Power system load frequency control method based on intelligent weight distribution mechanism

Abstract

The application discloses a power system load frequency control method based on an intelligent weight distribution mechanism and belongs to the technical field of power system load frequency control, which comprises the following steps: step 1, a power system load frequency control model containing traditional generator units and electric vehicles is established; step 2, based on the power system load frequency control model, an adaptive control responsibility sharing strategy is designed; step 3, a double-channel collaborative controller fusing algorithm optimization sampling and electric vehicle real-time feedback is constructed, and a state space equation of the power system is derived; and step 4, based on the state space equation, the power system stability is analyzed by using a Lyapunov stability method. Through the reconstruction of the system model, the design of the dynamic weight distribution strategy and the state-aware intelligent sampling mechanism, the application realizes the collaborative frequency control between GUs and EVs, improves the frequency regulation accuracy, effectively reduces the communication frequency and control redundancy, and improves the real-time performance and stability of system regulation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system load frequency control, in particular to a power system load frequency control method based on an intelligent distribution weight mechanism, which is suitable for a power grid operation scenario with high new energy penetration rate and large-scale access of electric vehicles, and belongs to the technical field of frequency modulation resource collaborative optimization and intelligent control. BACKGROUND

[0002] With the continuous promotion of clean energy transformation, the penetration rate of new energy generation (such as solar power generation and wind power generation) in the power system continues to rise. However, due to the characteristics of new energy such as strong volatility, high uncontrollability, and lack of inertia support, the frequency fluctuation of the power grid increases significantly, which puts higher requirements on the traditional power system load frequency control (LFC) mechanism. At the same time, the traditional load frequency control strategy mainly relies on centralized scheduling and traditional generator units (GUs) to provide adjustment support, and its frequency modulation performance and control strategy are challenged under the condition of high proportion of new energy access.

[0003] In recent years, electric vehicles (EVs) are gradually becoming potential assets to support power grid operation. In addition to helping reduce carbon emissions through clean transportation, electric vehicles also act as distributed energy storage units to improve the flexibility of the power system. Its rapid response capability enables the power grid demand to be quickly adjusted. Through the use of bidirectional charging technology (such as vehicle-to-grid V2G), electric vehicles can dynamically exchange power with the power grid, effectively alleviating frequency fluctuations and balancing real-time supply and demand.

[0004] Although the potential of EVs in frequency regulation has attracted widespread attention, existing research mainly focuses on various aspects of its integration. In some research, an event-based frequency regulation framework is proposed, which considers communication efficiency and aggregator dynamics under actual travel demand patterns. In addition, a hybrid repetitive control and dead-zone control scheme is proposed, which can achieve fast and accurate current regulation of the on-board V2G inverter even in the case of frequency changes. In some research, a power grid model with high solar photovoltaic penetration is constructed, and through time series simulation, it shows how electric vehicle charging and discharging can help frequency regulation. On the contrary, in existing technology, a delay LFC system using EVs integration is studied, and through new Lyapunov-Krasovskii functions (LKFs) and a bisection search-based algorithm, a more relaxed stability is derived. However, existing research mostly adopts static sampling strategies, and pays insufficient attention to the adaptive response of EVs in real-time system changes, resulting in problems such as uncoordinated response of frequency modulation resources, large communication overhead, and low regulation efficiency in the LFC system under the background of large-scale access of EVs. SUMMARY

[0005] The purpose of the present application is to overcome the problems existing in the prior art, and provide a power system load frequency control method based on an intelligent weight distribution mechanism, which realizes collaborative frequency control between traditional generator units (GUs) and electric vehicles (EVs) by reconstructing a system model, designing a dynamic weight distribution strategy and a state-aware intelligent sampling mechanism, effectively reduces communication frequency and control redundancy while improving frequency regulation accuracy, and improves real-time performance and stability of system regulation.

[0006] The purpose of the present application is to overcome the problems existing in the prior art, and provide a power system load frequency control method based on an intelligent weight distribution mechanism, which realizes collaborative frequency control between traditional generator units (GUs) and electric vehicles (EVs) by reconstructing a system model, designing a dynamic weight distribution strategy and a state-aware intelligent sampling mechanism, effectively reduces communication frequency and control redundancy while improving frequency regulation accuracy, and improves real-time performance and stability of system regulation.

[0007] The purpose of the present application is to overcome the problems existing in the prior art, and provide a power system load frequency control method based on an intelligent weight distribution mechanism, which realizes collaborative frequency control between traditional generator units (GUs) and electric vehicles (EVs) by reconstructing a system model, designing a dynamic weight distribution strategy and a state-aware intelligent sampling mechanism, effectively reduces communication frequency and control redundancy while improving frequency regulation accuracy, and improves real-time performance and stability of system regulation. Figure 1 The purpose of the present application is to overcome the problems existing in the prior art, and provide a power system load frequency control method based on an intelligent weight distribution mechanism, which realizes collaborative frequency control between traditional generator units (GUs) and electric vehicles (EVs) by reconstructing a system model, designing a dynamic weight distribution strategy and a state-aware intelligent sampling mechanism, effectively reduces communication frequency and control redundancy while improving frequency regulation accuracy, and improves real-time performance and stability of system regulation.

