Primary frequency modulation parameter optimization control method and system based on big data analysis

Abstract

The invention relates to the field of parameter optimization, and discloses a primary frequency modulation parameter optimization control method and system based on big data analysis, which are used for improving the frequency stability control level of a power system. Comprising the steps of collecting historical operation data of a power system, identifying frequency disturbance events to form an event set, extracting performance features in combination with system working conditions to construct a feature knowledge base, further generating a mapping rule of system working condition categories and optimization parameters, and judging the working condition categories in real time and matching parameter recommendation values during online operation. And adaptively adjusting the parameters of the frequency modulation unit. And triggering a revision process when the parameters do not meet the safety condition. The primary frequency modulation effect of the power system can be effectively improved, and the safe and stable operation capability of the system is enhanced.

Description

Technical Field

[0001] This invention relates to the field of parameter optimization, and in particular to a method and system for optimizing primary frequency regulation parameters based on big data analysis. Background Technology

[0002] With the continuous expansion of power system scale and the large-scale integration of new energy sources, the operating characteristics of power systems have undergone profound changes. The intermittent and fluctuating output of new energy sources poses more complex challenges to power system frequency regulation. Primary frequency regulation, as the first line of defense in power system frequency regulation, plays a crucial role in maintaining system frequency stability and ensuring the safe operation of the power grid. Traditional primary frequency regulation parameter setting methods are often based on experience or fixed patterns, making it difficult to adapt to the complex and ever-changing operating conditions of power systems. Under different operating conditions, the dynamic characteristics of the power system vary significantly. Fixed parameter settings cannot achieve optimal frequency regulation effects under different operating conditions, potentially leading to excessive frequency deviation, excessively long recovery times, or instability in the regulation process, thereby affecting the safe and stable operation of the power system.

[0003] Against the backdrop of the booming development of big data technology, power systems have accumulated massive amounts of operational data, which contain rich information and patterns about system operation. How to fully utilize these big data resources, tap their potential value, and achieve intelligent and adaptive optimization of primary frequency regulation parameters has become a key issue urgently needing to be addressed in the power system field. In-depth analysis of historical power system operational data can provide a more comprehensive understanding of the system's frequency response characteristics under different operating conditions, thereby providing a scientific basis for optimizing primary frequency regulation parameters and improving the power system's frequency regulation capability and safety and stability level.

[0004] Traditional methods typically keep parameters fixed after setting them, failing to dynamically adjust them based on the real-time operating status of the system. The operating status of a power system changes in real time, and the real-time operating conditions may differ significantly from those at the time of parameter setting. If parameters are not adjusted promptly, primary frequency regulating units may not be able to achieve optimal regulation performance under new operating conditions, failing to meet the system's real-time frequency regulation requirements and reducing system safety and reliability. In the process of optimizing frequency regulation parameters, existing technologies often focus on improving the speed and accuracy of frequency regulation, while not giving sufficient consideration to the safety and stability of the system. Parameter adjustments may affect the dynamic stability of the system. If the system's safety constraints are not considered, new safety hazards may emerge after parameter adjustments, threatening the safe and stable operation of the power system.

[0005] Therefore, we propose a primary frequency modulation parameter optimization control method and system based on big data analysis to solve the above problems. Summary of the Invention

[0006] This invention provides a primary frequency regulation parameter optimization control method and system based on big data analysis, which can be used to improve the frequency stability control level of power systems.

[0007] The first aspect of this invention provides a primary frequency regulation parameter optimization control method based on big data analysis. The method includes: collecting historical operating data of a power system, identifying frequency disturbance events within the data, and forming an event set; for each event in the event set, extracting performance characteristics describing its frequency response quality based on the system operating conditions at the time of the event, and constructing a feature knowledge base; based on the feature knowledge base, quantitatively evaluating the historical response effects of different parameter settings under various typical operating conditions, calculating the optimal primary frequency regulation parameter for each typical operating condition, and generating a mapping rule between system operating condition categories and optimized parameters; during online operation, determining the current system operating condition category in real time, matching the corresponding recommended value of the primary frequency regulation parameter according to the mapping rule, and adaptively adjusting the parameters of the frequency regulation unit based on the recommended value.

[0008] Optionally, in the first implementation of the first aspect of the present invention, the method includes: classifying the system operating conditions based on the total system load level, the proportion of new energy output, and the system spinning reserve rate at the time of the event, according to a preset classification threshold, and assigning an operating condition type label to each event; extracting numerical indicators from the frequency response process data of each event, including at least the absolute value of the maximum frequency deviation in the event, the time required for the frequency to recover to the stable dead zone, and the overshoot and oscillation count during the entire frequency regulation process, to form a set of performance indicators for the event; generating an operating condition-performance-parameter association record based on the operating condition type label of each event, its corresponding set of performance indicators, and the actual primary frequency regulation parameter settings used by each unit in the system at the time of the event; collecting all the operating condition-performance-parameter association records of all events, and classifying and organizing them according to their operating condition type labels to construct a feature knowledge base.

