A load frequency control method for a multi-region joint power system considering wind power access

Abstract

The application provides a load frequency control method of a multi-region joint power system considering wind power access, comprising the following steps: establishing a joint state space model of the multi-region joint power system, including a state space sub-model of a wind power region, a state space sub-model of a thermal power region and a tie line power deviation constraint relationship between the regions; establishing a joint control topology of the multi-region joint power system, including a sub-control topology of the wind power region, a sub-control topology of the thermal power region and a tie line power deviation feedback topology, wherein the tie line power deviation feedback topology comprises a feedback device, and the total number of the feedback device is equal to the total number of the wind power region and the thermal power region; and using control instructions output by the joint control topology to control devices of the wind power region and the thermal power region to perform grid load frequency regulation. The technical scheme of the application takes into account the dynamic response speed characteristics and comprehensive frequency modulation demand of the wind power and the thermal power, and optimally regulates the load frequency of the multi-region power system.

Description

Technical Field

[0001] This application belongs to the field of new energy control technology, and relates to wind power intelligent control technology. Specifically, it provides a load frequency control method for a multi-regional joint power system that takes wind power access into account. Background Technology

[0002] Load Frequency Control (LFC) is a dynamic control process that automatically adjusts the power of generators within a control area to maintain a dynamic balance with load changes and regional power exchange plans, while also keeping the interconnected grid frequency within a specified range. LFC outputs power commands to participating generators within the control area through data acquisition, processing, control decisions, and generator power allocation to meet control area deviation requirements. In traditional centralized power systems dominated by thermal power, LFC strategies are relatively mature. Common methods include classic controllers such as Proportional Integral (PI) and Proportional Integral Derivative (PID), which offer advantages such as simple structure and convenient implementation. However, these methods are mostly based on linearized models, making it difficult to cope with the frequently changing operating conditions and structural complexity of modern power systems.

[0003] With a large number of distributed wind power devices connected to multiple areas of the power system, not only can the power supply pressure in the area be effectively alleviated, but also, because electronic converters or energy storage can respond within hundreds of milliseconds to a few seconds, the adjustment speed is much faster than that of thermal power equipment. Therefore, by utilizing the flexible characteristics of wind power to participate in grid frequency regulation, a portion of the load dynamic response can be effectively diverted from thermal power, thereby enabling wind power to perform rapid frequency regulation while thermal power ensures base load or performs slow frequency regulation, allowing each device to operate in a more optimal range, achieving the comprehensive effect of extending lifespan and reducing regulation costs.

[0004] However, in multi-regional integrated power systems with wind power integration, wind power output fluctuates frequently due to the significant impact of weather conditions, making forecasting difficult. This introduces greater uncertainty and regulatory pressure to the joint frequency regulation. Furthermore, the reduced system inertia leads to rapid frequency drops during large disturbances, further reducing the buffer space available for thermal power regulation. This places higher demands on the associated frequency control and inertial support, resulting in substantial frequency dispatch pressure on the integrated power system. Therefore, a load frequency control method is needed that considers both the dynamic response characteristics of wind and thermal power units and the overall frequency regulation requirements, enabling coordinated operation and optimal control of multi-regional, multi-energy systems. Summary of the Invention

[0005] The purpose of this application is to provide a load frequency control method for a multi-regional joint power system, which achieves complementary advantages and overall system performance improvement by coordinating the control of wind power area and thermal power area to regulate grid load frequency.

[0006] The embodiments of this application can be implemented through the following technical solutions:

[0007] A load frequency control method for a multi-regional integrated power system considering wind power integration, wherein the multi-regional integrated power system includes at least one wind power region and at least one thermal power region, each wind power region includes at least one wind turbine unit, and each thermal power region includes at least one thermal power unit, the method comprising the following steps:

[0008] A joint state space model of a multi-regional power system is established, wherein the joint state space model includes a state space sub-model of the wind power region, a state space sub-model of the thermal power region, and the tie-line power deviation constraint relationship between the regions.

[0009] Based on the aforementioned joint state-space model, a joint control topology for a multi-regional joint power system is established. This joint control topology includes sub-control topologies for each wind power region, sub-control topologies for each thermal power region, and a tie-line power deviation feedback topology. The tie-line power deviation feedback topology includes... Each feedback unit is used to provide feedback on the tie-line power deviation to each wind power area or thermal power area. This represents the total number of wind power areas and thermal power areas.

[0010] The control commands output by the joint control topology are used to control the equipment in the wind power and thermal power areas to regulate the grid load frequency.

