Method for guaranteeing frequency safety in day-ahead plan arrangement of direct-current sending-end power grid

Abstract

The invention discloses a method for guaranteeing frequency safety in day-ahead plan arrangement of a direct-current sending-end power grid, and belongs to the technical field of power system dispatching operation. Comprising the following steps: firstly, establishing dynamic safety boundaries of a system rotational inertia demand, a power vacancy endurance capability and a frequency regulation demand based on historical operation data and simulation of a power grid; then, constructing a parallel architecture comprising three evaluation channels of inertia adequacy, power impact risk and adjustment rapidity based on the dynamic safety boundary, independently analyzing day-ahead plan safety for each channel, and outputting an evaluation result; a correction strategy is intelligently selected through a strategy matching engine, and an optimization algorithm is adopted to carry out safety correction on the day-ahead plan; and finally, performing safety margin quantitative analysis on the corrected plan, generating a real-time defense strategy packet, and establishing a collaborative defense system of the day-ahead plan and real-time control. The method solves the problems that in the prior art, the evaluation dimension is single, the boundary adaptability is poor, correction lacks intelligence, and planning and control are disjointed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power system dispatching operation, and particularly relates to a method for guaranteeing frequency safety in day-ahead planning arrangement of a direct-current sending-end power grid. BACKGROUND

[0002] With the deepening of energy transformation, the penetration rate of new energy represented by wind energy and solar energy in the direct-current sending-end power grid continues to rise, and the scale of extra-high voltage direct-current transmission is continuously expanding. On the one hand, this optimizes the energy configuration structure, and on the other hand, it brings unprecedented challenges to the frequency stability of the sending-end power grid. New energy units are connected to the grid through power electronic devices, and their inherent low inertia and weak anti-disturbance ability significantly weaken the equivalent rotational inertia and frequency response capability of the system; and when a single or multiple loop blocking faults occur in a large-capacity direct-current transmission system, a huge active power shortage will be instantly generated, which will easily cause the sending-end power grid frequency to drop sharply, and in severe cases, may lead to system splitting or large-area power outage accidents.

[0003] At present, the power grid dispatching and operation department mainly relies on traditional deterministic safety checking criteria, such as N-1 static safety constraints, when preparing the day-ahead generation plan. However, such methods are difficult to accurately quantify the dynamic frequency characteristics of the system under high proportion of new energy access. In the existing technology, although some research attempts to consider frequency safety in the planning stage, there are usually the following significant defects: first, the evaluation dimension is single, or only the system inertia is considered, or only the reserve capacity is considered, and there is a lack of coordinated analysis and comprehensive evaluation of the three key dimensions of inertia support, power impact resistance and regulation response, leading to incomplete risk assessment. Secondly, the safety boundaries used are mostly static thresholds, which cannot adapt to the volatility of new energy output and load, leading to either overly conservative planning arrangements that sacrifice economic efficiency or safety risks in extreme conditions. Finally, the existing methods generally lack a coordinated defense system that closely couples day-ahead plan evaluation, automatic correction and real-time defense strategies, making the day-ahead plan and real-time control disjointed and difficult to achieve integrated management of frequency risks through "prevention in advance and control in progress".

[0004] Therefore, there is an urgent need in the field for a comprehensive and accurate evaluation of the frequency safety risk of the direct-current sending-end power grid, and a forward-looking and adaptive correction in the day-ahead planning stage, as well as a coordinated cooperation with real-time control, to ensure the safe and stable operation of the power grid under high proportion of new energy access. SUMMARY

[0005] In view of the above deficiencies in the prior art, the purpose of the present application is to provide a method for guaranteeing frequency safety in day-ahead scheduling of a DC sending-end power grid, which solves the problems of single evaluation dimension, poor adaptability of static boundary, lack of intelligence of correction strategy, and disconnection between day-ahead scheduling and real-time control, realizes the transition from static passive checking to dynamic active defense, and comprehensively improves the frequency safety level of the power grid under high proportion of new energy access.

[0006] A method for guaranteeing frequency safety in day-ahead scheduling of a DC sending-end power grid, comprising the following steps:

[0007] Based on the historical operation data and simulation of the power grid, the key factors affecting frequency stability are analyzed, and a dynamic safety boundary is established;

[0008] Based on the dynamic safety boundary, a parallel architecture of evaluation channels is constructed, the safety of the day-ahead scheduling in different dimensions is independently analyzed for each channel, and the safety state evaluation results of each dimension are output;

[0009] The safety state evaluation results are intelligently selected for correction strategy through a strategy matching engine, and an optimization algorithm is used to perform safety correction on the day-ahead scheduling;

[0010] The plan after safety correction is subjected to safety margin quantitative analysis, and based on the analysis results, a real-time defense strategy package is generated to establish a cooperative defense system of day-ahead scheduling and real-time control.