[0008] Step 1, establishing a power system load frequency control model containing traditional generator units and electric vehicles;

[0009] Step 2, designing an adaptive control responsibility sharing strategy based on the power system load frequency control model;

[0010] Step 3, constructing a double-channel collaborative controller integrating algorithm optimization sampling and real-time feedback of electric vehicles, and deriving a state space equation of the power system;

[0011] Step 4, analyzing the stability of the power system using Lyapunov stability method based on the state space equation.

[0012] As a further improvement of the present application, the step 1 specifically comprises:

[0013] The following composite structure model is established:

[0014]

[0015] Wherein, X represents the state variable of the system; X represents the derivative of the state variable of the system at time t; Y represents the measured output of the system; U represents the control input signal at the traditional generator unit end; V represents the control input signal at the electric vehicle end; D represents the external disturbance; , , , and are parameter matrices.

[0016] As a further improvement of the present application, the step 2 specifically comprises:

[0017] To address the shortcomings of existing control strategies that rely on fixed adjustment ratios to effectively handle diverse dynamic characteristics, an adaptive control responsibility-sharing strategy based on system state is designed:

[0018]

[0019] Where t represents the current time and t+1 represents the next time. This represents the frequency regulation responsibility weight of traditional generator sets in the next moment. The weighting of the frequency modulation responsibility of electric vehicles in the next moment; , , and It is a given parameter; This represents the fifth state variable, namely Based on the proposed In the adaptive control responsibility-sharing strategy, the control task is flexibly allocated between GUs and EVs based on the real-time system state. When a significant deviation from equilibrium is detected, the GUs, with their powerful adjustment capabilities and high stability, will assume greater control responsibility. Conversely, when the system is close to equilibrium and fluctuations are small, the EVs, with their rapid response and flexibility, will take on more control tasks. This dynamic load-sharing mechanism not only improves response coordination but also ensures that all resources always operate within the optimal control range, thereby significantly enhancing the overall performance and robustness of frequency regulation.

[0020] As a further improvement of the present invention, step 3 specifically includes:

[0021] Given the significant mechanical inertia and slow dynamic response of traditional gate actuators (GUs), this invention further proposes a sampling-based control strategy to reduce the control update frequency during frequency regulation and effectively reduce communication resource consumption. This strategy avoids over-regulation of the GUs, allowing its control update rhythm to better align with its dynamic response characteristics, thereby improving overall resource utilization efficiency while achieving system frequency stability. The specific implementation formula of the control method is as follows:

[0022]

[0023] in It is the feedback gain matrix. This represents the k-th sampling time. These sampling times form a strictly increasing time series. . No. Next and first The interval between samples is defined as the sampling period. To ensure the stability and practical feasibility of the control, the sampling period... defined in the interval of two given parameters and , i.e. satisfying . Wherein, denotes the minimum allowable sampling period, denotes the maximum allowable sampling period.

[0024] To reduce the communication load and improve the system response efficiency, for the GUs unit with large inertia, a kind of intelligent sampling control strategy based on system state change is designed. The strategy dynamically adjusts the sampling interval according to the state evolution law, and realizes the adaptive triggering of control update. The specific method is as follows:

[0025] A performance function for measuring the current running state of the system is constructed , which takes the energy of the key state variable at the sampling time .

[0026]

[0027] Then, the change rate (gradient) of the performance function at the current time is calculated.

[0028] The gradient descent update strategy is adopted to adjust the next sampling period , and the update rule is as follows.

[0029]

[0030] Wherein, denotes the current sampling period, denotes the last sampling period, is the adjustment step size coefficient, indicates that the sampling period is limited in the given interval to ensure the stability of control.

[0031] Through the above mechanism, when the system state energy changes dramatically (such as the frequency deviation increases sharply), the sampling period is automatically shortened to better track the fast dynamics of the system; and when the system energy tends to be stable, the sampling period is automatically lengthened to reduce the calculation and communication overhead, realizing the dynamic balance of control accuracy and communication efficiency.

[0032] Considering that EVs have fast charging and discharging capability and high frequency response characteristics, in order to fully exert their flexibility, the present application proposes an EVs control strategy based on real-time state feedback. The strategy immediately adjusts the control signal when the system has frequency deviation, to realize fast and fine frequency regulation, and make up for the deficiency of the traditional GUs response lag.

