CN116505556B - Wind farm power control system and method based on primary frequency modulation - Google Patents
Wind farm power control system and method based on primary frequency modulation Download PDFInfo
- Publication number
- CN116505556B CN116505556B CN202310603777.6A CN202310603777A CN116505556B CN 116505556 B CN116505556 B CN 116505556B CN 202310603777 A CN202310603777 A CN 202310603777A CN 116505556 B CN116505556 B CN 116505556B
- Authority
- CN
- China
- Prior art keywords
- power output
- control
- wind
- power
- generating set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000005457 optimization Methods 0.000 claims abstract description 17
- 238000012423 maintenance Methods 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000013459 approach Methods 0.000 claims description 4
- 238000010248 power generation Methods 0.000 abstract description 8
- 238000004364 calculation method Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
- H02J3/241—The oscillation concerning frequency
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
Abstract
The invention belongs to the technical field of wind power plant control, and particularly relates to a wind power plant power control system and method based on primary frequency modulation. The system comprises a sensor module, a control module, a power output optimization module and a power classification and adjustment module. The control module adopts a primary frequency modulation controller and a fuzzy logic controller to calculate the maximum power output of the wind generating set, and realizes the control and the stability maintenance of the power output by adjusting the power output of the wind generating set. The power output optimization module classifies the optimized power output by using a classifier, and adjusts the control output of the wind driven generator according to the classification. The invention can improve the power output efficiency of the wind driven generator, and can reduce the running cost and the maintenance cost of the wind driven generator, thereby realizing more reliable and economic power generation.
Description
Technical Field
The invention belongs to the technical field of wind power plant control, and particularly relates to a wind power plant power control system and method based on primary frequency modulation.
Background
In recent years, wind energy is a clean and renewable energy resource, and is attracting attention and importance due to the increase of environmental problems and energy problems. Wind power generation has been widely used worldwide as an important clean energy source. In the field of wind power generation, wind farm power control technology is an important research direction.
The traditional wind power plant power control method mainly controls the output power of a wind power plant by adjusting the current or the rotating speed of a generator. However, this control method has many problems such as difficulty in adjusting power output in real time when wind speed changes, and thus a new control method is required to solve these problems.
In the prior art, us patent document US9,712,705B2 discloses a wind farm power control system which controls the output power of a wind farm by adjusting the blade angle of a wind turbine. However, this method can only control under the condition that the wind speed changes slowly, and once the wind speed changes too fast, it is difficult to adjust the blade angle in time, so that effective power control cannot be achieved.
Secondly, chinese patent document CN103574381a discloses a wind farm power control system based on fuzzy control, which controls the output power of the wind farm through a fuzzy controller. Compared with the traditional wind farm power control method, the method has certain improvement, but the method needs a large amount of real-time data input, and needs to study and test the conditions of various different wind farms, so that the cost is high and the application range is limited.
Finally, german patent document DE102013111817A1 discloses a wind farm power control method based on a genetic algorithm, which optimally controls the output power of the wind farm by means of the genetic algorithm. Although the method has a certain optimizing effect, the method has large calculation amount, is difficult to apply in real-time control, has high requirements on hardware and has relatively high cost.
In summary, the existing wind farm power control method has many problems, such as difficulty in adjusting output power in real time, requirement for a large amount of real-time data input, and large calculation amount. Therefore, a new wind farm power control approach is needed to address these issues.
Disclosure of Invention
The invention mainly aims to provide a wind farm power control system and method based on primary frequency modulation, which can realize the accurate control and optimization of a wind driven generator, improve the efficiency and reliability of wind power generation, reduce the running cost and maintenance cost and have important significance for promoting the development of clean energy and building low-carbon economy.
In order to solve the technical problems, the invention provides a wind farm power control system based on primary frequency modulation, comprising:
a primary frequency modulation based wind farm power control system, the system comprising: a sensor module for measuring the wind speed and the power output of the wind generating set; a control module for controlling a power output of a wind turbine, wherein the control module comprises: the primary frequency modulation controller is used for calculating the maximum power output of the wind generating set; a fuzzy logic controller for controlling the power output of the wind generating set according to the calculated maximum power output of the wind generating set to approach the maximum power output; the power output optimization module is used for optimizing the power output to obtain optimized power output; and the power classifying and adjusting module is configured to classify the power output by using a classifier, determine the category to which the current optimized power output belongs, and adjust the control output of the wind driven generator according to the category so as to realize the control and stability maintenance of the power output.
