Protection setting optimization method and system based on differential risk constraint

Abstract

This invention provides a method and system for optimizing protection settings based on differential risk constraints, belonging to the field of protection device setting technology. The method specifically involves: constructing a differential risk expectation function based on electrical operating parameters and fault characteristic parameters, combined with topological correlation and coordination relationships; using the minimum total protection operating time as the optimization objective and the differential risk expectation function as a penalty term to construct an optimized setting objective function; setting upper and lower limits of constraints based on topological information, electrical operating parameters, and fault characteristic parameters, and setting a differential margin correction coefficient in conjunction with operational influencing factors to obtain global constraint conditions; and performing optimization calculations on the optimized setting objective function under these constraints to obtain the optimal protection setting combination for protection setting adjustment. This invention can effectively alleviate the disturbance effect of extreme operating conditions on protection differential coordination, achieve synergistic optimization of relay protection speed and selectivity, and improve the accuracy and adaptability of setting parameters under all operating conditions.

Description

Technical Field

[0001] This invention relates to the field of protection device setting optimization technology, and in particular to a protection setting optimization method and system based on differential risk constraints. Background Technology

[0002] Speed ​​and selectivity are the main performance requirements that relay protection devices must meet in coordination during the setting of power system relay protection settings. Speed ​​is mainly reflected by minimizing the total operating time of power grid protection and relying on rapid fault clearing to reduce the risk of equipment impact and power grid disturbance. Selectivity is reflected in the hierarchical coordination relationship between protection devices, which can achieve accurate isolation of fault sections and avoid over-level tripping and protection failure to operate.

[0003] In actual power grid operation, the coordination of protection levels can remain basically stable under normal operating conditions. Therefore, current protection setting methods mostly use electrical operating parameters and basic topology information under normal power grid conditions as inputs, with the goal of minimizing the total protection operating time. After setting basic constraints, the setting adjustment is completed through a simple optimization algorithm. However, in extreme scenarios such as sudden changes in system operation mode, heavy overload, high-resistance faults, remote faults, and topology switching, the protection measurement parameters and fault current characteristics will deviate, the theoretical level difference margin between upper and lower level protections will be compressed, and the risk of level difference coordination will be significantly increased. If optimization continues to focus solely on speed, i.e., minimizing the total protection operating time, although it can meet the speed requirements in normal scenarios, the level difference margin will be insufficient in extreme scenarios, easily leading to loss of selectivity and cascading tripping.

[0004] It is evident that the current protection setting method is unable to cope with the disturbance impact of extreme scenarios on the protection level coordination, and cannot achieve a coordinated balance between speed in normal scenarios and selectivity in extreme scenarios. The accuracy of the setting parameters is low, and the operating condition adaptation effect is also poor, making it difficult to meet the complex and ever-changing operation setting requirements of the current power grid. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of current protection setting methods, which are unable to cope with the disturbance effects of extreme scenarios on protection level coordination, cannot achieve a coordinated balance between speed and selectivity, have low setting parameter accuracy, and poor operating condition adaptability. This invention provides a protection setting optimization method and system based on level risk constraints. By constructing a level risk expectation function, the level margin loss under extreme operating conditions is pre-set as a quantifiable penalty term. Simultaneously, a level margin correction coefficient and a global rigid constraint are introduced to automatically balance the speed requirements of normal scenarios and the selectivity of extreme scenarios during the optimization process. Finally, the optimal protection setting combination is output without scenario identification or mode switching, effectively mitigating the disturbance effects of extreme operating conditions on protection level coordination, achieving coordinated optimization of relay protection speed and selectivity, and improving the accuracy and adaptability of setting parameters under all operating conditions.

[0006] The objective of this invention is achieved through the following technical solution: The protection setting optimization method based on graded risk constraints includes: Electrical operating parameters and fault characteristic parameters of the entire target power grid are collected, and the differential risk expectation function of each region of the target power grid is constructed by combining the topological correlation and coordination relationship. The optimization objective is to minimize the total protection operating time of the target power grid, and the expected function of the differential risk is used as the penalty term to construct the corresponding optimization setting objective function. Based on the topology information, electrical operating parameters and fault characteristic parameters of the target power grid, set the upper and lower limits of constraints for each region, and set the corresponding differential margin correction coefficients in combination with the operating influencing factors. Construct global constraint conditions based on the upper and lower limits of constraints and differential margin correction coefficients for each region. Under the constraints of the entire domain, the optimization objective function is optimized to obtain the optimal protection setting combination of the target power grid; Based on the optimal protection setting combination, the protection settings of each protection device in each area of ​​the target power grid are set sequentially.

[0007] By collecting electrical operating parameters and fault characteristic parameters of the entire target power grid under various operating conditions, and combining them with regional topological correlations, a differential risk expectation function for each region is constructed to quantify potential protection differential coordination hazards under different operating scenarios and reflect the differential margin loss caused by various extreme operating conditions. Then, taking the minimum total protection action time as the optimization direction, the differential risk expectation function is simultaneously introduced into the optimization setting objective function as a penalty term. This effectively constrains differential coordination imbalance while ensuring basic instantaneous performance, avoiding performance imbalance caused by a single optimization index. Simultaneously, a differential margin correction coefficient is configured based on actual operating influencing factors, and a global constraint condition is established by combining upper and lower limit rigid constraints. This ensures that the constraint condition effectively fits the operating differences of each region, guaranteeing the reliability of subsequent setting results. Finally, the optimization setting objective function is solved within the boundary range of the global constraints, thereby selecting a setting combination with better overall performance within a controllable range. This balances optimization needs and operational safety constraints, balances protection instantaneous performance and selectivity, improves the accuracy of protection setting under all operating conditions, and reduces the impact of extreme operating conditions on differential coordination.

[0008] Furthermore, the topological association and coordination relationship includes at least the segment affiliation relationship of the power grid physical topology, the hierarchical subordinate coordination relationship between the main protection and backup protection, the importance association relationship of the protection devices within the region, and the hierarchical linkage constraint relationship of cross-regional protection.

[0009] Furthermore, the process of collecting electrical operating parameters and fault characteristic parameters of the entire target power grid and constructing the differential risk expectation function for each region of the target power grid based on topological correlation and coordination relationships includes: Based on the electrical operating parameters and fault characteristic parameters of the entire operating condition, as well as the action delay offset data of each protection device under extreme conditions, calculate the extreme value of the differential loss of each main and backup protection pair under various extreme conditions. Based on the actual setting level difference, level difference loss extreme value, minimum level difference limit value and ideal minimum level difference of each primary and backup protection pair, establish the corresponding over-level risk function and delay redundancy risk function. Based on the topological association and coordination relationship, determine the coordination weight of each primary and backup protection pair to its respective area and within the corresponding area; By combining the corresponding coordination weights, the risk functions of the cascading risk and the time-delay redundancy risk functions of each main and backup protection pair are weighted and aggregated according to region to construct the expected risk function of the differential risk for each region of the target power grid.

[0010] Furthermore, the step of establishing corresponding over-level risk functions and delay redundancy risk functions based on the actual setting level difference, level difference loss extreme value, minimum level difference limit, and ideal minimum level difference for each primary and backup protection pair includes: Based on the actual setting level difference, level difference loss extreme value and minimum level difference limit of each main and backup protection pair, determine the corresponding effective residual level difference under extreme working conditions; A mapping relationship is constructed based on the deviation characteristics of the effective remaining level difference relative to the minimum level difference limit of each primary and backup protection pair, forming a risk function for exceeding the level. Based on the difference range between the actual setting level difference and the ideal minimum level difference for each primary and backup protection pair, the corresponding time-limited redundancy is determined; A mapping relationship is constructed based on the changing characteristics of the time-limited redundancy of each primary and backup protection pair, forming a delay redundancy risk function.