[0011] As a further scheme, the establishment of the dynamic safety boundary comprises:

[0012] Data acquisition and feature extraction, a complete power grid operation data acquisition system is established, load data, new energy power generation, unit output, unit operation state, and DC transmission power key operation parameters are collected in real time through an energy management system (EMS) and a wide area measurement system (WAMS); a standardized frequency safety feature database is constructed, the collected data is subjected to quality checking, normalization and abnormal data cleaning, invalid data caused by communication interruption or measurement failure is eliminated, and the integrity and accuracy of the data are ensured;

[0013] S120. Rotational inertia safety boundary modeling, based on the historical operation data of the power grid, a multivariate nonlinear regression method is used to establish a rotational inertia safety boundary model, and the minimum inertia requirement for maintaining frequency stability under different load levels and different new energy penetration rates is analyzed;

[0014] S130. Power shortage tolerance boundary establishment, a power system time domain simulation analysis is performed to establish a power shortage tolerance capability evaluation model, the frequency dynamic process under single or multiple DC blocking faults in different operating modes is simulated, and the maximum allowable power shortage for ensuring system frequency stability is determined;

[0015] S140. Determine the frequency regulation capability boundary, analyze the characteristics of the system's frequency regulation resources, comprehensively consider the differences in regulation performance of different types of units, and establish a frequency regulation capability demand model.

[0016] Formula for frequency regulation capability requirement model:

[0017] ;

[0018] ;

[0019] for Minimum standby capacity required by the system during a given time period; for Minimum standby capacity required by the system during a given time period; This is the safety factor for reserve capacity; For adjusting the rate coefficient; The target time for the system frequency to recover to a safe level.

[0020] As a further option, the formula for the rotational inertia safety boundary model is expressed as:

[0021] ;

[0022] express Minimum permissible inertia of the system over a time period; for Total load over the time period; for Time period of new energy power output; function The regression coefficients are determined by fitting historical data using the least squares method. This is the model error term.

[0023] As a further option, the power deficit tolerance assessment model is expressed as follows:

[0024] ;

[0025] express Maximum permissible power deficit of the system during the time period; for The actual equivalent inertia of the system over a time period; These are the coefficients determined by fitting simulation data; This is the model error term;

[0026] As a further option, the parallel architecture includes: an inertia adequacy assessment channel, a power surge risk assessment channel, and a regulation speed assessment channel.

[0027] As a further option, the method for independently analyzing the security of each channel according to the planned security dimensions in different dimensions and outputting the security status assessment results for each dimension includes:

[0028] The inertia adequacy assessment channel, based on the unit combination scheme in the day-ahead power generation plan, calculates the system equivalent inertia for each time period according to the inertial time constant, rated capacity, and operating status of each unit; equivalent inertia The calculation formula is:

[0029] ;

[0030] For the unit The inertial time constant; For the unit Rated capacity; For the unit exist The operational status of a time period; This represents the total number of generating units.

[0031] The calculated equivalent inertia and Minimum permissible inertia for a time-limited system Compare the results and set an early warning threshold based on the relative difference. Output the evaluation results, and the judgment logic is as follows:

[0032] ;

[0033] The inertia adequacy warning threshold;

[0034] The power surge risk assessment channel identifies the maximum DC transmission power for each time period in the day-ahead plan, assesses the power surge risk in conjunction with the system operating status, considers the impact of multiple DC lines operating simultaneously, uses probabilistic analysis to quantify the risk level, and defines power surge risk indicators. :

[0035] ;

[0036] for Total DC transmission power over the entire time period; For the first DC return The equivalent failure rate for a time period can be determined based on historical failure statistics and the operating environment. This represents the maximum DC transmission power for each time period as planned in the current day.

[0037] Power surge risk indicators The risk level assessment result is output by comparing the power impact risk index with a preset threshold value.

[0038] The regulation rapidity assessment channel analyzes the spare capacity configuration in the day-ahead plan and the system equivalent regulation rate to construct a comprehensive assessment index The comprehensive assessment index is a short board function of the spare capacity and the regulation rate, and is expressed by the following formula:

[0039] ;

[0040] is the spare capacity configured in the day-ahead plan; is the system equivalent regulation rate;

[0041] The comprehensive assessment index is compared with the level threshold value, and an assessment result is output:

[0042] ;

[0043] The comprehensive assessment result is generated, the three assessment results are integrated, a unified safety state assessment system is established, and a safety state atlas with time as the horizontal axis and risk dimension as the vertical axis is generated.

[0044] As a further scheme, the step of correcting the day-ahead plan by using an optimization algorithm includes:

[0045] The strategy matching engine is designed to establish a rule-based strategy matching engine, and the input is a multi-dimensional safety assessment result, i.e., the assessment conclusions of the three channels of inertia adequacy, power impact risk and regulation rapidity; a complete strategy selection logic rule library is pre-stored in the engine, which can automatically and quickly match and recommend the optimal correction strategy sequence according to different risk combination modes;

[0046] S320. Correction strategy library construction, a structured and extensible correction strategy library is constructed to provide specific action plans for the strategy matching engine;

[0047] S330. Safety correction optimization model, an optimization model considering safety and economy is established, and the model introduces the frequency safety constraint derived from the dynamic safety boundary on the basis of the traditional unit commitment and economic dispatch model, and converts the frequency stability requirement into a hard constraint in the optimization problem;

[0048] The optimization model minimizes the total operation cost of the system, and the cost includes the active power generation cost of all started units, the start-up cost and the shutdown cost of the units, and the power impact risk index is also included in the cost form in the objective function to realize the trade-off between safety and economy; the mathematical expression is as follows:

[0049] ;

[0050] To ensure that the corrected plan meets the frequency safety requirements, the model introduces the following key constraints:

[0051] ;

[0052] The constraints respectively ensure that the system has sufficient rotational inertia at any time, the DC power impact is within the bearable range, the standby capacity meets the requirements, and the system has sufficient frequency regulation rate;

[0053] S340. Verification of the effect of correction, the day-ahead generation plan generated after optimization and correction is re-executed multi-dimensional frequency safety parallel evaluation process, by comparing the safety state of the system in each evaluation channel before and after correction, and calculating quantitative indicators such as safety constraint satisfaction rate, total cost change rate and risk index reduction rate, to comprehensively and objectively evaluate the safety improvement effect and economic impact of the correction strategy adopted;

[0054] Only the generation plan that has passed the verification and proved that its safety and economy have reached the expected standard will be finally adopted and applied to actual dispatch.