[0033]

[0034] wherein, is the gain matrix corresponding to EVs, is the current system state. The control signal of EVs is continuously updated at a high frequency, independent of the sampling mechanism, ensuring the immediacy and continuity of the frequency response. This strategy can provide dynamic support under various working conditions, improving the rapid response capability and frequency stability of the system.

[0035] This strategy complements the intelligent sampling mechanism of the aforementioned GUs, achieving a synergistic control effect of complementary resources, thereby enhancing the adaptability and robustness of the entire power grid system.

[0036] Based on the above analysis, the following state space equation is derived.

[0037]

[0038] As a further improvement of the present application, step 4 specifically includes:

[0039] In a power system described by a state space equation, for a given sampling interval and controller feedback gain matrix , if there exists a symmetric positive definite matrix , , , , and any matrix , , , , , such that the system can satisfy the LMI conditions described by the following formula, then the system can be guaranteed to have asymptotic stability and satisfy performance index .

[0040] ,

[0041] wherein:

[0042] ; ; ;

[0043] ; ;

[0044] ; ; ;

[0045] ; ; ;

[0046] ; , ;

[0047] ; ; ;

[0048] ; ; ; ;

[0049] ; ;

[0050] ; ;

[0051] ; , .

[0052] wherein, is the integrated main block matrix, which is summarized from the terms after the derivation of Lyapunov functional and independent of the sampling duration. denotes the main block matrix, denotes the number of the main block matrix, is the preceding duration correction term, which is obtained by processing the derivative / double integral term in the Lyapunov functional on the interval , and is used to characterize the influence of the elapsed duration from the current sampling start point to the current time on stability; in the formula, it appears in the form of , denotes the preceding duration correction term at different times, b denotes the number of the preceding duration correction term at different times; is the remaining duration correction term, which is obtained by processing on the interval , and is used to characterize the influence of the remaining duration from the current time to the next sampling on stability; in the formula, it appears in the form of , He denotes the Hermitian operation, col denotes the columnization or vectorization operation. is the base matrix, which denotes the extraction of the i th n-dimensional sub-block from the aggregated vector. The block matrix composed of the base matrix is used to rearrange / subtract / sum the sub-blocks in the required order and participate in the construction of the quadratic form. The matrix vector formed by stacking the base matrix array. and represents an intermediate matrix vector obtained by linear combination of system matrix , control gain , base matrix , etc.

[0053] To ensure the asymptotic stability of the system, the present application constructs LKFs as follows:

[0054]

[0055] The constructed LKFs are based on the sampled state of the system, the state evolution within the sampling interval, and the multiple sub-functions of the derivative characteristics, and the overall derivative derivation expression is as follows:

[0056]

[0057] wherein, represents the system state energy term, l represents the number of the system state energy term, when l=1, represents the current state energy term, when l=2, represents the historical state energy term, when l=3, represents the state change rate energy term, represents the derivative of the new Lyapunov-Krasovskii function, is a system state column vector, represents a comprehensive main block matrix composed of Lyapunov variables, represents a previous time length correction term, represents a remaining time length correction term, represents the coefficient of external disturbance. When the LMI condition is met, the system in the absence of external disturbance , the derivative of the LKFs satisfies , thereby guaranteeing the asymptotic stability of the system. Further, considering the disturbance signal , the above inequality is integrated on the interval , and the following is obtained:

[0058]

[0059] Under the initial condition , it can be deduced that the system output satisfies . It can be seen that when the system meets the LMI condition, not only the asymptotic stability is achieved, but also the preset performance index is realized.

[0060] As a further improvement of the present application, it further comprises:

[0061] In another power system described by a state space equation, for a given sampling interval , if there exist symmetric positive definite matrices , , , and an arbitrary matrix , , , , such that the system can satisfy the condition described by the following equation, then the system can be guaranteed to have asymptotic stability and satisfy performance index . Then, the controller feedback gain matrix can be derived.

[0062] ,

[0063] wherein has the same meaning as , has the same meaning as . The meanings of the remaining symbols are the same as those of the corresponding symbols in the previous state space equation, as follows:

[0064] ; ; ;

[0065] ; ;

[0066] ; ; ;

[0067] ; ; ;

[0068] ; , ;

[0069] ; ; ;

[0070] ; ; ; ;

[0071] ; ;

[0072] ; ;

[0073] .

[0074] It can be seen that unlike the previous state space equation, the variable transformation operation is introduced , and then the LMI conditions are multiplied by and its transpose on the left and right sides of the inequality respectively. At the same time, the equivalent matrix transformation relationship is constructed: 、 、 、 、 、 、 、 and .

[0075] Through the above variable substitution and matrix transformation, the original LMI criterion can be equivalently transformed into a more easily solvable form, thereby helping to calculate the controller gain matrix and realize the system stability judgment condition.

[0076] It should be further pointed out that the technical features corresponding to the above options can be combined or replaced with each other to form new technical solutions without conflict.