[0009] Optionally, in the second implementation of the first aspect of the present invention, multiple operating condition-performance-parameter associated records belonging to the same operating condition type tag are clustered based on the numerical distribution characteristics of each indicator in their performance indicator set. Event records with similar response characteristics are divided into the same response pattern cluster, and a pattern identifier is assigned to each response pattern cluster. For each response pattern cluster, the distribution of parameter setting values ​​of its internal event records is statistically analyzed, and the parameter setting value that appears most frequently and best matches the average performance indicator within the cluster is extracted as the representative parameter configuration of the response pattern cluster. Each operating condition type tag, the pattern identifiers of its subordinate response pattern clusters, and the representative parameter configurations corresponding to each pattern cluster are integrated to form a hierarchical knowledge base. When generating mapping rules in the future, the optimal parameter configuration of each different response pattern cluster under each operating condition type tag will be calculated based on the hierarchical knowledge base, so that the generated mapping rules can recommend parameters based on a more granular response pattern.

[0010] Optionally, in a third implementation of the first aspect of the present invention, the method includes: selecting all operating condition-performance-parameter association records belonging to the same operating condition type tag from the feature knowledge base to form a historical case set under that operating condition type; calculating a comprehensive performance score for each case in the historical case set based on its performance index set, wherein the comprehensive performance score comprehensively considers the maximum frequency deviation, recovery time, and regulation stability, and generating a quantitative evaluation result for each historical case; based on the quantitative evaluation result, identifying the historical case with the best comprehensive performance score in the historical case set, and extracting the primary frequency regulation parameter setting value corresponding to the occurrence of that case as... The initial reference parameters for this operating condition type are used; around these initial reference parameters, a set of discrete candidate parameter combinations are set within the allowable adjustment range of the parameters, and interpolation is performed using the historical performance of similar parameter combinations in the feature knowledge base to form a candidate parameter performance prediction table; based on the candidate parameter performance prediction table, the candidate parameters are ensured to meet the hard constraints of frequency security and stability, and the candidate parameters that meet the constraints are sorted according to the expected performance to determine the recommended parameters with the best overall performance for each type of operating condition; each typical operating condition is paired and bound to its determined set of recommended parameters, and the pairing relationship between all operating condition categories and recommended parameters is integrated to generate mapping rules.

[0011] Optionally, in the fourth implementation of the first aspect of the present invention, the method includes: periodically collecting real-time operating data of the power system and extracting real-time operating condition feature vectors of the same dimension as those in the feature knowledge base, wherein the real-time operating condition feature vectors include at least the real-time system total load factor, the proportion of renewable energy output, and the spinning reserve factor; comparing the real-time operating condition feature vectors with preset ranges of various typical operating conditions, calculating the comprehensive distance between the vectors and the center points of various typical operating conditions, and determining the operating condition category with the smallest comprehensive distance as the real-time operating condition category to which the current system belongs, thereby generating a real-time operating condition category identifier; and based on the real-time operating condition category identifier, querying the mapping between the system operating condition category and optimization parameters. The rules directly retrieve the optimized primary frequency regulation parameters bound to the real-time operating condition category as the recommended parameter values ​​for the current moment. These recommended values ​​are compared with the actual parameter settings of each frequency regulation unit. If the difference exceeds a set threshold, a parameter adjustment instruction sequence is generated. This sequence is then sent to each frequency regulation unit, controlling the unit to gradually adjust its primary frequency regulation parameters to the target parameter values ​​within the smooth transition time, completing the online adaptive switching of parameters. After a primary frequency regulation action is triggered, the actual frequency response data for that adjustment process is collected, its actual performance indicators are calculated, and compared with the expected performance of the operating condition category in the mapping rules to generate a performance evaluation report.

[0012] Optionally, in the fifth implementation of the first aspect of the present invention, before sending the parameter adjustment instruction sequence to each frequency modulation unit, the method further includes: assessing the current dynamic stability state of the system based on the real-time operating condition feature vector and the real-time frequency change trend, and generating a stability state rating; based on the stability state rating, matching the corresponding transition time coefficient set from a preset transition strategy library, and using the coefficient set to dynamically correct the smooth transition time requirement to obtain personalized smooth transition time parameters; based on the personalized smooth transition time parameters, regenerating an updated parameter adjustment instruction sequence containing precise time control requirements, and sending it to each frequency modulation unit instead of the original instruction sequence.

