Adaptive distributed coordinated control method for wind farms with fast grid frequency support
Through the wind farm adaptive distributed coordination control method, the prediction accuracy and response speed of the wind farm's grid frequency support are solved, the rapid stability of the grid frequency and the efficient utilization of wind energy resources are achieved, and the coordination between the wind farm and the grid is optimized. Cooperate.
Patent Information
- Application Number
- CN202411195104.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-29
AI Technical Summary
The prior art has problems such as insufficient prediction accuracy in wind farms for the power grid frequency support, slow control response speed, and in close coordination and coordination between the power grid, resulting in unstable grid frequency and waste of wind energy resources.
Adaptive distributed coordination control method of wind farm is adopted, and the wind farm's power output characteristics are determined by data collection, electrical characteristic analysis, wind farm output characteristics judgment, frequency voltage emergency control decision, adaptive control strategy formulation and distributed coordination control execution, the wind farm's grid frequency is realized. Quick response and support.
It improves the stability of the power grid frequency, reduces frequency deviation, optimizes the control strategy of the wind farm, ensures that the wind farm can quickly respond to the urgent needs of the power grid, reduces energy waste, and evaluates the support effect of the wind farm on the power grid frequency through the evaluation model.
Smart Images

Figure CN118713122B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of rapid grid frequency support, and in particular relates to a wind farm adaptive distributed coordinated control method for rapid grid frequency support. Background Art
[0002] In the current operation of the power grid, wind farms face many challenges in supporting the frequency of the power grid. In the existing technology, the output of wind farms is highly dependent on natural factors, especially changes in wind speed, which makes its output have great uncertainty and it is difficult to provide stable frequency support for the power grid. In order to deal with this problem, statistical analysis based on historical data or simple physical models are usually used to monitor and predict, trying to understand the output of wind farms in advance. However, these methods are often limited by factors such as data quality and model accuracy, and are not accurate and timely enough to effectively respond to rapid changes in wind farm output. For example, when the wind speed changes suddenly, the predicted results may differ greatly from the actual output, resulting in the inability of the power grid to adjust the backup capacity in time, affecting the stability of the power grid frequency.
[0003] In addition, when the grid frequency deviates, the existing control strategy mainly relies on preset thresholds and fixed control logic to determine whether control is needed. However, this approach may not be satisfactory in terms of response speed and adjustment effect. On the one hand, sensors that monitor parameters such as grid frequency and voltage may have sampling delays and errors, resulting in inaccurate basis for control decisions. On the other hand, the execution of control actions often requires coordination and transmission of multiple links, with obvious lags, and the grid frequency cannot be adjusted in a timely and effective manner, which may further expand the frequency deviation. For example, when the grid frequency suddenly drops, the control strategy may take a certain amount of time to start, and during this period, the frequency deviation may have adverse effects on other equipment in the grid.
[0004] In terms of coordination between wind farms and power grids, existing technical means have not yet been able to achieve adequate coordinated control. The information exchange between wind farms and power grids mainly relies on regular data reporting and manual intervention, which lacks real-time and initiative. At the same time, the collaborative work is not close enough. When the two sides formulate operation plans and respond to emergencies, they often lack effective communication and coordination mechanisms, and cannot give full play to the support role of wind farms for power grid frequency. For example, when the power grid load is at its peak, the wind farm may not be able to receive the peak load regulation instructions from the power grid in time, resulting in waste of wind energy resources and instability of the power grid frequency.
[0005] In summary, the existing technology has obvious deficiencies in terms of the prediction accuracy of wind farm output, the timeliness of response of control strategies, and the coordination and cooperation between wind farms and power grids, and these technologies need to be improved urgently. Summary of the invention
[0006] In view of the above deficiencies in the prior art, the purpose of the present invention is to provide an adaptive distributed coordinated control method for wind farms with rapid support for grid frequency, strengthen the coordinated control between wind farms and grids, realize the real-time and active nature of information interaction, improve the closeness of collaborative work, give full play to the supporting role of wind farms on grid frequency, and avoid waste of wind energy resources and instability of grid frequency.
[0007] In order to achieve the above objectives, the present invention adopts the following technical solutions:
[0008] The wind farm adaptive distributed coordinated control method for fast grid frequency support includes the following steps:
[0009] S1. Data collection: collect basic data of wind farm and power grid operation, including substation bus voltage, outgoing line interval data, current components of three-phase AC signals, and current and voltage signals of three-phase AC power, etc.;
[0010] S2. Electrical characteristic quantity analysis, based on the data collected in step S1, obtaining characteristic quantities such as voltage amplitude, frequency and phase;
[0011] S3. Determine the output characteristics of the wind farm, calculate the active power according to the electrical characteristic quantity obtained in step S2, and determine the output characteristics of the wind farm in combination with the calculation result of the active power and the power duration;
[0012] S4. Frequency and voltage emergency control action decision, based on the electrical characteristics such as frequency and voltage obtained in S2 and the output characteristics of the wind farm determined in step S3, determine whether the power grid is in a state requiring emergency control;
[0013] S5. Adaptive control strategy formulation: According to the judgment result of S4, when the power grid needs to reduce load, the load reduction calculation formula is used, and the appropriate load is selected for removal based on the outgoing line interval data and the priority of the load to achieve adaptive load reduction;
[0014] S6. Distributed coordinated control execution, generating control instructions for the wind turbines according to the adaptive control strategy developed in S5, and sending these instructions to the controllers of each wind turbine to achieve distributed coordinated control to ensure that the wind farm can quickly respond to the emergency control needs of the power grid;
[0015] S7. Frequency support capability assessment: Use the results of S1 to S6 to establish an assessment model, and use the assessment model to evaluate the support effect of wind farm output adjustment on grid frequency.
