Stability control system and control method of power system

Abstract

The invention provides a stability control system and a control method of an electric power system, the stability control system comprises a control quantity calculator, a calculation scheduler, a sample manager and a rolling manager, the control quantity calculator is configured to perform transient stability calculation on the electric power system according to predicted operation data of the electric power system in a preset time period and a preset fault set; generating a strategy sample according to a calculation result; the calculation scheduler is configured to predict calculation resource demands of different operation cycles of the power system so as to schedule the control quantity calculator according to the calculation resource demands; the sample manager is configured to collect all strategy samples generated by the control quantity calculator and generate a sample set; the rolling manager is configured to generate a control strategy required for controlling the operation of the power system in a preset time period according to the sample set. By sensing the state of the power grid in real time, a dynamic strategy updating mechanism is constructed to cope with the operation uncertainty of the power system, and safe, stable and efficient operation of the power grid is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of safe and stable control of power systems, and in particular to a stabilization control system and a control method for power systems. Background Art

[0002] The safety, stability, and control system (hereafter referred to as the "stability control system") uses the grid's operating mode and anticipated faults as input and provides stability control measures as output. This serves as a second line of defense for power system safety and stability and falls under the category of emergency control. The strategy table is a core element of the stability control system, and developing and updating the stability control strategy table is a key technology in its construction.

[0003] In related technologies, offline analysis methods are often used to formulate stabilization strategies. This involves summarizing the stabilization measures required for typical operating conditions and anticipated faults to form a strategy table. Operations and maintenance personnel then perform on-site upgrades of the stabilization strategies annually after the stabilization system is shut down. This offline analysis method is essentially based on a static matching mechanism between typical operating conditions (such as the "high in winter, low in summer" load curve) and pre-set fault scenarios. However, with the large-scale integration of high-proportion renewable energy and power electronic equipment, power system operating conditions exhibit multidimensional randomness, strong spatiotemporal coupling, and rapid dynamic evolution. Consequently, the stabilization strategies specified by this offline analysis method face three difficulties: First, the selection of typical scenarios fails to cover complex operating conditions such as fluctuating renewable energy output and flexible energy storage charging and discharging, resulting in an increased strategy mismatch rate; second, the empirically driven classification of characteristic indicators (such as inertia threshold and cross-sectional power) lacks dynamic correlation analysis, making it impossible to accurately capture the critical points of transient instability; third, the conservative margins calculated offline, combined with deviations from real-time operating conditions, can easily lead to over- or under-cutting, threatening economic efficiency and safety. That is, in related technologies, the security and stability of the power system's stabilization and control strategies are poor, making it difficult to meet the power system's operating requirements. Summary of the Invention

[0004] The embodiments of the present application provide a stabilization control system and a control method for an electric power system to solve the problem of poor security and stability of the control strategy of the electric power system in the related art.

[0005] To solve the above problems, the technical solutions provided by this application are as follows:

[0006] In the first aspect, the present application provides a stabilization and control system for an electric power system, comprising: a control quantity calculator, configured to perform transient stability calculations on the electric power system based on predicted operating data and a preset fault set of the electric power system in a preset time period, and generate a strategy sample based on the calculation results; a calculation scheduler, connected to the control quantity calculator, configured to predict the computing resource requirements of different operating cycles of the electric power system, so as to schedule the control quantity calculator based on the computing resource requirements; a sample manager, connected to the control quantity calculator, configured to collect all strategy samples generated by the control quantity calculator, and generate a sample set; a rolling manager, connected to the sample manager, configured to generate, based on the sample set, the control strategy required for controlling the operation of the electric power system in the preset time period.

[0007] In one embodiment, the control quantity calculator includes: a simulation module, configured to perform electromechanical transient time domain simulation according to the preset fault set to generate transient response data, wherein the transient response data at least includes power angle, voltage and frequency data of the power system; a trigger module, connected to the simulation module, configured to generate an instability signal when determining that the power system is unstable according to the transient response data; and a search module, connected to the trigger module, configured to search for control measure data for restoring stability of the power system in response to the instability signal to generate the strategy sample.

[0008] In one embodiment, the computing scheduler includes: a first computing cluster, configured to schedule the control quantity calculator to predict the operating state of the power system once every first time, and each prediction generates a first type of strategy sample corresponding to multiple moments in the future first time period that are evenly spaced according to the first time; a second computing cluster, configured to schedule the control quantity calculator to predict the operating state of the power system in the future second time period, the prediction period is equal to the length of the second time period, and each prediction generates a second type of strategy sample corresponding to multiple moments in the future second time period that are evenly spaced according to the second time; a third computing cluster, configured to schedule the control quantity calculator to predict the operating state of the power system in the future third time period, the prediction period is equal to the length of the third time period, and each prediction generates a third type of strategy sample corresponding to multiple moments in the future third time period that are evenly spaced according to the third time; wherein the length of the second time period is greater than the length of the first time period, the length of the third time period is greater than the length of the second time period, the second time is equal to the first time, and the length of the third time is greater than the length of the second time.

[0009] In one embodiment, the computing scheduler is further configured to: when the third computing cluster is in an idle state, call the control quantity calculator that performs calculations in the third computing cluster to the second computing cluster; and when the second computing cluster is in an idle state, call the control quantity calculator that performs calculations in the second computing cluster to the first computing cluster.