[0011] The load frequency control method for a multi-regional power system provided in this application firstly establishes targeted state-space models for wind power and thermal power regions, taking into account their different state-influencing factors and frequency response characteristics. Constraints are then established based on tie-line power deviation characteristics, forming a multi-regional joint state-space model. Next, the control architecture is optimized based on this joint state-space model. Each wind power and thermal power region designs a local sub-controller tailored to its specific characteristics. Multiple feedback loops generating tie-line power deviations are used to establish feedback mechanisms between regions, and different control strategies are applied to the wind power and thermal power regions. This interconnected topology, with each region controlling its own power, enables effective control within regions and information exchange and collaborative optimization between regions. It balances wind power output fluctuations, the dynamic response speed of thermal power units, and comprehensive frequency regulation requirements, thereby achieving coordinated operation and optimal control of multi-regional, multi-energy systems. Attached Figure Description

[0012] Figure 1 This is a flowchart of a load frequency control method for a multi-regional integrated power system provided according to an embodiment of this application;

[0013] Figure 2 This is the joint control topology of a multi-regional joint power system consisting of a wind power region and a thermal power region.

[0014] Figure 3 This is the joint control topology of a multi-regional joint power system consisting of one wind power region and two thermal power regions.

[0015] Figure 4 This is a schematic diagram of a disturbance signal applied in one specific embodiment;

[0016] Figure 5a This is a schematic diagram of the frequency response of a wind power area with a penetration rate of 50% in one specific embodiment;

[0017] Figure 5b This is a schematic diagram of the frequency response of a thermal power plant area with a penetration rate of 50% in one specific embodiment;

[0018] Figure 6a This is a schematic diagram of the frequency response of a wind power area with a penetration rate of 25% in one specific embodiment;

[0019] Figure 6b This is a schematic diagram of the frequency response of a thermal power plant area with a penetration rate of 25% in one specific embodiment;

[0020] Figure 6c This is a schematic diagram of the two-frequency response of a thermal power plant area with a penetration rate of 25% in one specific embodiment;

[0021] Figure 6d This is a schematic diagram of the three-frequency response of a thermal power plant area with a penetration rate of 25% in one specific embodiment;

[0022] Figure 7 This is a schematic diagram comparing the effects of different control strategies when the penetration rate is 50% in a specific embodiment;

[0023] Figure 8 This is a schematic diagram comparing the effects of different control strategies when the penetration rate is 25% in a specific embodiment. Detailed Implementation

[0024] The present application will now be further described based on preferred embodiments and with reference to the accompanying drawings.

[0025] In the description of the embodiments of this application, it should be noted that if terms such as "upper," "lower," "inner," or "outer" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this application is in use, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. In addition, in the description of this application, in order to distinguish different units, the terms "first," "second," etc. are used in this specification, but these are not limited by the manufacturing order, nor should they be construed as indicating or implying relative importance. Their names may differ in the detailed description and claims of this application.

[0026] The vocabulary used in this specification is for illustrative purposes and is not intended to limit the scope of this application. It should also be noted that, unless otherwise expressly specified and limited, the terms "set," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, a direct connection, or an indirect connection via an intermediate medium; or they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of these terms in this application.

[0027] This application provides a load frequency control method for a multi-regional integrated power system considering wind power access, wherein the multi-regional integrated power system includes at least one wind power region and at least one thermal power region, and each wind power region includes at least one wind turbine unit, and each thermal power region includes at least one thermal power unit.

[0028] In some embodiments, such as Figure 1 As shown, the method includes the following steps:

[0029] Step 100: Establish a joint state space model of a multi-regional power system. The joint state space model includes a state space sub-model of the wind power region, a state space sub-model of the thermal power region, and the tie-line power deviation constraint relationship between the regions.

[0030] Step 200: Based on the joint state-space model, establish a joint control topology for the multi-regional joint power system. The joint control topology includes sub-control topologies for each wind power region, sub-control topologies for each thermal power region, and a tie-line power deviation feedback topology. The tie-line power deviation feedback topology includes... Each feedback unit is used to provide feedback on the tie-line power deviation to each wind power area or thermal power area. This represents the total number of wind power areas and thermal power areas.

[0031] Step 300: Use the control commands output by the joint control topology to control the equipment in the wind power area and the thermal power area to regulate the grid load frequency.

[0032] The specific implementation methods of steps 100, 200, and 300 are described in detail below.

[0033] <Step 100>

[0034] Due to differences in regional structure, diverse power sources, and complex tie-line coupling, multi-regional power systems exhibit significant dynamic differences and nonlinear characteristics in their frequency response. Especially with the integration of a high proportion of wind power, the system inertia level drops significantly, and conventional frequency regulation methods are insufficient to ensure the rapid recovery and stable operation of the system frequency. Therefore, constructing an accurate and representative regional state-space model is a prerequisite for designing high-performance controllers.