[0011] Furthermore, based on the electrical operating parameters and fault characteristic parameters of the entire operating range, as well as the action delay offset data of each protection device under extreme operating conditions, the extreme values ​​of the differential loss of each main and backup protection pair under various extreme operating conditions are calculated, including: Based on the electrical operating parameters and fault characteristic parameters of the entire operating range, determine the typical characteristic parameter combinations of various extreme operating conditions, and determine the action delay offset calibration value of each protection device under the corresponding extreme operating conditions based on the typical characteristic parameter combinations. The corresponding candidate extreme operating condition category is determined based on the installation location, protection type, and status of the associated circuit breaker for each main and backup protection pair. Based on the corresponding action delay offset calibration value, the total differential loss of each main and backup protection pair under each corresponding candidate extreme condition is calculated by superimposing the values. The maximum total differential loss of each main and backup protection pair under all corresponding candidate extreme conditions is taken as the corresponding extreme value of differential loss.

[0012] Furthermore, the setting of upper and lower limits for constraints in each region based on the topology information, electrical operating parameters, and fault characteristic parameters of the target power grid includes: Based on the topology information of the target power grid, the physical wiring structure, node distribution status and line power supply coverage of each region are obtained, and the grid coupling strength and line-level coupling degree of each region are calculated according to the corresponding physical wiring structure and node distribution status. The corresponding electrical disturbance fluctuation range is obtained based on the electrical operating parameters of each region, and the fault impact range is determined based on the fault characteristic parameters. Based on the corresponding primary and backup protection for the extreme values ​​of the corresponding level loss, the regional loss reference value of each region is selected. Combined with the corresponding network coupling strength, line level coupling degree and electrical disturbance fluctuation range, the minimum limit of the protection level difference of each region is determined. Based on the corresponding fault impact range and the power supply coverage of the line, the maximum limit of the protection level difference for each area is determined; Based on the minimum and maximum limits of the corresponding protection level, the level constraint boundaries for each region are formed.

[0013] Furthermore, the step of setting corresponding differential margin correction coefficients in conjunction with operational influencing factors includes: Based on fault characteristic parameters, the fault fluctuation distribution characteristics of each region are extracted, and corresponding adjustment conditions are extracted based on load change data and power grid operation mode switching data of each region. Based on the characteristics of fault fluctuation distribution and the extracted adjustment conditions, the corresponding operational influencing factors for each region are determined; Based on the corresponding grid coupling strength, electrical disturbance fluctuation range and operation influencing factors, the power grid operation stability conditions of each region are matched, and the adjustable range of the level difference parameter is determined according to the regional loss reference amount and the limit range of the protection level difference in each region. Based on operational influencing factors and grid operation stability conditions, the adjustment requirements of protection level differences in each region are matched by regional loss reference values. Under the constraint of the adjustable range of the grade difference parameter, the grade difference margin correction coefficient of each region is set according to the corresponding adjustment requirements.

[0014] Furthermore, the optimization calculation of the objective function under global constraints includes: Using the setting parameters of protection at all levels as the decision variables to be optimized, the decision variables to be optimized are initialized to obtain the initial variable set; Substitute the initial set of variables into the optimization tuning objective function to obtain the initial objective function value; Based on the initial objective function value, and combined with the preset optimization algorithm, the decision variables to be optimized are updated and combined iteratively in a dimension-by-dimensional manner to obtain several combinations of variables; Iterate through the global constraints in turn, and select the feasible solution set that is within the range of the global constraints from the obtained variable combinations; Calculate the objective function value corresponding to each combination of variables in the feasible solution set, and compare them to obtain the current optimal variable set; Based on the objective function value corresponding to the current optimal variable set, the next round of updates and iterations is carried out in combination with the preset optimization algorithm until the iteration termination condition is met, and the optimal protection setting parameter combination is obtained.

[0015] Furthermore, before performing the next round of updates and iterations based on the objective function value corresponding to the current optimal variable set and in conjunction with the preset optimization algorithm, the following is also executed: Based on the objective function values ​​of the remaining variable groups in the current iterative feasible solution, excluding the current optimal variable group, obtain the variable combination distribution characteristics of the current iteration; Based on the distribution characteristics of the variable combinations in the current iteration, the search range of decision variables for the next round of update iteration is set.

[0016] A protection setting optimization tuning system based on differential risk constraints, used to execute any of the above optimization tuning methods, including: The differential risk analysis module is used to collect electrical operating parameters and fault characteristic parameters of the entire target power grid, and construct the differential risk expectation function of each region of the target power grid by combining the topological correlation and coordination relationship. The optimization setting planning module is used to construct the corresponding optimization setting objective function with the goal of minimizing the total protection action time of the target power grid and the expected function of differential risk as the penalty term. At the same time, it sets the upper and lower limits of constraints for each region and the corresponding differential margin correction coefficient to construct global constraints. The optimization setting calculation module is used to perform optimization calculations on the objective function under global constraints to obtain the optimal protection setting combination for the target power grid. The setting module is used to set the protection settings of each protection device in each area of ​​the target power grid in sequence based on the optimal protection setting combination.

[0017] The beneficial effects of this invention are: By collecting electrical operating parameters and fault characteristic parameters of the entire target power grid under various operating conditions, and combining them with regional topological correlations, a differential risk expectation function for each region is constructed to quantify potential protection differential coordination hazards under different operating scenarios and reflect the differential margin loss caused by various extreme operating conditions. Then, taking the minimum total protection action time as the optimization direction, the differential risk expectation function is simultaneously introduced into the optimization setting objective function as a penalty term. This effectively constrains differential coordination imbalance while ensuring basic instantaneous performance, avoiding performance imbalance caused by a single optimization index. Simultaneously, a differential margin correction coefficient is configured based on actual operating influencing factors, and a global constraint condition is established by combining upper and lower limit rigid constraints. This ensures that the constraint condition effectively fits the operating differences of each region, guaranteeing the reliability of subsequent setting results. Finally, the optimization setting objective function is solved within the boundary range of the global constraints, thereby selecting a setting combination with better overall performance within a controllable range. This balances optimization needs and operational safety constraints, balances protection instantaneous performance and selectivity, improves the accuracy of protection setting under all operating conditions, and reduces the impact of extreme operating conditions on differential coordination. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of a process of the present invention; Figure 2 This is a schematic diagram illustrating the construction process of an optimized tuning objective function according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the optimization calculation process for the objective function of the present invention. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0020] Example: A protection setting optimization method based on differential risk constraints, such as... Figure 1 As shown, it includes: Electrical operating parameters and fault characteristic parameters of the entire target power grid are collected, and the differential risk expectation function of each region of the target power grid is constructed by combining the topological correlation and coordination relationship. The optimization objective is to minimize the total protection operating time of the target power grid, and the expected function of the differential risk is used as the penalty term to construct the corresponding optimization setting objective function. Based on the topology information, electrical operating parameters and fault characteristic parameters of the target power grid, set the upper and lower limits of constraints for each region, and set the corresponding differential margin correction coefficients in combination with the operating influencing factors. Construct global constraint conditions based on the upper and lower limits of constraints and differential margin correction coefficients for each region. Under the constraints of the entire domain, the optimization objective function is optimized to obtain the optimal protection setting combination of the target power grid; Based on the optimal protection setting combination, the protection settings of each protection device in each area of ​​the target power grid are set sequentially.

[0021] By collecting electrical operating parameters and fault characteristic parameters of the entire operating range, and combining them with topological correlation and coordination relationships, the potential risks of differential coordination scattered under different operating conditions are integrated to establish a regional differential risk expectation function that covers all anticipated extreme operating conditions and corresponds to the protection coordination logic. This function not only reflects the delay loss that may be caused by differential redundancy under normal operating conditions, but also quantifies the risk of over-level tripping that may be caused by insufficient differential coordination under extreme operating conditions.

[0022] The expected function of the grade difference risk is then placed in the same objective function as the main objective of minimizing the total protection action time, in the form of a penalty term. This means that while pursuing speed, the optimization algorithm must bear the cost of the risk of exceeding the grade due to the compression of grade differences. Thus, within a single objective function, the speed requirements of normal scenarios and the safety requirements of extreme scenarios are integrated, achieving an automatic balance between speed and selectivity.

[0023] Meanwhile, based on the topology information and electrical operating parameters under normal operating conditions, as well as the fault characteristic parameters under extreme operating conditions, the upper and lower limits of the basic rigid constraints for the entire operating range are set, and the differential margin correction coefficient is used to differentiate the influencing factors such as operating mode switching, load fluctuation, and equipment aging, so that the constraints have the ability to adapt to actual operating disturbances and ensure the reliability of the subsequent optimization results under the entire operating range.