[0010] In one embodiment, the sample manager includes: a removal module, configured to remove the policy samples in the sample set that are generated earlier than the current moment; a replacement module, connected to the removal module, configured to replace the policy samples in the sample set that overlap in time with the existing policy samples in the sample set with the policy samples generated at the current moment; and a supplement module, connected to the replacement module, configured to supplement the policy samples generated at the current moment to the sample set when the policy samples generated at the current moment are different in time from the existing policy samples in the sample set.

[0011] In one embodiment, the rolling manager includes: a first strategy generation module, configured to generate a first strategy based on the sample set when the calculation of the first strategy sample is completed or the calculation time reaches a preset time length, so as to update the first strategy every preset time length.

[0012] In one embodiment, the rolling manager further includes: a verification module configured to perform a verification simulation on the first type of strategy currently applied to the power system, and update the first type of strategy when the verification simulation result indicates that the power system is unstable.

[0013] In one embodiment, the rolling manager also includes: a second strategy generation module, configured to generate a second type of strategy based on the sample set when the calculation of the second type of strategy samples is completed; wherein the second type of strategy is configured to replace the first type of strategy when the first type of strategy fails.

[0014] In one embodiment, the rolling manager also includes: a third strategy generation module, configured to generate a third type of strategy based on the sample set when the calculation of the third type of strategy samples is completed; the third type of strategy is configured to be executed when both the first type of strategy and the second type of strategy fail.

[0015] In one embodiment, the strategy sample includes at least one of the following: power system operating status data, fault data, and control measure data required to resolve a preset fault at a target time in a preset operating period of the power system.

[0016] In the second aspect, the present application provides a control method for an electric power system, comprising: performing transient stability calculation on the electric power system based on the predicted operating data of the electric power system in a preset time period and a preset fault set, and generating strategy samples based on the calculation results; predicting the strategy samples required to deal with preset faults in different operating cycles of the electric power system; collecting all strategy samples and generating a sample set; and generating the control strategy required to control the operation of the electric power system in the preset time period based on the sample set.

[0017] In one embodiment, the transient stability calculation of the power system is performed based on the predicted operating data of the power system in a preset time period and a preset fault set, and the strategy sample is generated based on the calculation result, including: performing electromechanical transient time domain simulation based on the preset fault set to generate transient response data, the transient response data including at least power angle, voltage and frequency data of the power system; when it is determined that the power system is unstable based on the transient response data, searching for control measure data for restoring stability of the power system to generate the strategy sample.

[0018] In one embodiment, the strategy samples required for predicting the preset faults in different operation cycles of the power system include: predicting the operation status of the power system once every first time, and generating a first type of strategy samples corresponding to multiple moments in the first time period in the future at equal intervals each time; predicting the operation status of the power system in the second time period in the future, with a prediction period equal to the length of the second time period, and generating a second type of strategy samples corresponding to multiple moments in the second time period in the future at equal intervals each time; predicting the operation status of the power system in the third time period in the future, with a prediction period equal to the length of the third time period, and generating a third type of strategy samples corresponding to multiple moments in the third time period in the future at equal intervals each time; wherein the length of the second time period is greater than the length of the first time period, the length of the third time period is greater than the length of the second time period, the second time is equal to the first time, and the length of the third time is greater than the length of the second time.

[0019] In one embodiment, the collecting of all policy samples and generating a sample set includes: eliminating the policy samples in the sample set that are generated earlier than the current moment; when the policy samples generated at the current moment overlap in time with the existing policy samples in the sample set, replacing the policy samples in the sample set that overlap in time with the policy samples generated at the current moment; when the policy samples generated at the current moment differ in time from the existing policy samples in the sample set, adding the policy samples generated at the current moment to the sample set.

[0020] In one embodiment, the control strategy required for controlling the operation of the power system during a preset time period is generated based on the sample set, including: when the calculation of the first type of strategy samples is completed or the calculation time reaches a preset time length, the first type of strategy is generated based on the sample set to achieve updating of the first type of strategy every preset time length.

[0021] In one embodiment, generating the control strategy required for controlling the operation of the power system in a preset time period based on the sample set includes: performing a verification simulation on the first type of strategy currently applied to the power system, and updating the first type of strategy when the verification simulation result shows that the power system is unstable.

[0022] In one embodiment, generating a control strategy required for controlling the operation of the power system during a preset time period based on the sample set includes: generating a second type of strategy based on the sample set when the calculation of the second type of strategy samples is completed; wherein the second type of strategy is configured to replace the first type of strategy when the first type of strategy fails.

[0023] In one embodiment, generating a control strategy required for controlling the operation of the power system during a preset time period based on the sample set includes: generating a third type of strategy based on the sample set when the calculation of the third type of strategy samples is completed; and configuring the third type of strategy to be executed when both the first type of strategy and the second type of strategy fail.