[0013] Optionally, in the sixth implementation of the first aspect of the present invention, the method further includes: calculating the performance stability metric of each recommended parameter combination under the same operating condition based on the performance of multiple disturbance events under the same operating condition category in historical data, and forming a parameter stability evaluation table; when making real-time parameter recommendations, combining the parameter recommendation value obtained by querying the mapping rule and the stability metric of the parameter retrieved from the parameter stability evaluation table under the corresponding operating condition to generate an enhanced recommendation instruction; establishing and maintaining an abnormal operating condition registration library to record the operating states where the matching degree between the real-time identified operating condition category and the preset typical operating condition feature is lower than a set threshold, and simultaneously recording the frequency modulation parameters actually applied under the state and the performance data of subsequent disturbance events; periodically analyzing the newly accumulated event data in the abnormal operating condition registration library, and when a certain type of atypical operating condition accumulates enough effective event records, summarizing its features and defining it as a new operating condition category, and simultaneously performing parameter optimization calculation based on its historical data to generate optimized parameters for the new operating condition category, and dynamically expanding the new category and its optimized parameters into the mapping rule between the system operating condition category and the optimized parameters.

[0014] Optionally, in the seventh implementation of the first aspect of the present invention, based on real-time power flow and key section power data, the frequency safety margin of the system under a preset N-1 fault after applying the recommended parameter values ​​is calculated, and a safety assessment result is generated; the safety assessment result is compared with the frequency safety boundary values ​​specified in the power grid operation procedure to determine whether the system meets all safety boundary conditions after applying the new parameters, and a binary permission judgment is generated accordingly; when the binary permission judgment is allowed, the enhanced recommendation instruction is directly converted into a parameter adjustment instruction sequence to be issued; when the judgment is not allowed, a parameter revision process is triggered: in the mapping rule, a parameter combination with suboptimal performance that meets all safety boundary conditions is found for the current real-time operating condition category as an alternative recommendation value, and an enhanced recommendation instruction and parameter adjustment instruction sequence are regenerated based on the alternative recommendation value.

[0015] A second aspect of this invention provides a primary frequency regulation parameter optimization control system based on big data analysis. The system includes: a data acquisition module for acquiring historical operating data of the power system, identifying frequency disturbance events, and forming an event set; an extraction module for extracting performance characteristics describing the frequency response quality of each event in the event set, combined with the system operating conditions at the time of the event, and constructing a feature knowledge base; a setting module for calculating the optimal primary frequency regulation parameter for each typical operating condition by quantitatively evaluating the historical response effects of different parameter settings under various typical operating conditions based on the feature knowledge base, and generating a mapping rule between system operating condition categories and optimized parameters; and an allocation module for determining the current system operating condition category in real time during online operation, matching the corresponding recommended value of the primary frequency regulation parameter according to the mapping rule, and adaptively adjusting the parameters of the frequency regulation unit based on the recommended value.

[0016] Beneficial effects: By accurately matching parameters according to the actual operating conditions of the system, frequency fluctuations are effectively suppressed, frequency deviations are reduced, frequency recovery time is shortened, the accuracy and speed of frequency regulation are improved, the primary frequency regulation effect is significantly enhanced, and the frequency stability of the power system is strengthened; The primary frequency regulation parameters can be adjusted in a timely manner according to the real-time operating status of the system, ensuring that the optimal regulation performance can be achieved under different operating conditions, meeting the real-time requirements of the system for frequency regulation, and improving the system's adaptability to dynamic changes and its safety and reliability. While ensuring frequency regulation performance, we must fully consider system safety and stability, avoid new safety hazards caused by parameter adjustments, ensure that the system can still operate safely and stably after parameter adjustments, and improve the overall safety level of the power system. Establish an abnormal operating condition registry to record atypical operating condition data. When enough valid event records are accumulated, define new operating condition categories and generate optimization parameters, and dynamically expand the mapping rules. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of an embodiment of the primary frequency modulation parameter optimization control method based on big data analysis in this invention. Figure 2 This is a schematic diagram illustrating the acquisition of historical data on frequency, generator power, and system load from SCADA and WAMS systems. Figure 3 This is a schematic diagram of an embodiment of the primary frequency modulation parameter optimization control system based on big data analysis in this invention. Figure 4 This is a schematic diagram of an embodiment of a primary frequency modulation parameter optimization control device based on big data analysis in this invention. Detailed Implementation

[0018] This invention provides a primary frequency regulation parameter optimization control method and system based on big data analysis to improve the frequency stability control level of power systems. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figures 1-2 One embodiment of the primary frequency modulation parameter optimization control method based on big data analysis in this invention includes: 101. Collect historical operating data of the power system, identify frequency disturbance events, and form an event set; It is understood that the executing entity of this invention can be a primary frequency regulation parameter optimization control system based on big data analysis, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0020] Specifically, historical time-series data of frequency, unit active power, and total system load are obtained from the dispatching system and the wide-area measurement system; The historical time series data of frequency, unit active power and total system load are time-aligned and synchronized to generate aligned multi-source time series data; Based on aligned multi-source time-series data, potential frequency disturbance start points are captured by monitoring whether the frequency change rate exceeds the first threshold. In conjunction with monitoring whether a single unit power change exceeds the second threshold or the total system load change exceeds the third threshold, potential start points are correlated and confirmed, thereby identifying the start timestamp of the disturbance event. Based on the identified start timestamp, a preset historical time window is traced backward and a preset future time window is extended backward. A complete data segment containing the steady state before the disturbance, the disturbance occurrence process, and the entire system adjustment process is extracted from the aligned multi-source time series data to form a preliminary event record corresponding to this event. For each preliminary event record, a set of causal features are calculated and bound. The causal features include at least: the basic operating conditions of the system before the disturbance, the changes in the unit or load identified as the trigger source, and the power response sequence of each unit during the event. A quality assessment is performed on all preliminary event records with causal characteristics, and event records with severe data loss or unclear perturbation characteristics are removed to form a final event set with multi-dimensional quality labels.