[0077] Compared with the prior art, the present application has the beneficial effects that:

[0078] The present application proposes a coordinated control method that integrates adaptive control responsibility allocation and sampling optimization for the load frequency control problem under the auxiliary power grid framework of electric vehicles. The present application makes full use of the dynamic heterogeneity of GUs and EVs in response speed and regulation ability, realizes flexible allocation of regulation tasks and efficient use of communication resources. In the control strategy design, an adaptive sampling mechanism based on gradient descent is introduced for inertia-dominated GUs, and a real-time feedback control strategy is designed for fast-response EVs. At the same time, a stability criterion is constructed combining LKF theory to ensure the asymptotic stability of the system under sampling constraints and performance. Numerical simulation results verify that the proposed strategy is superior to traditional methods in system stability, regulation accuracy and communication efficiency, and shows good engineering application prospects. The specific contributions include:

[0079] (1) A composite system model integrating GUs and EVs is constructed, which improves the accuracy and scalability of system modeling by distinguishing the dynamic characteristics and control paths of GUs and EVs, and provides a unified mathematical framework support for subsequent control strategy design and performance analysis.

[0080] (2) An adaptive control responsibility allocation strategy based on system state changes is proposed, which can adjust the adjustment weights of GUs and EVs in real time, improving the control response coordination and system resource utilization efficiency.

[0081] (3) For GUs with large inertia, a sampling-driven control mechanism is designed, and a gradient descent algorithm is introduced to dynamically adjust the sampling interval, effectively reducing unnecessary control updates and reducing communication burden.

[0082] (4) For EVs with fast dynamic characteristics, a real-time feedback control strategy is adopted to quickly track frequency deviation and provide timely and effective auxiliary adjustment capability.

[0083] (5) Based on the sampling state and interval state evolution information, LKFs with ring connection characteristics are constructed. This method ensures the robust stability of the control system while effectively reducing the conservatism of stability analysis.

[0084] (6) Through numerical simulation verification, the proposed control strategy is superior to existing methods in terms of frequency regulation effect, communication resource utilization, and multi-source regulation coordination, and has good engineering adaptability and promotion value. BRIEF DESCRIPTION OF DRAWINGS

[0085] Figure 1 The flowchart of the power system load frequency control method based on the intelligent allocation weight mechanism of the present application;

[0086] Figure 2 The GUs control signal trajectory graph under the fixed control weight allocation strategy in embodiment 1 of the present application;

[0087] Figure 3 The GUs control signal trajectory graph under the fixed control weight allocation strategy in embodiment 1 of the present application;

[0088] Figure 4 The interval change graph under the supervision optimization and the random strategy of the present application;

[0089] Figure 5 The output trajectory of the GUs and EVs control signal under the intelligent allocation weight mechanism of the present application. DETAILED DESCRIPTION

[0090] The technical solutions of the present application will be described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various configurations. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0091] It should be noted that the defects of the above prior art solutions are the results obtained by the inventors after practice and careful study, therefore, the discovery process of the above problems and the solutions proposed by the embodiments of the present application to the above problems should be the contributions made by the inventors to the present application in the process of invention and creation, and should not be understood as technical contents known to those skilled in the art.

[0092] The application constructs a coordinated regulation model based on traditional generator units and electric vehicles, proposes an adaptive control responsibility allocation strategy dynamically adjusted according to system frequency deviation, and realizes reasonable allocation and response matching of frequency regulation tasks. On this basis, an optimized sampling mechanism based on system state evolution is designed to dynamically adjust the control signal update frequency and reduce the communication load. Further, a controller design criterion is constructed based on Lyapunov stability theory and LMI method to ensure the stability and performance index constraints of the system under various control weight configurations. The simulation results show that the method effectively reduces the triggering frequency of the control channel and the consumption of communication resources while achieving rapid suppression of frequency disturbance and improving frequency regulation accuracy, and has good robustness and engineering application value.

[0093] Embodiment 1: coordinated LFC method based on fixed control weight allocation

[0094] This embodiment designs a frequency regulation strategy under a fixed control weight allocation mechanism for an LFC system in which GUs and EVs participate in coordination. The main features of this embodiment are as follows:

[0095] 1. System structure:

[0096] A power system model composed of GUs and EVs is constructed. The system takes the frequency deviation state as the main feedback basis, and outputs are used to generate the regulation instructions of the GUs and EVs control channels and to achieve the frequency regulation goal together. The system also considers the influence of external disturbance on dynamic behavior. Specifically as follows:

[0097]

[0098] wherein, represents the state variable of the system; represents the derivative of the state variable of the system at time t; represents the measured output of the system; represents the control input signal at the GUs end; represents the control input signal at the EVs end; represents the external disturbance; , , , and is a parameter matrix.

