Frequency emergency control method and system based on event unlocking-response driving
By adopting an event-based unlock-response driven frequency emergency control method, and through causal analysis and online decision-making, the method can quickly respond to the frequency change rate, thus solving the problem of rapid frequency drop in new energy power grids caused by traditional control methods, and improving frequency security and grid stability.
Patent Information
- Application Number
- CN202511014085.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-22
- Publication Date
- 2025-10-28
AI Technical Summary
Traditional emergency frequency control methods are ineffective in dealing with the rapid frequency drop caused by temporary power surges triggered by new energy fault ride-throughs. This is especially true in grids with a high proportion of new energy sources, where existing control modes that rely solely on event triggering are ill-equipped to handle uncertain power surges, leading to an increased risk of frequency instability.
An event-driven frequency emergency control method is adopted, which identifies the dominant fault event through causal offline analysis and formulates a control strategy by combining the frequency change rate information collected by the PMU. This method complements and enhances traditional frequency emergency control, including a causal offline analysis module, a dispatch center-side decision-making module, and a stability control plant-side strategy matching module. The control quantity is dynamically matched to cope with uncertain power impacts.
It improves frequency safety levels, prevents excessively rapid and deep frequency drops, reduces the risk of low-frequency load shedding and major power outages, and enhances the grid's ability to cope with temporary power surges.
Smart Images
Figure CN120855398A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power technology, and in particular to a frequency emergency control method and system based on event unlocking-response driving. Background Art
[0002] With the high proportion of new energy and other power electronic equipment being connected, a large number of synchronous generators are being replaced, reducing the strength of the AC system and its dynamic reactive power support capacity. The range of low voltage ride-through caused by new energy after grid short-circuit faults will further expand, and the recovery time will also become longer. In other words, the power surge caused by large-scale fault ride-through will further increase. In addition, due to the increased proportion of new energy, the system's rotational inertia and frequency regulation capability are declining, and the grid's ability to resist power disturbances is also decreasing. Recoverable faults such as multiple DC commutation failures and large-scale low voltage ride-through of new energy may cause a short-term sharp drop in grid frequency, or even trigger low-frequency load shedding. After the action, it may cause high-frequency problems. Traditional defense configurations have low adaptability to such new problems, and frequency stability, especially low-frequency problems, has become a new control challenge.
[0003] Existing frequency stability control measures after faults include emergency control to prevent frequency instability and corrective control to prevent frequency collapse. Frequency corrective control generally uses response triggering as a "safety net" measure, but it has disadvantages such as large control input and easy loss of residential load. Frequency emergency control, on the other hand, has the advantages of rapid action and relatively small control input. It generally uses event triggering and is the priority control measure after a high-power imbalance fault. However, since traditional frequency emergency control is often designed for deterministic and permanent high-power imbalance faults, it is insufficient to prevent rapid frequency drops caused by temporary power surges due to renewable energy fault ride-throughs. Major blackouts have shown that high-proportion renewable energy grids are prone to rapid frequency drops due to temporary high-power surges, a typical frequency instability scenario. The existing frequency emergency control mode, which relies solely on event triggering, is inadequate for this situation and urgently needs optimization and upgrading.
[0004] In view of the limitations of traditional emergency control schemes caused by rapid frequency drops under high-power surges in the context of inter-regional DC development and high-proportion renewable energy power systems, it is necessary to propose a frequency emergency control method and system based on event unlocking-response driving. Summary of the Invention
[0005] Purpose of the invention: The purpose of this invention is to provide a frequency emergency control method and system based on event unlocking-response driving.
[0006] Technical solution: The frequency emergency control method based on event unlocking-response driving according to the present invention includes the following steps:
[0007] (1) Offline causal analysis of events and responses: The causal quantitative analysis of the response information of rapid frequency drop and the fault events that may cause rapid frequency drop is carried out through time domain simulation. The dominant causal event that causes rapid frequency drop is obtained and the event is associated with the frequency drop response information.