[0024] Then, optimization calculations are performed within the global rigid constraint boundary to search for the combination of protection setpoints that simultaneously satisfies all lower and upper constraints and makes the objective function globally optimal, without distinguishing between normal and extreme scenarios. Under the combined effect of the differential risk penalty term and the differential margin correction coefficient, the final output setpoint can effectively resist differential coordination disturbances under extreme operating conditions and avoid the risk of over-level tripping, while automatically compressing unnecessary differential redundancy under normal operating conditions, thus possessing optimal speed and achieving a coordinated balance between speed and selectivity.

[0025] Finally, the output protection setting combination is sent to the protection devices in each area. The target power grid can cover all expected operating conditions with the same set of protection settings. There is no need to design multiple sets of setting combinations for different operating states, nor is there any need to perform any scenario identification or setting group switching operations during operation. This avoids the risk of protection maloperation and failure to operate caused by misjudgment of operating mode, switching logic error or inconsistency of setting groups, and can effectively improve the full operating condition adaptability of protection settings.

[0026] The actual operation of the power grid includes various routine operation scenarios and complex fault scenarios. The frequency of occurrence and operating characteristics of different scenarios vary, and the coordination status of protection levels will fluctuate continuously with the switching of scenarios. The risk of level coordination presents a multi-scenario coupled distribution characteristic. At the same time, the target power grid covers multiple power supply areas, which are interconnected through the power grid topology. The level coordination of protection devices is also related to the topological connections between areas.

[0027] Therefore, by collecting electrical operating parameters and fault characteristic parameters of the entire target power grid under different operating conditions, and obtaining the parameter changes and fault characteristics under different operating scenarios, the differential coordination logic of protection devices in different areas is determined by combining the topological correlation between each area. By analyzing the parameter changes and fault characteristics under different operating scenarios, and the corresponding differential coordination logic of the protection devices, the differential coordination status of protection devices in each area is determined, thereby obtaining the differential change pattern under different operating conditions.

[0028] Based on the obtained differential change pattern, a differential risk expectation function is constructed to integrate the differential fluctuation impact caused by different operating conditions, fault types and topological coordination relationships. The comprehensive fluctuation level of differential coordination in each region is uniformly represented by the function form.

[0029] Specifically, the electrical operating parameters of the target power grid under all operating conditions include electrical quantity data of the target power grid under various anticipated operating conditions such as normal operation, maximum operating mode, minimum operating mode, islanded operation, high distributed power source penetration, near-area faults, and weak feeder lines, which can be obtained through the historical operating data of the target power grid.

[0030] Specifically, the electrical operating parameters include the amplitude and phase of the bus voltage in each region, the transmission line current, the transformer operating parameters, the load power in each region, the system frequency, the line impedance parameters, and the dynamic change data of parameters under various operating conditions.

[0031] The fault characteristic parameters of the target power grid under the entire operating conditions are the electrical change laws and protection action offset characteristics that reflect the inherent characteristics of the fault and the response characteristics of the protection action under various fault scenarios and extreme disturbance conditions of the target power grid. These include the differentiated characteristic information of various typical faults such as high resistance faults and remote faults, as well as fault scenarios involving topology switching and sudden changes in operating mode. Specifically, these parameters can be obtained by integrating historical fault records of the target power grid, simulation data, and measured statistical data under extreme operating conditions.

[0032] The fault characteristic parameters specifically include fault type, fault location, fault current attenuation characteristics, and action delay offset data of main and backup protection devices under various extreme operating conditions.

[0033] The topological association and coordination relationship refers to the overall association logic of regional division, hierarchical subordination, importance matching and cross-regional linkage formed by various power equipment and relay protection devices in the target power grid according to the corresponding physical grid architecture and relay protection configuration rules. It can be specifically determined according to the grid structure layout, equipment zoning planning and protection configuration rules of the target power grid.

[0034] The topological association and coordination relationship includes at least the segment affiliation relationship of the power grid physical topology, the hierarchical subordinate coordination relationship between the main protection and backup protection, the importance association relationship of the protection devices in the region, and the hierarchical linkage constraint relationship of cross-regional protection.

[0035] Specifically, based on the actual grid layout of the target power grid, the power operation sections, such as transmission lines, busbars, distribution equipment, and transformers, are divided into regions and their boundaries are defined. Based on the divided regions and corresponding boundaries, the control areas to which each power device and its corresponding protection device belong are determined, thereby obtaining the segment affiliation relationships of the power grid's physical topology.

[0036] Based on the hierarchical correspondence rules in the relay protection configuration rules, the hierarchical order of protection on the upstream power supply side and the downstream load side is defined, thereby determining the pairing relationship between the main protection device, near backup protection device, and far backup protection device corresponding to each line and equipment, and thus obtaining the hierarchical subordinate relationship between the main protection and backup protection. The relay protection configuration rules can be set according to actual needs.

[0037] Based on the power supply attributes, load capacity, and pre-defined importance of power grid hubs in each region of the target power grid, the importance levels of different protection devices within each region are defined. The coordination constraints and risk transmission relationships between protection devices of different importance are determined, thereby obtaining the importance correlation of protection devices within the region. The rules for classifying the importance levels of protection devices can be set according to the actual operational needs of the target power grid and equipment priority planning.

[0038] Based on the target power grid's inter-regional interconnection architecture, boundary tie equipment configuration, and fault propagation characteristics across regions, the coordination criteria and differential linkage constraints for inter-regional protection devices are defined. The differential coordination logic and action linkage rules between boundary protection and tie line protection in each region are determined, thereby obtaining the differential linkage constraint relationship for inter-regional protection. The coordination rules for inter-regional protection can be set according to the target power grid's topology and fault prevention requirements.

[0039] Based on this, such as Figure 2 As shown, the process involves collecting electrical operating parameters and fault characteristic parameters of the entire target power grid, and constructing a differential risk expectation function for each region of the target power grid based on topological correlation and coordination relationships. This includes: Based on the electrical operating parameters and fault characteristic parameters of the entire operating condition, as well as the action delay offset data of each protection device under extreme conditions, calculate the extreme value of the differential loss of each main and backup protection pair under various extreme conditions. Based on the actual setting level difference, level difference loss extreme value, minimum level difference limit value and ideal minimum level difference of each primary and backup protection pair, establish the corresponding over-level risk function and delay redundancy risk function. Based on the topological association and coordination relationship, determine the coordination weight of each primary and backup protection pair to its respective area and within the corresponding area; By combining the corresponding coordination weights, the risk functions of the cascading risk and the time-delay redundancy risk functions of each main and backup protection pair are weighted and aggregated according to region to construct the expected risk function of the differential risk for each region of the target power grid.

[0040] Current protection settings are typically based solely on short-circuit current amplitudes under typical operating conditions to set operating thresholds and time limits, without considering potential deviations in the protection device's own operating delay under extreme conditions. These deviations include increased delays in digital protection filtering algorithms due to high harmonic distortion, extended start-up confirmation windows under weak feeder conditions, measurement lag caused by CT saturation, and prolonged tripping times due to circuit breaker aging. These deviations caused by extreme conditions directly compress the effective time difference between the main and backup protection, making settings configured according to conventional operating conditions highly susceptible to cascading tripping in extreme scenarios.

[0041] Therefore, by pre-collecting electrical operating parameters and fault characteristic parameters under all operating conditions, including various extreme conditions, and combining the action delay offset calibration data of various types of protection devices under given extreme conditions and the circuit breaker tripping delay data, the total differential loss of each main and backup protection pair under all possible extreme conditions is calculated, and the maximum value is taken as the corresponding differential loss extreme value. This transforms the uncertainty disturbance of differential coordination caused by extreme conditions into the compensation margin corresponding to each main and backup protection pair, i.e., the differential loss extreme value. Then, based on this fixed extreme value, the subsequent differential risk expectation function is established. Thus, during the optimization process, more differential margin is automatically reserved for protection pairs severely affected by extreme conditions, while protection pairs with minor impact do not need to be over-compensated.