[0016] Furthermore, in S2, the method for obtaining the voltage amplitude characteristic quantity includes:
[0017] S21. Coordinate transformation, convert the three-phase AC signal collected by S1 from the three-phase stationary coordinate system (abc coordinate system) to the vector rotating space coordinate system (dq coordinate system), the formula is:
[0018] ;
[0019] In the formula, and Respectively expressed in and The current component on the axis; , , The three-phase current is , , The weight of the phase; is the rotation angle, which is related to the phase of the grid;
[0020] S22. Use a low-pass filter to filter the coordinates after the transformation. (straight axis) and The voltage component on the axis (quadrature axis) is filtered to obtain the voltage amplitude, the formula is:
[0021] ;
[0022] ;
[0023] ;
[0024] ;
[0025] ;
[0026] ;
[0027] is the voltage amplitude obtained after filtering, is the rotation angle, and After coordinate transformation, (straight axis) and Voltage component on the axis (quadrature axis) , , To convert the components of the voltage in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system;
[0028] Furthermore, in S2, the method for obtaining the frequency feature quantity is:
[0029] The voltage signal is processed by a proportional-integral controller to extract the frequency. The formula is:
[0030] ;
[0031] in, is the frequency extracted by the proportional-integral controller; The error signal is the difference between the actual voltage and the set voltage. It reflects the degree of deviation of the system. The controller adjusts the output according to the error signal to reduce the deviation. is the proportional coefficient, which determines the instantaneous response of the controller to the error signal. A value of will make the controller more sensitive to the error signal and be able to adjust the output more quickly, but it may cause system instability and oscillation; is the integral coefficient, which is used to eliminate the steady-state error of the system. The value will make the integral effect weaker, and the larger the The excessive value may lead to integral saturation, causing the system to respond more slowly or even lose stability.
[0032] Furthermore, in S2, the method for obtaining the phase feature quantity is:
[0033] The integrated phase is obtained by integrating the frequency signal, and the formula is:
[0034] ;
[0035] is the frequency extracted by the proportional-integral controller, The integrated phase is obtained by integration, and the integrated phase signal reflects the phase information of the voltage signal.
[0036] Furthermore, in S3, determining the output characteristics of the wind farm includes the following steps:
[0037] S31. Calculate the active power. The calculation formula for active power is:
[0038] ;
[0039] ;
[0040] ;
[0041] ;
[0042] ;
[0043] ;
[0044] ;
[0045] ;
[0046] ;
[0047] ;
[0048] is the voltage amplitude obtained after filtering; , , is to convert the components of the voltage in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system; and After coordinate transformation, and Voltage component on the axis; , , To convert the components of the current in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system; and is the component of the current in the vector rotation space coordinate system; is the rotation angle; is the voltage phase; is the current phase; is the phase difference between voltage and current;
[0049] S32. Determine the output characteristics of the wind farm based on the calculation results of active power and the duration of power,
[0050] like And remain stable for a period of time , then it is determined to be a power supply characteristic; if And keep this state for the same time , it is determined to be a load characteristic; during the judgment process, the change of active power is continuously monitored. ;
[0051] Furthermore, in step S4, determining whether the power grid is in a state requiring emergency control includes:
[0052] By comparing the absolute value of the difference between the frequency extracted by the proportional-integral controller and the standard frequency to see if it exceeds the preset threshold, When , it indicates that the absolute value of the difference between the frequency extracted by the proportional-integral controller and the standard frequency exceeds the preset threshold, indicating that the frequency extracted by the proportional-integral controller has a large deviation, and emergency control actions need to be taken to adjust the grid frequency; among them, is the frequency extracted by the proportional-integral controller, To set the standard frequency, is the preset frequency deviation threshold;
[0053] Whether the absolute value of the difference between the voltage amplitude obtained after filtering and the set standard voltage amplitude is greater than the preset voltage amplitude deviation threshold, When , it indicates that the absolute value of the difference between the voltage amplitude obtained after filtering and the set standard voltage amplitude is greater than the preset voltage amplitude deviation threshold, indicating that the voltage amplitude obtained after filtering is abnormal and emergency control action is required to ensure that the voltage is stable within a reasonable range; among them, is the voltage amplitude obtained after filtering, The set standard voltage amplitude, is a preset voltage amplitude deviation threshold;
[0054] Comparison of the rate of change of frequency and voltage amplitude. and , indicating that the change rates of frequency and voltage have exceeded the corresponding preset change rate thresholds, indicating that the state of the power grid changes rapidly and that emergency control actions are required to stabilize the power grid; and are the rates of change of frequency and voltage respectively, and is a corresponding preset change rate threshold;
[0055] Emergency control measures may include, but are not limited to: adjusting the generation of wind farms to provide necessary frequency support; activating backup generation resources to quickly respond to grid needs; and implementing load management strategies, such as demand response or load shedding, to reduce grid load.