[0008] (2) Online decision-making at the dispatch center: By distinguishing between permanent and temporary power surge events, different control strategies are formulated online;
[0009] (3) Stability control plant side strategy matching: By matching the frequency emergency control strategy to unlock the expected fault event, on the basis of the existing plant side stability control system, the frequency change rate response information collected by the PMU is added, and the corresponding control quantity is matched according to the strategy table to assist in matching the control strategy.
[0010] Furthermore, in step (1), inverse information entropy causal inference is used to quantitatively analyze the causal degree between multiple fault events and the system frequency response.
[0011] The frequency emergency control system based on event unlocking and response driving described in this invention includes a causal offline analysis module, a dispatch center-side decision-making module, and a plant-side strategy matching module for stability control. The dispatch center-side decision-making module shares future-state data generation and verification information with the traditional frequency emergency control decision-making module, and adopts different decision-making methods according to different operating conditions and fault types, ultimately issuing the decision to the plant-side stability control device for refresh execution. The plant-side strategy matching module for stability control, based on the existing plant-side stability control system, unlocks the frequency emergency control strategy for anticipated fault events and uses the frequency change rate information collected by the received PMU for strategy-assisted matching in rapid frequency drop scenarios.
[0012] Furthermore, during the online pre-decision process of the dispatch center side decision module, when the system supply and demand are relatively sufficient, if the faulty component is not equipped with a stability control information monitoring substation, or if the expected fault is an uncertain power imbalance fault, then based on the causal model of the fault event and frequency response information given by the offline pre-analysis, the dominant event causing the frequency drop is identified, and an event-response driven frequency emergency control strategy is formulated.
[0013] Furthermore, the decision-making module on the dispatch center side obtains the amount of unbalanced power the system is subjected to based on the system response information. In a high-proportion renewable energy grid, the amount of unbalanced power is considered in relation to the power output P of the faulty component. d Transient grid disconnection power P of new energy sources caused by fault disturbance c Power loss P during low-voltage operation of new energy sources LVRT Power lost during DC commutation failure P CF The corresponding event-response driven frequency emergency control power PE-RDFEC Represented as P E-RDFEC =±P d +P c -P ac
[0014] Among them, P d The sign before P is determined by the positive or negative sign of the power imbalance caused by the tripping of the faulty component; a power deficit is positive, and vice versa. ac This represents the maximum power deficit that the power grid can withstand.
[0015] Furthermore, the amount of unbalanced power borne by the system is expressed as:
[0016]
[0017] Among them, E MwS f is the total kinetic energy of the system. CoI f is the system's inertial center frequency; N This is the frequency rating.
[0018] Furthermore, the event-response driven frequency emergency control strategy includes the control power amount and its duration, which can be expressed as:
[0019]
[0020] Where: t MFD To control the frequency response trajectory below f min The moment; Δt E-RDFEC To control the duration of power application; when the frequency drops rapidly below the allowable normal operating value f. min In this scenario, the underlying mechanism includes: when the frequency drops to f min At that time, the power ΔP released by the system's rotational inertia m Primary frequency modulation power ΔP r The sum of the total power that hinders the frequency decrease is less than ΔP u Record the frequency trajectory below f min The duration is T MFD ,
[0021] The control power must be greater than the difference between the equivalent unbalanced power, the system inertia power, and the primary frequency modulation power, and the duration of action must be greater than the time during which the frequency response trajectory falls below the minimum allowable value for normal operation.
[0022] Furthermore, in the stability control plant-side strategy matching module, when the frequency change rate meets the rapid frequency drop scenario, for deterministic unbalanced power impacts and faulty component configurations of the stability control information monitoring substation faults, the traditional frequency emergency control strategy is selected; for uncertain unbalanced power impacts, it is necessary to combine the fault information given by the PMU to determine whether it is a temporary power impact, and match the corresponding control quantity from the event-response driven frequency emergency control strategy table according to the magnitude of the frequency change rate.
[0023] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: The present invention can improve the frequency emergency control effect in the scenario of rapid frequency drop caused by high power impact. By unlocking the frequency emergency control strategy through key fault events and matching the corresponding control quantity of the action with frequency dynamic response information, the unbalanced power can be rapidly reduced in the early stage of frequency drop. This completes the supplementary emergency control with "event" unlocking first and "response" information matching later, preventing the frequency drop from being too fast or too deep, and improving the frequency safety level. Attached Figure Description
[0024] Figure 1 This is a block diagram illustrating the principle of the frequency emergency control method based on event unlocking and response driven by the present invention.