[0042] The step of calculating the extreme values ​​of the differential loss of each main and backup protection pair under various extreme conditions, based on the electrical operating parameters and fault characteristic parameters of the entire operating range, as well as the action delay offset data of each protection device under extreme conditions, includes: Based on the electrical operating parameters and fault characteristic parameters of the entire operating range, determine the typical characteristic parameter combinations of various extreme operating conditions, and determine the action delay offset calibration value of each protection device under the corresponding extreme operating conditions based on the typical characteristic parameter combinations. The corresponding candidate extreme operating condition category is determined based on the installation location, protection type, and status of the associated circuit breaker for each main and backup protection pair. Based on the corresponding action delay offset calibration value, the total differential loss of each main and backup protection pair under each corresponding candidate extreme condition is calculated by superimposing the values. The maximum total differential loss of each main and backup protection pair under all corresponding candidate extreme conditions is taken as the corresponding extreme value of differential loss.

[0043] First, cluster analysis is performed on the collected electrical operating parameters and fault characteristic parameters of the entire operating range to determine the characteristic parameter combinations corresponding to standard operating conditions and each type of extreme operating condition, forming an operating condition feature library. Based on the formed operating condition feature library, the action delay offset values ​​of each type of protection device under each type of extreme operating condition are pre-calibrated through offline testing or digital simulation, forming a protection device feature database. The formed protection device feature database stores a fixed increment between the action time of each protection device under standard operating conditions and the action time under each extreme operating condition. The standard operating condition is the target power grid being in a state of no distortion, sufficient short-circuit current, unsaturated current transformers (CTs), and normal operation of circuit breakers.

[0044] Then, for each main and backup protection pair, the candidate extreme operating conditions it may encounter are determined based on its installation location, protection type, and the status of the associated circuit breakers. Specifically, installation location data includes whether the protection device is located at the beginning or end of the line, whether it is close to the distributed power supply access point, and the short-circuit capacity level of the line. Protection type data includes equipment protection type data such as overcurrent protection, distance protection, or differential protection. Associated circuit breaker status data includes the circuit breaker's service life, cumulative number of operations, and the difference between the measured tripping time and the rated value.

[0045] Then, based on the preset extreme condition mapping rules, the candidate extreme conditions that each main and backup protection pair may encounter are determined by feature matching. The extreme condition mapping rules can be set according to actual needs. For example, if the installation location is at the end of the line and there is a high proportion of distributed power supply access, the candidate extreme conditions are determined to include high harmonic distortion conditions. If the installation location is at the end of the line and the system short-circuit capacity of the feeder is less than the preset threshold, the candidate extreme conditions are determined to include weak feeder conditions, etc.

[0046] The candidate extreme operating conditions identified by each primary and backup protection pair are merged and deduplicated to obtain the final list of candidate extreme operating condition categories. Then, for each candidate extreme operating condition, the operating delay offset calibration values ​​of the primary and backup protection under that condition are retrieved from the pre-established protection device characteristic database.

[0047] The effective step difference between the main and backup protection depends on the duration of the fault detected by the backup protection from the occurrence of the fault to its actual clearing. After the main protection operates, the circuit breaker needs time to disconnect the current. If the main protection operation itself is delayed due to extreme operating conditions, and the circuit breaker's disconnection is also delayed due to aging or other reasons, then the duration of the fault current will be the sum of the main protection operation delay and the circuit breaker's tripping delay—that is, the time from the main protection issuing the command to the actual current disconnection. The backup protection starts timing from the very beginning of the fault, so the later the fault is cleared, the easier it is for the backup protection to reach its operating time limit, meaning the effective step difference is compressed more.

[0048] The obtained action delay offset calibration value is actually only the additional delay between the occurrence of the fault and the issuance of the trip command by the main protection. Based on the fact that the circuit breaker delay also causes an effective level difference deviation, the tripping delay offset is further obtained according to the current aging degree of the circuit breaker. The main protection delay offset is added to the circuit breaker tripping delay offset to obtain the total level difference loss under the corresponding operating condition. The associated circuit breaker used in judging the candidate extreme operating condition type and the circuit breaker used in calculating the circuit breaker tripping delay are both circuit breakers controlled by the main protection, that is, the switching equipment that actually performs the fault current disconnection operation after the main protection issues the trip command.

[0049] After calculating the total differential loss for all candidate operating conditions, the maximum value among them is taken as the extreme value of differential loss for the corresponding main and backup protection pair. The obtained extreme value of differential loss represents the additional compensation margin that needs to be reserved in the setting differential to resist extreme operating condition disturbances.

[0050] The extreme value of the differential loss for each primary and backup protection pair represents the maximum compression of the effective differential under extreme operating conditions, and is a fixed compensation value. However, the extreme value of the differential loss itself does not change with the setting differential, but the setting differential is a variable that needs to be continuously adjusted in the actual optimization process. In order to dynamically evaluate the differential coordination risk corresponding to each set of candidate setting differentials in the subsequent optimization calculation, and to automatically balance the selectivity under extreme operating conditions and the speed under normal operating conditions, a risk function with the actual setting differential as the independent variable is established for each protection pair based on the obtained extreme value of differential loss. The established risk function specifically includes a step-over risk function and a delay redundancy risk function. The step-over risk function is used to quantify the safety cost of step-over tripping that may be caused by insufficient effective remaining differential under extreme operating conditions, while the delay redundancy risk function is used to quantify the speed loss caused by excessive setting differential under normal operating conditions.

[0051] Specifically, the step of establishing corresponding over-level risk functions and delay redundancy risk functions based on the actual setting level difference, level difference loss extreme value, minimum level difference limit, and ideal minimum level difference for each primary and backup protection pair includes: Based on the actual setting level difference, level difference loss extreme value and minimum level difference limit of each main and backup protection pair, determine the corresponding effective residual level difference under extreme working conditions; A mapping relationship is constructed based on the deviation characteristics of the effective remaining level difference relative to the minimum level difference limit of each primary and backup protection pair, forming a risk function for exceeding the level. Based on the difference range between the actual setting level difference and the ideal minimum level difference for each primary and backup protection pair, the corresponding time-limited redundancy is determined; A mapping relationship is constructed based on the changing characteristics of the time-limited redundancy of each primary and backup protection pair, forming a delay redundancy risk function.

[0052] First, based on the actual setting level difference and level loss extreme value of each main and backup protection pair, determine the corresponding effective residual level difference under extreme operating conditions. That is, subtract the level loss extreme value from the actual setting level difference, and the resulting difference is the actual residual level difference that can be used to ensure selectivity under extreme operating conditions.

[0053] If the calculated effective residual grade difference is greater than or equal to the minimum grade difference limit, it is considered that the grade difference matching is still safe even if the disturbance of extreme working conditions is taken into account. If the effective residual grade difference is less than the minimum grade difference limit, there is a risk of exceeding the grade limit, and the greater the deficiency, the higher the risk.

[0054] Based on the deviation characteristics of the effective residual difference relative to the minimum difference limit, i.e., the mapping relationship between the difference between the minimum difference limit and the effective residual difference and the change in risk value, a risk function for exceeding the risk level is formed, the expression of which is: ; in, This is a risk function that bypasses the standard level. This refers to the actual tuning level, which is the decision variable to be optimized in the subsequent optimization process. This is the minimum differential limit. For effective residual difference, and , This represents the extreme value of the differential loss. This is the preset risk gain coefficient for exceeding the limit.

[0055] The constructed risk function takes the value of zero when the effective residual difference reaches or exceeds the minimum difference limit. Once the effective residual difference is lower than the minimum difference limit, the risk value increases linearly with the shortfall, and its growth slope is determined by the preset risk gain coefficient.

[0056] Based on the difference between the actual setting level difference and the ideal minimum level difference for each primary and backup protection pair, the corresponding time-limit redundancy is determined. This is achieved by subtracting the ideal minimum level difference from the actual setting level difference, and the resulting difference represents the excess of the actual setting level difference compared to the ideal optimal level difference. The ideal minimum level difference is a target value that balances speed and slight fit margin under normal operating conditions, and is typically less than the minimum level difference limit.

[0057] If the actual setting level difference is not greater than the ideal minimum level difference, it is considered that there is no delay redundancy and the delay redundancy risk is zero. If the actual setting level difference is greater than the ideal minimum level difference, there is a positive time limit redundancy. In this case, the level difference delay redundancy is too large, which will significantly increase the fault clearing time and weaken the protection speed. Moreover, the larger the redundancy, the more serious the loss of speed.