[0024] The embodiment of the present application provides a stabilization control system and control method for an electric power system, which is used to control the electric power system. The stabilization control system includes a control quantity calculator, which is configured to perform transient stability calculations on the electric power system based on the operation prediction data of the electric power system in a preset time period and a preset fault set, and generate, based on the calculation results, a strategy sample required to solve the preset fault at a specified time in the preset time period of the electric power system; a calculation scheduler, which is configured to schedule the control quantity calculator based on the computing resources required for predicting the operation status of different electric power systems; a sample manager, which is configured to manage all the strategy samples generated by the control quantity calculator and generate a sample set; and a rolling manager, which is configured to generate, based on the sample set, the control instructions required for controlling the operation of the electric power system in the preset time period. The present application constructs a dynamic strategy update mechanism to cope with the uncertainty of the operation of the electric power system by sensing the state of the power grid in real time, which is conducive to ensuring the safe, stable and efficient operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0026] Attachment Figure 1 This is a schematic diagram of the module structure of the stabilization control system in the embodiment of the present application;

[0027] Attachment Figure 2 This is a schematic diagram of the module structure of the control quantity calculator in an embodiment of the present application;

[0028] Attachment Figure 3 This is a rolling timing diagram of the stabilization control system in an embodiment of the present application.

[0029] 1. Stable control system;

[0030] 100, control quantity calculator; 110, simulation module; 120, trigger module; 130, search module;

[0031] 200, computing scheduler; 210, first computing cluster; 220, second computing cluster; 230, third computing cluster;

[0032] 300, sample manager; 310, rejection module; 320, replacement module; 330, replenishment module;

[0033] 400, scroll manager; 410, first strategy generation module; 420, second strategy generation module; 430, third strategy generation module; 440, verification module. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0035] It will be understood that although the terms "first", "second", etc. may be used herein to describe various components, these components should not be limited by these terms. These components are only used to distinguish one component from another.

[0036] It will also be understood that the terms “comprises” and / or “comprising” as used herein specify the presence of features or components, but do not preclude the presence or addition of one or more other features or components.

[0037] Reference Figure 1 As shown, the present application provides a stabilization control system 1 for an electric power system, which includes a control variable calculator 100 , a calculation scheduler 200 , a sample manager 300 and a rolling manager 400 .

[0038] The control variable calculator 100 is configured to perform transient stability calculations on the power system based on the predicted operating data of the power system in a preset period and a preset fault set, and generate strategy samples based on the calculation results.

[0039] It should be noted that this policy sample refers to a system-level stability control strategy for a specific fault, based on a specific grid operating mode. The control strategy ultimately generated by the rolling manager is a device-level stability control strategy. Specifically, the components of a device-level stability control strategy can be composed of two parts: a stability limit component and a stability control strategy component.

[0040] Among them, the components of the stability limit include operating mode, control object and stability limit value. Specifically, the components of the stability limit operating mode include: the start-up and shutdown status of the primary equipment of the power system, the operating status and control mode of the DC system, the active power or current value of the primary equipment of the power system or its combination (section), the busbar operating mode, mainly the combined busbar operating mode and the denominator operating mode, the number or capacity of the started units (including the number of conventional units started and the capacity of new energy units started), the rotating standby unit, the busbar voltage, the start-up and shutdown status of the stabilization device or the stabilization system, the start-up and shutdown status of the stabilization device pressure plate and the equipment maintenance pressure plate, the number and capacity of the units that can be switched, the load that can be switched and the load level of the power grid. The control object refers to the primary equipment that needs to be power controlled to meet the safety and stability requirements, specifically including the equipment composition, flow direction and weight coefficient. The stability limit value includes the upper or lower operating limits of the power, current, voltage, number of started units, rotating standby units, etc. of the primary equipment.

[0041] The stabilization strategy includes elements such as operating mode, fault components and types, cross-section power flow, and control measures. Specifically, the stabilization strategy operating mode is used to describe the corresponding operating mode constraints in the stabilization strategy. Its components include: the start and stop status of the power system primary equipment (lines, main transformers, units, DC, switches, etc.) or their combinations; the active power of the power system primary equipment; the start and stop status of the operation mode pressure plate or component maintenance pressure plate of the power grid stabilization device; and the power station busbar operating mode, including the combined busbar operation mode and the denominator operation mode.

[0042] Sectional power flow is used to describe the power of important transmission lines, transformers, or tie-line transformer combinations in the power system corresponding to the stabilization strategy. Its components include: section composition (the set of transmission lines and transformers and their combinations that constitute the transmission section), power direction (the active power direction of the primary equipment related to the section), and power gear (section power gear setting).

[0043] The control measures describe the control objects, control quantities, and control principles (including control object selection methods and control quantity allocation principles) included in the stabilization control strategy. Control objects are categorized into the following types: disconnecting units (including conventional generators, pumped storage units, and renewable energy units); disconnecting loads; disconnecting main transformers; DC power modulation (including DC power ramp-down, ramp-up, and DC blocking); disconnecting or closing lines and busbars; disconnecting low-voltage capacitors and reactors; adjusting the gear position of controllable high-voltage reactors; and adjusting the output of renewable energy units.

[0044] The control quantity supports specifying according to the adjustment quantity or the reserved quantity. The specific methods include: specifying the generator control measures according to the active power size or the number of disconnected units; specifying the pumped storage units according to the active power size or the number of disconnected units based on the pumping or power generation status; specifying the load control measures according to the load active power size or proportion; specifying the main transformer control measures according to whether to cut off; specifying the DC control measures according to power modulation, active power gear adjustment or locking; specifying the line control measures according to whether to de-energize; specifying the low-voltage capacitors and reactors according to whether to be put into use or cut off; specifying the controllable high-voltage reactor gear according to absolute or relative value adjustment; and specifying the new energy units according to the power adjustment amount.