[0099] 2. Controller design and sampling optimization:

[0100] A sampling control strategy is proposed to reduce the control update frequency, reduce communication resource consumption, avoid excessive regulation, and improve system frequency stability and resource utilization efficiency. Specifically as follows:

[0101]

[0102] where is the feedback gain matrix, denotes the k-th sampling time. The sampling time constitutes a strictly increasing time sequence .

[0103] To reduce communication load and improve system response efficiency, an intelligent sampling control strategy based on system state change is designed for GUs units with large inertia. This strategy dynamically adjusts the sampling interval according to the state evolution law to achieve adaptive triggering of control updates. The specific method is as follows:

[0104] A performance function is constructed to measure the current operating state of the system , whose value is the energy of the key state variable at the sampling time . The performance function is as follows:

[0105]

[0106] Then, the rate of change (gradient) of this performance function at the current time is calculated, as follows:

[0107]

[0108] The gradient descent update strategy is used to adjust the next sampling period , and the update rule is as follows:

[0109]

[0110] where is the adjustment step size coefficient, denotes the restriction of the sampling period to a given interval to ensure control stability.

[0111] Through the above mechanism, when the system state energy changes dramatically (such as the frequency deviation increases sharply), the sampling period is automatically shortened to better track the rapid dynamics of the system; and when the system energy tends to be stable, the sampling period is automatically lengthened to reduce the calculation and communication overhead, achieving a dynamic balance between control accuracy and communication efficiency.

[0112] Considering that EVs have fast charging and discharging capabilities and high frequency response characteristics, in order to fully exert their flexibility, the present application proposes an EVs control strategy based on real-time state feedback. This strategy immediately adjusts the control signal when the system frequency deviates, to achieve fast and fine frequency regulation, making up for the lack of response lag of traditional GUs. Specifically as follows:

[0113]

[0114] wherein, is the gain matrix corresponding to the EVs, is the current system state. The control signal of the EVs is continuously updated at a high frequency and does not depend on the sampling mechanism, ensuring the immediacy and continuity of the frequency response. This strategy can provide dynamic support under various operating conditions, improving the rapid response capability and frequency stability of the system.

[0115] Based on the above analysis, the following state space equation is derived:

[0116]

[0117] 3、 Main results:

[0118] This embodiment is aimed at the load frequency control system of traditional GUs and EVs described by the above state space equation. The stability and performance analysis under fixed control weight is carried out. To achieve this goal, the present application proposes two theorems: Theorem 1 proposes a set of sufficient conditions that can guarantee the asymptotic stability and performance guarantee of the system under a given sampling strategy; Theorem 2 gives a feasible solution scheme for solving the controller feedback gain matrix by LMI method.

[0119] Theorem 1: In the power system described by the above state space equation, for the given sampling interval and the controller feedback gain matrix , if there exist symmetric positive definite matrices , , , , and any matrix , , , , ​such that the system satisfies the LMIs conditions described by Performance index The following LMI inequalities are constructed:

[0120] ,

[0121] where:

[0122] ; ; ;

[0123] ; ;

[0124] ; ; ;

[0125] ; ; ;

[0126] ; , ;

[0127] ; ; ;

[0128] ; ; ; ;

[0129] ; ;

[0130] ; ;

[0131] ; , .

[0132] Theorem 2: In the power system described by the state space equation above, for a given sampling interval , if there exist a symmetric positive definite matrix , , , , and an arbitrary matrix , , 、 、 If the system can satisfy the condition described by the following equation, then the asymptotic stability of the system can be guaranteed, and the performance index can be satisfied Performance index Then, the controller feedback gain matrix can be derived, and the following LMI inequality can be constructed:

[0133] ,

[0134] The calculation of each symbol in Theorem 2 is similar to that in Theorem 1, and is not described here. The difference is as follows:

[0135] .

[0136] Proof: By introducing a variable transformation operation , then multiplying the LMI of Theorem 2 by on the left side and its transpose on the right side, respectively. At the same time, the equivalent matrix transformation relationship is constructed as follows: , , , , , , , and .

[0137] This embodiment is based on the MATLAB platform to build a single-area load frequency control system with the participation of GUs and EVs, to verify the frequency regulation performance and controller design effect under the fixed control weight mechanism. The model is established according to existing research results, and the main structure and parameters are set according to the standard power system configuration, and the key parameters are listed in Table 1.

[0138] Table 1 Parameter values of EVs integrated power system

[0139]

[0140] Then, to analyze the influence of different frequency regulation task allocation ratios on the system performance and controller structure, this embodiment sets four groups of fixed control weight schemes, corresponding to different participation ratios of GUs and EVs. Under each group of weight configuration, LKFs are constructed based on Theorem 2, and the corresponding state feedback controller gain matrix is obtained by solving the inequality.