[0055] As a further solution, strict strategy priority rules are also defined in the strategy matching engine, ensuring that when multiple risks occur simultaneously, the risks can be ordered according to the degree of harm to frequency stability, and the optimization of system safety is realized.

[0056] As a further solution, the correction strategy library core includes inertia enhancement strategy, power impact mitigation strategy, and regulation capacity enhancement strategy;

[0057] The core execution measure of the inertia enhancement strategy is to adjust the unit combination, in the period when the safety evaluation is dangerous or warning, the dispatching algorithm will start or increase the output of high inertia units in priority under the premise of meeting other constraints, and correspondingly reduce or shut down the units with small inertia contribution;

[0058] The core execution measure of the power impact mitigation strategy is to optimize the DC power plan, in the period when the power impact risk is high, by coordinating with the DC dispatching center, actively and moderately reducing the transmission power of the key DC loop, or adjusting its power curve to make it change more smoothly, reducing the maximum power shortage that may be caused by a single fault; Its control parameter is the DC power safety limit value calculated according to the system tolerance capacity;

[0059] The core execution measure of the regulation capacity enhancement strategy is to optimize the distribution and configuration of standby capacity.

[0060] As a further solution, the step of establishing a coordinated defense system of day-ahead plan and real-time control includes:

[0061] Safety margin quantitative analysis, by calculating the safety margin of each period in the three dimensions of inertia adequacy, power impact tolerance and regulation speed, the safety level of the system is accurately quantified;

[0062] Inertia margin reflects the relative proportion of the actual inertia of the system exceeding the minimum demand;

[0063] Power impact tolerance margin reflects the degree of surplus of the maximum allowable power shortage of the system relative to the maximum potential power impact;

[0064] The regulation capacity margin reflects the excess level of the actual regulation capacity of the system relative to the minimum requirement;

[0065] According to the margin value, the safety state of each period of the system is finely divided into four levels of loose, normal, tense and critical, so as to accurately identify the weak links and key risk periods in the system.

[0066] Real-time defense strategy design, based on the safety margin quantitative results, design hierarchical and classified real-time defense strategy, and closely link with the risk level prediction results in the day-ahead plan, the defense system starts different defense measures according to different margin levels;

[0067] Strategy package verification and deployment, to ensure the reliability and effectiveness of the designed real-time defense strategy.

[0068] As a further scheme, the differentiated defense measures include:

[0069] When the system is in loose or normal level, execute the routine operation monitoring, the defense system is in low alert standby state;

[0070] When the system enters the tense level, the preparatory control strategy is automatically activated, and the strategy does not immediately affect the grid operation, but is ready to execute at any time;

[0071] When the system is considered to be in critical level, the emergency control strategy is prepared to be started; once the real-time monitoring system detects the preset large power shortage fault occurs, the system will automatically execute emergency control measures such as accurate load shedding, DC power speed reduction, etc., to prevent frequency collapse.

[0072] As a further scheme, the method for ensuring the reliability and effectiveness of the designed real-time defense strategy includes:

[0073] First, through the digital simulation platform, the defense strategy package is strictly closed-loop tested, various possible expected accident scenarios are simulated, the correctness and effectiveness of the strategy action and the coordination between different control measures are verified; according to the simulation results, the parameters of the strategy are finely optimized;

[0074] Subsequently, the well-verified strategy package and its trigger logic are formally deployed into the production control system, and it is ensured that it can be called and updated online and rolling according to the actual day-ahead plan and real-time operation mode.

[0075] To sum up, due to the adoption of the technical scheme, the beneficial technical effects of the invention are:

[0076] Multi-dimensional accurate evaluation of frequency safety is realized: by constructing parallel evaluation channels containing inertia adequacy, power impact risk and regulation rapidity, the limitations of traditional single-dimensional evaluation are overcome, and the weak links of the day-ahead plan in different operation periods and different risk dimensions can be comprehensively and accurately identified, thereby significantly improving the comprehensiveness and accuracy of the frequency safety risk evaluation.

[0077] The adaptive ability of the day-ahead plan to the change of the operation mode is improved: the dynamic safety boundary established by data driving can adapt to the fluctuations of new energy output and load, so that the safety standard is closely related to the real-time operation state of the power grid, and the problem of poor adaptability of the static boundary is effectively solved, while ensuring safety and avoiding economic loss caused by excessive conservatism.

[0078] Intelligentization and optimization of safety correction are realized: through the strategy matching engine and the correction strategy library, the correction strategy can be intelligently matched and generated according to the multi-dimensional evaluation results; and through the optimization model embedded with the frequency safety constraint, the economic optimal generation plan is automatically solved under the hard requirement of ensuring the stability of system frequency, thereby realizing the best balance between safety and economy.