[0025] Figure 2 This is a real-world power grid fault evolution process of the frequency emergency control method based on event unlocking-response driven by the present invention;
[0026] Figure 3 This is a diagram of the frequency emergency control system based on event unlocking-response driven by the present invention;
[0027] Figure 4 This is a flowchart of the online pre-decision process for frequency emergency control based on event unlocking-response driven, as described in this invention.
[0028] Figure 5 This is a flowchart of the frequency emergency control strategy matching process based on event unlocking-response driven by the present invention. Detailed Implementation
[0029] The technical solution of the present invention will be further described below with reference to the accompanying drawings.
[0030] like Figure 1 As shown, traditional frequency emergency control is a feedforward control system. The frequency control strategy is triggered by a fault event and issued to the power system to offset the impact of power imbalance disturbances on the system frequency, thus maintaining the frequency at the set value. Based on traditional frequency emergency control, dynamic estimation of the disturbance based on the frequency response is added.
[0031] The frequency emergency control strategy can be expressed as:
[0032]
[0033] In theory, when the disturbance F(s) is known and the models of Gd(s) and G1(s) are sufficiently accurate, error-free frequency control can be achieved. However, in rapid frequency drop (RFD) scenarios, a series of successive faults may occur, and the event that truly causes the rapid frequency drop may be unclear. Compared to traditional large-frequency-difference disturbance events where the disturbance quantities are known, such as DC blocking or large unit tripping, the time-varying nature of renewable energy output and control nonlinearity in RFD scenarios make it difficult to accurately obtain the unbalanced power disturbance quantities causing the rapid frequency drop, making it difficult to formulate traditional emergency frequency control measures.
[0034] In RFD scenarios, rapid frequency drops are essentially frequency safety issues, requiring decisive action shortly after a fault occurs. One approach is to unlock emergency frequency control strategies through critical fault events. By matching the dynamic frequency response (e.g., rate of change, frequency deviation) with the corresponding control input, unbalanced power can be rapidly reduced in the early stages of a frequency drop, preventing excessively rapid or deep drops. Because this control method prioritizes "event" unlocking followed by "response" information matching, it is called Event-Response Driven Frequency Emergency Control (E-RDFEC) and is suitable for RFD scenarios. The fundamental difference between E-RDFEC and traditional emergency frequency control is that the events triggering traditional emergency control actions are deterministic and easily obtainable unbalanced faults (e.g., DC blockage, large power plant failures), while the unbalanced power caused by faults triggering E-RDFEC is uncertain and difficult to obtain directly (e.g., AC component short circuits causing widespread low-voltage ride-through in renewable energy sources). E-RDFEC requires estimation based on system response information and matching with appropriate control strategies.
[0035] Taking an AC short-circuit fault in a real UHVDC receiving-end power grid as an example, a causal analysis of the event and response is conducted, demonstrating that multiple fault events are causally related to the decline in system frequency response. This power grid feeds into several UHVDC lines and simultaneously connects to tens of millions of kilowatts of renewable energy generation, with DC plus renewable energy power accounting for more than 50% of the total load. After an AC short-circuit fault occurs at a weak node, a series of events will occur, including renewable energy entering the low-voltage range, DC commutation failure followed by blocking, and distributed renewable energy disconnection caused by voltage drops during power flow transfer. Simultaneously, this leads to a rapid decline in system frequency response, posing a risk of low-frequency load shedding. The accident evolution process is as follows: Figure 2As shown, multiple fault events are causally related to the drop in system frequency response. However, in engineering practice, it is neither possible nor necessary to implement emergency frequency control measures for all fault events. In fact, controlling the fault event with the strongest causal relationship to the frequency drop is a more cost-effective approach. In simple scenarios, the causal relationship between fault events and response information such as the rate of change of frequency drop and the minimum point can be qualitatively determined based on engineering experience. However, in complex scenarios, a quantitative analysis of the degree of causation is required.