[0141] Table 2 Controller gains of four groups of fixed control weight schemes

[0142]

[0143] As shown in Table 2, the designed controller gain shows an upward trend with the increase of the participation ratio of EVs. This phenomenon indicates that the overall inertia level of the system decreases, and the dominant role of GUs in frequency regulation is gradually replaced by EVs with faster response speed. Since the decrease in inertia will lead to an increase in the sensitivity of the system to disturbances, in order to maintain stability and control effect, the gain level of the controller needs to be correspondingly increased to enhance the regulation ability. At the same time, higher controller gain helps to fully release the fast response advantage of EVs under small frequency fluctuations, achieving more precise dynamic compensation. Therefore, the trend of the change of the controller gain is highly consistent with the essential difference in the dynamic characteristics of GUs and EVs, further verifying the adaptability and rationality of the proposed control strategy.

[0144] In order to intuitively show the influence of different control weight distribution on the performance of the system, the dynamic response trajectory of the system over time under four groups of fixed weight configurations is simulated, and the frequency deviation change process of the system under the action of load disturbance is understood, which can be used to evaluate the frequency modulation speed, stable time and dynamic characteristics under each distribution scheme. The simulation results show that with the increase of the participation ratio of EVs, the frequency response speed of the system is accelerated, the fluctuation amplitude is reduced, the steady-state recovery time is shortened, and the frequency modulation performance is obviously improved.

[0145] In order to further observe the sampling and adjustment behavior of the control channel during the disturbance process, the sampling interval change of the system during operation under four types of weight distribution is analyzed. In the initial stage of disturbance, the system actively shortens the sampling interval to increase the control update frequency in response to the rapid frequency shift; while in the process of gradually restoring the frequency, the sampling interval increases, which reflects the ability of dynamic communication resource saving. This optimization process is derived from the judgment mechanism of the system state evolution trend in the control strategy, which can automatically adjust the sampling period according to the state energy gradient, so as to realize the coordination of "fast-slow" matching communication scheduling and control rhythm. The proposed method still has strong sampling flexibility and regulation efficiency under the fixed control weight mechanism, effectively balancing the system response speed and communication load control.

[0146] Figure 2 The output signal of the GUs control channel is shown under the fixed control weight configuration, the horizontal coordinate represents time (unit: second), and the vertical coordinate represents the output control signal (power) change trajectory, Figure 2 Fig. (a) is the change trajectory of the output control signal of GUs under the type 1 weight configuration in Table 2, Figure 2 Fig. (b) is the change trajectory of the output control signal of GUs under the type 2 weight configuration in Table 2, Figure 2 Fig. (c) is the change trajectory of the output control signal of GUs under the type 3 weight configuration in Table 2, Figure 2(d) shows the trajectory of the GUs output control signal change under weight configuration type 4 in Table 2. Figure 2 As can be seen, the GUs control signal exhibits a certain amplitude adjustment in the initial stage of disturbance, but the overall response is relatively stable without violent oscillations. This is consistent with the inherent mechanical inertia characteristics of the GUs, whose control signal is mainly used to provide initial inertial support and basic stability assurance for the system. Thanks to the sampling control strategy designed in this embodiment, the update frequency of the GUs control signal can be dynamically adjusted according to state changes, avoiding frequent invalid updates when the disturbance is small, thereby improving communication efficiency and the coordination of control output.

[0147] Figure 3 This diagram illustrates the evolution of the output signal of the EVs control channel over time under a fixed control weight configuration. The horizontal axis represents time (in seconds), and the vertical axis represents the evolution of the output control signal (power). Figure 3 (a) shows the trajectory of the EVs output control signal change under the type 1 weight configuration in Table 2. Figure 3 (b) shows the trajectory of the EVs output control signal change under the type 2 weight configuration in Table 2. Figure 3 (c) shows the trajectory of the EVs output control signal change under the type 3 weight configuration in Table 2. Figure 3 Figure (d) shows the trajectory of the EVs output control signal under weight configuration type 4 in Table 2. It can be observed that the EVs control signal reacts rapidly after a disturbance occurs and continuously outputs a regulating signal throughout the entire system regulation process. This fast response characteristic effectively compensates for the slow response of traditional GUs. Especially when the system experiences small frequency fluctuations, the EVs control channel can achieve continuous, real-time power regulation, enhancing the system's fine-grained frequency modulation control capability. Figure 3 The control signals shown change frequently but with moderate amplitude, demonstrating the advantages of EVs in high-frequency dynamic compensation and providing key support for the stable operation of the overall system. Figure 2 and Figure 3 The study jointly verified the complementarity of the control channels GUs and EVs in response characteristics under the proposed fixed control weight mechanism. GUs mainly provides basic stability support, while EVs realize dynamic and rapid adjustment. The coordinated effect of the two effectively improves the frequency control performance and dynamic stability of the system.