[0079] A coordinated defense system of day-ahead plan and real-time control is constructed: the safety state of the corrected plan is quantified into fine margin levels, and the hierarchical and classified real-time defense strategy package is generated accordingly, thereby realizing the close coordination between the pre-planning and the in-process control. This enables the real-time control system to quickly and accurately start the corresponding defense measures according to the risk level predicted by the day-ahead plan, thereby greatly improving the defense speed and effectiveness of the power grid in response to large power shortage faults, and forming a frequency safety active defense capability throughout the whole process of dispatching and operation. BRIEF DESCRIPTION OF DRAWINGS

[0080] Figure 1 A flowchart of a method for guaranteeing frequency safety in day-ahead planning of a DC sending-end power grid;

[0081] Figure 2 A flowchart of the method S100 for guaranteeing frequency safety in day-ahead planning of a DC sending-end power grid;

[0082] Figure 3 A flowchart of the method S200 for guaranteeing frequency safety in day-ahead planning of a DC sending-end power grid;

[0083] Figure 4 A flow chart of a method S300 for guaranteeing frequency security in day-ahead scheduling of a DC sending-end power grid;

[0084] Figure 5 A flow chart of a method S400 for guaranteeing frequency security in day-ahead scheduling of a DC sending-end power grid. DETAILED DESCRIPTION

[0085] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, circuits, and

[0086] In order to make the purpose, technical scheme and advantages of the present application clearer, the following further describes the present application in combination with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0087] The core logic of the method is to build a frequency security active defense system that runs through the whole chain of dynamic boundary establishment, multi-dimensional parallel evaluation, intelligent correction optimization, and security quantification and defense coordination. The system establishes accurate dynamic security boundaries through data-driven methods, performs multi-dimensional and parallel frequency security physical examination on the day-ahead plan based on the boundaries, intelligently generates correction schemes with safety and economy based on the physical examination results, quantifies the security state of the corrected plan as a navigation chart for real-time defense, and thus realizes a fundamental change from static passive correction to dynamic active defense.

[0088] Referring to Figure 1 which shows a method for guaranteeing frequency security in day-ahead scheduling of a DC sending-end power grid according to an embodiment of the present application, including the following steps:

[0089] A method for guaranteeing frequency security in day-ahead scheduling of a DC sending-end power grid, including the following steps:

[0090] S100. Frequency security feature analysis and dynamic boundary establishment, based on historical operation data and simulation results of the power grid, analyzing key factors affecting frequency stability, and establishing dynamic security boundaries of system rotational inertia demand, power shortage tolerance capacity, and frequency regulation demand;

[0091] S200. Multi-dimensional frequency security parallel assessment, based on dynamic security boundary, a parallel architecture containing three assessment channels of inertia adequacy, power impact risk and regulation rapidity is constructed, the security of day-ahead plan in different dimensions is independently analyzed, and the security state assessment results of each dimension are output;

[0092] S300. Intelligent correction strategy generation and plan optimization, based on the security assessment results, the correction strategy is intelligently selected through the strategy matching engine, and the day-ahead plan is corrected by using the optimization algorithm;

[0093] S400. Security margin quantification and real-time defense strategy construction, the security margin of the corrected plan is quantitatively analyzed, the real-time defense strategy package is generated based on the analysis results, and the cooperative defense system of day-ahead plan and real-time control is established.

[0094] Please refer to Figure 2 , which shows a flowchart of a method S100 for guaranteeing frequency security in a day-ahead plan of a direct current sending terminal power grid, which aims to lay a precise quantitative and perceptual foundation for the entire method. By systematically collecting and processing historical and real-time data of the power grid, the complex physical characteristics (inertia, tolerance, and regulation capacity) affecting frequency stability are converted into a series of calculable and predictable dynamic mathematical models (security boundaries). This is equivalent to establishing dynamic standards for frequency health for the system, so that subsequent assessments no longer rely on fixed and conservative empirical values, but are closely related to the real-time operating state of the power grid. The specific steps include:

[0095] S110. Data acquisition and feature extraction, establish a complete power grid operation data acquisition system, collect load data, new energy power generation, unit output, unit operation state, and direct current transmission power key operation parameters in real time through energy management system (EMS) and wide area measurement system (WAMS); construct a standardized frequency security feature database, perform quality checking, normalization, and abnormal data cleaning on the collected data, eliminate invalid data due to communication interruption or measurement failure, and ensure the integrity and accuracy of the data;

[0096] S120. Rotational inertia safety boundary modeling, based on the historical data of the power grid, a multi-element nonlinear regression method is used to establish a rotational inertia safety boundary model, and the minimum inertia required to maintain frequency stability under different load levels and different new energy penetration rates is analyzed; the model comprehensively considers the influence of operating mode changes on inertia demand, and establishes a quantitative relationship between inertia demand and operating state; the formula of the rotational inertia safety boundary model is:

[0097] ;

[0098] represents Minimum allowed inertia of the system in the time period; For Total load in the time period; For New energy output in the time period; Function is the regression coefficient, which is determined by fitting the historical data by the least square method; is the model error term;