[0056] Furthermore, in S5, the method for formulating the adaptive load reduction strategy includes:
[0057] S51. Load reduction calculation, considering the impact of power loss and surplus, using weighted coefficient to calculate load reduction:
[0058] ;
[0059] is the amount of load to be removed; and is the weight coefficient used to adjust the impact of power loss and surplus; is the current power loss of the grid; is the current power surplus of the grid;
[0060] S52. Load selection:
[0061] S521. Exclude the distribution network branch with power supply connected to the grid, exclude the distribution network branch with power supply connected to the grid from all the distribution network branches of the substation, and use the remaining distribution network branches as candidate distribution network branches. The specific method includes:
[0062] Assume that the set of all distribution network branches of the substation is: ,in, Indicates Strip distribution network branch;
[0063] The distribution network branches for power supply to the Internet are as follows: ;
[0064] The candidate distribution network branch set is: ;
[0065] S522. Initialize loop variables and accumulate load power, set loop variables , initialize the loop variable , set Article The load power corresponding to the distribution network branch is , then the accumulated distribution network branch load power is ;
[0066] S523. Determine whether the accumulated distribution network branch load power is greater than the load reduction amount:
[0067] Is it established? If , then execute the next step, loop variable Add 1 to form a new loop variable, that is ; The new Load power corresponding to the distribution network branch , added to Repeat the judgment until Established;
[0068] During the process, once If established, then record the first to the Distribution network branch collection and sends tripping commands to these distribution network branches to cut off the load.
[0069] Further, in S6, the distributed coordinated control method includes:
[0070] S61. Determine the control target. According to the emergency control requirements of the power grid and the current output of the wind turbines, determine the new output target that each wind turbine should achieve to reduce load or meet other requirements of the power grid. The calculation formula of the new output target is:
[0071] ;
[0072] In the formula, For the new output goal, is the current output of the wind turbine, The amount of load to be removed;
[0073] S62. Calculate the adjustment amount. According to the performance parameters and load reduction requirements of the wind turbine, calculate the output amount that needs to be adjusted for each wind turbine to ensure that the wind turbine can accurately respond to the control target. The calculation formula for the output amount that needs to be adjusted for the wind turbine is:
[0074] ;
[0075] In the formula, The output of the wind turbine needs to be adjusted. The amount of load to be removed, It is a coefficient determined according to the performance parameters of the wind turbine. The type, capacity and efficiency of wind turbines were considered comprehensively;
[0076] S63 generates a control instruction, and generates a specific control instruction according to the adjustment amount calculated by S62, so that the wind turbine adjusts the output to meet the control requirements;
[0077] The control instructions include pitch angle adjustment and speed adjustment, etc.
[0078] For a fixed pitch wind turbine, the pitch angle adjustment is:
[0079] ;
[0080] For variable pitch wind turbines, the speed adjustment is:
[0081] ;
[0082] in, is the pitch angle adjustment; The output of the wind turbine needs to be adjusted; is the speed adjustment amount; and are proportional coefficients, It reflects the effect of pitch angle change on power output. It reflects the effect of speed change on power output.
[0083] Furthermore, in S7, the method of evaluating the effect of wind farm output adjustment on grid frequency support through an evaluation model includes:
[0084] S71. Establish a model, which is represented by:
[0085] ;
[0086] ;
[0087] in, It is an accumulated sequence of original data; It is the original data, including the new output target , the amount of load to be removed , the frequency extracted by the proportional-integral controller And the voltage amplitude after filtering ; and are the parameters of the model;
[0088] Using raw data Train the model. and It is estimated by the least squares method. The specific method is:
[0089] make, ;
[0090] ;
[0091] Then the parameter and The estimated value of is:
[0092] ;
[0093] is the vector associated with the original data; is the matrix associated with the accumulated data; for The inverse matrix of
[0094] S72. Predict the grid frequency. Predict the grid frequency at a future time based on the original data. The specific prediction formula is:
[0095] ;
[0096] in, To predict the grid frequency at future times, is a cumulative sequence obtained based on the model prediction;
[0097] S73. Evaluate model accuracy. Use mean square error and absolute mean error to evaluate prediction accuracy.
[0098] The formula for calculating the mean square error is:
[0099] ;
[0100] is the frequency extracted by the proportional-integral controller, The predicted grid frequency at time The value of is the number of samples; the mean square error reflects the average square error between the predicted value and the actual value. The smaller the value, the higher the prediction accuracy;
[0101] The absolute mean error is calculated as:
[0102] ;
[0103] The absolute mean error reflects the average absolute error between the predicted value and the actual value. The smaller the value, the smaller the deviation between the predicted value and the actual value, and the higher the prediction accuracy. is the frequency extracted by the proportional-integral controller, The predicted grid frequency at time The value of is the sample size;
[0104] S74. Analyze the supporting effect, and analyze the supporting effect of the wind farm on the grid frequency through the deviation reduction degree and recovery time;
[0105] The degree of deviation reduction is to evaluate the effect of wind farm output adjustment on reducing frequency deviation by comparing the grid frequency deviation before and after wind farm output adjustment. The greater the degree of deviation reduction, the better the wind farm's support effect on grid frequency, and the more effectively it can stabilize grid frequency.