[0035] In step 100 of this application, based on the above analysis, in terms of wind power area modeling, considering its low inertia characteristics and virtual inertia control strategy, a dynamic model including factors such as wind speed disturbance, pitch angle control, and energy storage system response is established to capture its key influence mechanism in frequency regulation. For thermal power areas, based on the classic LFC model, the tie-line power interaction with the wind power area is introduced to establish a multi-area joint model.

[0036] 1) Wind power area state space sub-model:

[0037] The dynamic process of a wind turbine (such as a wind turbine) can be represented as follows:

[0038] (1).

[0039] (1) The definitions of each variable in the formula are shown in Table 1.

[0040] Table 1 Definition of variables in the dynamic equation of wind turbine

[0041]

[0042] When measuring wind speed, the actual wind speed may differ from the measured speed due to disturbances. Therefore, the actual wind speed can be... Decomposed into two parts:

[0043] (2),

[0044] in, For the measurement interval, in the formula To measure wind speed, The disturbance error is bounded.

[0045] Next, we need to derive the frequency response of the wind power area. Similar to the process of thermal power plants participating in load frequency control, the frequency response of the wind power area can also be expressed using tie-line power. The output power of wind turbines and disturbance power However, due to differences in physical properties, the frequency response of wind power areas needs to be represented by other variables.

[0046] After the introduction of wind turbines, they will undertake some of the frequency regulation tasks of traditional thermal power plants. However, due to the lower inertia of wind turbines, the inertial response capability of wind power areas cannot support frequency regulation needs compared to thermal power plants. Therefore, virtual inertial control strategies can be incorporated into LFC (Low-Frequency Control) to compensate for the inertia of wind turbines. Virtual inertial control (VIC) is a control strategy that simulates the rotational mass inertial power of wind turbine blades and provides an inertial response based on the rate of frequency change. Its implementation relies on energy storage systems (ESS) configured in wind farms. For multiple wind power areas... For a wind power region (generally, each wind power region includes at least one wind turbine), considering virtual inertia control and velocity disturbances, its frequency response characteristics can be expressed as:

[0047] (3),

[0048] in, Frequency deviation (i.e., the difference between the actual frequency and the rated frequency of the power grid in the wind power area) (e.g., a deviation of 50Hz). , , , These are the output power deviation of the wind turbine, the virtual inertial support power deviation of the energy storage system (representing the deviation of the injected power component output by the energy storage system configured in the wind power area), the tie line deviation, and the disturbance power deviation. , These are the frequency regulation proportional coefficient and frequency response time constant for the wind power area, respectively. , These are the damping coefficient and inertial constant of the wind power area, respectively.

[0049] Linearize equation (1) (ignore the nonlinear terms) (higher-order minor quantities), and then combined with equation (3), we can obtain the first... The spatial state equations of a wind power region (i.e., the state space sub-model of a wind power region):

[0050] (4),

[0051] in, For wind power area State variables, To control variables, As a distractor, The boundary for the interference terms. This indicates the power variation of the tie lines between this region and other regions. For area control error (can be used) express); , , , , The coefficient matrix, in some specific embodiments, takes the form shown below:

[0052] ,

[0053] ,

[0054] ,

[0055] ,

[0056] ,

[0057] in, For wind power area Frequency offset coefficient, For the region For the region The gain coefficient of the connector, , Wind power areas The virtual inertial control gain and virtual damping coefficient, The time constant of the energy storage system , , exist Domain satisfies , For the Laplace operator; To linearize the operating point, , They are respectively , exist The value at that location.

[0058] In addition, through The expression can be obtained as follows: Right now Include as well as part.

[0059] 2) State-space sub-model of thermal power plant area:

[0060] The state-space sub-model of a thermal power region can refer to that of a wind power region. For example, in some embodiments, the thermal power units in the thermal power region are regarded as a system consisting of a regulator, a prime mover (such as various turbines such as gas turbines), and a generator. By combining its equations of motion with its frequency response characteristics, the following equation can be obtained for the thermal power region: State-space sub-model:

[0061] (5),

[0062] In the above formula, This refers to the generator's output power deviation. This refers to the position deviation of the regulating valve in the regulator. The deviation in the output power of the prime mover. For the control commands of the regulator, , , , thermal power area The state variables, control variables, disturbance terms, and output terms. , , , , The coefficient matrix, in some specific embodiments, takes the form shown below:

[0063] ,

[0064] ,

[0065] ,

[0066] ,

[0067] ,

[0068] in, , These are the reheat gain and reheat time constant of the prime mover, respectively. , These represent the system gain and system time constant of the generator, respectively. The time constant of the regulator; The turbine time constant of the prime mover.