[0099] S130. Power shortage tolerance boundary establishment, through time domain simulation analysis of the power system (such as using PSS / E, DigSILENT, etc.), establish the power shortage tolerance capability evaluation model, simulate the frequency dynamic process when single or multi-loop DC blocking fault occurs under different operation modes, determine the maximum allowed power shortage to ensure the stability of system frequency; establish the mapping relationship between power shortage tolerance capability and system net load:

[0100] ;

[0101] Indicates Maximum allowed power shortage of the system in the time period; For Actual equivalent inertia of the system in the time period; is the coefficient determined by fitting the simulation data; is the model error term;

[0102] S140. Frequency regulation capability boundary determination, analyze the characteristics of system frequency regulation resources, comprehensively consider the regulation performance difference of different types of units, establish the frequency regulation capability demand model; determine the minimum reserve capacity and minimum regulation rate requirement required by the system to respond to power shortage; Frequency regulation capability demand model formula:

[0103] ;

[0104] ;

[0105] For Minimum reserve capacity required by the system in the time period; For Minimum reserve capacity required by the system in the time period; is the reserve capacity safety factor; is the regulation rate coefficient; is the target time for system frequency to recover to a safe level.

[0106] Please refer to Figure 3, which shows a method S200 for ensuring frequency security in a day-ahead scheduling of a DC sending-end power grid according to an example of the present application, aiming at conducting a comprehensive, efficient and in-depth frequency security scanning on the day-ahead generation scheduling. By constructing three independent and parallel evaluation channels, the weak points of the scheduling in the three key dimensions of inertia defense capability, impact resistance capability and rapid recovery capability are diagnosed. Only the evaluation efficiency is improved, more importantly, the one-sidedness of single index evaluation is avoided, ensuring the comprehensiveness and accuracy of risk identification, and providing clear targeting targets for subsequent accurate correction; the specific steps include:

[0107] S210. Inertia adequacy evaluation channel, based on the unit combination scheme in the day-ahead generation scheduling, the equivalent inertia of the system in each period is calculated according to the inertia time constant, rated capacity and operating state of each unit; the equivalent inertia The calculation formula is:

[0108] ;

[0109] is the inertia time constant of the unit ; is the rated capacity of the unit ; is the operating state of the unit in the period; is the total number of units;

[0110] The calculated equivalent inertia is compared with the minimum allowable inertia of the system in the period, and a warning threshold is set according to the relative difference, the evaluation result is output, and the judgment logic is:

[0111] ;

[0112] is the inertia adequacy warning threshold;

[0113] S220. Power impact risk evaluation channel, identify the maximum DC transmission power in each period in the day-ahead scheduling, evaluate the power impact risk combined with the system operating state, consider the influence of multiple DCs running simultaneously, use probability analysis method to quantify the risk level, and define the power impact risk index :

[0114] ;

[0115] is the total DC transmission power in the period; is the back-to-back dc The equivalent failure rate of the time period can be determined according to historical failure statistics and operating environment; The maximum DC transmission power in the day-ahead plan for each time period;

[0116] The power impact risk index Compare the power impact risk index with the preset threshold value, and output the risk level evaluation result;

[0117] S230. Adjust the rapidity evaluation channel, analyze the reserve capacity configuration in the day-ahead plan and the equivalent regulation rate of the system, and build a comprehensive evaluation index The comprehensive evaluation index is the short board function of the reserve adequacy and the regulation rate, and the formula is:

[0118] ;

[0119] The reserve capacity configured in the day-ahead plan; The equivalent regulation rate of the system;

[0120] Compare the comprehensive evaluation index with the grade threshold value, and output the evaluation result:

[0121] ;

[0122] S240. Comprehensive evaluation result generation, integrate the three evaluation results, establish a unified safety state evaluation system, and generate a safety state map with time as the horizontal axis and risk dimension as the vertical axis.

[0123] Please refer to Figure 4 , which shows a method S300 for ensuring frequency safety in the day-ahead plan arrangement of a DC sending end power grid according to the present application, which aims to: convert the safety risks found in the evaluation into specific, executable and economically optimal correction actions; through the strategy matching engine, intelligently diagnose the risk source and recommend the best treatment scheme (correction strategy), and then through the safety correction optimization model, under the premise that all hard requirements for frequency safety are met, taking into account the economy, solve the optimal generation plan. Finally, through correction effect verification to ensure that the treatment scheme is foolproof, forming an optimization process from diagnosis to prescription to review. The specific steps include:

[0124] S310. The strategy matching engine is designed to establish a rule-based strategy matching engine, the input of which is the multi-dimensional safety evaluation results generated in step S200, i.e. the evaluation conclusions of the three channels of inertia adequacy, power shock risk and regulation rapidity; the engine internally pre-stores a complete strategy selection logic rule library, which can automatically and quickly match and recommend the optimal correction strategy sequence according to different risk combination modes. For example, when the system appears an inertia danger alarm, the engine will preferentially recommend an inertia promotion strategy; when the system simultaneously faces high power shock risk and insufficient regulation capacity, the engine will recommend a coordinated strategy of power shock mitigation and regulation capacity enhancement.

[0125] In addition, strict strategy priority rules are defined in the engine to ensure that when multiple risks occur simultaneously, the hazards to frequency stability can be ordered and managed according to the risk, achieving optimal recovery of system safety.