[0106] The recovery time is measured by observing the time required for the frequency to return to the normal range. The shorter the recovery time, the faster the wind farm can restore the grid frequency to stability, which means that the wind farm can provide better support for the grid frequency.
[0107] In summary, due to the adoption of the above technical solution, the beneficial technical effects of the invention are:
[0108] The adaptive distributed coordinated control method of wind farms with rapid grid frequency support improves the stability of grid frequency. By collecting basic data on wind farm and grid operation, including substation bus voltage, outgoing line interval status, etc., the operating status of the grid can be fully understood.
[0109] The electrical characteristic quantities are analyzed to obtain characteristic quantities such as voltage amplitude, frequency and phase, providing accurate data support for subsequent judgment of the output characteristics of the wind farm and the operating status of the power grid.
[0110] The judgment of wind farm output characteristics and the decision on frequency and voltage emergency control actions can timely detect the deviation of grid frequency and take corresponding control measures, such as adjusting the power generation of wind farms, starting backup power generation resources or implementing load management strategies, thereby effectively reducing frequency deviation and improving the stability of grid frequency.
[0111] The adaptive distributed coordinated control method for wind farms with fast grid frequency support realizes adaptive load shedding and distributed coordinated control. The adaptive control strategy is formulated according to the needs of the grid, and the load shedding calculation formula and load selection method are used to realize adaptive load shedding. By excluding the distribution network branches from the power supply to the grid and reasonably selecting the load to be removed, it can ensure that the grid can operate quickly and accurately when load shedding is needed, reducing unnecessary energy waste.
[0112] Distributed coordinated control execution generates control instructions for wind turbines based on the adaptive load reduction strategy and sends these instructions to the controllers of each wind turbine, thus realizing distributed coordinated control. This control method can ensure that the wind farm can quickly respond to the emergency control needs of the power grid and improve the wind farm's support capacity for the power grid.
[0113] The adaptive distributed coordinated control method for wind farms with fast grid frequency support optimizes the control strategy of wind farms. The frequency support capability evaluation uses the results of steps S1 to S6 to establish an evaluation model to evaluate the supporting effect of wind farm output adjustment on grid frequency. By using mean square error and absolute mean error to evaluate the prediction accuracy, the supporting effect of wind farms on grid frequency can be accurately understood.
[0114] Based on the frequency support capacity assessment results, the control strategy optimization and adjustment analyzes possible problems in the control strategy and adjusts relevant parameters. For example, the control parameters of the wind turbine (such as the parameters related to the pitch angle adjustment and the speed adjustment), the strategy parameters of the load shedding (such as the weight coefficient, etc.), and the threshold parameters related to the grid frequency and voltage are adjusted according to the assessment results. The adjusted control strategy is applied to the distributed coordinated control of the wind farm, which optimizes and updates the control strategy and enables the wind farm to better adapt to the changes and needs of the grid operation. BRIEF DESCRIPTION OF THE DRAWINGS
[0115] Figure 1 Flowchart of the adaptive distributed coordinated control method for wind farms to provide rapid support for grid frequency;
[0116] Figure 2 is a flow chart of a method for obtaining voltage amplitude characteristic quantities;
[0117] Figure 3 A flow chart of a method for determining output characteristics of a wind farm;
[0118] Figure 4 Develop a method flow chart for adaptive load shedding strategy;
[0119] Figure 5 A flow chart of a method for distributed coordinated control;
[0120] Figure 6A flow chart of a method for evaluating the supporting effect of wind farm output adjustment on grid frequency; DETAILED DESCRIPTION
[0121] In order to make the purpose, technical solution and advantages of the invention more clear, the invention is further described in detail below in combination with the embodiments. It should be understood that the specific embodiments described here are only used to explain the invention and are not used to limit the invention.
[0122] like Figure 1 As shown, the wind farm adaptive distributed coordinated control method with fast grid frequency support includes the following steps:
[0123] S1. Collect basic data of wind farm and power grid operation, including substation bus voltage, outgoing line bay status data, current component of three-phase AC signal, and current and voltage signals of three-phase AC power, etc.;
[0124] S2. Electrical characteristic quantity analysis: Based on the data collected in S1, characteristic quantities such as voltage amplitude, frequency and phase are obtained. These characteristic quantities will be used to judge the output characteristics of the wind farm and the operating status of the power grid in the future;
[0125] like Figure 2 As shown, in S2, the method for obtaining the voltage amplitude characteristic quantity includes:
[0126] S21. Coordinate transformation, convert the three-phase AC signal collected by S1 from the three-phase stationary coordinate system (abc coordinate system) to the vector rotating space coordinate system (dq coordinate system), the formula is:
[0127] ;
[0128] In the formula, and Respectively expressed in (straight axis) and The current component on the axis (quadrature axis); , , The three-phase current is , , The weight of the phase; is the rotation angle, which is related to the phase of the grid;
[0129] S22. Use a low-pass filter to filter the coordinates after the transformation. (straight axis) and The voltage component on the axis (quadrature axis) is filtered to obtain the voltage amplitude, the formula is:
[0130] ;
[0131] ;
[0132] ;
[0133] ;
[0134] ;
[0135] ;
[0136] is the voltage amplitude obtained after filtering; is the rotation angle; and After coordinate transformation, (straight axis) and The voltage component on the axis (quadrature axis); , , To convert the components of the voltage in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system;
[0137] In S2, the method for obtaining the frequency characteristic quantity is:
[0138] The voltage signal is processed by a proportional-integral controller to extract the frequency. The formula is:
[0139] ;
[0140] in, is the frequency extracted by the proportional-integral controller; is the error signal (e.g. the difference between the actual voltage and the set voltage), which reflects the degree of deviation of the system. The controller adjusts the output according to the error signal to reduce the deviation; is the proportional coefficient, which determines the instantaneous response of the controller to the error signal. A value of will make the controller more sensitive to the error signal and be able to adjust the output more quickly, but it may cause system instability and oscillation; is the integral coefficient, which is used to eliminate the steady-state error of the system. The value will make the integral effect weaker, and a larger The excessive value may lead to integral saturation, causing the system to respond more slowly or even lose stability.