[0069] 3) Constraints between wind power areas and thermal power areas

[0070] Equations (4) and (5) above are independent state space sub-models of wind power and thermal power regions, respectively. When wind power and thermal power regions exist simultaneously in the system, the two sub-models can be combined and the subscripts can be removed. , Thus, the joint state-space model of the multi-regional joint power system is obtained as shown in the following equation.

[0071] (6),

[0072] In formula (6), each region The state variables, control variables, disturbance terms, output terms, and coefficient matrices can be selected based on wind power and thermal power. It can be observed that different wind power and thermal power regions are interconnected through tie lines and by merging... and From the corresponding terms in the table, we can obtain the following tie-line power deviation constraint relationship:

[0073] a. For any wind power region, its tie-line power deviation is determined based on the difference between the load frequency deviation of this region and the load frequency deviation of each other wind power region, and the difference between the load frequency deviation of this region and the load frequency deviation of each thermal power region;

[0074] b. For any thermal power region, its tie-line power deviation is determined based on the difference between the load frequency deviation of this region and the load frequency deviation of each other thermal power region, and the difference between the load frequency deviation of this region and the load frequency deviation of each wind power region.

[0075] With system memory individual wind power areas and thermal power area ( For example, let the sets of wind power regions and thermal power regions be respectively... , For any wind power area Its tie line power deviation exist The expression for the domain is shown below:

[0076] (7),

[0077] For any thermal power region Its tie line power deviation exist The expression for the domain is determined by the following formula:

[0078] (8).

[0079] Through the above formulas, the constraints that wind power areas and thermal power areas need to meet in the joint frequency regulation process can be established, thereby linking the various wind power areas and thermal power areas.

[0080] <Step 200>

[0081] Using the joint state-space model established in step 100 and the obtained tie-line power deviation constraints, a joint control topology for a multi-regional joint power system can be established. In the embodiments of this application, the joint control topology includes sub-control topologies for each wind power region and each thermal power region. Furthermore, based on the tie-line power deviation constraints obtained in step 100, the joint control topology also includes a tie-line power deviation feedback topology, wherein the tie-line power deviation feedback topology includes... One feedback device ( (This is the total number of wind power areas and thermal power areas), used to feed back the tie-line power deviation to each wind power area or thermal power area.

[0082] Figure 2 A specific embodiment of a joint control topology for a multi-regional power system is shown. In this embodiment, the power system includes a wind power region (let's call it region 1) and a thermal power region (let's call it region 2), meaning the wind power penetration rate is 50%. Figure 2 As shown, the joint control topology includes a sub-control topology for a wind power area and a sub-control topology for a thermal power area.

[0083] Specifically, the upper part of the wind electronic control topology includes the MPC controller, wind turbine model, virtual inertial control model, and frequency response model.

[0084] Specifically, the MPC controller is based on feedback inputs in the wind power area. The optimal control commands for each wind turbine in the wind power area are generated through the MPC optimization strategy. The feedback input for the wind power area... Based on the load frequency deviation of wind power area 1 and tie line power deviation Determined, its specific expression is as follows: The wind turbine model determines the output power deviation of the wind turbine based on the optimal control commands generated by MPC. The input to the virtual inertial control model is ,based on , , The constructed transfer function outputs the virtual inertial support power deviation of the wind turbine. The frequency response model is based on the output power deviation. Virtual inertial support power deviation Power deviation of connecting lines and disturbance power deviation (Various forms of disturbance known to those skilled in the art, such as random signals, quasi-periodic disturbances, step or abrupt disturbances, etc., can be used) to determine the load frequency deviation of the wind power area. .

[0085] The lower half of the thermal power control sub-topology includes a PI controller, a governor model, a turbine model, and a generator model. Specifically, the PI controller is based on the feedback input of the thermal power region. The rate adjustment is achieved through a PI control strategy, combined with feedback-derived data. This generates control commands for the thermal power unit, wherein the feedback input of the thermal power area 2 is... Based on the load frequency deviation of thermal power area 2 and tie line power deviation The control commands are used to obtain the output power deviation of the thermal power unit after passing through the governor model and turbine model. The generator model is based on the output power deviation. Power deviation of connecting lines and disturbance power deviation Determine the load frequency deviation of the thermal power area. .

[0086] Furthermore, a tie-line power deviation feedback topology is set between the sub-control topology of wind power area 1 and the sub-control topology of thermal power area 2. Since there is only one wind power area and one thermal power area in this embodiment, there is only one tie line. Accordingly, the tie-line power deviation feedback topology includes only one feedback unit, which is used to feed back the tie-line power deviation to wind power area 1 and thermal power area 2. and .

[0087] Specifically, , It can be obtained from equations (7) and (8). Obviously, for the case of a single connecting line, Therefore, see Figure 2 Feedback input , Then, take the difference between the two, and then multiply by . and Then, multiply by 1 and -1 respectively, and use as... , Feedback is sent to the wind power and thermal power areas.