[0045] The control object selection method includes: whether the control object can be cut or retained; screening according to the priority of the control object; selecting according to the optimality of the control object; and matching according to the controllable quantity of the control object.

[0046] The control quantity allocation principles are divided into: the maximum undercutting principle according to the control quantity size; the minimum overcutting principle according to the control quantity size; the closest principle according to the control quantity size; and other specific allocation principles.

[0047] Based on the above, in some embodiments of the present application, the strategy sample includes at least one of the following: power system operating status data, fault data, and control measure data required to solve the preset fault at the target time in the preset operating period of the power system.

[0048] The calculation scheduler 200 is connected to the control quantity calculator 100, and the calculation scheduler 200 is configured to predict the computing resource requirements of different operation cycles of the power system, so as to schedule the control quantity calculator 100 according to the computing resource requirements;

[0049] The sample manager 300 is also connected to the control amount calculator 100 , and the sample manager 300 is configured to collect all policy samples generated by the control amount calculator 100 and generate a sample set;

[0050] The rolling manager 400 is connected to the sample manager 300 , and the rolling manager 400 is configured to generate a control strategy required for controlling the operation of the power system in a preset period according to the sample set.

[0051] The stabilization control system 1 of the present application utilizes the aforementioned control variable calculator 100, computation scheduler 200, sample manager 300, and rolling manager 400 to establish a dynamic strategy update mechanism to address the multi-dimensional randomness, strong spatiotemporal coupling, and rapid dynamic evolution characteristics of power grid operation. This makes the stabilization control system 1 of the present application more adaptable and more precise than power system stabilization control systems 1 in related technologies, thus facilitating stable operation of the power system.

[0052] Reference Figure 2 As shown, in some embodiments of the present application, the control amount calculator 100 includes a simulation module 110 , a trigger module 120 and a search module 130 .

[0053] The simulation module 110 is configured to perform electromechanical transient time-domain simulation based on the preset fault set and generate transient response data. The transient response data includes at least power angle, voltage, and frequency data of the power system. It should be noted that under normal steady-state operation, the electromagnetic torque output by each generator set in the power system is balanced with the mechanical torque input by the prime mover, thus maintaining a constant rotor speed for all generators. However, the power system is inevitably subject to major disturbances, such as the switching on and off of large-capacity generators or large loads due to various short-circuit faults. After experiencing a major disturbance, the power system undergoes not only electromagnetic transients but also electromechanical transients. During electromechanical transients, due to significant changes in the system structure, the system power flow and the output electromagnetic power of each generator also change, disrupting the power balance between the prime mover and the generator, generating unbalanced torque on the generator shaft, and causing the generator rotor to accelerate or decelerate. Typically, after a major disturbance, the power imbalance between each generator set varies. Furthermore, the moment of inertia of each generator rotor also varies, resulting in different speed changes for each unit. This generates relative motion between the generator rotors, causing the relative power angle between the rotors to change. This change in the relative power angle between the rotors, in turn, affects the output power of each generator, causing the power, speed, and relative power angle between the rotors to continue to change. Simultaneously, changes in the generator terminal voltage and stator current trigger excitation regulation, changes in unit speed trigger adjustments in the speed control system, changes in bus voltage in the power system cause changes in composite power, and changes in power network currents trigger adjustments in other control devices. All of these changes directly or indirectly affect the power balance on the generator shaft. These changes interact, forming an electromechanical transient process dominated by the mechanical motion of the generator rotor and changes in electromagnetic power.

[0054] In some more specific examples, simulation module 110 calculates electromechanical transients in a power system using a time-domain simulation method. Based on this, control variable calculator 100 uses an electromechanical transient time-domain simulation program to calculate the control type and control variable that can achieve stable control based on the power system's future operating mode and anticipated faults. This control type and control variable can be considered a sample point on a stable control strategy table curve (or multidimensional surface). The time-domain simulation method utilizes numerical methods for solving initial value problems to solve the differential equations that describe the dynamic characteristics of the power system and determines system stability based on the rotor power angle differences between generator sets. This method is more intuitive and accurate, and is more conducive to handling the impact of various complex models and complex faults.

[0055] Based on the above, the trigger module 120 is connected to the simulation module 110 and is configured to determine the stability of the power angle, voltage, and frequency of the power system operation using a preset fault set. The trigger module 120 is configured to generate an instability signal when the power system is determined to be unstable based on the transient response data. The search module 130 is connected to the trigger module 120 and is configured to search for control measure data that can restore the stability of the power system in response to the instability signal to generate a strategy sample. It should be noted that power system instability refers to the situation where the power system is unstable due to a disturbance or the operating parameters seriously exceed the specified range.

[0056] Reference Figure 1 As shown, in some embodiments of the present application, the computing scheduler 200 includes a first computing cluster 210, a second computing cluster 220, and a third computing cluster 230. It should be noted that the first computing cluster 210, the second computing cluster 220, and the third computing cluster 230 are hardware resources managed by the computing scheduler 200, such as the computing power of a central processing unit (CPU) or a graphics processing unit (GPU), which are responsible for providing computing power support for tasks of different cycles, while the control quantity calculator 100 is a software functional module running on each cluster. The control quantity calculator 100 performs transient stability calculations and control quantity generation in the corresponding cluster according to the resource allocation instructions of the computing scheduler 200.