[0021] 102. For each event in the event set, extract the performance characteristics describing its frequency response quality based on the system operating conditions at the time the event occurs, and construct a feature knowledge base that associates specific operating conditions with corresponding response performance. Specifically, based on the total system load level, the proportion of new energy output, and the system spinning reserve rate at the time of the event, the system operating conditions are classified according to preset classification thresholds, and each event is assigned an operating condition type label that represents its background. From the frequency response process data of each event, a set of numerical indicators for quantitatively evaluating the frequency modulation effect is extracted. The indicators include at least the absolute value of the maximum frequency deviation in the event, the time required for the frequency to recover to the stable dead zone, and the overshoot and number of oscillations in the entire frequency adjustment process, forming a set of performance indicators for the event. The operating condition type label of each event, its corresponding set of performance indicators, and the actual primary frequency regulation parameter settings used by each unit in the system when the event occurs are associated and integrated to generate a structured operating condition-performance-parameter association record. The system compiles all structured records of chemical conditions, performance, and parameters related to events, and categorizes and organizes them according to their operating condition type tags. This creates a feature knowledge base indexed by operating condition type, allowing users to query corresponding historical parameter settings and their resulting actual response performance.

[0022] Furthermore, the construction of the feature knowledge base also includes: For multiple structural working condition-performance-parameter association records belonging to the same working condition type label, cluster analysis is performed based on the numerical distribution characteristics of each indicator in its performance index set. Event records with similar response characteristics are divided into the same response pattern cluster, and a pattern identifier is assigned to each response pattern cluster. For each response pattern cluster, the distribution of parameter settings in its internal event records is statistically analyzed, and the parameter settings that appear most frequently and best match the average performance index within the cluster are extracted as the representative parameter configurations for that response pattern cluster. The label of each working condition type, the mode identifier of each response mode cluster under it, and the representative parameter configuration corresponding to each mode cluster are integrated to form a structured hierarchical knowledge base. This hierarchical knowledge base further distinguishes different typical response modes and corresponding parameters under the working condition type. When generating mapping rules in the future, the optimal parameter configuration will be calculated for each different response mode cluster under each working condition type label based on the hierarchical knowledge base, so that the generated mapping rules can recommend parameters based on more granular response modes.

[0023] 103. Based on the feature knowledge base, the historical response effects of different parameter settings under various typical operating conditions are evaluated by quantitative assessment. A set of primary frequency regulation parameters with the best overall effect is calculated for each type of typical operating condition, and a set of mapping rules between system operating condition categories and optimization parameters is generated. Specifically, from the feature knowledge base, all working condition-performance-parameter association records belonging to the same working condition type label are selected to form a set of historical cases under that working condition type; For each case in the historical case set, a comprehensive performance score is calculated based on its set of performance indicators. The score takes into account the maximum frequency deviation, recovery time, and adjustment stability, thereby generating a quantitative evaluation result for each historical case that represents the overall performance of its frequency modulation. Based on the quantitative evaluation results, the historical case with the best comprehensive performance score is identified in the historical case set, and the primary frequency regulation parameter setting value corresponding to the occurrence of the case is extracted as the initial reference parameter for this working condition type. Based on the initial reference parameters, a set of discrete candidate parameter combinations are set within the allowable adjustment range of the parameters. Interpolation is performed using the historical performance of similar parameter combinations in the feature knowledge base to predict the expected performance that each candidate parameter combination may achieve under the working condition, forming a candidate parameter performance prediction table containing multiple sets of predicted performance. Based on the candidate parameter performance prediction table, a step-by-step screening strategy is adopted. First, the candidate parameters are ensured to meet the hard constraints of frequency safety and stability. Then, the candidate parameters that meet the constraints are sorted according to the expected performance. Finally, a set of recommended parameters with the best overall performance is determined for each type of working condition. Each typical operating condition is paired and bound to a set of recommended parameters, integrating the pairing relationships between all operating condition categories and recommended parameters, thereby generating a set of mapping rules between system operating condition categories and optimization parameters that can be directly used for online querying and matching.

[0024] 104. During online operation, the system determines the current operating condition category in real time, matches the corresponding recommended value of primary frequency regulation parameters according to the mapping rules, and adaptively adjusts the parameters of the frequency regulation unit based on the recommended value.