[0148] Embodiment 2: This embodiment is optimized on the basis of Embodiment 1, and further introduces an adaptive control responsibility allocation strategy based on system state, to dynamically adjust the frequency regulation task allocation proportion between GUs and EVs. Compared with fixed control weight configuration, the adaptive mechanism can flexibly allocate resources according to the size of frequency deviation, so that GUs mainly respond to system severe disturbance, and EVs undertake rapid regulation task in the small amplitude fluctuation stage, thereby improving the overall frequency regulation efficiency and resource coordination ability of the system. The adaptive control responsibility allocation strategy based on system state is as follows:

[0149]

[0150] wherein t represents the current time, t+1 represents the next time, represents the frequency regulation responsibility participation weight of the traditional generator unit at the next time, represents the frequency regulation responsibility participation weight of the electric vehicle at the next time; and is a given parameter; represents the fifth state variable, i.e. .

[0151] In the proposed adaptive control responsibility allocation strategy based on , the control task is flexibly allocated between GUs and EVs according to the real-time system state. When a significant deviation from the equilibrium state is detected, GUs will undertake more control responsibility due to their strong regulation ability and high stability. On the contrary, in the case of approaching equilibrium state and small fluctuation, EVs undertake more control task through their fast response and flexible characteristics. This dynamic allocation mechanism not only improves the response coordination, but also ensures that each resource always operates within the optimal control range, thereby significantly enhancing the overall performance and system robustness of frequency regulation.

[0152] Further verify the frequency regulation response effect of the proposed adaptive control responsibility allocation strategy under different system states. Compared with fixed control weight configuration, the adaptive mechanism can adjust the regulation task allocation proportion of GUs and EVs according to the change of frequency deviation, so as to realize more sensitive and coordinated control response. In the initial disturbance stage, GUs undertake the main control task to provide inertia support; while in the frequency tends to be stable, EVs realize rapid compensation through high frequency small amplitude regulation, effectively improving the frequency regulation accuracy and recovery speed of the system. The response process shows that the proposed control weight allocation strategy has good dynamic adaptability and control performance advantage, and can fully play the complementary regulation role of GUs and EVs in different stages of system operation.

[0153] ​​Further, the adaptive control strategy is disclosed in the task adjustment mechanism in different operating stages. In the initial stage of disturbance, the control weight ratio of GUs is high, and the frequency regulation task is preferentially undertaken to provide the inertia support and stability guarantee required by the system. As the system state gradually recovers and tends to be stable, the participation weight of EVs gradually rises, and fine control and dynamic compensation are realized through rapid feedback regulation. The trajectory clearly reflects the characteristics of dynamic adjustment of control weight according to the evolution trend of frequency deviation, and embodies the ability of the proposed strategy to reasonably allocate GUs and EVs resources and enhance the coordination of frequency response in different periods.

[0154] To verify the effect of the proposed optimization sampling mechanism in the frequency adjustment of control signal update frequency, Figure 4 The comparison shows the sampling interval change process of GUs control channel under the optimization sampling strategy and the random sampling strategy. The abscissa represents time (unit: second), and the ordinate represents the sampling interval. Figure 4 In Fig. (a), the sampling interval change under the optimization sampling strategy is shown. Figure 4 In Fig. (b), the sampling interval change under the random sampling strategy is shown. Figure 4 As can be seen from Fig., the sampling interval under the optimization strategy presents the characteristics of dynamic adjustment according to the evolution trend of the system state: it is automatically shortened in the disturbance stage and gradually lengthened in the steady state, showing clear response coordination and rhythm adaptability. Under the random sampling strategy, the sampling interval presents irregular fluctuations, which cannot ensure timely updating of the control signal at critical moments, and may cause redundant communication when the system is stable, leading to frequency regulation performance fluctuations and resource waste. The comparison result further embodies that the proposed sampling strategy achieves a good balance between control accuracy and communication efficiency, and has obvious engineering practicability superior to traditional non-perception sampling methods.

[0155] Table 3 Comparison of sampling characteristics under different control responsibility allocation

[0156]

[0157] To further quantify the influence of different control responsibility allocation strategies on sampling frequency and system performance, Table 3 statistically compares the sampling characteristics indicators under the six strategies, including total trigger times, average sampling interval and performance improvement ratio. The results show that the traditional random sampling strategy has the characteristics of high trigger frequency (100 times) and short sampling interval (0.300s), which leads to heavy communication burden and limited frequency modulation performance. In contrast, the fixed control weight strategy can significantly reduce the trigger times while ensuring performance under the premise of clear control task allocation, and the average sampling interval can be extended to 0.405s, bringing up to 26.00% performance improvement. Especially noteworthy is that after using the adaptive control responsibility allocation strategy, the total trigger times of the system are further reduced to 65 times, the average sampling interval is increased to 0.462s, and up to 35.00% of the overall performance improvement is obtained. This result fully verifies the ability of the proposed method to reduce communication overhead while significantly enhancing the system frequency modulation performance, showing high engineering practicability and strategy optimization potential.