[0057] More specifically, in some embodiments of the present application, the first computing cluster 210 is configured such that the dispatch control quantity calculator 100 predicts the power system operating state once every first time period, and each prediction generates first-class strategy samples corresponding to multiple time periods evenly spaced according to the first time period within a first future time period. In some specific examples, the first computing cluster 210 is an ultra-short-period computing cluster that predicts the power system operating state once every 15 minutes, and each prediction generates first-class strategy samples corresponding to 16 time periods evenly spaced according to 15 minutes within the next four hours. It should be noted that the first computing cluster 210 uses a rolling update method, that is, at time T, 16 first-class strategy samples corresponding to the period from time T to T+4 hours are predicted, and then at time T+15 minutes, 16 first-class strategy samples corresponding to the period from time T+15 minutes to T+4 hours+15 minutes are predicted. The first-class strategy samples obtained twice are repeated, thereby enabling continuous prediction and calculation of the power system operating state within a short period of less than four hours, ensuring the timeliness of the calculations and enabling timely response to uncertainties in power system operation.

[0058] The second computing cluster 220 is configured as a dispatch control quantity calculator 100 to predict the operating state of the power system in the second time period in the future. The prediction period is equal to the second time period. Each prediction generates a second type of strategy samples corresponding to multiple moments in the second time period in the future, which are equally spaced at the second time. In some specific examples, the second computing cluster 220 is a daily planning computing cluster with a prediction period of 24 hours. That is, it predicts the power grid operation mode for the next day and generates a calculation file that can be used by the electromechanical transient timing simulation calculation program. At intervals of 15 minutes, a total of 96 second type strategy samples corresponding to the moments are generated. To ensure easy control operation, these 96 moments can be composed of the 0th minute, the 15th minute, the 30th minute, and the 45th minute of each hour. The specific generation time can be to generate strategy samples corresponding to 96 moments on the N+1 day at a certain moment on the Nth day (such as 22:00).

[0059] The third computing cluster 230 is configured as the dispatch control quantity calculator 100 to predict the power system operating state within a third future time period. The prediction period is equal to the third future time period, and each prediction generates a plurality of third-type strategy samples corresponding to a plurality of time points equally spaced at third time intervals within the third future time period. In some specific examples, the third computing cluster 230 is a long-term computing cluster that predicts the power system operating state within a week to a month. The prediction period can be one week, ten days, or one month. Each prediction generates a plurality of third-type strategy samples discretely spaced at four-hour intervals within the prediction period.

[0060] On this basis, within the same time scale, the computing power required by the second type of computing cluster will be significantly higher than that of the third type of computing cluster, and the update frequency of the first type of computing cluster is also significantly higher than that of the second type of computing cluster 220 and the third type of computing cluster. After the second type of computing cluster and the third type of computing cluster complete the calculation, the corresponding called control quantity calculator 100 will be in an idle state. Therefore, the computing scheduler 200 of the present application is also configured to call the control quantity calculator 100 performing calculations in the third computing cluster 230 to the second computing cluster 220 when the third computing cluster 230 is in an idle state; and to call the control quantity calculator 100 performing calculations in the second computing cluster 220 to the first computing cluster 210 when the second computing cluster 220 is in an idle state. Through this setting, the computing resources in the stabilization control system 1 can be fully utilized to ensure that each computing cluster has sufficient computing power to ensure the accuracy of the prediction calculation.

[0061] Reference Figure 3 As shown, in some embodiments of the present application, the data generation of the first type of strategy is to generate data at intervals of 15 minutes for the next 4 hours at a time that is divisible by 15 minutes (i.e. Figure 3The ultra-short-term method shown above) has a total of 16 moments. Assuming that the number of the first type of strategies generated at moment n is m n , then the total number of first-class strategies N short for:

[0062]

[0063] The data generation of the second type of strategy is to generate the future plan of Day 2 at a certain moment before the end of Day 1, such as 22:00 on Day 1. The second type of strategy corresponds to 96 moments on Day 1 that are divisible by 15 minutes. Each moment generates a result based on the degree of multidimensional uncertainty or an indefinite number of predictions. Assume that the number of second type strategies generated at moment n is m n , then the total number of the second type of strategies N day for:

[0064]

[0065] The data generation of the third type of strategy is different from the time interval scale of the previous two cycles. It uses a 4-hour interval to predict the weekly, ten-day, and monthly power grid operation mode and form a long-term mode file. The number of sample points for the third type of strategy is:

[0066]

[0067] Where D is the number of days for the third type of strategy. Because the interval period of the third type of strategy is inconsistent with that of the second and first types of strategies, the third type of strategy cannot be used as a backup for the above two types of strategies, but can only be used as a supplement to the sample set.

[0068] On this basis, refer to Figure 1 As shown, in some embodiments of the present application, the sample manager 300 includes a rejection module 310 , a replacement module 320 and a supplement module 330 .

[0069] Specifically, the elimination module 310 is configured to eliminate strategy samples whose centralized generation time is earlier than the current time. It should be noted that during the continuous operation of the power system, the state also changes in real time. Therefore, historical samples will not be able to reflect the current transient stability requirements. If they continue to be retained, the strategy samples will become invalid. Therefore, the samples whose strategy sample time is earlier than the current time will be automatically eliminated.