Further, the power output optimizing module is configured to optimize power output, and the process of optimizing power output includes: the power output is optimized using the following formula:
wherein P is s (t) represents the optimized power output, N w The optimization window size is represented, and V is the time within the optimization window.
Further, the power classifying and adjusting module classifies the optimized power output by using a classifier, and the method for determining the category to which the current optimized power output belongs comprises the following steps: extracting the characteristics of the power output of the wind speed and wind generating set to obtain a characteristic vector reflecting the power output change; marking the data according to the classification standard of the feature vector to obtain corresponding category information; training the marked data by using a classifier to obtain model parameters of the classifier; classifying the power output data acquired in real time by adopting a trained classifier, and determining the category to which the current power output belongs; and adjusting the control output of the wind driven generator according to the classification result of the power output so as to realize the control and stability maintenance of the power output.
Further, the classifier is expressed using the following formula:
wherein P is state (t) represents the power state at the current time, W is a weight vector, φ (P) s (t)) is a feature vector, b is a bias term, and argmax represents a state corresponding to the maximum value.
A wind farm power control method based on primary frequency modulation comprises the following steps:
step 1: acquiring current wind speed and power output of a wind generating set from a sensor module;
step 2: calculating the maximum power output of the wind generating set by using the primary frequency modulation controller;
step 3: calculating the control quantity of the next moment by using a preset multivariable controller based on the current wind speed and power error;
step 4: taking the calculated control quantity as output, and adjusting the power output of the wind generating set;
step 5: and (5) continuously repeating the steps 1-4 to realize the power control of the wind generating set.
Further, the method for calculating the maximum power output of the wind generating set by using the primary frequency modulation controller in the step 2 includes: the maximum power output of the wind turbine is calculated using the following formula:
wherein P is max For the maximum power output of the wind generating set, ρ is the air density, A is the blade area, V rated For rated wind speed, C p Is the power coefficient lambda opt Is the optimal torque coefficient.
Further, the expression of the multivariable controller is:
wherein P (t+1) is the power output at the next moment, u (t) is the control quantity at the current moment, N is the length of the control time domain, and alpha s And beta s Is a weight coefficient; s is the time step.
Further, the optimization objective of the multi-variable controller is to minimize the following cost function:
wherein P is ref (t+k) is the reference power output at the next time, P (t+k) is the power output at the predicted time, u (t+k) is the control amount, N is the length of the control time domain, N a The length of the time domain is the control quantity; k is a time index in a control time domain window; the first term of the cost function (P ref (t+k)-P(t+k)) 2 For describing the error between the predicted power output and the reference power output, the smaller the error, the smaller the cost; second term of cost functionFor describing the magnitude of the control quantity, if the control quantity is too large or too small, the cost is correspondingly increased, and the control quantity constraint of the multivariable controller is as follows: u (u) min ≤u(t)≤u max
Wherein u is min And u max Minimum and maximum values for the control quantity; the state constraints of the multivariable controller are: p (P) min ≤P(t)≤P max ；
Wherein P is min And P max Is the minimum and maximum of the power output.
Further, the method for adjusting the control quantity by the multivariable controller according to the prediction error comprises the following steps:
the control amount adjusted according to the error is calculated using the following formula:
wherein u (t) represents the control amount at the current time, γ represents the learning rate, P (t-k) and P ref (t-k) represents the predicted power and the reference power within the kth window, respectively.
Further, the method for predicting the wind speed by the multivariable controller through the sliding window comprises the following steps:
the predicted wind speed is calculated using the following formula:
wherein V is forecast (t + k) represents the predicted wind speed at time k,for the current wind speed, p is the autoregressive order, phi i For the autoregressive coefficients, q is the moving average order, θ j For the moving average coefficient +.>For the residual in the ith window, +.>Is the noise in the j-th window.