[0036] Various causal inference methods exist for different problem scenarios, mainly falling into two categories: those based on conditional independence between variables and those based on data-driven causal inference. The former requires some prior knowledge and cannot solve the Markov equivalence chain problem, while the latter extends structural equation modeling to discover causal relationships from data distributions. Examples include linear non-Gaussian non-cyclic model algorithms, information causal inference algorithms, and their improved Inverse Information Entropy Causal Inference (RIECI) method. RIECI is a causal inference method based on the orthogonality of information between variables, capable of quantifying the strength of causality. This method requires no prior knowledge and can directly perform causal inference from data. It can be used for the quantitative analysis of causal factors causing a rapid decline in frequency under multiple fault events.
[0037] RIECI causality index between variables x and y as follows:
[0038]
[0039] Where: p(y Ns ), p(x Ns ) are the probability density functions of variables y and x after normalization and sorting, respectively; p(y) and p(x) are the probability density functions of the original variables y and x, respectively; S(·) is the differential entropy, which is a measure of the information content (complexity of change) of random variables.
[0040] Taking the probability density function p(x) as an example, its differential entropy S(p(x)) is:
[0041] S(p(x))=∫p(x)logp(x)dy
[0042] When the variable data is finite, S(p(x)) can be estimated using an entropy estimator, i.e.:
[0043]
[0044] Where ψ(·) is the double gamma function; N is the dimension of the data x.
[0045] After identifying the dominant event causing the frequency drop through causal analysis, corresponding E-RDFEC control strategies can be formulated to address the unbalanced power caused by the dominant event. For high-proportion renewable energy grids, the unbalanced power includes the power output P from the faulty component itself. d Transient grid disconnection power P of new energy sources caused by fault disturbance c Power loss P during low-voltage operation of new energy sources LVRT Power lost during DC commutation failure P CF , where P d 、P c All of these are permanent power surges, and their power values are deterministic, which can be directly obtained through measurement and control devices. The corresponding E-RDFEC frequency emergency control strategy is as follows:
[0046] P E-RDFEC =±P d +P c -P ac
[0047] Among them, P d The sign before P is determined by the positive or negative sign of the power imbalance caused by the tripping of the faulty component; a power deficit is positive, and vice versa. ac This represents the maximum power deficit that the power grid can withstand.
[0048] And P LVRT 、P CF As a temporary power surge, and due to its strong nonlinear characteristics, the magnitude of the unbalanced power surge is uncertain and cannot be reflected by simple measurement. However, the equivalent unbalanced power can be estimated by the trajectory of the system's initial transient frequency response, as shown in the following formula:
[0049]
[0050] Among them, E MWS f is the total kinetic energy of the system. COI f is the system's inertial center frequency; N This is the frequency rating.
[0051] The frequency dropped rapidly, falling below the allowable normal operating value f. min In this scenario, the essential mechanism lies in the fact that when the frequency drops to f min At that time, the power ΔP released by the system's rotational inertia m Primary frequency modulation power ΔP r The sum of the total power that hinders the frequency decrease is still less than ΔP. u Record the frequency trajectory below f min The duration is T MFD The control strategy of E-RDFEC is as follows:
[0052]
[0053] Where: t MFD To control the frequency response trajectory below f min The moment; Δt E-RDFEC To control the duration of power application, since the power surge is temporary, the control of E-RDFEC is also temporary.
[0054] The function of event-response driven frequency emergency control is to expand and consolidate the original frequency security defense system. It is to supplement and strengthen the original traditional online frequency emergency control decision-making without changing it. The purpose is to improve the power grid's ability to cope with new problems and challenges such as low-frequency load reduction caused by rapid frequency drops under short-term high-power disturbances, and reduce the risk of low-frequency load reduction and loss of residential load.