[0058] Based on the changing characteristics of this time-limited redundancy, i.e., the time-limited redundancy increases monotonically from zero, and its mapping relationship with the change in risk value, a time-delayed redundancy risk function is formed, the expression of which is: ; in, For the delay redundancy risk function, For the actual setting level difference, This is a preset delay redundancy risk gain coefficient, which is usually smaller than the over-level risk gain coefficient. This is a time-limited redundancy, and , This represents the ideal minimum difference.

[0059] The constructed delay redundancy risk function takes a value of zero when the actual set difference does not exceed the ideal minimum difference. When the actual set difference exceeds the ideal minimum difference, the risk value increases linearly with the excess, reflecting the loss of speed caused by excessive difference under normal operating conditions.

[0060] Since the target power grid consists of multiple power supply areas, the grid coupling strength, electrical disturbance characteristics, and fault impact range vary between different areas. Furthermore, the load levels served by different primary and backup protection pairs within the same area, as well as the power outage losses due to faults, also differ significantly. Therefore, to distinguish the decoupling relationships between areas and the importance hierarchy of protection pairs within an area when constructing the differential risk expectation function based on the cascading risk function and the time-delay redundancy risk function, the region to which each primary and backup protection pair belongs is further determined based on the topological correlation and coordination relationship. Corresponding coordination weights are then assigned based on the hierarchical subordinate coordination relationship and importance correlation between the primary and backup protection within the region. This ensures that the final output differential risk expectation function can accurately reflect the objective impact of power grid topology and operational importance on differential coordination risk.

[0061] Specifically, the regional affiliation of each protection pair is initially determined using the segment affiliation relationship, which records the primary equipment controlled by each protection device and its corresponding power supply area. For each primary and backup protection pair, the regional identifiers corresponding to the primary and backup protections are queried separately. If they are the same, the region is temporarily recorded as the region to which the protection pair belongs; if they are different, the primary and backup protection pair is marked as cross-region pending. Then, the step direction of each primary and backup protection pair is confirmed using the hierarchical coordination relationship.

[0062] For primary and backup protection pairs marked as undetermined across regions, their final region of affiliation is determined using cross-regional differential linkage constraints. These constraints include the coordination criteria and differential linkage limitations for boundary protection devices, as well as the affiliation rules for protection pairs on both sides of the tie line. For determining the affiliation of primary and backup protection pairs across regions, a single dominant region rule is preferentially adopted. This involves determining a dominant region based on the fault current direction or the location on the power supply side, thus consolidating cross-regional protection pairs into a single region and avoiding cross-regional coupling during risk aggregation.

[0063] After determining the regions to which all protection pairs belong, for each protection pair, the importance level of its primary and backup protection is determined based on the importance correlation. These importance levels are then converted into coefficients between 0 and 1 according to a preset mapping rule. The coefficients of all primary and backup protection pairs within the corresponding region are normalized so that the sum of the coordination weights of all protection pairs within that region is 1, thereby determining the coordination weights of the corresponding primary and backup protection pairs in each region. The mapping rule can be to take the higher level of the primary and backup protection and then perform coefficient conversion using empirical formulas or the analytic hierarchy process (AHP).

[0064] After obtaining the cascading risk function and delay redundancy risk function for each primary and backup protection pair, and determining the region to which each protection pair belongs and its coordination weight within the region based on topological association and coordination relationships, the risk functions of all protection pairs within the same region are further weighted and aggregated to form the expected differential risk function for the corresponding region, the expression of which is: ; in, Let be the expected value function of the differential risk in one of the regions. This represents the set of actual setting differences for all primary and backup protection systems within the region. , This refers to the number of primary and backup protection pairs in this area. For the first The coordination weight of primary and backup protection in this area For the first The over-level risk function of each primary and backup protection pair For the first Delay redundancy risk function for each primary and backup protection pair For the first The actual setting level difference between the main and backup protection pairs.

[0065] The expected value of the grade difference risk function reflects the overall grade difference coordination risk level of all primary and backup protection pairs within the region under the set of grade differences. The cascading risk component reflects the potential safety hazard of cascading tripping under extreme operating conditions, while the delay redundancy risk component reflects the loss of speed due to excessive grade differences under normal operating conditions. Since the coordination weights are determined based on the importance correlation between primary and backup protection pairs and cross-regional linkage constraints, primary and backup protection pairs with larger weights occupy a stronger dominant position in the regional risk function.

[0066] After determining the differential risk expectation function for each region, the differential risk expectation functions for all regions are summed to obtain the differential risk expectation function for the entire target power grid.

[0067] Then, the action time of each level of protection device is set as the decision variable, the total action time of all protection devices in the target power grid is minimized as the main optimization objective, and the constructed differential risk expectation function is used as the corresponding penalty term to construct the final optimization setting objective function. The total action time is usually defined as the sum of the time from the occurrence of the fault to the issuance of the trip command by the main protection of the line when a metallic short circuit occurs at a preset typical fault point.

[0068] The final expression for the optimized tuning objective function is: ; in, To optimize the tuning objective function, For the set of decision variables, , The number of protection devices within the target power grid. For the first The set of actual setting differences for all primary and backup protection pairs within a given region. The number of areas within the target power grid. For the first The setpoint for the operating time of each protection device is obtained by subtracting the operating time of the main protection device from the operating time of the backup protection device in the corresponding main and backup protection pair, based on the actual setting difference in the expected function of the differential risk.

[0069] However, relying solely on the risk penalty term in the objective function to guide optimization may still result in solutions that fail to meet basic safety requirements or exceed the device's performance limits. For example, under extreme operating conditions, even with constraints imposed on the risk penalty term, situations may still arise where the effective step difference is slightly greater than the minimum step difference limit, but the absolute operating time exceeds the maximum allowable time limit of the protection device, or the sensitivity is lower than the equipment requirements, or the current setting exceeds the circuit breaker's breaking capacity. Therefore, it is necessary to further set corresponding upper and lower limits for each region based on the power grid topology information, electrical operating parameters, and fault characteristic parameters to ensure the reliability of the final output protection setting.

[0070] Specifically, setting upper and lower limits for constraints in each region based on the topology information, electrical operating parameters, and fault characteristic parameters of the target power grid includes: Based on the topology information of the target power grid, the physical wiring structure, node distribution status and line power supply coverage of each region are obtained, and the grid coupling strength and line-level coupling degree of each region are calculated according to the corresponding physical wiring structure and node distribution status. The corresponding electrical disturbance fluctuation range is obtained based on the electrical operating parameters of each region, and the fault impact range is determined based on the fault characteristic parameters. Based on the corresponding primary and backup protection for the extreme values ​​of the corresponding level loss, the regional loss reference value of each region is selected. Combined with the corresponding network coupling strength, line level coupling degree and electrical disturbance fluctuation range, the minimum limit of the protection level difference of each region is determined. Based on the corresponding fault impact range and the power supply coverage of the line, the maximum limit of the protection level difference for each area is determined; Based on the minimum and maximum limits of the corresponding protection level, the level constraint boundaries for each region are formed.

[0071] First, extract the physical wiring structure, node distribution status, and line power supply coverage of each area based on the topology information of the target power grid. The physical wiring structure includes the radial or ring connection of the lines in the area and the configuration relationship between transformers and busbars. The node distribution status includes the electrical distance between each protection installation point and the power supply path. The line power supply coverage includes the load area powered by each line.

[0072] Then, based on the corresponding physical wiring structure type, the network coupling strength between different lines within the area is assigned a value. For a pure radial network, a smaller coupling strength value is used; for a network with a single tie line, a medium value is used; and for a multi-circuit parallel or ring network structure, a larger value is used. The specific assignment rules can be set according to actual needs. The larger the value of the network coupling strength, the tighter the electrical connection between lines, the greater the impact of a fault on the protection measurement of multiple lines, and the higher the risk of differential coordination failure.

[0073] Simultaneously, starting from the regional power supply side, the search proceeds downwards along the power supply path. Based on the node distribution status, the number of protection device levels traversed on the longest path is counted. This maximum number of levels is taken as the line-level coupling degree. The larger the line-level coupling degree value, the more line levels there are. The more times the main and backup coordination is required during a fault, the more the cumulative error of each coordination level and the dispersion of circuit breaker operation time will be amplified step by step, resulting in a more severe compression of the effective level difference of the end protection.