[0088] In addition, it is also possible to and The difference multiplied by and Then multiply by 1 and -1 respectively, and use the result as... , Feedback is sent to both the wind power and thermal power areas. Clearly, the feedback values ​​should remain consistent in both cases. = .

[0089] Figure 3 A schematic diagram of a joint control topology for controlling a multi-regional power system consisting of one wind power region (numbered 1) and two thermal power regions (numbered 2 and 3) is shown, as follows. Figure 3 As shown, this control topology, in addition to the sub-control topologies corresponding to the wind power and thermal power areas, also includes a tie-line power deviation feedback topology consisting of three feedback units, as shown in the figure. The tie-line power deviations fed back to areas 1, 2, and 3 are respectively... , , .

[0090] pass Figure 2 , 3 The illustrated embodiment determines that when the sum of the number of wind power regions and thermal power regions included in a multi-regional integrated power system is At that time, the tie-line power deviation feedback topology includes A feedback unit is used to provide feedback on the tie-line power deviation between any two regions, wherein... Indicates from The number of combinations of choosing any two numbers from a given number is obviously... Furthermore, for any wind power region or thermal power region, the tie-line power deviation can be uniformly determined by the following formula:

[0091] (10).

[0092] <Step 300>

[0093] In step 300, the joint control topology established in step 200 can be used to output optimized control commands to the wind turbines and thermal power units in the respective regions via the MPC controller in the wind power region and the PI controller in the thermal power region, respectively. This achieves load frequency regulation of the power grid, while continuously sampling the changes in the power grid load frequency in real time. Combined with the power deviation of the tie lines between the regions, the corresponding... It then feeds back to the controllers in each area, thus forming a closed-loop feedback.

[0094] Specifically, such as Figure 2As shown, closed-loop feedback control for thermal power areas is performed using a PI controller, while closed-loop feedback control for wind power areas is performed using an MPC controller. Because the PI controller has a simple structure and mature parameter tuning, it can stably and quickly eliminate frequency deviations in traditional thermal power units, meeting their control requirements at a relatively low cost. However, wind power output is greatly affected by wind speed fluctuations, offering strong adjustment flexibility, but also exhibiting high dynamic uncertainty and significant nonlinear / multivariable coupling within the system itself. Therefore, using MPC (Model Predictive Control) allows for the utilization of real-time measurement / predictive models to optimize decisions regarding wind farm dynamics and constrained multi-objectives, achieving more flexible, intelligent, and personalized frequency allocation and inertia / damping control.

[0095] Meanwhile, the constraint of feedback quantity formed by the power deviation of the tie line between wind power areas, between thermal power areas, and between wind power areas and thermal power areas can effectively ensure the safety of energy exchange between different types of power generation areas, and improve the coordination efficiency and safety robustness of multi-energy complementarity and multi-regional power systems.

[0096] Specifically, the parameter tuning and control methods of the PI controller are known to those skilled in the art. In some specific embodiments, the proportional coefficient and integral time constant can be optimized and adjusted using empirical formulas, the critical proportional method, and the Ziegler-Nichols tuning method, based on the dynamic characteristics, response speed, and stability requirements of the thermal power unit, through simulation analysis or field testing, to meet the control performance requirements under different operating conditions. Furthermore, the PI controller can be combined with other control strategies (such as feedforward compensation and adaptive control) to further improve the system's regulation accuracy and robustness.

[0097] The process of MPC obtaining the optimal control command is to use a preset objective function as the optimization objective, use a prediction model to make rolling predictions on the dynamic response of the system within a certain prediction time domain, and obtain the optimal control command at the current moment in real time by solving a quadratic optimization problem that includes inputs, outputs and physical constraints.

[0098] The prediction model of the MPC controller can adopt the state-space sub-model of the wind power area established above. Its parameters and coefficient matrix can be determined in various ways known to those skilled in the art. For example, a preliminary physical model can be established based on the aerodynamic theory of wind turbines and the electrical characteristics of generators and converters. Then, system identification algorithms (such as least squares method, recursive least squares, online identification, etc.) can be used to fit and optimize the dynamic response data of the wind power system under different wind speeds, loads and regulation commands, thereby obtaining the parameters in the state space and transfer function model. In addition, historical operating data of wind turbines can be used to estimate and correct the coefficient matrix of the model through data-driven methods (such as minimum mean square error, particle swarm optimization, genetic algorithm, etc.) to improve the modeling accuracy and robustness of wind farms under different operating environments.