[0141] In S2, the method for obtaining the phase characteristic quantity is:
[0142] The integrated phase is obtained by integrating the frequency signal, and the formula is:
[0143] ;
[0144] is the frequency extracted by the proportional-integral controller, The integrated phase is obtained by integration, and the integrated phase signal reflects the phase information of the voltage signal.
[0145] S3. Determine the output characteristics of the wind farm, calculate the active power according to the electrical characteristic quantity extracted in S2, and determine the output characteristics of the wind farm by combining the calculation result of the active power and the power duration;
[0146] like Figure 3 As shown, in S3, determining the output characteristics of the wind farm includes the following steps:
[0147] S31. Calculate the active power. The calculation formula for active power is:
[0148] ;
[0149] ;
[0150] ;
[0151] ;
[0152] ;
[0153] ;
[0154] ;
[0155] ;
[0156] ;
[0157] ;
[0158] ;
[0159] is the voltage amplitude obtained after filtering; is the direct axis of the current in the vector rotation space coordinate system The amount of , , To convert the components of the voltage in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system; and After coordinate transformation, and Voltage component on the axis; , , To convert the components of the current in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system; and is the component of the current in the vector rotation space coordinate system; is the rotation angle; is the voltage phase; is the current phase; is the phase difference between voltage and current;
[0160] S32. Determine the output characteristics of the wind farm based on the calculation results of active power and the duration of power,
[0161] like And remain stable for a period of time , then it is determined to be a power supply characteristic; if And keep this state for the same time , it is determined as a load characteristic; during the judgment process, the changes in active power are continuously monitored to ensure that the duration requirements are met. ;
[0162] S4. Frequency and voltage emergency control action decision, based on the electrical characteristics such as frequency and voltage obtained in S2 and the output characteristics of the wind farm determined in S3, determine whether the power grid is in a state requiring emergency control;
[0163] In S4, determining whether the power grid is in a state requiring emergency control includes:
[0164] By comparing the absolute value of the difference between the frequency extracted by the proportional-integral controller and the standard frequency to see if it exceeds the preset threshold, When , it indicates that the absolute value of the difference between the frequency extracted by the proportional-integral controller and the standard frequency exceeds the preset threshold, indicating that the frequency extracted by the proportional-integral controller has a large deviation, and emergency control actions need to be taken to adjust the grid frequency; among them, is the frequency extracted by the proportional-integral controller, To set the standard frequency, is the preset frequency deviation threshold;
[0165] Whether the absolute value of the difference between the voltage amplitude obtained after filtering and the set standard voltage amplitude is greater than the preset voltage amplitude deviation threshold, When , it indicates that the absolute value of the difference between the voltage amplitude obtained after filtering and the set standard voltage amplitude is greater than the preset voltage amplitude deviation threshold, indicating that the voltage amplitude obtained after filtering is abnormal and emergency control action is required to ensure that the voltage is stable within a reasonable range; among them, is the voltage amplitude obtained after filtering, The set standard voltage amplitude, is a preset voltage amplitude deviation threshold;
[0166] Comparison of the rate of change of frequency and voltage amplitude. and , indicating that the change rates of frequency and voltage have exceeded the corresponding preset change rate thresholds, indicating that the state of the power grid changes rapidly and that emergency control actions are required to stabilize the power grid; and are the rates of change of frequency and voltage respectively, and is a corresponding preset change rate threshold;
[0167] Emergency control measures may include, but are not limited to: adjusting the generation of wind farms to provide necessary frequency support; activating backup generation resources to quickly respond to grid needs; and implementing load management strategies, such as demand response or load shedding, to reduce grid load.
[0168] S5. Adaptive control strategy formulation: According to the judgment result of step 4, when the power grid needs to reduce load, the load reduction calculation formula is used, and according to the outgoing line interval data and the priority of the load, the appropriate load is selected for removal to achieve adaptive load reduction;
[0169] like Figure 4 As shown, the adaptive load shedding strategy formulation method includes:
[0170] S51. Load reduction calculation, considering the impact of power loss and surplus, using weighted coefficient to calculate load reduction:
[0171] ;
[0172] is the amount of load to be removed; and is the weight coefficient used to adjust the impact of power loss and surplus; is the current power loss of the grid; is the current power surplus of the grid;
[0173] S52. Load selection:
[0174] S521. Exclude the distribution network branch with power supply to the Internet, exclude the distribution network branch with power supply to the Internet from all the distribution network branches of the substation, and use the remaining distribution network branches as candidate distribution network branches;
[0175] Assume that the set of all distribution network branches of the substation is: ,in, Indicates Strip distribution network branch;
[0176] The distribution network branches for power supply to the Internet are as follows: ;
[0177] The candidate distribution network branch set is: ;
[0178] S522. Initialize loop variables and accumulate load power, set loop variables , initialize the loop variable , set Article The load power corresponding to the distribution network branch is , then the accumulated distribution network branch load power is ;
[0179] S523. Determine whether the accumulated distribution network branch load power is greater than the load reduction amount:
[0180] Is it established? If , then execute the next step, loop variable Add 1 to form a new loop variable, that is ; The new Load power corresponding to the distribution network branch , added to Repeat the judgment until Established;
[0181] During the process, once If established, then record the first to the Distribution network branch collection and sends tripping commands to these distribution network branches to cut off the load.