[0055] This solution is divided into two main parts: decision-making at the dispatch center and strategy matching at the stability control plant / station side. The functional architecture diagram is shown below. Figure 3 As shown, the overall approach adopts an "online pre-decision, real-time matching" technical route. Regarding online decision-making at the dispatch center, the E-RDFEC control decision module shares future-state data generation and verification information with the traditional frequency emergency control decision module, and adopts different decision-making methods based on different operating conditions and fault types, ultimately issuing the decision to the plant-side stability control device for refresh and execution. Regarding plant-side strategy matching, existing frequency emergency control systems are designed for deterministic power imbalance faults and require pre-configuration of corresponding stability control information monitoring substations on faulty components. This results in poor adaptability to uncertain power imbalance faults and faults in components without configured information monitoring substations. With the widespread application of wide-area measurement systems (WAMS) based on synchronous phasor measurement units (PMUs), system fault information and frequency-related electrical quantity information can be collected in near real-time. Therefore, this scheme, based on the existing plant-side stability control system, receives df data collected by the PMU. COI Information such as / dt is used for policy-assisted matching in RFD scenarios.
[0056] Online pre-decision flowchart as follows Figure 4As shown, the E-RDFEC and traditional online frequency emergency control decision-making processes are represented in blue and green, respectively. The online dispatching technical support system generates a typical operating mode for the next decision cycle based on load forecasting and renewable energy output forecasting, and determines the sufficiency of power supply and demand for this operating mode. If the system supply and demand are relatively sufficient, it enters the E-RDFEC and traditional online frequency emergency control decision-making stages. When the anticipated fault is a deterministic power imbalance fault and the faulty component has been configured with a stability control information monitoring substation, meeting the conditions for traditional stability control actions, time-domain simulation and strategy formulation are performed according to the traditional online frequency emergency control decision-making process. When the faulty component has not been configured with a stability control information monitoring substation, or the anticipated fault is an uncertain power imbalance fault, the dominant event causing the frequency drop is identified based on the causal model of the fault event and frequency response information provided in the offline pre-analysis, and an E-RDFEC control strategy is formulated.
[0057] The process flow diagram for matching strategies on the plant side is as follows: Figure 5 As shown, after receiving the E-RDFEC and traditional frequency emergency control strategies issued by the dispatch and measurement decision system, the stability control system makes a strategy matching judgment by receiving information such as frequency change rate and fault events collected by the PMU and stability monitoring substation in real time.
[0058] (1) When df COI When / dt<α, the system is considered to be in normal operation, with high supply and demand adequacy and no faults causing large power imbalance have occurred. α can be determined based on the historical maximum value of RoCoF under normal operation of the power grid in this region.
[0059] (2) When df COI When / dt>γ, it is considered that RoCoF has an information acquisition error and is ignored. γ can be determined based on the maximum value of RoCoF in the expected extreme scenario of the power grid in this area.
[0060] (4) When α≤df COI When / dt≤γ, it is considered that the current operation is in an RFD scenario, but the causes of rapid frequency drop can be divided into two categories: deterministic and uncertain unbalanced power surges. For the former, and the faulty component is configured in the stability monitoring substation, the traditional frequency emergency control strategy is selected; for the latter, it is necessary to combine the fault information given by the PMU to determine whether it is a temporary power surge, and then according to dt≤γ, the frequency drop is determined. COI The size of / dt is matched with the corresponding control quantity from the E-RDFEC policy table.
[0061] To ensure the stable reception and processing of information from the PMU, the main control station also needs to address issues such as signal protocol interoperability and hardware interface channel expansion.
[0062] In summary, this invention can improve the emergency frequency control effect in scenarios of rapid frequency drop caused by high-power impacts. By unlocking the emergency frequency control strategy through key fault events and matching the corresponding control quantities of the action with frequency dynamic response information, it can quickly reduce unbalanced power in the early stage of frequency drop, completing supplementary emergency control with "event" unlocking first and "response" information matching later, preventing the frequency drop from being too fast or too deep, and improving the frequency safety level.
Claims
1. A frequency emergency control method based on event unlocking-response driven, characterized in that, Includes the following steps: (1) Offline causal analysis of events and responses: The causal quantitative analysis of the response information of rapid frequency drop and the fault events that may cause rapid frequency drop is carried out through time domain simulation. The dominant causal event that causes rapid frequency drop is obtained and the event is associated with the frequency drop response information. (2) Online decision-making at the dispatch center: By distinguishing between permanent and temporary power surge events, different control strategies are formulated online; (3) Stability control plant side strategy matching: By matching the frequency emergency control strategy to unlock the expected fault event, on the basis of the existing plant side stability control system, the frequency change rate response information collected by the PMU is added, and the corresponding control quantity is matched according to the strategy table to assist in matching the control strategy.