[0070] The replacement module 320 is connected to the elimination module 310. When the policy sample generated at the current moment overlaps with an existing policy sample in the sample set, the policy sample generated at the current moment replaces the policy sample in the sample set with the policy sample. Specifically, for policy samples that overlap in multiple periods, the newly generated policy sample replaces the old policy sample. It should be noted that the first computing cluster 210, the second computing cluster 220, and the third computing cluster 230 have different calculation accuracies due to their different update intervals or update durations. The first computing cluster 210 has the highest prediction accuracy. When policy samples from multiple periods exist at the same moment, new samples from higher-precision sources are prioritized. In this case, the policy samples are updated via the replacement module 320.

[0071] The supplementation module 330 is connected to the replacement module 320 and is configured to supplement the strategy samples generated at the current moment into the sample set when the time of the strategy samples generated at the current moment is different from the time of the strategy samples already in the sample set, that is, to supplement the newly formed strategy samples that do not appear in the sample set into the sample set. In some embodiments of the present application, the third type of strategy samples are generated at 4-hour intervals, which differ from the time points of the first and second type of strategy samples generated at 15-minute intervals. Therefore, supplementing the samples at the missing moments by the supplementation module 330 is conducive to constructing a strategy surface with global time coverage, thereby improving the accuracy and stability of power system control.

[0072] Reference Figure 1 As shown, in some embodiments of the present application, the scroll manager 400 includes a first strategy generating module 410 , a first strategy generating module 410 , a first strategy generating module 410 , and a checking module 440 .

[0073] Specifically, the first strategy generation module 410 is configured to generate a first strategy based on the sample set when the calculation of the first strategy sample is completed or the calculation time reaches a preset time, so as to update the first strategy every preset time. It should be noted that the premise for whether the first strategy is to be rolled out is to perform a strategy verification simulation on the first strategy currently running in the power system. If the power system is found to be unstable after verification, the first strategy needs to be rolled out. Therefore, a verification module 440 is provided, which is configured to perform a verification simulation on the first strategy currently applied in the power system and update the first strategy in the event of power system instability.

[0074] In some more specific examples, the strategy generation method of the first strategy generation module 410 is rolling generation, which automatically rolls every 15 minutes. It should be noted that rolling is a dynamic iterative update mechanism, the core of which is to achieve periodic refresh and incremental optimization of the strategy through a sliding time window. Specifically in this embodiment, the prediction window is advanced by one step every 15 minutes, and the strategy of the new window can inherit the valid data of the previous window to avoid repeated calculations. This setting can prioritize the real-time control, but since the prediction is a short-term data volume, there may be certain coverage blind spots. In addition, the rolling mechanism of the first strategy generation module 410 allows the time consumption to reach an upper limit, which can avoid the stagnation of strategy sample updates due to delays in individual sample calculations.

[0075] On this basis, the second strategy generation module 420 is configured to generate a second strategy based on the sample set after the second strategy sample calculation is complete. Compared to the first strategy generation module 410, the second strategy generation module 420 provides a longer-term periodic strategy generation, thus serving as a backup solution for the first strategy generation module 410, which helps improve the reliability of global control. The first strategy generation module 410 and the second strategy generation module 420 are nested and complementary, facilitating the formation of a high-precision control system with high-frequency rolling optimization and low-frequency global calibration.

[0076] It should be noted that the above-mentioned first strategy generation module 410 and second strategy generation module 420 are both modules that run in real time online, but there is only one set of strategies running in real time online at the same time. During multi-period rolling decision-making, if the generation of the first type of strategy fails and the original online strategy is not suitable for the current control, the second type of strategy will be switched to the online strategy. When the second type of strategy also fails to meet the requirements, another set of strategies will be needed as a backup. Therefore, a third strategy generation module 430 is set up, which is configured to generate a third type of strategy as a backup strategy based on the sample set when the calculation of the third type of strategy samples is completed. In some more specific examples, the third type of strategy can also be a backup strategy with a large margin generated by manual offline calculation.

[0077] According to a second aspect of the present application, a control method for a power system is further provided. The control method is implemented by the stabilization control system of the aforementioned embodiment. The control method can achieve all the beneficial effects of the aforementioned stabilization control system. The present application will not elaborate on this. Specifically, the control method includes:

[0078] S10: Calculate the transient stability of the power system based on the predicted operation data of the power system in a preset period and a preset fault set, and generate a strategy sample based on the calculation result.

[0079] In this step S10, specifically, electromechanical transient time domain simulation is performed according to the preset fault set to generate transient response data, and the transient response data at least includes power angle, voltage and frequency data of the power system; when it is determined that the power system is unstable according to the transient response data, control measure data for restoring stability of the power system is searched to generate the strategy sample.

[0080] S20: Predict the strategy samples required to deal with preset faults in different operation cycles of the power system.

[0081] In step S20, specifically, the power system operating state is predicted once every first time period, and each prediction generates first type strategy samples corresponding to multiple moments in the future first time period that are equally spaced according to the first time period.

[0082] The operating state of the power system in the second time period in the future is predicted, and the prediction period is equal to the second time period. Each prediction generates second type strategy samples corresponding to multiple moments in the second time period in the future that are equally spaced according to the second time.

[0083] The operating state of the power system in the third future time period is predicted, and the prediction period is equal to the third time period. Each prediction generates third type strategy samples corresponding to multiple moments in the third future time period that are equally spaced according to the third time.

[0084] The length of the second period is greater than that of the first period, the length of the third period is greater than that of the second period, the second time is equal to the first time, and the length of the third time is greater than that of the second time.