[0182] S6. Distributed coordinated control execution, generating control instructions for the wind turbines according to the adaptive load reduction strategy formulated in S5, and sending these instructions to the controllers of each wind turbine to implement distributed coordinated control to ensure that the wind farm can quickly respond to the emergency control needs of the power grid;
[0183] like Figure 5 As shown, in S6, the distributed coordinated control method includes:
[0184] S61. Determine the control target. According to the emergency control requirements of the power grid and the current output of the wind turbines, determine the new output target that each wind turbine should achieve to reduce load or meet other requirements of the power grid; the calculation formula for the new output target is:
[0185] ;
[0186] In the formula, For the new output goal, is the current output of the wind turbine, is the load amount to be removed;
[0187] S62. Calculate the adjustment amount. Based on the performance parameters and load reduction requirements of the wind turbine, calculate the output amount that needs to be adjusted for each wind turbine to ensure that the wind turbine can accurately respond to the control target. The calculation formula for the output amount that needs to be adjusted for the wind turbine is:
[0188] ;
[0189] In the formula, The output of the wind turbine needs to be adjusted; is the load amount to be removed; It is a coefficient determined according to the performance parameters of the wind turbine. The type, capacity, efficiency and other factors of the wind turbine are comprehensively considered to ensure that the calculated adjustment matches the actual capacity of the wind turbine;
[0190] S63. Generate control instructions. Generate specific control instructions based on the adjustment amount calculated by S62, so that the wind turbine can adjust the output to meet the control requirements;
[0191] The control instructions include pitch angle adjustment and speed adjustment, etc.
[0192] For a fixed pitch wind turbine, the pitch angle adjustment is:
[0193] ;
[0194] For variable pitch wind turbines, the speed adjustment is:
[0195] ;
[0196] in, is the pitch angle adjustment amount; The output of the wind turbine needs to be adjusted; is the speed adjustment amount; and are proportional coefficients, It reflects the effect of pitch angle change on power output. It reflects the effect of speed change on power output.
[0197] S7. Frequency support capability assessment: Use the results of S1 to S6 to establish an assessment model, and use the assessment model to assess the support effect of wind farm output adjustment on grid frequency;
[0198] like Figure 6 As shown, the method for evaluating the effect of wind farm output adjustment on grid frequency support through an evaluation model includes:
[0199] S71. Establish a model, which is represented by:
[0200] ;
[0201] ;
[0202] in, It is an accumulated sequence of original data; It is the original data, including the new output target , the amount of load to be removed , the frequency extracted by the proportional-integral controller And the voltage amplitude after filtering ; and are the parameters of the model;
[0203] Using raw data Train the model. and It is estimated by the least squares method. The specific method is:
[0204] make, ;
[0205] ;
[0206] Then the parameter and The estimated value of is:
[0207] ;
[0208] is the vector associated with the original data; is the matrix associated with the accumulated data; for The inverse matrix of
[0209] S72. Predict the grid frequency. Predict the grid frequency at a future time based on the original data. The specific prediction formula is:
[0210] ;
[0211] in, To predict the grid frequency at future times, is a cumulative sequence obtained based on the model prediction;
[0212] S73. Evaluate model accuracy. Use mean square error and absolute mean error to evaluate prediction accuracy.
[0213] The formula for calculating the mean square error is:
[0214] ;
[0215] is the frequency extracted by the proportional-integral controller, The predicted grid frequency at time The value of is the number of samples; the mean square error reflects the average square error between the predicted value and the actual value. The smaller the value, the higher the prediction accuracy;
[0216] The absolute mean error is calculated as:
[0217] ;
[0218] The absolute mean error reflects the average absolute error between the predicted value and the actual value. The smaller the value, the smaller the deviation between the predicted value and the actual value, and the higher the prediction accuracy. is the frequency extracted by the proportional-integral controller, The predicted grid frequency at time The value of is the sample size;
[0219] S74. Analyze the supporting effect, and analyze the supporting effect of the wind farm on the grid frequency through the deviation reduction degree and recovery time;
[0220] The degree of deviation reduction is to evaluate the effect of wind farm output adjustment on reducing frequency deviation by comparing the grid frequency deviation before and after wind farm output adjustment. The greater the degree of deviation reduction, the better the wind farm's support effect on grid frequency, and the more effectively it can stabilize grid frequency.
[0221] The recovery time is measured by observing the time required for the frequency to return to the normal range. The shorter the recovery time, the faster the wind farm can restore the grid frequency to stability, which means that the wind farm's support effect on the grid frequency is better.