[0099] In some preferred embodiments, to minimize system frequency deviation and reduce tie-line power fluctuations at any given time... The MPC controller is determined based on the following objective function. Optimal control commands at each time point after time step:

[0100] (11),

[0101] in, Let be the objective function. For the number of state prediction steps, To optimize the number of control steps and , , , These are the times. Load frequency deviation, tie line power deviation, and control command deviation; , , This is a weighting coefficient, and its specific value can be initially set based on historical experience values, and further adjusted according to the actual operating results during the actual operation.

[0102] Furthermore, MPC typically requires rolling system state prediction and optimal control command output, during which the state prediction step size... The value can be determined by comprehensively considering factors such as the system's dynamic characteristics, sampling period, and the predictability of future disturbances; for example, it can be set to 10-50. In practical applications, in the optimized control command sequence, usually only the first control command is executed, i.e. The value is 1, and then at the next sampling time, prediction and optimization are performed again to achieve rolling optimization control.

[0103] The constraints during the optimization process mainly include the system's physical and operational constraints, such as generator active power output limits, tie-line power transmission limits, and allowable frequency deviation ranges. In some specific embodiments, the constraints for the MPC controller in determining the optimal control command are:

[0104] (12)

[0105] in, , Wind power areas Output power The lower and upper limits, , The respective tie-line power of the wind power area The lower and upper limits, For the maximum allowable frequency deviation, For wind power area Control command deviation, This represents the maximum permissible change in control commands.

[0106] Solving the optimization problem of MPC can be done using various methods known to those skilled in the art, including commonly used quadratic programming (QP) algorithms, sequential quadratic programming (SQP), interior point methods, etc. For example, in some embodiments, the optimization problem can be transformed into a standard quadratic programming form:

[0107] (13)

[0108] in, The sequence of control commands to be optimized is typically represented as follows: , The weight matrix is ​​a quadratic term. The state weighting matrix, A weighting matrix to control the changes in input. Here is the state transition matrix. , The coefficient matrix and vector of the constraint relationship. The state evolution matrix is... For state vectors, The perturbation transfer matrix, This is the perturbation vector.

[0109] The solution to the optimization problem described by equation (13) can be achieved either by writing an executable program or by directly using commercial optimization solvers such as CPLEX, Gurobi, MOSEK, MATLAB OptimizationToolbox, etc.

[0110] Because power systems have model errors, parameter variations, and various uncertainties, relying solely on predictive models for open-loop control is insufficient to guarantee system control performance. Therefore, in some preferred embodiments, the sub-control topology of the wind power area employs a state estimation algorithm based on Kalman filtering to dynamically correct the input signal of the MPC controller, thereby achieving feedback correction.

[0111] Specifically, the feedback correction mechanism monitors the actual output of the system in real time and compares it with the output of the MPC prediction model to obtain the prediction error. This prediction error is then used to correct the prediction model, improving its accuracy and enabling the control strategy to better adapt to the actual operating conditions of the system. Specific Implementation Example 1

[0113] Specific embodiment one is used to verify the effectiveness of the method provided in this application. First, a multi-regional power system experimental test platform is built (the multi-regional power system includes one wind power region and three thermal power regions). Then, the load frequency control method provided in this application is used for simulation testing. The specific simulation test parameters are described below:

[0114] 1) Wind power area system parameters

[0115] For typical wind farms, the controlled object is the wind turbine. Considering its complex internal structure, it is assumed to consist of a turbine and a governor. This satisfies both the power generation and wind-driven speed regulation needs of the wind farm, while also simplifying the design and calculations. The transfer functions of the turbine and the governor are as follows: and , and These represent their response times. The parameters required for wind power areas are shown in Table 2 below.

[0116] Table 2 Wind Power Area System Parameters

[0117]

[0118] The interconnection gain coefficient between wind power areas and other wind power areas is 0.1, and the interconnection gain coefficient with thermal power areas is 1.3989.

[0119] 2) Thermal power area system parameters

[0120] Table 3 below shows the system parameters for the thermal power area.

[0121]

[0122] The interconnection gain between each thermal power plant is equal, with a value of 0.5, and the interconnection gain between each plant and the wind power area is also equal, with a value of 1.3989.

[0123] 3) MPC parameters

[0124] For the MPC controller, the main parameters that need to be set are: sampling time. Prediction range Control range and the weight matrix of the controlled object parameters and .

[0125] Prediction Time Domain The prediction time domain determines the range of model predictive control's predictions for the system's future state. A longer prediction time domain allows for consideration of longer-term system trends, providing more comprehensive information and facilitating the development of more forward-looking control strategies. When the prediction time domain is sufficiently long, model predictive control can detect significant load changes or fluctuations in renewable energy generation in advance, thereby adjusting generator output ahead of time and better maintaining system frequency stability. If the prediction time domain is too short, model predictive control can only make decisions based on the system's recent state, which may fail to address potential problems in the system in a timely manner, resulting in poor control performance. When the system load grows rapidly or renewable energy generation experiences drastic fluctuations, an excessively short prediction time domain may prevent the controller from making timely adjustments, leading to increased system frequency deviation.