[0025] Specifically, real-time operating data of the power system is collected periodically, and real-time operating condition feature vectors of the same dimension as those in the feature knowledge base are extracted from it. The feature vectors include at least the real-time total system load factor, the proportion of new energy output, and the spinning reserve factor. The real-time operating condition feature vector is compared with the preset range of typical operating conditions feature vectors item by item, the comprehensive distance between the vector and the center point of each typical operating condition feature vector is calculated, and the operating condition category with the smallest comprehensive distance is determined as the real-time operating condition category to which the current system belongs, generating a real-time operating condition category identifier. Based on the real-time operating condition category identifier, query the mapping rules between the system operating condition category and the optimized parameters, and directly retrieve a set of optimized first-order frequency regulation parameters bound to the real-time operating condition category as the recommended parameter values ​​for the current moment; The recommended parameter values ​​are compared with the actual parameter settings of each frequency regulation unit. If the difference exceeds the set threshold, a parameter adjustment instruction sequence containing the target parameter value and the smooth transition time requirement is generated. The parameter adjustment command sequence is sent to each frequency regulation unit, and the control unit gradually adjusts its primary frequency regulation parameters to the target parameter value within a smooth transition time, thus completing the online adaptive switching of parameters; After a frequency modulation action is triggered, the actual frequency response data of the modulation process is collected, its actual performance index is calculated, and it is compared with the expected performance of the working condition category in the mapping rule. A performance evaluation report is generated to evaluate the effectiveness of the parameter settings. The report will serve as the feedback basis for subsequent periodic iterative optimization of the mapping rule.

[0026] Furthermore, before issuing the parameter adjustment command sequence to each frequency regulation unit, the process also includes: Based on the real-time operating condition feature vector and the real-time frequency change trend, the current dynamic stable state of the system is assessed, and a stable state rating characterizing the system's sensitivity to parameter changes is generated. Based on the stable state rating, the corresponding transition time coefficient set is matched from the preset transition strategy library, and the smooth transition time requirement is dynamically modified using this coefficient set to obtain a personalized smooth transition time parameter that adapts to the current system dynamics. Based on the personalized smooth transition time parameters, an updated parameter adjustment instruction sequence containing precise time control requirements is regenerated to replace the original instruction sequence and be sent to each frequency regulation unit.

[0027] 105. Based on the performance of multiple disturbance events under the same working condition category in historical data, calculate the performance stability metric value of each recommended parameter combination under the working condition, and form a parameter stability evaluation table with the stability of different parameter combinations under each working condition category. When making real-time parameter recommendations, the recommended parameter values ​​obtained from the query mapping rules and the stability measurement values ​​of the parameter under the corresponding working conditions retrieved from the parameter stability evaluation table are combined to generate an enhanced recommendation instruction that includes the parameter value and its confidence level. Establish and maintain an abnormal operating condition registry to record operating states where the real-time identified operating condition category matches the preset typical operating condition characteristics with a lower than set threshold. At the same time, record the actual frequency regulation parameters applied in this state and the performance data of subsequent disturbance events. The system periodically analyzes newly accumulated event data in the abnormal operating condition registry. When a certain type of atypical operating condition accumulates enough valid event records, its characteristics are summarized and defined as a new operating condition category. At the same time, parameter optimization calculations are performed based on its historical data to generate optimized parameters for the new operating condition category. The new category and its optimized parameters are then dynamically expanded into the mapping rules between system operating condition categories and optimized parameters.

[0028] Specifically, after generating the enhanced recommendation instruction and before issuing the parameter adjustment instruction sequence, the process also includes: Based on real-time power flow and key section power data, the frequency safety margin of the system under the preset N-1 fault is calculated after applying the recommended values ​​of the parameters, and a safety assessment result containing the lowest frequency point under various anticipated faults is generated. The safety assessment results are compared with the frequency safety boundary values ​​specified in the power grid operation procedures to determine whether the system meets all safety boundary conditions after applying the new parameters, and a binary permission judgment is generated based on this to determine whether the parameter adjustment is allowed. When the binary permission is determined to be allowed, the enhanced recommended instruction is directly converted into a sequence of parameter adjustment instructions to be issued; when it is determined to be disallowed, a parameter revision process is triggered: in the mapping rules, the parameter combination that has the second-best performance and meets all safety boundary conditions for the current real-time operating condition category is found as an alternative recommended value, and the enhanced recommended instruction and parameter adjustment instruction sequence are regenerated based on the alternative recommended value.