[0222] The above description is a preferred embodiment of the invention and is not intended to limit the invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the invention should be included in the protection scope of the invention.
Claims
1. A wind farm adaptive distributed coordinated control method for fast grid frequency support, characterized in that: The following steps are involved: S1. Data collection: collecting basic data of wind farm and power grid operation; S2. Electrical characteristic quantity analysis, based on the data collected in S1, obtain the voltage amplitude, frequency and phase characteristic quantities; S3. Determine the output characteristics of the wind farm, calculate the active power according to the electrical characteristic quantity obtained in S2, and determine the output characteristics of the wind farm in combination with the calculation result of the active power and the power duration; S4. Frequency and voltage emergency control action decision, based on the electrical characteristic quantity obtained in S2 and the output characteristics of the wind farm determined in S3, determine whether the power grid is in a state requiring emergency control; S5. Adaptive control strategy formulation: According to the judgment result of S4, when the power grid needs to reduce load, the load reduction calculation formula is used, and the appropriate load is selected for removal based on the outgoing line interval data and the priority of the load to achieve adaptive load reduction; S6. Distributed coordinated control execution, according to the adaptive control strategy formulated in step S5, generates control instructions for the wind turbines, and sends these instructions to the controllers of each wind turbine to achieve distributed coordinated control; S7. Frequency support capacity assessment: Use the results of S1 to S6 to establish an assessment model, and use the assessment model to assess the impact of wind farm output adjustment on power grid The supporting effect of network frequency; In S5, the method for formulating the adaptive load reduction strategy includes: S51. Load reduction calculation, considering the impact of power loss and surplus, using weighted coefficient to calculate load reduction: ; is the amount of load to be removed; and is the weight coefficient used to adjust the impact of power loss and surplus; is the current power loss of the grid; is the current power surplus of the grid; S52. Load selection: S521. Exclude the distribution network branch with power supply connected to the grid, exclude the distribution network branch with power supply connected to the grid from all the distribution network branches of the substation, and use the remaining distribution network branches as candidate distribution network branches. The specific method includes: Assume that the set of all distribution network branches of the substation is: ,in, Indicates Strip distribution network branch; The distribution network branches for power supply to the Internet are as follows: ; The candidate distribution network branch set is: ; S522. Initialize loop variables and accumulate load power, set loop variables , initialize the loop variable , set Article The load power corresponding to the distribution network branch is , then the accumulated distribution network branch load power is ; S523. Determine whether the accumulated distribution network branch load power is greater than the load reduction amount: Is it established? If , then execute the next step, loop variable Add 1 to form a new loop variable, that is ; The new Load power corresponding to the distribution network branch , added to Repeat the judgment until Established; During the process, once If established, then record the first to the Distribution network branch collection and sends tripping commands to these distribution network branches to cut off the load.
2. The wind farm adaptive distributed coordinated control method for grid frequency rapid support according to claim 1 is characterized in that: In S2, the method for obtaining the voltage amplitude characteristic quantity includes: S21. Coordinate transformation, convert the three-phase AC signal collected by S1 from the three-phase stationary coordinate system to the vector rotating space coordinate system. The formula is: ; In the formula, and Respectively expressed in and The current component on the axis; , , The three-phase current is , , The weight of the phase; is the rotation angle, which is related to the phase of the grid; S22. Use a low-pass filter to filter the coordinates after the transformation. and The voltage component on the axis is filtered to obtain the voltage amplitude, the formula is: ; ; ; ; ; ; is the voltage amplitude obtained after filtering; is the rotation angle; and After coordinate transformation, and Voltage component on the axis; , , To convert the components of the voltage in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system.
3. The wind farm adaptive distributed coordinated control method for rapid grid frequency support according to claim 1 is characterized in that: In S2, the method for obtaining the frequency characteristic quantity is: The voltage signal is processed by a proportional-integral controller to extract the frequency. The formula is: ; in, is the frequency extracted by the proportional-integral controller; is the error signal, which is the difference between the actual voltage and the set voltage; is the proportionality coefficient; is the integration coefficient.
4. The wind farm adaptive distributed coordinated control method for rapid grid frequency support according to claim 1 is characterized in that: In S2, the method for obtaining the phase characteristic quantity is: The integrated phase is obtained by integrating the frequency signal, and the formula is: ; is the frequency extracted by the proportional-integral controller, The integrated phase is obtained by integration, and the integrated phase signal reflects the phase information of the voltage signal.
5. The wind farm adaptive distributed coordinated control method for rapid grid frequency support according to claim 1 is characterized in that: In S3, determining the output characteristics of the wind farm includes the following steps: S31. Calculate the active power. The calculation formula for active power is: ; ; ; ; ; ; ; ; ; ; ; is the voltage amplitude obtained after filtering; is the direct axis of the current in the vector rotation space coordinate system The weight of , , To convert the components of the voltage in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system; and After coordinate transformation, and Voltage component on the axis; , , To convert the components of the current in the vector rotating space coordinate system back to the components in the three-phase stationary coordinate system; and is the component of the current in the vector rotation space coordinate system; is the rotation angle; is the voltage phase; is the current phase; is the phase difference between voltage and current; S32. Determine the output characteristics of the wind farm based on the calculation results of active power and the duration of power, like And remain stable for a period of time , then it is determined to be a power supply characteristic; if And keep this state for the same time , it is determined to be a load characteristic. .