[0126] Control Time Domain This determines the length of the control sequence obtained from the optimization calculation. A larger control time domain means that more future control actions can be considered at each sampling time, allowing for more full utilization of the system's dynamic information and achieving more precise control. Increasing the control time domain can improve the system's control accuracy and stability to some extent. However, if the control time domain is too large, it will increase the computational load and the difficulty of solving the optimization problem, leading to prolonged computation time and affecting the real-time performance of the control. In practical applications, it is necessary to reasonably control the computational load while ensuring control effectiveness to meet the requirements of real-time control.

[0127] To determine suitable prediction and control time-domain parameters, it is necessary to comprehensively consider factors such as the dynamic characteristics of multi-regional power systems, computational resources, and control requirements. A common approach is to evaluate the control effect under different parameter combinations through simulation experiments. During the simulation, various operating scenarios are set up, such as random load fluctuations and intermittent changes in renewable energy generation, and the changes in indicators such as system frequency deviation and tie-line power fluctuations under different parameter combinations are compared.

[0128] Based on the system simulation results, set , , When faced with various disturbances, the system can keep the frequency deviation and tie-line power fluctuation within a small range, and the control effect is good.

[0129] For the weight matrix and Considering that frequency variation is considered the most influential state variable in a wind power region, followed by interconnection gains between regions, and then generator speed variation, while the effects of wind power region inertia compensation, actual speed, and pitch angle can be ignored, a weight matrix for the wind power system is set. As shown in the following formula:

[0130] .

[0131] Similarly, for control variables, the pitch angle plays a nearly decisive role, but its impact is smaller than that of inertial compensation compared to state variables. Therefore, the weight matrix of the control variables... The settings are as follows:

[0132] .

[0133] The load frequency control capability was evaluated under the conditions of 50% wind power penetration (i.e., including one wind power area and one thermal power area) and 25% wind power penetration (i.e., one wind power area and three thermal power areas).

[0134] Figure 4 The diagram shows a disturbance signal applied to the power system with an amplitude of 2 MW and a duration of 6 seconds. Figure 5a , Figure 5b The frequency response of wind power areas and thermal power areas is shown respectively when the wind power penetration rate is 50%. Figures 6a to 6d The frequency response of the wind power area and the three thermal power areas are shown respectively when the wind power penetration rate is 25%.

[0135] As shown in the figures above, in multi-regional power systems with wind power penetration rates of 50% and 25%, both employing MPC controllers, after a momentary frequency drop of approximately 1 Hz, the MPC controller intervenes, causing the frequency to recover rapidly. The required overshoot for stabilization is then approximately 0.3 Hz, within the system's allowable range. This demonstrates that the employed MPC controller can effectively regulate the frequency.

[0136] In thermal power areas, systems with a wind power penetration rate of 25% obviously require more time to reach stability than systems with a wind power penetration rate of 50%, because in systems with lower wind power penetration rates, the frequency regulation capability of wind power areas cannot be fully utilized, and thermal power still holds a relatively dominant position.

[0137] Meanwhile, to verify the frequency characteristics of the MPC controller compared to the traditional PI controller, simulations were performed to compare the frequency responses of systems using PI control and those using MPC control. Figure 7 , Figure 8 The comparison between penetration rates of 50% and 25% is presented separately.

[0138] As can be seen, regardless of the wind power penetration rate, the system using MPC control exhibits better response speed and steady-state conditions than the system using traditional PI control when faced with the same disturbance. In a system with a wind power penetration rate of 50%, with similar overshoot, the system using MPC control reaches stability significantly faster and exhibits smaller steady-state fluctuations. In a system with a wind power penetration rate of 25%, the systems using both control methods reach stability at similar speeds, but the instantaneous frequency drop of the system using MPC control is significantly smaller than that of the system using PI control, demonstrating stronger resistance to disturbances.