[0029] In this embodiment of the invention, the system operating conditions are meticulously categorized, and quantitative performance indicators are extracted to form a performance indicator set. These are then linked and integrated to generate structured records of operating conditions, performance, and parameters, constructing a feature knowledge base. Further clustering analysis forms a hierarchical knowledge base, distinguishing different typical response modes and corresponding parameters. This more accurately reflects the relationship between system response characteristics and parameters under different operating conditions, providing a more detailed and accurate basis for parameter optimization. Compared to traditional knowledge bases, this approach better adapts to complex and ever-changing power system operating conditions. Based on the feature knowledge base, the historical response effects of different parameter settings are quantitatively evaluated. Through steps such as comprehensive performance scoring, initial reference parameter selection, candidate parameter combination setting, interpolation and performance prediction, and step-by-step selection to determine recommended parameters, mapping rules between system operating condition categories and optimized parameters are generated. This ensures that the generated mapping rules provide the optimal primary frequency regulation parameters for different operating conditions, improving system frequency regulation performance and enhancing power system stability. During online operation, the system identifies operating condition categories in real time, matches recommended parameter values ​​according to mapping rules, and adaptively adjusts frequency regulation unit parameters. Simultaneously, it collects actual frequency response data to calculate performance indicators and generates performance evaluation reports as feedback, ensuring the system remains in optimal frequency regulation state. Furthermore, performance evaluation feedback continuously optimizes mapping rules, creating a virtuous cycle that effectively enhances the system's ability to respond to real-time changes and guarantees the safe and stable operation of the power system. The system calculates performance stability metrics for each recommended parameter combination to form a parameter stability evaluation table. Combining recommended parameter values ​​with stability metrics generates enhanced recommendation instructions, considering not only the optimization effect of parameters but also their stability under different operating conditions. This provides more comprehensive and reliable information for parameter adjustment, helping to improve the reliability and stability of system frequency regulation and reduce the frequency regulation risks caused by parameter instability. An abnormal operating condition registry is established to record atypical operating condition data. Accumulated event data is periodically analyzed, new operating condition categories are defined and optimized parameters are generated, and mapping rules are dynamically expanded. This allows the system to adapt to the constantly changing power system operating environment, promptly detect and handle newly emerging operating conditions, and continuously improve mapping rules, thereby enhancing the system's adaptability to various operating conditions and its frequency regulation performance. Before parameter adjustment, the frequency safety margin is calculated based on real-time grid data and compared with the safety boundary value to generate a binary permissive judgment. Based on the judgment result, it is determined whether to execute parameter adjustment or trigger the parameter revision process, ensuring that parameter adjustment will not adversely affect system frequency safety. While pursuing frequency regulation performance optimization, the system strictly guarantees the safe and stable operation of the power system and effectively reduces safety risks.

[0030] The above describes the primary frequency regulation parameter optimization control method based on big data analysis in the embodiments of the present invention. The following describes the primary frequency regulation parameter optimization control system based on big data analysis in the embodiments of the present invention. Please refer to [link / reference]. Figure 3An embodiment of the primary frequency regulation parameter optimization control system based on big data analysis in this invention includes: a data acquisition module 201, used to acquire historical operating data of the power system, identify frequency disturbance events therein, and form an event set; an extraction module 202, used to extract performance characteristics describing the frequency response quality of each event in the event set, combined with the system operating conditions at the time of the event, and construct a feature knowledge base; a setting module 203, used to calculate the optimal primary frequency regulation parameter for each type of typical operating condition by quantitatively evaluating the historical response effects of different parameter settings under various typical operating conditions based on the feature knowledge base, and generate a mapping rule between system operating condition categories and optimization parameters; and an allocation module 204, used to determine the current system operating condition category in real time during online operation, match the corresponding recommended value of primary frequency regulation parameter according to the mapping rule, and adaptively adjust the parameters of the frequency regulation unit based on the recommended value.

[0031] In this embodiment of the invention, the operating conditions of power systems are complex, diverse, and constantly changing. Traditional fixed parameter settings are difficult to adapt to different operating conditions. By constructing a mapping rule between system operating condition categories and optimized parameters, the operating condition category is identified in real time during online operation, and the parameters of the frequency regulation unit are adaptively adjusted. This enables the system to respond quickly to changes in different operating conditions, effectively enhancing the adaptability of the primary frequency regulation system to complex and ever-changing power systems. The precisely optimized parameters and good adaptability allow the frequency regulation unit to operate with more reasonable parameters under different operating conditions, thereby responding to frequency disturbances more promptly and accurately, effectively improving the primary frequency regulation effect, reducing the range and duration of frequency fluctuations, enhancing the stability of the power system, and ensuring the safe and reliable supply of power. By making full use of the rich historical operating data of the power system and deeply mining the value of the data through big data analysis technology, the originally scattered and complex data is transformed into a feature knowledge base and mapping rules, providing a scientific basis for the optimization of primary frequency regulation parameters.

[0032] Figure 4 This is a schematic diagram of a primary frequency modulation parameter optimization control device 300 based on big data analysis, provided in an embodiment of the present invention. The device 300 may include a processor 301 and a memory 302. The memory 302 stores program instructions and / or data, and the processor 301 executes the program instructions stored in the memory 302 to implement the method described in the above embodiment.

[0033] Optionally, the memory 302 and the processor 301 are coupled. The coupling is an indirect coupling or communication connection between devices, units, or modules, and can be electrical, mechanical, or other forms, for information interaction between devices, units, or modules.