[0074] The electrical disturbance fluctuation range is obtained based on the electrical operating parameters of each region, including voltage fluctuation range, extreme values ​​of load change rate, and extreme values ​​of frequency offset. These electrical disturbance fluctuations affect the measurement accuracy and action delay of protection devices. The larger the fluctuation, the greater the step margin is needed to maintain the reliability of the coordination. At the same time, the fault impact range is determined based on fault characteristic parameters, that is, the number of protection devices that may be affected when a fault occurs at a certain point in the region. The larger this number, the tighter the upper limit of the fault clearing time should be to prevent the fault from spreading.

[0075] Then, based on the extreme values ​​of the differential losses of all primary and backup protection pairs in each region, the maximum value is selected as the corresponding regional loss reference. The basic minimum differential loss limit is added to the regional loss reference, and then corrected according to the preset correction rules, combined with the network coupling strength and the line level coupling degree. Specifically, the preset correction rules can be set according to actual needs. For example, for every 0.1 increase in network coupling strength, the minimum limit increases by 0.02 seconds; for every 1 increase in line level, the minimum limit increases by 0.03 seconds. The corrected minimum differential loss limit is then corrected again according to the electrical disturbance fluctuation range. The specific rules for this second correction can also be set according to actual needs. For example, for every 0.1 increase in voltage fluctuation width, the minimum differential loss limit increases by 0.01 seconds. For areas with large load change rates or frequency offsets, the increase in the minimum differential loss limit can be further increased.

[0076] After linearly superimposing the regional loss reference value, the grid coupling strength, the line level coupling degree, and the electrical disturbance fluctuation range, the minimum limit of the protection level difference for the corresponding region can be obtained. This limit ensures that the effective level difference of all main and backup protection pairs in the region is not lower than the safety requirements under extreme operating conditions.

[0077] Starting from the basic maximum allowable difference in protection levels, the limits are then reduced based on the fault impact range and the line power supply coverage area. The rules for this reduction can be set according to actual needs; for example, for every increase of 1 in the fault impact range coefficient, the maximum limit is reduced by 0.05 seconds; for every increase of 0.1 in the normalized value of the line power supply coverage area, the maximum limit is reduced by 0.2 seconds, and so on. After reduction and correction, the maximum limit for the protection level difference in this area is obtained to prevent excessively long operating times from compromising system stability or the power supply to critical loads.

[0078] Finally, the calculated minimum and maximum limits are used to form the grade constraint boundary of the region, meaning that the setting grade difference of all protection pairs in the region must be within the interval formed by the minimum and maximum limits.

[0079] The constructed upper and lower limits of constraints are actually static fixed boundaries based on the expected operating conditions, which can only ensure that the protection settings do not exceed the limits under the pre-enumerated full-domain operating conditions. However, in actual power grid operation, there are many dynamic factors that change in real time, such as random fluctuations in the output of distributed power sources, switching of large impact loads, temporary switching of operating modes, slow increase in tripping time caused by circuit breaker aging, and the influence of ambient temperature on the operating accuracy of protection devices.

[0080] These factors are difficult to fully enumerate and solidify into the static constraint boundary during the offline phase. Therefore, a grade margin correction coefficient determined by such dynamic factors is further introduced to dynamically fine-tune the already determined static grade boundary. Without re-optimizing the entire domain, the settings are periodically updated according to short-term changes in the power grid operation, so that the final output protection settings can not only meet the rigid requirements of the extreme operating conditions expected offline, but also flexibly adapt to changes in the online real-time operating environment.

[0081] Specifically, the setting of corresponding differential margin correction coefficients in conjunction with operational influencing factors includes: Based on fault characteristic parameters, the fault fluctuation distribution characteristics of each region are extracted, and corresponding adjustment conditions are extracted based on load change data and power grid operation mode switching data of each region. Based on the characteristics of fault fluctuation distribution and the extracted adjustment conditions, the corresponding operational influencing factors for each region are determined; Based on the corresponding grid coupling strength, electrical disturbance fluctuation range and operation influencing factors, the power grid operation stability conditions of each region are matched, and the adjustable range of the level difference parameter is determined according to the regional loss reference amount and the limit range of the protection level difference in each region. Based on operational influencing factors and grid operation stability conditions, the adjustment requirements of protection level differences in each region are matched by regional loss reference values. Under the constraint of the adjustable range of the grade difference parameter, the grade difference margin correction coefficient of each region is set according to the corresponding adjustment requirements.

[0082] For each region, the frequency of occurrence of various faults, the distribution of fault duration, and the range of fault current fluctuation amplitude are statistically analyzed. Combined with the fault disturbance intensity under extreme operating conditions, the fault fluctuation probability distribution curve of each region is fitted to determine the concentration range, peak value, and dispersion of fault fluctuations in each region, forming the fault fluctuation distribution characteristics of each region. The fault fluctuation distribution characteristics can reflect the degree of influence of fault disturbances on the protection differential coordination within the region.

[0083] Load change data and grid operation mode switching data from various regions of the target power grid are collected synchronously. After noise reduction processing of both types of data, corresponding adjustment conditions are extracted. The load change data includes real-time changes, frequency, and amplitude of active and reactive power. The grid operation mode switching data includes topology interconnection switch switching records, system operation mode change timing, and inter-regional power exchange adjustment data. The extracted adjustment conditions specifically include load surge / descent thresholds, load fluctuation duration thresholds, topology switching timing nodes, and electrical parameter offset ranges.

[0084] The correlation coefficients between each parameter in the fault fluctuation distribution characteristics and adjustment conditions and the grade difference margin fluctuation are calculated. Parameters whose correlation coefficients with the grade difference margin fluctuation exceed the preset correlation threshold are selected as operational influencing factors. Finally, a set of operational influencing factors for each region is formed, and the influence weights of various factors on the grade difference margin are allocated by the analytic hierarchy process.

[0085] Based on the grid coupling strength, electrical disturbance fluctuation range and operational influencing factors of each region, the grid operation status under different operating conditions is simulated, the inter-regional linkage stability threshold is calculated, the electrical quantity stability boundary is defined, and the grid operation stability range of each region is defined through multiple rounds of simulation iteration, constraining the grid operation status boundary during the differential adjustment process.

[0086] Meanwhile, based on the limit range of the protection level difference, the grid coupling strength, electrical disturbance fluctuation range and various operational influencing factors are substituted into the range, and the grid stability simulation verification is carried out for each level difference value in the range, eliminating the value range in the static constraint range that cannot adapt to the stable operation conditions of the grid.

[0087] For the remaining grade difference values, a correlation verification is performed with the regional loss reference values ​​of each region. Grade difference values ​​that would cause regional loss to exceed the limit are screened out, and the final retained continuous sub-intervals are used as the adjustable range of the grade difference parameters for the corresponding regions.

[0088] By comparing the intensity of disturbances caused by operational influencing factors, the boundary conditions of grid operation stability, and the threshold of regional loss reference values, the adjustment direction and adjustment range of protection level differences in each region are determined. For example, when the values ​​of fault fluctuations and load fluctuations exceed the corresponding thresholds, the disturbance effect is strong, and the grid stability simulation leaves room for adjustment, the level difference margin gap is calculated, and the adjustment range is matched to increase the level difference margin. When the values ​​of fault and load fluctuations are stable, the disturbance effect is weak, and the regional loss statistics exceed the preset threshold, the level difference redundancy is calculated, and the adjustment range is matched to reduce the level difference margin.

[0089] Determine the baseline tuning value for the differential margin correction coefficient. This value corresponds to a differential margin that remains at the baseline configuration without any amplification or reduction. When the adjustment requirement is to amplify the differential margin, the baseline tuning value is increased according to the matching adjustment range under the coefficient threshold constraint corresponding to the upper limit of the adjustable range of the corresponding differential parameter. If the adjustment requirement is to reduce the differential margin, the baseline tuning value is reduced according to the matching adjustment range under the coefficient threshold constraint corresponding to the lower limit of the adjustable range of the corresponding differential parameter, thus determining the final differential margin coefficient.