[0126] S320. Correction strategy library construction, a structured and extensible correction strategy library is constructed to provide specific action plans for the strategy matching engine, and the strategy library core includes three types of strategies:

[0127] Inertia promotion strategy: the core execution measure of this strategy is to adjust the unit combination, specifically, during the period when the safety evaluation is dangerous or warning, the dispatching algorithm will start or increase the output of high-inertia units and correspondingly reduce or shut down units with small inertia contribution under the premise of meeting other constraints; the control target is to promote the system equivalent inertia to above the safety boundary and retain a certain margin;

[0128] Power shock mitigation strategy: the core execution measure of this strategy is to optimize the direct current power plan, specifically, during the period when the power shock risk is high, by cooperating with the direct current dispatching center, the transmission power of the key direct current loop is actively and moderately reduced, or the power curve is adjusted to make it change more gently, thereby reducing the maximum power shortage that may be caused by a single fault; the control parameter is the direct current power safety limit value calculated based on the system tolerance capacity.

[0129] Regulation capacity enhancement strategy: the core execution measure of this strategy is to optimize the distribution and configuration of standby capacity, specifically, the standby tasks of each unit are redistributed, and the proportion of standby capacity assumed by units with fast response capability is increased to ensure that there is sufficient standby capacity and fast power recovery speed after the disturbance occurs.

[0130] S330. An optimization model considering safety and economy is established, the model introduces the frequency safety constraint derived from the dynamic safety boundary on the basis of the traditional unit combination and economic dispatching model, and converts the frequency stability requirement into a hard constraint in the optimization problem.

[0131] The optimization model aims to minimize the total operation cost of the system, including the active power generation cost of all on-line units, the start-up cost and shut-down cost of the units, and the power impulse risk index in the form of cost in the objective function, to achieve the trade-off between safety and economy; the mathematical expression is:

[0132] ;

[0133] To ensure that the corrected plan meets the frequency safety requirements, the model introduces the following key constraints:

[0134] ;

[0135] The constraints ensure that the system has sufficient rotational inertia, the DC power impulse is within the acceptable range, the reserve capacity meets the requirements, and the system has sufficient frequency regulation rate at any time;

[0136] The optimization model is a typical mixed integer programming problem, which is solved by using a mature commercial mathematical optimization solver, and finally outputs a day-ahead generation plan that meets all frequency safety constraints and is as economical as possible;

[0137] S340. Verification of the effect of correction, the day-ahead generation plan generated after optimization and correction in S330, re-executes the multi-dimensional frequency safety parallel evaluation process in S200, compares the safety state of the system in each evaluation channel before and after correction, and calculates quantitative indicators such as safety constraint satisfaction rate, total cost change rate (total cost change caused by correction) and risk index reduction rate, to comprehensively and objectively evaluate the safety improvement effect and economic impact of the correction strategy adopted;

[0138] Only the generation plan that has passed the verification and proved to have safety and economy up to the expected standard will be finally adopted and applied to actual dispatch.

[0139] Please refer to Figure 5 , which shows a flowchart of a method S400 for ensuring frequency safety in day-ahead planning of a DC sending-end power grid, aiming to: break through the last mile between day-ahead planning and real-time control, and build a coordinated defense system. First, the final determined day-ahead plan is subjected to safety margin quantitative analysis, like drawing a defense map with different risk levels for the system. Based on this map, hierarchical and classified real-time defense strategies are designed in advance, so that the real-time control system can quickly and accurately start the corresponding level of defense measures (from pre-warning standby to emergency control) according to the risk level of the current operating state. Finally, through strategy package verification, it is ensured that the entire defense system is effective and reliable. The specific steps include:

[0140] S410. Security margin quantitative analysis, the security level of the system is accurately quantified by calculating the security margin of each period in the three dimensions of inertia adequacy, power impact tolerance and regulation speed; Specifically:

[0141] Inertia margin reflects the relative proportion of the actual inertia of the system exceeding the minimum demand;

[0142] Power impact tolerance margin reflects the degree of surplus of the maximum allowable power shortage of the system relative to the maximum potential power impact;

[0143] The regulation capability margin reflects the excess level of the actual regulation capability of the system relative to the minimum requirement;

[0144] According to the margin value, the security state of each period of the system is finely divided into four levels of loose, normal, tense and critical, so as to accurately identify the weak links and key risk periods in the system.

[0145] S420. Real-time defense strategy design, based on the security margin quantitative results, design hierarchical and classified real-time defense strategy, and closely link with the risk level prediction results in the day-ahead plan, the defense system starts different defense measures according to different margin levels:

[0146] When the system is in a loose or normal level, execute the routine operation monitoring, and the defense system is in a low alert standby state;

[0147] When the system enters the tense level, the preparatory control strategy is automatically activated. Such strategy does not immediately affect the power grid operation, but is ready to execute at any time, such as: pre-setting and verifying the load shedding scheme for a specific area, enabling the DC power modulation function to be in standby state, instructing the fast start unit to enter the hot standby, etc.

[0148] When the system is considered to be in a critical level, the emergency control strategy is prepared to be started. Once the real-time monitoring system detects the occurrence of a preset large power shortage fault, the system will automatically execute emergency control measures such as precise load shedding and DC power speed reduction to prevent frequency collapse.