The wind farm power control system and method based on primary frequency modulation have the following beneficial effects: firstly, by adopting a power control method based on model predictive control, the power control method can accurately predict and control the output power of the wind driven generator. By establishing a mathematical model and using a prediction algorithm, the relation between the wind speed and the power output can be modeled and predicted, so that the power output of the wind driven generator is optimized, the power fluctuation and loss are reduced, and the power generation efficiency and stability are improved. Secondly, by adopting a power output classification method based on a classifier, the power output monitoring and classifying device can realize real-time monitoring and classification of power output. The sensor data of the wind driven generator is collected and processed, the characteristic vector of the power output is extracted, and is classified by a classifier, so that the current power output state can be rapidly and accurately identified, and corresponding control output adjustment is performed. The power output efficiency of the wind driven generator can be improved, and the running cost and the maintenance cost of the wind driven generator can be reduced, so that more reliable and economical power generation can be realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a system structure of a wind farm power control system based on primary frequency modulation according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a control module of a wind farm power control system based on primary frequency modulation according to an embodiment of the present invention
Fig. 3 is a schematic flow chart of a method for controlling power of a wind farm based on primary frequency modulation according to an embodiment of the present invention.
Detailed Description
The method of the present invention will be described in further detail with reference to the accompanying drawings.
Referring to fig. 1, 2 and 3, a primary frequency modulation based wind farm power control system, the system comprising: a sensor module for measuring the wind speed and the power output of the wind generating set; a control module for controlling a power output of a wind turbine, wherein the control module comprises: the primary frequency modulation controller is used for calculating the maximum power output of the wind generating set; a fuzzy logic controller for controlling the power output of the wind generating set according to the calculated maximum power output of the wind generating set to approach the maximum power output; the power output optimization module is used for optimizing the power output to obtain optimized power output; and the power classifying and adjusting module is configured to classify the power output by using a classifier, determine the category to which the current optimized power output belongs, and adjust the control output of the wind driven generator according to the category so as to realize the control and stability maintenance of the power output.
Specifically, the power output optimizing module is configured to optimize power output, where the process includes: the power output is optimized using the following formula:
wherein P is s (t) represents the optimized power output, N w The optimization window size is represented, and V is the time within the optimization window.
Specifically, the power classifying and adjusting module classifies the optimized power output by using a classifier, and the method for determining the category to which the current optimized power output belongs comprises the following steps: extracting the characteristics of the power output of the wind speed and wind generating set to obtain a characteristic vector reflecting the power output change; marking the data according to the classification standard of the feature vector to obtain corresponding category information; training the marked data by using a classifier to obtain model parameters of the classifier; classifying the power output data acquired in real time by adopting a trained classifier, and determining the category to which the current power output belongs; and adjusting the control output of the wind driven generator according to the classification result of the power output so as to realize the control and stability maintenance of the power output.
Specifically, the classifier expression is expressed using the following formula:
wherein P is state (t) represents the power state at the current time, W is a weight vector, φ (P) s (t)) is a feature vector, b is a bias term, and argmax represents a state corresponding to the maximum value.
Specifically, P in the formula state (t) represents the current power state, is a discretized variable, and can take a plurality of values.Indicating that in all possible power states, the selection enables W T φ(P s (t)) + b maximum state as the current power state. W is a weight vector used for representing the influence degree of different characteristics on the power state, and is one of core parameters of the classifier. Phi (P) s (t)) is a feature vector for mapping the power output into a high-dimensional space, thereby facilitating classification by the classifier. b is an offset term used to adjust the position and offset of the classification hyperplane, and is also a parameter of the classifier.
Therefore, the expression of the classifier means that the power state at the current moment is obtained through feature extraction and calculation of the classifier according to the power output at the current moment. According to the classification result of the power state, the control output of the wind driven generator can be adjusted so as to realize the control and stability maintenance of the power output. Specifically, according to the current power state, a corresponding control strategy, such as adjusting the rotating speed, the pitch angle and the like, can be adopted to realize the control and optimization of the power output.
A wind farm power control method based on primary frequency modulation comprises the following steps:
step 1: acquiring current wind speed and power output of a wind generating set from a sensor module;
step 2: calculating the maximum power output of the wind generating set by using the primary frequency modulation controller;
step 3: calculating the control quantity of the next moment by using a preset multivariable controller based on the current wind speed and power error;
step 4: taking the calculated control quantity as output, and adjusting the power output of the wind generating set;
step 5: and (5) continuously repeating the steps 1-4 to realize the power control of the wind generating set.
The multivariable controller of the present invention is an advanced control strategy commonly used to address multivariable, nonlinear and time-varying control problems. In a multivariable controller, the controller predicts the controlled object using a mathematical model, and then develops an optimal control strategy based on the predictions. Such a strategy would typically optimize an objective function to achieve the desired control effect while meeting various constraints.
