing time. 3) Bad data processing Since the pressure plate activation and deactivation strategy is based on the main transformer oil temperature calibration and real-time N-1 calibration, the following bad data situations need to be considered: a) If the oil temperature of the relevant transformer is abnormal, and the standby automatic start setting takes the oil temperature factor into consideration, the current operating state of the standby automatic start pressure plate shall be maintained unchanged; b) In the state estimation calculation, if bad data is detected in the 220kV main transformer related to the backup automatic switch, the 220kV main transformer will not be checked; c) In the state estimation calculation, if bad data is detected in the 110kV main transformer or 110kV line, the current operating state of the backup automatic voltage board is maintained unchanged; 4) Handling abnormal situations During the execution of the pressure plate launch and withdrawal of the standby automatic start-up, the following abnormal situations need to be considered: 1) SCADA telemetry data is not refreshed; 2) state estimation does not converge; 3) the maximum number of remote control times per hour for the standby automatic start-up is limited; 4) the maximum number of remote control times per day for the standby automatic start-up is limited.

3. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 2 is characterized in that: In step 1), the platen insertion and withdrawal strategy taking into account the oil temperature and initial load rate is as follows: 1) 220kV main transformer oil temperature and initial load rate detection; a) For a 220kV main transformer, if the initial load rate exceeds 0.9, the corresponding standby automatic switch should be locked; b) For a 220kV main transformer, if the initial load rate exceeds 0.85 and the oil temperature exceeds 85℃ or the winding temperature exceeds 105℃, the corresponding standby automatic switch should also be locked; 2) 110kV main transformer oil temperature and initial load rate detection a) For a 110kV main transformer, if the initial load rate exceeds 0.85, the corresponding backup automatic switch should be locked. b) For a 110kV main transformer, if the initial load rate exceeds 0.85, and at the same time the oil temperature exceeds 85°C (adjustable) or the winding temperature exceeds 105°C, the corresponding standby automatic switching should also be locked.

4. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 2 is characterized in that: In step 2), the fault analysis and processing method is as follows: 1) Determine the backup automatic switch caused by N-1 fault based on power supply path search When the backup automatic transfer system meets the charging conditions, if the working bus loses power but the backup power supply is energized, the backup automatic transfer system will start; 2) Detailed processing based on power supply path search When searching for power supply paths, we start with the 220kV bus and generator nodes as the root nodes. The specific logic is as follows: simulate disconnecting the 220kV line switch to establish the relationship between the node and the branch; use depth-first downward search to record the low-voltage bus power supply path encountered; establish a mapping relationship between the equipment and the backup automatic re-start bus; identify loop information, and loop equipment failure will not cause the backup automatic re-start action; if the equipment has a mapping relationship with both the working bus and the backup bus, it is regarded as a homologous device, and its failure will not trigger the backup automatic re-start; record the number of power supply points related to the backup automatic re-start, and determine the key power supply path equipment, whose failure will trigger the backup automatic re-start action; 3) Processing of standby automatic investment and withdrawal strategies based on multiple objectives When a specific N-1 fault will trigger N backup automatic switching operations, if all the backup automatic switching operations lead to overload of power grid components, and when less than all the backup automatic switching operations occur, no other power grid components are overloaded, it is obviously not practical to fully block all the backup automatic switching operations in this case. In fact, only some of the backup automatic switching operations need to be blocked. Therefore, a priority is set for each backup automatic switching operation. V pri =10000×V run +1000×V pid -int(V bus )+4000-V vip ×3000 Where: V pri is the calculated priority of standby automatic switching, the smaller the value, the higher the priority; V run It is the function start and end mark, the local exit value is 2, the automatic control value is 1, and the other values ​​are 0; V pid It is the value of the automatic pressure plate factor. If there is a pressure plate ID, the value is 1, otherwise it is 0; V bus The active power of the automatic operation busbar; V vip To prepare for the self-investment of important user factors, the value is 1 when there is an important user, otherwise it is 0.

5. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 1, characterized in that: The control right switching system realizes seamless switching of control rights between the main control system and the backup control system; The system is equipped with two control right marking points, each of which has two status values, representing control right and no control right respectively, and the status of these two marking points always remain mutually exclusive. The main dispatching system is responsible for real-time monitoring of the status of the main dispatching control right marking point, while the backup dispatching system is responsible for real-time detection of the status of the backup dispatching control right marking point. The switching operation of the system is essentially completed by exchanging the status values ​​of the two control right marking points, thereby realizing seamless handover of control rights; The calculation process of the system can be divided into two major links: strategy calculation and strategy execution. The strategy calculation link is responsible for calculating the start and stop strategies of each standby automatic start-up according to the current operating conditions of the power grid, while the strategy execution link uses the results of the strategy calculation to remotely start and stop the standby automatic start-up soft pressure plate. In the main and standby dispatching systems, only the strategy execution link will detect their respective control right mark points. When the mark point shows "control right", remote control will be executed, otherwise it will not be executed.

6. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 1, characterized in that: The fault detection system is a well-designed fault detection mechanism; The main dispatching system and the backup dispatching system use a heartbeat detection mechanism. These two systems periodically send heartbeat messages to the message queue server to detect the running status of the server. If the main or backup system cannot receive the heartbeat information sent by itself or the other party within the set time window, the system will immediately determine it as a communication failure and take corresponding troubleshooting measures; When the system receives its own and the other party's heartbeat information normally again, it means that the communication link has been restored and data synchronization can continue.

7. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 1 is characterized in that: The steps for the data synchronization system to achieve data backup and synchronization are as follows; 1) Authentication before synchronization The control right of the system is determined and obtained through the SCADA detection point. For this purpose, the platform provides two detection points: the main control right detection point (marked as: A1) and the backup control right detection point (marked as: B1). These two detection points are mutually exclusive in the same system, which means that at the same time, only one detection point can indicate the control right. At the same time, these two detection points correspond to each other in two different systems and have the same function and meaning. 2) Data backup Each strategy calculation uses the current operation mode and status to calculate the investment and withdrawal status of the standby automatic investment; 3) Data synchronization.

8. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 7 is characterized in that: The steps for data synchronization are as follows: 1) Data classification It is mainly divided into the following categories: a System interface operation data: After the user performs system parameter maintenance through the human-machine interface, the relevant maintenance information will be synchronized to the other party's system to ensure the consistency of the system parameters of both parties; b. Operation data of the standby automatic start interface: If the user modifies the standby automatic start status through the human-machine interface, the corresponding control information must be synchronized to the other party's system in real time to maintain the synchronous update of the standby automatic start status; c System calculation data: The two systems perform strategy calculations independently, but the calculation interval is set to 10 minutes. To avoid large deviations in strategy due to differences in calculation time points, the system with control will send the calculation time to the other system after completing the calculation. After receiving it, the other system will check it against its own time and directly correct the local timer if it exceeds the allowable error range; d. All system data: The ABC computing system has a small amount of data, which only contains one entry to record various computing information. Currently, every 6 hours, the system with control rights will synchronize all data to the system without control rights; e All data of backup self-investment: The amount of data of ABC backup self-investment is moderate, with about more than 2,000 records. Every 6 hours, the system with control rights will synchronize all the calculation data of backup self-investment to the system without control rights; f Heartbeat data: This data is sent periodically to detect whether the connection with the server is disconnected. When the connection is disconnected and reconnected successfully, it will trigger the synchronization of all data in different systems and all data in the backup system. 2) Communication method In the active-active system, we use JMS message queues to implement the mechanism of message transmission and data exchange. Specifically, we set up message queues between the primary system and the backup system, and use the API provided by JMS to realize message production and consumption, thereby achieving two-way message transmission and data exchange; 3) Synchronization direction Except for heartbeat, other data synchronization is always synchronized from the authorized party to the unauthorized party.

9. The automatic launching and retreating system of the backup pressure plate based on the active-active architecture as claimed in claim 1, characterized in that: The data verification system integrates a variety of autonomous judgment and self-rescue functions, including the following parts: 1) Data consistency check The consistency check of active-active data synchronization requires an independent module to perform. This module is responsible for checking three types of data: system global parameters, standby automatic start-up control parameters, and standby automatic start-up strategies. The system will periodically export these three types of data in the form of files and send them to the comparison server for comparison. The comparison server reads the content in the file and makes a detailed comparison of the data of the two systems. The comparison results are displayed in a graphical interface, including the number of data comparisons, the consistency ratio, and the specific difference details. At the same time, according to different data types, a corresponding consistency ratio threshold will be set. If the threshold is exceeded, the switching of the active and standby systems is not allowed. 2) Comprehensive parameter synchronization after communication is restored When the heartbeat detection communication is restored, the system will perform comprehensive parameter synchronization, including global parameters and standby automatic control parameters; 3) Automatic restart strategy for abnormal process exit When it is detected that a process exits abnormally, the system will immediately try to automatically restart the process. If the automatic restart fails, the system will continue to try, but the number of attempts is limited. If multiple consecutive attempts fail to successfully restart the process, it is considered that the problem may be permanent and have a fatal impact on system operation.

10. A method for automatically launching and withdrawing a standby automatic pressure plate based on an active-active architecture, using the automatic launching and withdrawing system for a standby automatic pressure plate according to claims 1 to 9, characterized in that: The steps include: aIn a dual-active system, the status of the other party and its own system is periodically monitored through heartbeat testing; b. If the system loses connection or fails, the control right switching process needs to be started. The control right is switched according to the preset strategy and mechanism. The system control right can also be switched manually when the system is normal. Regardless of the switching method, only one system has control right. c In the active-active system, both have the function of periodically collecting data and calculating the activation and deactivation strategies; d. Use a master-slave mechanism, where the one with control is the master and the other is the backup; The integrity, correctness and consistency of the synchronized data and the respective calculation result data are checked periodically, so that problems can be discovered early and corrected and updated in a timely manner.

Images