[0149] In this way, a real-time defense mechanism of monitoring-evaluation-decision-execution integration seamlessly connected with the day-ahead plan is established;

[0150] S430. Strategy package verification and deployment, to ensure the reliability and effectiveness of the designed real-time defense strategy:

[0151] First, the defense strategy package is strictly tested in a closed loop by a digital simulation platform (such as RTDS, ADPSS), various possible expected accident scenarios are simulated, and the correctness and effectiveness of the strategy action and the coordination between different control measures are verified; according to the simulation results, the parameters of the strategy are finely optimized.

[0152] Subsequently, the fully verified strategy package and its trigger logic are formally deployed to the production control system, and it is ensured that it can be called and updated online and rolling according to the actual day-ahead plan and real-time operation mode.

[0153] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for frequency security in day-ahead scheduling of a direct current sending-end power grid, characterized in that, The method comprises the following steps: Based on the historical operation data and simulation of the power grid, the key factors affecting frequency stability are analyzed, and a dynamic security boundary is established; Based on the dynamic security boundary, a parallel architecture for evaluating the channel is constructed, and the safety of the day-ahead plan in different dimensions is independently analyzed for each channel, and the safety state evaluation results of each dimension are output; The safety state evaluation results are intelligently selected for correction strategies through a strategy matching engine, and an optimization algorithm is used to correct the safety of the day-ahead plan; The safety of the corrected plan is quantitatively analyzed, and a real-time defense strategy package is generated based on the analysis results, and a coordinated defense system of day-ahead plan and real-time control is established.

2. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 1, characterized in that, The establishment of the dynamic security boundary comprises: Data acquisition and feature extraction, a complete power grid operation data acquisition system is established, the load data, new energy power generation, unit output, unit operation state, DC power transmission key operation parameters are collected in real time through energy management system EMS and wide area measurement system WAMS; a standardized frequency safety feature database is constructed, the collected data is quality checked, normalized and abnormal data is cleaned, invalid data caused by communication interruption or measurement failure is eliminated, and the integrity and accuracy of the data are ensured; Rotational inertia safety boundary modeling, based on the historical operation data of the power grid, a rotational inertia safety boundary model is established by using a multivariate nonlinear regression method, the minimum inertia requirement for maintaining frequency stability under different load levels and different new energy penetration rates is analyzed; Power shortage tolerance boundary establishment, a power shortage tolerance capacity evaluation model is established through power system time domain simulation analysis; Frequency regulation capability boundary determination, the characteristics of frequency regulation resources are analyzed, and the differences in regulation performance of different types of units are considered, and a frequency regulation capability demand model is established; The formula of the frequency regulation capability demand model is: ； ； To Minimum reserve capacity required by the time period system; To Minimum reserve capacity required by the time period system; To reserve capacity safety factor; To adjust the rate factor; To target time for system frequency recovery to a safe level.

3. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 2, characterized in that, The formula of the rotational inertia safety boundary model is: ； denotes minimum allowed inertia of the time period system; is total load of the time period; is new energy output of the time period; function is a regression coefficient, determined by fitting historical data by least squares method; is a model error term; The formula of the power shortage tolerance capacity evaluation model is: ； denotes time period system maximum allowed power deficit; is time period system actual equivalent inertia; is a coefficient determined by fitting simulation data; is a model error term.

4. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 1, characterized in that, The step of constructing the parallel architecture for evaluating the channel, independently analyzing the safety of the day-ahead plan in different dimensions for each channel, and outputting the safety state evaluation results of each dimension comprises: The inertia adequacy evaluation channel calculates the equivalent inertia of the system in each period based on the unit combination scheme in the day-ahead generation plan, according to the inertia time constant, rated capacity and operating state of each unit; the equivalent inertia The calculation formula is: ； inertia time constant of the machine set; rated capacity of the machine set; operating state of the machine set in time period; total number of machine sets;​​​ The calculated equivalent inertia With The minimum allowable inertia of the time period system The comparison is made, and a warning threshold is set according to the relative difference value The evaluation result is output, and the judgment logic is: ； an alert threshold for inertia sufficiency; The power impact risk assessment channel identifies the maximum DC transmission power of each period in the day-ahead plan, evaluates the power impact risk in combination with the system operation state, considers the influence of simultaneous operation of multiple DCs, quantifies the risk level by using a probability analysis method, and defines a power impact risk index : ； For The total DC transmission power in the time period; For the The DC transmission power in the time period; The equivalent failure rate in the time period can be determined according to historical failure statistics and operating environment; The maximum DC transmission power in the time period in the day-ahead plan; Power surge risk indicators The risk level assessment result is output by comparing it with the preset threshold of the power impact risk index. The adjustment rapidity evaluation channel analyzes the spare capacity configuration in the day-ahead plan and the equivalent adjustment rate of the system, and constructs a comprehensive evaluation index The comprehensive evaluation index is a short board function of the spare adequacy and the adjustment rate, and is expressed by a formula. ； reserve capacity configured in the day-ahead schedule; system equivalent regulation rate; Comparing the comprehensive evaluation index with the grade threshold value to output the evaluation results: ； Comprehensive evaluation result generation, integrating the three evaluation results, establishing a unified safety state evaluation system, and generating a safety state atlas with time as the horizontal axis and risk dimension as the vertical axis.

5. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 1, characterized in that, The step of using an optimization algorithm to correct the safety of the day-ahead plan comprises: Strategy matching engine design, a rule-based strategy matching engine is established, the input of which is the multi-dimensional safety evaluation results, i.e. the evaluation conclusions of the inertia adequacy, power impact risk and regulation rapidity three channels; a complete strategy selection logic rule library is pre-stored in the engine, which can automatically and quickly match and recommend the optimal correction strategy sequence according to different risk combination modes; Correction strategy library construction, a structured and extensible correction strategy library is constructed to provide specific action plans for the strategy matching engine; The safety correction optimization model is an optimization model considering safety and economy comprehensively. The model introduces the frequency safety constraint derived from the dynamic safety boundary on the basis of the traditional unit commitment and economic dispatch model, and converts the frequency stability requirement into a hard constraint in the optimization problem. The optimization model takes the minimization of the total system operation cost as the objective, and the cost includes the active power generation cost of all online units, the start-up cost and shutdown cost of units. Meanwhile, the power impact risk index is also included in the objective function in the form of cost to realize the trade-off between safety and economy. The mathematical expression of the objective function is: ； To ensure that the corrected plan meets the frequency safety requirement, the model introduces the following key constraints: ； The constraints ensure that the system has sufficient rotational inertia, the DC power impact is within the tolerable range, the reserve capacity meets the requirements, and the system has sufficient frequency regulation rate at any time. The correction effect verification re-executes the multi-dimensional frequency safety parallel evaluation process for the day-ahead generation plan generated after optimization correction. By comparing the safety status of the system in each evaluation channel before and after correction, and calculating quantitative indicators such as safety constraint satisfaction rate, total cost change rate, and risk index reduction rate, the safety improvement effect and economic impact of the correction strategy are comprehensively and objectively evaluated. Only the generation plan that has passed the verification and proved to meet the expected standards of safety and economy will be finally adopted and applied to actual dispatch.

6. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 5, characterized in that, The strategy matching engine also defines strict strategy priority rules to ensure that when multiple risks occur simultaneously, the system safety can be restored in an orderly manner according to the degree of risk to frequency stability, achieving the optimization of system safety.

7. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 5, characterized in that, The correction strategy library core includes inertia enhancement strategy, power impact mitigation strategy, and regulation capacity enhancement strategy. The core execution measure of the inertia enhancement strategy is to adjust the unit commitment. In the time period where the safety evaluation is dangerous or warning, the dispatch algorithm will start or increase the output of high-inertia units and reduce or shut down the units with small inertia contribution under the premise of meeting other constraints. The core execution measure of the power impact mitigation strategy is to optimize the DC power plan. In the time period with high power impact risk, by coordinating with the DC dispatch center, the key DC loop power is actively and moderately reduced, or the power curve is adjusted to make it change more smoothly, reducing the maximum power shortage that may be caused by a single fault. The control parameter is the DC power safety limit calculated based on the system tolerance capacity. The core execution measure of the regulation capacity enhancement strategy is to optimize the distribution and configuration of reserve capacity.

8. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 1, characterized in that, The steps of establishing the coordinated defense system of day-ahead plan and real-time control include: Safety margin quantitative analysis accurately quantifies the safety level of the system by calculating the safety margin in three dimensions of inertia adequacy, power impact tolerance, and regulation rapidity. The inertia margin reflects the relative proportion of the actual inertia of the system exceeding the minimum demand. The power impact tolerance margin reflects the surplus degree of the maximum allowable power shortage relative to the maximum potential power impact. The regulation capability margin reflects the excess level of the actual regulation capability of the system relative to the minimum requirement; According to the margin value, the safety state of each period of the system is finely divided into four levels of relaxed, normal, tense and critical, so as to accurately identify the weak links and key risk periods in the system; Real-time defense strategy design, based on the safety margin quantification result, designs hierarchical and classified real-time defense strategy, and closely links with the risk level prediction result in the day-ahead plan, and the defense system starts differential defense measures according to different margin levels; Strategy package verification and deployment, to ensure the reliability and effectiveness of the designed real-time defense strategy.

9. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 8, characterized in that, The differential defense measures include: When the system is in the relaxed or normal level, the routine operation monitoring is performed, and the defense system is in the low alert standby state; When the system enters the tense level, the preparatory control strategy is automatically activated, and the strategy does not immediately affect the power grid operation, but is ready to execute at any time; When the system is considered to be in the critical level, the emergency control strategy is prepared to be started; once the real-time monitoring system detects the preset large power shortage fault, the system will automatically execute emergency control measures such as accurate load shedding, DC power speed reduction, etc., to prevent frequency collapse.

10. The method for frequency security in day-ahead scheduling of a DC sending-end power grid according to claim 9, characterized in that, The method for ensuring the reliability and effectiveness of the designed real-time defense strategy includes: Firstly, the defense strategy package is strictly tested in a closed loop through a digital simulation platform, various possible expected accident scenarios are simulated, the correctness and effectiveness of the strategy action and the coordination between different control measures are verified, and the parameters of the strategy are finely optimized according to the simulation result; Subsequently, the fully verified strategy package and its trigger logic are formally deployed to the production control system, and it is ensured that it can be called and updated online and rolling according to the actual day-ahead plan and real-time operation mode.

Images