[0158] To further reveal the adjustment effect of the adaptive control responsibility allocation strategy on the dynamic characteristics of the control signal, Figure 5 The output trajectories of the GUs and EVs control signals under the action of the strategy are plotted, where the horizontal coordinate represents time and the vertical coordinate represents the control signal (power) trajectory, and Figure 5 (a) in FIG. 6 represents the output trajectory of the GUs control signal, Figure 5 (b) in FIG. 6 represents the output trajectory of the EVs control signal. As can be seen from Figure 5 at the beginning of the system disturbance, the control strategy allocates tasks to GUs first, making them quickly output large control signals to provide inertia support and quickly suppress frequency deviation; as the system state gradually stabilizes, the adjustment task is gradually transferred to EVs, which complete fine repair and fluctuation suppression of the frequency through more frequent small control operations. The whole control process realizes the orderly switching of master and slave and the cooperation of high and low frequency adjustment, showing clear division of responsibilities and dynamic coordination ability between GUs and EVs. This result further verifies the practicability and effectiveness of the proposed adaptive allocation mechanism in improving system response sensitivity and optimizing control output rhythm.

[0159] This embodiment focuses on the intelligent load frequency control problem of electric vehicles collaborative participation. First, a frequency regulation system structure containing GUs and EVs is established, and a collaborative regulation model with adjustable control weight is introduced. Second, an adaptive control responsibility allocation mechanism based on frequency deviation state is proposed, which realizes the dynamic switching and coordination of GUs and EVs frequency regulation tasks. Then, a robust controller is designed by constructing a linear matrix inequality condition, and a state-driven optimization sampling strategy is introduced, which effectively improves the control accuracy and communication efficiency of the system. Finally, simulation experiments show that the proposed strategy has faster response speed, higher steady-state accuracy and lower communication overhead under frequency disturbance, which embodies good engineering practicability and robust reliability. This embodiment provides theoretical support and application value for building an efficient coordinated frequency regulation method for new power systems.

[0160] The above detailed description is a detailed description of the present application, which cannot be considered as limiting the specific embodiments of the present application only to these descriptions. For ordinary skilled persons in the technical field to which the present application belongs, some simple deductions and substitutions can be made without departing from the concept of the present application, which should be considered as belonging to the protection scope of the present application.

Claims

1. A method for power system load frequency control based on intelligent weight allocation mechanism, characterized in that, The method comprises the following steps: Step 1, establishing a power system load frequency control model comprising a traditional generator set and an electric vehicle; Step 2, designing an adaptive control responsibility sharing strategy based on the power system load frequency control model; The adaptive control responsibility sharing strategy comprises: where t represents the current time, t+1 represents the next time, represents the frequency modulation responsibility participation weight of the traditional generator set at the next time, represents the frequency modulation responsibility participation weight of the electric vehicle at the next time; , , and is a given parameter; represents the fifth state variable; Step 3, constructing a dual-channel cooperative controller integrating algorithm optimization sampling and real-time feedback of the electric vehicle, and deducing a state space equation of the power system; Step 4, analyzing the stability of the power system by using a Lyapunov stability method based on the state space equation.

2. The power system load frequency control method based on intelligent weight distribution mechanism according to claim 1, wherein, The step 1 specifically comprises: A composite structure model is established as follows: wherein, denotes a state variable of the system; denotes the derivative of a state variable of the system with respect to time t; denotes a system measurement output; denotes a conventional generator set end control input signal; denotes an electric vehicle end control input signal; denotes an external disturbance; , , , and is a parameter matrix. 3.The power system load frequency control method based on the intelligent weight distribution mechanism of claim 1, wherein, The step 3 specifically comprises: An adaptive sampling mechanism based on gradient descent is introduced for the traditional generator set: wherein, denotes the current sampling period, denotes the previous sampling period, , denotes the minimum allowed sampling period, denotes the maximum allowed sampling period, is the adjustment step size factor, denotes the restriction of the sampling period to a given interval, denotes the rate of change of the performance function at the current sampling instant; A real-time feedback control strategy is designed for the electric vehicle; the state space equation is as follows: wherein, K denotes the gain matrix, denotes the k-th sampling instant.

4. The power system load frequency control method based on intelligent weight distribution mechanism according to claim 3, wherein, The step 4 specifically comprises: A new Lyapunov-Krasovskii function is constructed as follows: The overall derivative derivation is expressed as follows: wherein, represents a system state energy term, l represents the number of the system state energy term, l = 1, represents a current state energy term, l = 2, represents a history state energy term, l = 3, represents a state change rate energy term, represents a derivative of a new Lyapunov-Krasovskii function, is a system state column vector, represents a comprehensive main block matrix composed of Lyapunov variables, represents a preceding time length correction term, represents a remaining time length correction term, represents a coefficient of an external disturbance.

Images