6. The wind farm adaptive distributed coordinated control method for grid frequency rapid support according to claim 1 is characterized in that: In S4, determining whether the power grid is in a state requiring emergency control includes: By comparing the absolute value of the difference between the frequency extracted by the proportional-integral controller and the standard frequency to see if it exceeds the preset threshold, When , it indicates that the absolute value of the difference between the frequency extracted by the proportional-integral controller and the standard frequency exceeds the preset threshold, indicating that the frequency extracted by the proportional-integral controller has a large deviation, and emergency control actions need to be taken to adjust the grid frequency; among them, is the frequency extracted by the proportional-integral controller, To set the standard frequency, is the preset frequency deviation threshold; Whether the absolute value of the difference between the voltage amplitude obtained after filtering and the set standard voltage amplitude is greater than the preset voltage amplitude deviation threshold, When , it indicates that the absolute value of the difference between the voltage amplitude obtained after filtering and the set standard voltage amplitude is greater than the preset voltage amplitude deviation threshold, indicating that the voltage amplitude obtained after filtering is abnormal and emergency control action is required to ensure that the voltage is stable within a reasonable range; among them, is the voltage amplitude obtained after filtering, The set standard voltage amplitude, is a preset voltage amplitude deviation threshold; Comparison of the rate of change of frequency and voltage amplitude. and , indicating that the change rates of frequency and voltage have exceeded the corresponding preset change rate thresholds, indicating that the state of the power grid changes rapidly and that emergency control actions are required to stabilize the power grid; and are the rates of change of frequency and voltage respectively, and is the corresponding preset change rate threshold.
7. The wind farm adaptive distributed coordinated control method for grid frequency rapid support according to claim 1 is characterized in that: In S6, the distributed coordinated control method includes: S61. Determine the control target. According to the emergency control requirements of the power grid and the current output of the wind turbines, determine the new output target that each wind turbine should achieve; wherein the calculation formula for the new output target is: ; In the formula, For the new output goal, is the current output of the wind turbine, The amount of load to be removed; S62. Calculate the adjustment amount. According to the performance parameters and load reduction requirements of the wind turbine generator set, calculate the output amount that needs to be adjusted for each wind turbine generator set. The calculation formula for the output amount that needs to be adjusted for the wind turbine generator set is: ; In the formula, The output of the wind turbine needs to be adjusted. is the amount of load to be removed, It is a coefficient determined according to the performance parameters of the wind turbine. The type, capacity and efficiency of wind turbines were considered comprehensively; S63. Generate control instructions. Generate specific control instructions based on the adjustment amount calculated by S62, so that the wind turbine can adjust the output to meet the control requirements; The control instructions include pitch angle adjustment and speed adjustment. For a fixed pitch wind turbine, the pitch angle adjustment is: ; For variable pitch wind turbines, the speed adjustment is: ; in, is the pitch angle adjustment amount; The output of the wind turbine needs to be adjusted; is the speed adjustment amount; and All are proportional coefficients.
8. The wind farm adaptive distributed coordinated control method for grid frequency rapid support according to claim 1 is characterized in that: The method for evaluating the supporting effect of wind farm output adjustment on grid frequency through an evaluation model includes: S71. Establish a model, which is represented by: ; ; in, It is an accumulated sequence of original data; It is the original data, including the new output target , the amount of load to be removed , the frequency extracted by the proportional-integral controller And the voltage amplitude after filtering ; and are the parameters of the model; Using raw data Train the model. and It is estimated by the least squares method. The specific method is: make, ; ; Then the parameter and The estimated value of is: ; is the vector associated with the original data; is the matrix associated with the accumulated data; for The inverse matrix of S72. Predict the grid frequency. Predict the grid frequency at a future time based on the original data. The specific prediction formula is: ; in, To predict the grid frequency at future times, is a cumulative sequence obtained based on the model prediction; S73. Evaluate model accuracy. Use mean square error and absolute mean error to evaluate prediction accuracy. The formula for calculating the mean square error is: ; is the frequency extracted by the proportional-integral controller, is the predicted value of the grid frequency at time, is the number of samples; the mean square error reflects the average square error between the predicted value and the actual value. The smaller the value, the higher the prediction accuracy; The absolute mean error is calculated as: ; The absolute mean error reflects the average absolute error between the predicted value and the actual value. The smaller the value, the smaller the deviation between the predicted value and the actual value, and the higher the prediction accuracy. is the frequency extracted by the proportional-integral controller, is the predicted grid frequency at time The value of is the sample size; S74. Analyze the supporting effect, and analyze the supporting effect of the wind farm on the grid frequency through the deviation reduction degree and recovery time; The degree of deviation reduction is to evaluate the effect of wind farm output adjustment on reducing frequency deviation by comparing the grid frequency deviation before and after wind farm output adjustment. The greater the degree of deviation reduction, the better the wind farm's support effect on grid frequency, and the more effectively it can stabilize grid frequency. The recovery time is measured by observing the time required for the frequency to return to the normal range. The shorter the recovery time, the faster the wind farm can restore the grid frequency to stability, indicating that the wind farm has a better support effect on the grid frequency.