[0034] Optionally, the primary frequency modulation parameter optimization control device 300 based on big data analysis may further include a communication interface 303. The communication interface 303 is used to communicate with other devices through a transmission medium, for example, transmitting received signals from other communication devices to the processor 301, or transmitting signals from the processor 301 to other communication devices. The communication interface 303 may be a transceiver or an interface circuit, such as a transceiver circuit or a transceiver chip.

[0035] This application embodiment does not limit the specific connection medium between the processor 301, memory 302, and communication interface 303. This application embodiment... Figure 4 The processor 301, memory 302, and communication interface 303 are connected via a bus 304. Figure 4 The connections between other components are shown in bold and are for illustrative purposes only, not as limiting information. The bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0036] The present invention also provides a primary frequency modulation parameter optimization control device based on big data analysis. The primary frequency modulation parameter optimization control device based on big data analysis includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the primary frequency modulation parameter optimization control method based on big data analysis in the above embodiments.

[0037] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the primary frequency modulation parameter optimization control method based on big data analysis.

[0038] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0039] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0040] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A primary frequency regulation parameter optimization control method based on big data analysis, characterized in that, include: Collect historical operating data of the power system, identify frequency disturbance events, and form an event set; For each event in the event set, performance characteristics describing its frequency response quality are extracted based on the system operating conditions at the time the event occurs, and a feature knowledge base is constructed. Based on the feature knowledge base, the historical response effects of different parameter settings under various typical operating conditions are quantitatively evaluated, and the optimal frequency regulation parameter with the best overall effect is calculated for each type of typical operating condition, generating a mapping rule between system operating condition categories and optimization parameters. During online operation, the system determines the current operating condition category in real time, matches the corresponding recommended value of primary frequency regulation parameters according to the mapping rules, and adaptively adjusts the parameters of the frequency regulation unit based on the recommended value.

2. The primary frequency modulation parameter optimization control method based on big data analysis according to claim 1, characterized in that, include: Based on the total system load level, the proportion of new energy output, and the system spinning reserve rate at the time of the event, the system operating conditions are classified according to a preset classification threshold, and each event is assigned an operating condition type label. Numerical indicators are extracted from the frequency response process data of each event, including at least the absolute value of the maximum frequency deviation in the event, the time required for the frequency to recover to the stable dead zone, and the overshoot and number of oscillations during the entire frequency adjustment process, to form a set of performance indicators for the event. Based on the operating condition type label of each event, its corresponding set of performance indicators, and the actual primary frequency regulation parameter settings used by each unit in the system when the event occurs, an operating condition-performance-parameter association record is generated. Collect all event-performance-parameter association records, and categorize and organize them according to their operating condition type tags to build a feature knowledge base.

3. The primary frequency modulation parameter optimization control method based on big data analysis according to claim 2, characterized in that, For multiple working condition-performance-parameter associated records belonging to the same working condition type label, cluster analysis is performed based on the numerical distribution characteristics of each indicator in its performance indicator set. Event records with similar response characteristics are divided into the same response pattern cluster, and a pattern identifier is assigned to each response pattern cluster. For each response pattern cluster, the distribution of parameter setting values ​​of its internal event records is statistically analyzed, and the parameter setting value that appears most frequently and best matches the average performance index within the cluster is extracted as the representative parameter configuration of the response pattern cluster. The hierarchical knowledge base is formed by integrating each working condition type label, the mode identifier of each response mode cluster under it, and the representative parameter configurations corresponding to each mode cluster. When generating mapping rules in the future, the optimal parameter configuration will be calculated for each different response mode cluster under each working condition type label based on the hierarchical knowledge base, so that the generated mapping rules can recommend parameters based on more granular response modes.

4. The primary frequency modulation parameter optimization control method based on big data analysis according to claim 2, characterized in that, include: From the feature knowledge base, all working condition-performance-parameter association records belonging to the same working condition type tag are selected to form a historical case set under that working condition type; For each case in the historical case set, a comprehensive performance score is calculated based on its performance index set. The comprehensive performance score takes into account the maximum frequency deviation, recovery time, and regulation stability, and generates a quantitative evaluation result for each historical case. Based on the quantitative evaluation results, the historical case with the best comprehensive performance score is identified in the historical case set, and the primary frequency regulation parameter setting value corresponding to the occurrence of the case is extracted as the initial reference parameter for this working condition type. Based on the initial reference parameters, a set of discrete candidate parameter combinations are set within the allowable adjustment range of the parameters, and interpolation is performed using the historical performance of similar parameter combinations in the feature knowledge base to form a candidate parameter performance prediction table. Based on the candidate parameter performance prediction table, ensure that the candidate parameters meet the hard constraints of frequency safety and stability. Sort the candidate parameters that meet the constraints according to the expected performance and determine the recommended parameters with the best overall performance for each type of working condition. Each typical working condition is paired and bound to a set of recommended parameters, and the pairing relationship between all working condition categories and recommended parameters is integrated to generate mapping rules.