[0085] S30: Collect all strategy samples and generate a sample set.

[0086] In step S30, the strategy samples need to be processed. Specifically, the strategy samples generated earlier than the current time in the sample set need to be removed.

[0087] It is also necessary to replace the policy sample in the sample set with the policy sample generated at the current moment when the policy sample in the sample set overlaps in time with the policy sample in the sample set;

[0088] And when the time of the policy sample generated at the current moment is different from that of the existing policy samples in the sample set, the policy sample generated at the current moment is added to the sample set.

[0089] S40: generating a control strategy required for controlling the operation of the power system during a preset period according to the sample set.

[0090] In step S40, different control strategies need to be set according to different control requirements, specifically including:

[0091] When the calculation of the first type of strategy samples is completed or the calculation time reaches the preset time, the first type of strategy is generated based on the sample set to update the first type of strategy every preset time. It should be noted that the premise of whether the first type of strategy is rolled out is to perform a verification simulation on the first type of strategy currently applied to the power system, and to update the first type of strategy when the verification simulation result shows that the power system is unstable. In some embodiments, the first type of strategy samples are used to make a rolling prediction of the power grid operation mode for the next 4 hours at intervals of 15 minutes, and each rolling generates a calculation file of 16 moments in the next 4 hours.

[0092] After the calculation of the second-type strategy samples is completed, a second-type strategy is generated based on the sample set, wherein the second-type strategy is configured to replace the first-type strategy if the first-type strategy fails. In some embodiments, the second-type strategy samples are calculation files generated by predicting the grid operation mode for the next day and can be used by the electromechanical transient time-domain simulation calculation program. These files are generated at 15-minute intervals, totaling 96 files corresponding to time points.

[0093] When the calculation of the third-type strategy samples is completed, a third-type strategy is generated based on the sample set, and the third-type strategy is configured to be executed if both the first-type strategy and the second-type strategy fail. In some embodiments, the third-type strategy samples are used to predict the power grid operation mode for a week, ten-day period, or month at a 4-hour interval, and the calculation file is generated.

[0094] This application uses a multi-cycle nested intelligent rolling decision-making method to transform the strategy update from a manual offline method to an automatic online update by the program. While reducing the manual workload, with the help of the multi-cycle nested intelligent rolling strategy mode, it can better balance the two performance indicators of strategy reliability and accuracy, and fully ensure the safe, stable and efficient operation of the power grid. At the same time, the main content of the stabilization strategy comes from the transient stability simulation analysis under ultra-short-term, day-ahead or long-cycle methods. By perceiving the power grid status in a relatively real-time manner, combined with simulation and distributed computing platforms, a "perception-decision-control" closed loop is realized. Its core lies in building a dynamic strategy update mechanism to deal with uncertainty. Compared with traditional offline strategies, its strategy adaptability is stronger and the control accuracy is also improved.

[0095] In summary, although the present application has been disclosed as above with preferred embodiments, the above preferred embodiments are not intended to limit the present application. Ordinary technicians in this field can make various changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims.

Claims

1. A stabilization control system for an electric power system, characterized in that: include: a control quantity calculator configured to calculate transient stability of the power system based on predicted operating data of the power system in a preset period and a preset fault set, and generate a strategy sample based on the calculation result; a computation scheduler connected to the control quantity calculator and configured to predict computation resource requirements of different operation cycles of the power system so as to schedule the control quantity calculator according to the computation resource requirements; a sample manager connected to the control quantity calculator and configured to collect all policy samples generated by the control quantity calculator and generate a sample set; The rolling manager is connected to the sample manager and is configured to generate a control strategy required for controlling the operation of the power system during the preset time period according to the sample set.

2. The stabilization control system according to claim 1, characterized in that: The control quantity calculator includes: a simulation module configured to perform electromechanical transient time-domain simulation according to the preset fault set to generate transient response data, wherein the transient response data at least includes power angle, voltage, and frequency data of the power system; a trigger module connected to the simulation module and configured to generate an instability signal when it is determined that the power system is unstable according to the transient response data; The search module is connected to the trigger module and configured to search for control measure data for restoring stability of the power system in response to the instability signal, so as to generate the strategy sample.

3. The stabilization control system according to claim 1, characterized in that: The calculation scheduler includes: A first computing cluster is configured to dispatch the control quantity calculator to predict the operating state of the power system once every first time, and each prediction generates first-type strategy samples corresponding to multiple moments in the future first time period that are evenly spaced according to the first time; The second computing cluster is configured to dispatch the control quantity calculator to predict the operating state of the power system in a second time period in the future, where the prediction period is equal to the second time period, and each prediction generates second type strategy samples corresponding to multiple moments distributed at equal intervals of a second time in the second time period in the future; a third computing cluster configured to dispatch the control quantity calculator to predict the operating state of the power system within a third time period in the future, wherein the prediction period is equal to the third time period, and each prediction generates a third type of strategy samples corresponding to a plurality of moments in the third time period that are equally spaced at a third time interval; The length of the second period is greater than that of the first period, the length of the third period is greater than that of the second period, the second time is equal to the first time, and the length of the third time is greater than that of the second time.