[0090] After defining the minimum and maximum limits of protection level differences for each region and determining the level difference margin correction coefficient for each region, these two pieces of information are integrated to form a unified rigid constraint on the setting level difference for each primary and backup protection pair. Specifically, for any region, its level difference margin correction coefficient is superimposed on the original minimum limit for that region to obtain the corrected lower limit. Then, for each primary and backup protection pair within that region, its actual setting level difference must simultaneously satisfy the restriction of being no less than the corrected lower limit and no greater than the original upper limit. For cross-regional boundary protection pairs, if they have been designated to belong to a dominant region, the constraints of that region are applied. If they are simultaneously constrained by multiple regions, the maximum value of the corrected lower limits of each region is taken as the lower limit, and the minimum value of the upper limits of each region is taken as the upper limit. This rule determines the constraint range for each region after integrating the limit range and the level difference margin correction coefficient. The constraint ranges of all regions are summarized to obtain the final global constraint conditions.

[0091] Under the constraints of the constructed global domain, the optimization objective function is further optimized and solved to finally converge to obtain the unique set of protection settings that minimizes the sum of the total action time and the differential risk penalty.

[0092] Specifically, the optimization calculation of the objective function under global constraints includes: Using the setting parameters of protection at all levels as the decision variables to be optimized, the decision variables to be optimized are initialized to obtain the initial variable set; Substitute the initial set of variables into the optimization tuning objective function to obtain the initial objective function value; Based on the initial objective function value, and combined with the preset optimization algorithm, the decision variables to be optimized are updated and combined iteratively in a dimension-by-dimensional manner to obtain several combinations of variables; Iterate through the global constraints in turn, and select the feasible solution set that is within the range of the global constraints from the obtained variable combinations; Calculate the objective function value corresponding to each combination of variables in the feasible solution set, and compare them to obtain the current optimal variable set; Based on the objective function value corresponding to the current optimal variable set, the next round of updates and iterations is carried out in combination with the preset optimization algorithm until the iteration termination condition is met, and the optimal protection setting parameter combination is obtained.

[0093] In this embodiment, the protection setting parameters at each level are specifically the operating time settings of the protection devices at each level. The preset optimization algorithm is an evolutionary algorithm, such as differential evolution or particle swarm optimization, and its optimization calculation process is as follows: Figure 3 As shown.

[0094] First, based on the physical limitations and operational experience of the protection devices, an initial search range is set for each decision variable. For example, the lower limit of the action time setpoint is 0.05 seconds and the upper limit is 1.5 seconds. Within this range, several initial solutions are randomly generated to form an initial population. Each solution set contains the action time setpoints of all protection devices.

[0095] Substitute each set of initial solutions into the optimization tuning objective function, calculate the corresponding initial objective function value, and record the solution with the smallest objective function in the initial population as the initial optimal solution.

[0096] The iterative optimization process then begins. In each iteration, based on the distribution of the objective function value of the current population, a new set of candidate solutions is generated according to a preset evolutionary strategy, such as mutation and crossover operations in differential evolution. For each candidate solution generated, it is substituted into the global constraints for verification. First, the actual tuning difference for each primary and backup protection pair is calculated from the action times in the candidate solution. Then, it is checked whether this difference falls between the corrected lower and upper limits of the corresponding region. If all protection pairs satisfy the constraints, the candidate solution is retained as a feasible solution; otherwise, it is discarded. After constraint screening of all generated candidate solutions, the set of feasible solutions for the current generation is obtained.

[0097] Calculate the objective function value for each solution in the feasible solution set, and select the set of solutions with the smallest objective function value as the optimal solution for the current generation. At the same time, in order to maintain the diversity of the population, randomly select a portion of the remaining solutions, or use a clustering method based on Euclidean distance to screen out those solutions that are significantly different from other existing solutions. The screened solutions and the optimal solution of the current generation together constitute the next generation of the population.

[0098] After each update of the optimal solution, it is determined whether the iteration termination condition is met, such as the rate of change of the optimal objective function value for several consecutive generations being less than a preset threshold for the rate of change of the function value, or the number of iterations reaching a preset maximum value.

[0099] If the conditions are met, the iteration terminates and the current optimal solution is output as the optimal protection tuning parameter combination. If the conditions are not met, the next round of mutation, crossover, constraint screening, and target evaluation continues.

[0100] To avoid getting stuck in local optima and improve the convergence speed, before performing the next round of updates based on the objective function value corresponding to the current optimal variable set and the preset optimization algorithm, the following is also executed: Based on the objective function values ​​of the remaining variable groups in the current iterative feasible solution, excluding the current optimal variable group, obtain the variable combination distribution characteristics of the current iteration; Based on the distribution characteristics of the variable combinations in the current iteration, the search range of decision variables for the next round of update iteration is set.

[0101] The distribution characteristics of the variable combinations include the mean, variance, extreme values, and value concentration intervals of the objective function values ​​of the remaining variable groups in different dimensions. Based on these distribution characteristics, the degree of clustering and the direction of dispersion of the current population in the decision variable space can be inferred. If feasible solutions are concentrated in a relatively small sub-region, it indicates that the local development stage has been entered, and the search range can be appropriately narrowed to accelerate convergence. If feasible solutions are dispersed and the objective function values ​​are not significantly different, it indicates that the global exploration stage is still underway, and the search range should be maintained or expanded to avoid missing potential optimal regions. The specific adjustment range for narrowing and expanding can be set according to actual needs.

[0102] The final output of the optimal protection setting combination minimizes the sum of the total operating time and the expected risk level, ensuring that the actual setting level difference of all main and backup protection pairs is not lower than the modified lower limit, that is, the safety margin is sufficient under extreme conditions. The actual setting level difference of all main and backup protection pairs is not higher than the upper limit, that is, the operating time limit is not exceeded under normal conditions, thus achieving optimal performance under all operating conditions.

[0103] It should be noted that the decision variables to be optimized in this embodiment only include the operating time settings of each protection device. Other setting parameters, such as the current setting and curve type of the protection devices, are predetermined and fixed using preset rules independent of the optimization process. The differential coordination refers to the difference in operating time between the main protection and the backup protection. Under extreme conditions, the main disturbances to the differential coordination are directly reflected in changes on the time axis. The current setting mainly undertakes fault identification and sensitivity assurance functions, and can be used through traditional methods, such as setting it according to the reliability coefficient of the maximum short-circuit current, to reserve sufficient margin and cover current measurement deviations under extreme conditions. Therefore, decoupling the current setting from the time setting and setting them independently in advance ensures the selectivity and speed of protection while reducing the complexity of the optimization problem and improving solution efficiency.

[0104] Finally, the optimal action time combination obtained through optimization is combined with the predetermined current setting and other parameters to form a complete protection setting combination, which is then sent to each protection device.

[0105] Another aspect of this embodiment provides a protection setting optimization tuning system based on graded risk constraints, including: The differential risk analysis module is used to collect electrical operating parameters and fault characteristic parameters of the entire target power grid, and construct the differential risk expectation function of each region of the target power grid by combining the topological correlation and coordination relationship. The optimization setting planning module is used to construct the corresponding optimization setting objective function with the goal of minimizing the total protection action time of the target power grid and the expected function of differential risk as the penalty term. At the same time, it sets the upper and lower limits of constraints for each region and the corresponding differential margin correction coefficient to construct global constraints. The optimization setting calculation module is used to perform optimization calculations on the objective function under global constraints to obtain the optimal protection setting combination for the target power grid. The setting module is used to set the protection settings of each protection device in each area of ​​the target power grid in sequence based on the optimal protection setting combination.

[0106] The differential risk analysis module is connected to the optimization tuning planning module, the optimization tuning calculation module is connected to the optimization tuning planning module, and the tuning module is connected to the optimization tuning calculation module.

[0107] The differential risk analysis module, optimization tuning planning module, optimization tuning calculation module, and tuning module are all data processing devices such as microprocessors and computers with corresponding data processing capabilities, and have corresponding communication ports, which can receive the required data and transmit the calculation results.

[0108] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.