[0139] The specific embodiments of this application have been described in detail above. For those skilled in the art, several improvements and modifications can be made to this application without departing from the principle of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A load frequency control method for a multi-regional integrated power system considering wind power integration, wherein the multi-regional integrated power system includes at least one wind power region and at least one thermal power region, each wind power region includes at least one wind turbine unit, and each thermal power region includes at least one thermal power turbine unit, characterized in that, Includes the following steps: A joint state-space model of a multi-regional power system is established, wherein the joint state-space model includes a state-space sub-model of the wind power region, a state-space sub-model of the thermal power region, and the tie-line power deviation constraint relationship between the regions. Based on the aforementioned joint state-space model, a joint control topology for a multi-regional joint power system is established. This joint control topology includes sub-control topologies for each wind power region, sub-control topologies for each thermal power region, and a tie-line power deviation feedback topology. The tie-line power deviation feedback topology includes... Each feedback unit is used to provide feedback on the tie-line power deviation to each wind power area or thermal power area. This represents the total number of wind power areas and thermal power areas. The control commands output by the joint control topology are used to control the equipment in the wind power area and the thermal power area to regulate the grid load frequency. The control sub-topology of the wind power area includes: MPC controller, wind turbine model, virtual inertial control model, and frequency response model; The MPC controller generates optimal control commands for each wind turbine in the wind power area based on the regional control error of the wind power area through the MPC optimization strategy. The regional control error of the wind power area is determined based on the load frequency deviation and tie line power deviation of the wind power area. The wind turbine model determines the output power deviation of the wind turbine based on the optimal control command; The virtual inertial control model is used to output the virtual inertial support power deviation of the wind turbine. The frequency response model determines the load frequency deviation of the wind power area based on the output power deviation, virtual inertial support power deviation, tie line power deviation, and disturbance power deviation. The thermal power units in the thermal power area consist of a speed governor, a turbine, and a generator. The control sub-topology of the thermal power area includes: PI controller, speed governor model, turbine model, generator model; The PI controller generates control commands for the thermal power unit based on the regional control error of the thermal power area as input feedback, through a PI adjustment strategy and in combination with the rate adjustment amount obtained from the feedback. The regional control error of the thermal power area as input feedback is determined based on the load frequency deviation of the thermal power area and the tie line power deviation. The rate adjustment amount obtained from the feedback is the ratio of the load frequency deviation of the thermal power area to the droop parameter of the speed governor. The control command is passed through the governor model and the turbine model to obtain the output power deviation of the thermal power unit; The generator model determines the load frequency deviation of the thermal power area based on the output power deviation, tie line power deviation, and disturbance power deviation.

2. The load frequency control method for a multi-regional joint power system considering wind power integration according to claim 1, characterized in that: The state space sub-model of the wind power area is determined based on the operating state equations of the wind turbines and their frequency response characteristics under the virtual inertial control strategy.

3. The load frequency control method for a multi-regional joint power system considering wind power integration according to claim 1, characterized in that, The tie-line power deviation constraint relationship between the various regions is as follows: For any wind power region, its tie-line power deviation is determined based on the difference between the load frequency deviation of this region and the load frequency deviation of each other wind power region, as well as the difference between the load frequency deviation of this region and the load frequency deviation of each thermal power region. For any thermal power region, its tie-line power deviation is determined based on the difference between the load frequency deviation of this region and the load frequency deviation of other thermal power regions, as well as the difference between the load frequency deviation of this region and the load frequency deviation of each wind power region.

4. The load frequency control method for a multi-regional joint power system considering wind power integration according to claim 1, characterized in that, For any wind power region or thermal power region, the tie-line power deviation is determined by the following formula: ， in, For the Laplace transform operator, , This is the designation for either a wind power zone or a thermal power zone. For the region The power deviation of the connecting line, , They are respectively regions , Load frequency deviation, For the region , Interconnection gain between For the region , Interconnection gain between them.

5. The load frequency control method for a multi-regional joint power system considering wind power integration according to claim 4, characterized in that, Each feedback unit is connected to the transmission lines of any two different areas of the wind power area and the thermal power area to obtain the load frequency deviation between the two areas and to feed back the tie line power deviation to the two areas.

6. The load frequency control method for a multi-regional joint power system considering wind power integration according to claim 1, characterized in that, At any time The MPC controller determines based on the following objective function. Optimal control commands at each time point after time step: ， in, Let be the objective function. For the number of state prediction steps, To optimize the number of control steps and , , , These are the times. Load frequency deviation, tie line power deviation, and control command deviation; , , These are the weighting coefficients.

7. The load frequency control method for a multi-regional joint power system considering wind power integration according to claim 6, characterized in that, The constraints of the MPC controller in determining the optimal control command are as follows: ， in, For wind power area 'output power' , These are its lower limit and upper limit, respectively. For the tie-line power of the wind power area, , These are its lower limit and upper limit, respectively. For wind power area Load frequency deviation, For the maximum allowable frequency deviation, For wind power area Control command deviation, This represents the maximum permissible change in control commands.

8. The load frequency control method for a multi-regional joint power system considering wind power integration according to claim 1, characterized in that, The sub-control topology of the wind power area adopts a state estimation algorithm based on Kalman filtering to dynamically correct the input signal of the MPC controller.

Images