5. The primary frequency modulation parameter optimization control method based on big data analysis according to claim 4, characterized in that, include: Real-time operating data of the power system is periodically collected, and real-time operating condition feature vectors of the same dimension as those in the feature knowledge base are extracted from it. The real-time operating condition feature vectors include at least the real-time total system load factor, the proportion of new energy output, and the spinning reserve factor. The real-time operating condition feature vector is compared with the preset range of various typical operating conditions feature items one by one, the comprehensive distance between it and the center point of various typical operating conditions feature is calculated, and the operating condition category with the smallest comprehensive distance is determined as the real-time operating condition category to which the current system belongs, and a real-time operating condition category identifier is generated. Based on the real-time operating condition category identifier, the mapping rule between the system operating condition category and the optimization parameters is queried, and the optimized first frequency regulation parameter bound to the real-time operating condition category is directly retrieved as the recommended parameter value at the current moment. The recommended parameter values ​​are compared with the actual parameter settings of each frequency regulation unit. If the difference exceeds a set threshold, a parameter adjustment instruction sequence is generated. The parameter adjustment instruction sequence is sent to each frequency modulation unit, and the control unit gradually adjusts its primary frequency modulation parameters to the target parameter value within the smooth transition time, thus completing the online adaptive switching of parameters; After a frequency modulation action is triggered, the actual frequency response data of the modulation process is collected, its actual performance index is calculated, and it is compared with the expected performance of the working condition category in the mapping rules to generate a performance evaluation report.

6. The primary frequency modulation parameter optimization control method based on big data analysis according to claim 5, characterized in that, Before sending the parameter adjustment command sequence to each frequency modulation unit, the process also includes: Based on the real-time operating condition feature vector and the real-time frequency change trend, the current dynamic stability state of the system is assessed, and a stability state rating is generated. Based on the stable state rating, a set of corresponding transition time coefficients is matched from a preset transition strategy library, and the smooth transition time requirement is dynamically corrected using this set of coefficients to obtain personalized smooth transition time parameters. Based on the personalized smooth transition time parameters, an updated parameter adjustment instruction sequence containing precise time control requirements is regenerated and sent to each frequency regulation unit to replace the original instruction sequence.

7. The primary frequency modulation parameter optimization control method based on big data analysis according to claim 5, characterized in that, Also includes: Based on the performance of multiple disturbance events under the same working condition category in historical data, the performance stability metric of each recommended parameter combination under the working condition is calculated to form a parameter stability evaluation table. When making real-time parameter recommendations, the recommended parameter values ​​obtained by querying the mapping rules and the stability measurement values ​​of the parameter under the corresponding working conditions retrieved from the parameter stability evaluation table are combined to generate an enhanced recommendation instruction. Establish and maintain an abnormal operating condition registry to record operating states where the real-time identified operating condition category matches the preset typical operating condition characteristics with a lower than set threshold. At the same time, record the actual frequency regulation parameters applied in this state and the performance data of subsequent disturbance events. The newly accumulated event data in the abnormal operating condition registration library is periodically analyzed. When a certain type of atypical operating condition accumulates enough valid event records, its characteristics are summarized and defined as a new operating condition category. At the same time, parameter optimization calculation is performed based on its historical data to generate optimized parameters for the new operating condition category. The new category and its optimized parameters are then dynamically expanded into the mapping rules between system operating condition categories and optimized parameters.

8. The primary frequency modulation parameter optimization control method based on big data analysis according to claim 7, characterized in that, Based on real-time power flow and key section power data, the frequency safety margin of the system under a preset N-1 fault is calculated after applying the recommended values ​​of the parameters, and a safety assessment result is generated. The safety assessment results are compared with the frequency safety boundary values ​​specified in the power grid operation procedures to determine whether the system meets all safety boundary conditions after applying the new parameters, and a binary permission judgment is generated accordingly. When the binary permission is determined to be allowed, the enhanced recommendation instruction is directly converted into a sequence of parameter adjustment instructions to be issued; when it is determined to be disallowed, a parameter revision process is triggered: in the mapping rule, a parameter combination with suboptimal performance that meets all safety boundary conditions is found for the current real-time operating condition category as an alternative recommendation value, and the enhanced recommendation instruction and parameter adjustment instruction sequence are regenerated based on the alternative recommendation value.

9. A primary frequency regulation parameter optimization control system based on big data analysis, characterized in that, The primary frequency regulation parameter optimization control system based on big data analysis includes: The data acquisition module is used to collect historical operating data of the power system, identify frequency disturbance events, and form an event set. The extraction module is used to extract performance features describing the frequency response quality of each event in the event set, combined with the system operating conditions at the time of the event, and to build a feature knowledge base. The setting module is used to, based on the feature knowledge base, quantitatively evaluate the historical response effects of different parameter settings under various typical operating conditions, calculate the optimal primary frequency regulation parameter for each type of typical operating condition, and generate a mapping rule between system operating condition categories and optimization parameters. The allocation module is used to determine the current operating condition category of the system in real time during online operation, match the corresponding recommended value of primary frequency regulation parameters according to the mapping rules, and adaptively adjust the parameters of the frequency regulation unit based on the recommended value.

Images