Claims

1. A method for optimizing protection settings based on differential risk constraints, characterized in that, include: Electrical operating parameters and fault characteristic parameters of the entire target power grid are collected, and the differential risk expectation function of each region of the target power grid is constructed by combining the topological correlation and coordination relationship. The optimization objective is to minimize the total protection operating time of the target power grid, and the expected function of the differential risk is used as the penalty term to construct the corresponding optimization setting objective function. Based on the topology information, electrical operating parameters and fault characteristic parameters of the target power grid, set the upper and lower limits of constraints for each region, and set the corresponding differential margin correction coefficients in combination with the operating influencing factors. Construct global constraint conditions based on the upper and lower limits of constraints and differential margin correction coefficients for each region. Under the constraints of the entire domain, the optimization objective function is optimized to obtain the optimal protection setting combination of the target power grid; Based on the optimal protection setting combination, the protection settings of each protection device in each area of ​​the target power grid are set sequentially.

2. The protection setting optimization method based on differential risk constraints according to claim 1, characterized in that, The topological association and coordination relationships include at least the segment affiliation relationship of the power grid physical topology, the hierarchical subordinate coordination relationship between the main protection and backup protection, the importance association relationship of protection devices within the region, and the hierarchical linkage constraint relationship of cross-regional protection.

3. The protection setting optimization method based on differential risk constraints according to claim 2, characterized in that, The process involves collecting electrical operating parameters and fault characteristic parameters across the entire target power grid, and constructing a differential risk expectation function for each region of the target power grid based on topological correlations and coordination relationships. This includes: Based on the electrical operating parameters and fault characteristic parameters of the entire operating range, as well as the action delay offset data of each protection device under extreme operating conditions, calculate the extreme value of the differential loss of each main and backup protection pair under various extreme operating conditions. Based on the actual setting level difference, level difference loss extreme value, minimum level difference limit value and ideal minimum level difference of each primary and backup protection pair, establish the corresponding over-level risk function and delay redundancy risk function. Based on the topological association and coordination relationship, determine the coordination weight of each primary and backup protection pair to its respective area and within the corresponding area; By combining the corresponding coordination weights, the risk functions of the cascading risk and the time-delay redundancy risk functions of each main and backup protection pair are weighted and aggregated according to region to construct the expected risk function of the differential risk for each region of the target power grid.

4. The protection setting optimization method based on differential risk constraints according to claim 3, characterized in that, The process of establishing corresponding over-level risk functions and delay redundancy risk functions based on the actual setting level difference, level difference loss extreme value, minimum level difference limit, and ideal minimum level difference for each primary and backup protection pair includes: Based on the actual setting level difference, level difference loss extreme value and minimum level difference limit of each main and backup protection pair, determine the corresponding effective residual level difference under extreme working conditions; A mapping relationship is constructed based on the deviation characteristics of the effective remaining level difference relative to the minimum level difference limit of each primary and backup protection pair, forming a risk function for exceeding the level. Based on the difference range between the actual setting level difference and the ideal minimum level difference for each primary and backup protection pair, the corresponding time-limited redundancy is determined; A mapping relationship is constructed based on the changing characteristics of the time-limited redundancy of each primary and backup protection pair, forming a delay redundancy risk function.

5. The protection setting optimization method based on differential risk constraints according to claim 3, characterized in that, The calculation of the extreme values ​​of the differential loss for each main and backup protection pair under various extreme conditions is based on the electrical operating parameters and fault characteristic parameters of the entire operating range, as well as the action delay offset data of each protection device under extreme conditions. Based on the electrical operating parameters and fault characteristic parameters of the entire operating range, determine the typical characteristic parameter combinations of various extreme operating conditions, and determine the action delay offset calibration value of each protection device under the corresponding extreme operating conditions based on the typical characteristic parameter combinations. The corresponding candidate extreme operating condition category is determined based on the installation location, protection type, and status of the associated circuit breaker for each main and backup protection pair. Based on the corresponding action delay offset calibration value, the total differential loss of each main and backup protection pair under each corresponding candidate extreme condition is calculated by superimposing the values. The maximum total differential loss of each main and backup protection pair under all corresponding candidate extreme conditions is taken as the corresponding extreme value of differential loss.

6. The protection setting optimization method based on differential risk constraints according to claim 1, characterized in that, The setting of upper and lower limits for constraints in each region based on the topology information, electrical operating parameters, and fault characteristic parameters of the target power grid includes: Based on the topology information of the target power grid, the physical wiring structure, node distribution status and line power supply coverage of each region are obtained, and the grid coupling strength and line-level coupling degree of each region are calculated according to the corresponding physical wiring structure and node distribution status. The corresponding electrical disturbance fluctuation range is obtained based on the electrical operating parameters of each region, and the fault impact range is determined based on the fault characteristic parameters. Based on the corresponding primary and backup protection for the extreme values ​​of the corresponding level loss, the regional loss reference value of each region is selected. Combined with the corresponding network coupling strength, line level coupling degree and electrical disturbance fluctuation range, the minimum limit of the protection level difference of each region is determined. Based on the corresponding fault impact range and the power supply coverage of the line, the maximum limit of the protection level difference for each area is determined; Based on the minimum and maximum limits of the corresponding protection level, the level constraint boundaries for each region are formed.

7. The protection setting optimization method based on differential risk constraints according to claim 6, characterized in that, The method of setting corresponding differential margin correction coefficients based on operational influencing factors includes: Based on fault characteristic parameters, the fault fluctuation distribution characteristics of each region are extracted, and corresponding adjustment conditions are extracted based on load change data and power grid operation mode switching data of each region. Based on the characteristics of fault fluctuation distribution and the extracted adjustment conditions, the corresponding operational influencing factors for each region are determined; Based on the corresponding grid coupling strength, electrical disturbance fluctuation range and operation influencing factors, the power grid operation stability conditions of each region are matched, and the adjustable range of the level difference parameter is determined according to the regional loss reference amount and the limit range of the protection level difference in each region. Based on operational influencing factors and grid operation stability conditions, the adjustment requirements of protection level differences in each region are matched by regional loss reference values. Under the constraint of the adjustable range of the grade difference parameter, the grade difference margin correction coefficient of each region is set according to the corresponding adjustment requirements.

8. The protection setting optimization method based on differential risk constraints according to claim 1, characterized in that, The optimization calculation of the objective function under global constraints includes: Using the setting parameters of protection at all levels as the decision variables to be optimized, the decision variables to be optimized are initialized to obtain the initial variable set; Substitute the initial set of variables into the optimization tuning objective function to obtain the initial objective function value; Based on the initial objective function value, and combined with the preset optimization algorithm, the decision variables to be optimized are updated and combined iteratively in a dimension-by-dimensional manner to obtain several combinations of variables; Iterate through the global constraints in turn, and select the feasible solution set that is within the range of the global constraints from the obtained variable combinations; Calculate the objective function value corresponding to each combination of variables in the feasible solution set, and compare them to obtain the current optimal variable set; Based on the objective function value corresponding to the current optimal variable set, the next round of updates and iterations is carried out in combination with the preset optimization algorithm until the iteration termination condition is met, and the optimal protection setting parameter combination is obtained.

9. The protection setting optimization method based on differential risk constraints according to claim 8, characterized in that, Before performing the next round of updates and iterations based on the objective function value corresponding to the current optimal variable set and in conjunction with the preset optimization algorithm, the following steps are also executed: Based on the objective function values ​​of the remaining variable groups in the current iterative feasible solution, excluding the current optimal variable group, obtain the variable combination distribution characteristics of the current iteration; Based on the distribution characteristics of the variable combinations in the current iteration, the search range of decision variables for the next round of update iteration is set.

10. A protection setting optimization tuning system based on differential risk constraints, used to execute the optimization tuning method according to any one of claims 1 to 9, characterized in that, include: The differential risk analysis module is used to collect electrical operating parameters and fault characteristic parameters of the entire target power grid, and construct the differential risk expectation function of each region of the target power grid by combining the topological correlation and coordination relationship. The optimization setting planning module is used to construct the corresponding optimization setting objective function with the goal of minimizing the total protection action time of the target power grid and the expected function of differential risk as the penalty term. At the same time, it sets the upper and lower limits of constraints for each region and the corresponding differential margin correction coefficient to construct global constraints. The optimization setting calculation module is used to perform optimization calculations on the objective function under global constraints to obtain the optimal protection setting combination for the target power grid. The setting module is used to set the protection settings of each protection device in each area of ​​the target power grid in sequence based on the optimal protection setting combination.

Images