4. The stabilization control system according to claim 3, characterized in that: The computing scheduler is further configured to: When the third computing cluster is in an idle state, calling the control quantity calculator that performs calculations in the third computing cluster to the second computing cluster; as well as When the second computing cluster is in an idle state, the control amount calculator executing calculations in the second computing cluster is called to the first computing cluster.

5. The stabilization control system according to claim 1, characterized in that: The sample manager includes: A removal module configured to remove the strategy samples whose generation time in the sample set is earlier than the current time; a replacement module connected to the elimination module and configured to, when a policy sample generated at a current moment overlaps with an existing policy sample in the sample set, replace the policy sample in the sample set with the policy sample generated at the current moment; The supplement module is connected to the replacement module and is configured to supplement the policy sample generated at the current moment into the sample set when the time of the policy sample generated at the current moment is different from that of the existing policy samples in the sample set.

6. The stabilization control system according to claim 3, characterized in that: The scroll manager includes: The first strategy generating module is configured to generate the first strategy according to the sample set when the calculation of the first strategy sample is completed or the calculation time reaches a preset time, so as to update the first strategy every preset time.

7. The stabilization control system according to claim 6, characterized in that: The scroll manager also includes: The verification module is connected to the first strategy generation module and is configured to perform a verification simulation on the first type of strategy currently applied to the power system, and update the first type of strategy when the verification simulation result shows that the power system is unstable.

8. The stabilization control system according to claim 6, characterized in that: The scroll manager also includes: A second strategy generating module is configured to generate a second strategy according to the sample set when the calculation of the second strategy sample is completed; The second type of policy is configured to replace the first type of policy when the first type of policy fails.

9. The stabilization control system according to claim 8, characterized in that: The scroll manager also includes: A third strategy generating module is configured to generate a third type of strategy according to the sample set when the calculation of the third type of strategy samples is completed; The third type of policy is configured to be executed when both the first type of policy and the second type of policy fail.

10. The stabilization control system according to any one of claims 1 to 9, characterized in that: The strategy sample includes at least one of the following: power system operating state data, fault data, and control measure data required to solve a preset fault at a target time in a preset operating period of the power system.

11. A method for controlling a power system, characterized in that: include: Calculate the transient stability of the power system based on the predicted operating data of the power system in a preset period and a preset fault set, and generate a strategy sample based on the calculation results; Predicting the strategy samples required to deal with pre-set faults in different operation cycles of the power system; Collect all strategy samples and generate a sample set; A control strategy required for controlling the operation of the power system during a preset time period is generated according to the sample set.

12. The control method according to claim 11, characterized in that: The performing of transient stability calculation on the power system according to the predicted operation data of the power system in a preset period and a preset fault set, and generating a strategy sample according to the calculation result includes: Performing electromechanical transient time-domain simulation according to the preset fault set to generate transient response data, wherein the transient response data at least includes power angle, voltage and frequency data of the power system; When it is determined according to the transient response data that the power system is unstable, control measure data for restoring stability to the power system is searched to generate the strategy sample.

13. The control method according to claim 11, characterized in that: The strategy samples required to cope with preset faults in different operation cycles of the predicted power system include: Predicting the power system operation state once every first time period, and generating first-type strategy samples corresponding to a plurality of moments in a first future time period that are equally spaced according to the first time period during each prediction; Predicting the operating state of the power system in a second future time period, where the prediction period is equal to the second future time period, and generating second type strategy samples corresponding to a plurality of moments in the second future time period that are equally spaced at second time intervals for each prediction; Predicting the operating state of the power system within a third future time period, where the prediction period is equal to the third future time period, and generating third type strategy samples corresponding to a plurality of moments equally spaced at a third time within the third future time period for each prediction; The length of the second period is greater than that of the first period, the length of the third period is greater than that of the second period, the second time is equal to the first time, and the length of the third time is greater than that of the second time.

14. The control method according to claim 11, characterized in that: Collecting all policy samples and generating a sample set includes: Eliminate the strategy samples whose generation time is earlier than the current time in the sample set; When the policy sample generated at the current moment overlaps with the existing policy sample in the sample set, the policy sample generated at the current moment replaces the policy sample in the sample set that overlaps with the existing policy sample in the sample set; When the policy sample generated at the current moment is different from the time of the existing policy samples in the sample set, the policy sample generated at the current moment is added to the sample set.

15. The control method according to claim 13, characterized in that: Generating a control strategy required for controlling the operation of the power system during a preset period according to the sample set includes: When the calculation of the first type of policy samples is completed or the calculation time reaches a preset time, the first type of policy is generated according to the sample set to achieve updating of the first type of policy every preset time.

16. The control method according to claim 15, characterized in that: Generating a control strategy required for controlling the operation of the power system during a preset period according to the sample set includes: In order to perform a verification simulation on the first type of strategy currently applied to the power system, the first type of strategy is updated when the verification simulation result shows that the power system is unstable.

17. The control method according to claim 15, characterized in that: Generating a control strategy required for controlling the operation of the power system during a preset period according to the sample set includes: When the calculation of the second type of strategy samples is completed, generating the second type of strategy according to the sample set; The second type of policy is configured to replace the first type of policy when the first type of policy fails.

18. The control method according to claim 17, characterized in that: Generating a control strategy required for controlling the operation of the power system during a preset period according to the sample set includes: When the calculation of the third type of strategy samples is completed, generating the third type of strategy according to the sample set; The third type of policy is configured to be executed when both the first type of policy and the second type of policy fail.

Images