2. The frequency emergency control method based on event unlocking-response driven according to claim 1, characterized in that, In step (1), inverse information entropy causal inference is used to quantify the causal degree between multiple fault events and the system frequency response.
3. A frequency emergency control system based on event-unlocked-response driven operation, characterized in that, It includes a causal offline analysis module, a dispatch center-side decision-making module, and a stability control plant-side strategy matching module. The dispatch center-side decision-making module shares future-state data generation and verification information with the traditional frequency emergency control decision-making module, and adopts different decision-making methods according to different operating conditions and fault types, and finally issues the decision to the plant-side stability control device for refresh execution. The stability control plant-side strategy matching module, based on the existing plant-side stability control system, unlocks the frequency emergency control strategy by anticipating fault events, and uses the frequency change rate information collected by the received PMU for strategy-assisted matching in rapid frequency drop scenarios.
4. The frequency emergency control system based on event unlocking-response driving according to claim 3, characterized in that, During the online pre-decision process of the dispatch center side decision module, when the system supply and demand are relatively sufficient, if the faulty component is not equipped with a stability control information monitoring substation, or if the expected fault is an uncertain power imbalance fault, then based on the causal model of the fault event and frequency response information given by the offline pre-analysis, the dominant event causing the frequency drop is identified, and an event-response driven frequency emergency control strategy is formulated.
5. The frequency emergency control system based on event unlocking-response driving according to claim 3, characterized in that, The decision-making module at the dispatch center side obtains the amount of unbalanced power the system is experiencing based on the system response information. In a high-proportion renewable energy grid, the amount of unbalanced power is considered in relation to the power output P of the faulty component. d Transient grid disconnection power P of new energy sources caused by fault disturbance c Power loss P during low-voltage operation of new energy sources LVRT Power lost during DC commutation failure P CF The corresponding event-response driven frequency emergency control power P E-RDFEC Expressed as P E-RDFEC =±P d +P c -P ac Among them, P d The sign before P is determined by the positive or negative sign of the power imbalance caused by the tripping of the faulty component; a power deficit is positive, and vice versa. ac This represents the maximum power deficit that the power grid can withstand.
6. The frequency emergency control system based on event unlocking-response driving according to claim 5, characterized in that, The amount of unbalanced power borne by the system is expressed as: Among them, E MWS f is the total kinetic energy of the system. COI f is the system's inertial center frequency; N This is the frequency rating.
7. The frequency emergency control system based on event unlocking-response driven according to claim 4, characterized in that, The event-response driven frequency emergency control strategy includes the control power amount and its duration, which can be expressed as: Where: t MFD To control the frequency response trajectory below f min The moment; Δt E-RDFEC To control the duration of power application; when the frequency drops rapidly below the allowable normal operating value f. min In this scenario, the underlying mechanism includes: when the frequency drops to f min At that time, the power ΔP released by the system's rotational inertia m Primary frequency modulation power ΔP r The sum of the total power that hinders the frequency decrease is less than ΔP u Record the frequency trajectory below f min The duration is T MFD , The control power must be greater than the difference between the equivalent unbalanced power, the system inertia power, and the primary frequency modulation power, and the duration of action must be greater than the time during which the frequency response trajectory falls below the minimum allowable value for normal operation.
8. The frequency emergency control system based on event unlocking-response driven according to claim 3, characterized in that, In the stability control plant-side strategy matching module, when the frequency change rate meets the rapid frequency drop scenario, for deterministic unbalanced power impacts and faulty component configurations of the stability control information monitoring substation faults, the traditional frequency emergency control strategy is selected; for uncertain unbalanced power impacts, it is necessary to combine the fault information given by the PMU to determine whether it is a temporary power impact, and match the corresponding control quantity from the event-response driven frequency emergency control strategy table according to the magnitude of the frequency change rate.