Advantages of a multivariable controller include: strong control capability to multivariable, nonlinear and time-varying systems; various uncertainties and disturbances can be handled; the system response can be predicted, and an optimal control decision is made for the prediction result; the calculated amount and the control effect can be weighed by adjusting the prediction time domain and the control time domain; various constraints, such as output constraints, control quantity constraints, state constraints, etc., may be conveniently introduced.
Specifically, the method for calculating the maximum power output of the wind generating set by using the primary frequency modulation controller in the step 2 includes: the maximum power output of the wind turbine is calculated using the following formula:
wherein P is max For the maximum power output of the wind generating set, ρ is the air density, A is the blade area, V rated For rated wind speed, C p Is the power coefficient lambda opt Is the optimal torque coefficient.
Specifically, the expression of the multivariable controller is:
wherein P (t+1) is the power at the next timeOutput, u (t) is the control quantity at the current time, N is the length of the control time domain, alpha s And beta s Is a weight coefficient; s is the time step.
α s And beta s Are weight coefficients in a multivariable controller that are used to calculate the power output at the next time, as well as the magnitude of the control quantity. The manner in which these weight coefficients are calculated is generally based on a mathematical model of the system and the length of the predicted time domain and may be obtained in a number of ways, including:
linear regression: and fitting a weight coefficient by using the past data and the length of the prediction time domain through a linear regression method.
Least squares method: and solving the weight coefficient by using the past data and the length of the prediction time domain through a least square method.
Optimization-based methods: and obtaining the optimal weight coefficient by optimizing the cost function.
Whichever method is used, factors such as characteristics of the system, length of the prediction time domain, and computational complexity need to be considered. In practical applications, the optimal weight coefficient is usually obtained by performing multiple experiments and adjustments on the model and the data.
It should be noted that the calculation of the weight coefficient is a key element in a multivariable controller, and directly affects the control effect and stability of the controller. Therefore, in practical application, careful design and verification of the calculation method of the weight coefficient is required to ensure the reliability and performance of the controller.
The following is a specific process of calculating the weight coefficient:
modeling a system: and establishing a model prediction model, namely a model in a fourth formula, according to the characteristics and parameters of the actual system.
And (3) data acquisition: data in the actual running process is collected, including wind speed, power output, control quantity and the like.
And (3) data processing: the acquired data is processed and preprocessed, such as denoising, smoothing, normalizing, etc.
Dividing data: the processed data is divided into a training data set and a test data set.
Training a model: inputting the training data set into a model, and fitting out weight coefficients by using a least square method, namely:
wherein P (t) is the power output at the predicted time, u (t) is the control quantity, N is the length of the control time domain, and alpha k And beta k Is a weight coefficient.
The left matrix in the above equation is the design matrix, the right matrix is the response matrix, and the objective of the least squares method is to minimize the error between them.
Test model: the test dataset is input into the model, the predicted power output and control quantity are calculated and compared with the actual output to evaluate the performance and accuracy of the model.
A finite time domain window is typically used to predict system behavior over a period of time in the future. The length of this window is generally referred to as the control time domain length and is generally denoted by the symbol N. In this window, assuming that the current time is s, the time t+s represents a time that is s time steps from the current time, where s is a non-negative integer representing the time difference between time t and time t+s, also called the time step of the prediction horizon.
Typically, the time step s of the prediction horizon will be determined according to specific problems and systems, for example, according to factors such as response time of the system, complexity of the model, computational resources, and the like. In the model of a multivariable controller, the time step s of the prediction time domain is usually an adjustable parameter, and the optimal value can be determined through experiments or simulation so as to achieve better control effect and system performance.
Specifically, the optimization objective of the multi-variable controller is to minimize the following cost function:
wherein P is ref (t+k) is the reference power output at the next time, P (t+k) is the power output at the predicted time, u (t+k) is the control amount, N is the length of the control time domain, N u The length of the time domain is the control quantity; k is a time index in a control time domain window; the first term of the cost function (P ref (t+k)-P(t+k)) 2 For describing the error between the predicted power output and the reference power output, the smaller the error, the smaller the cost; second term of cost functionFor describing the magnitude of the control quantity, if the control quantity is too large or too small, the cost is correspondingly increased, and the control quantity constraint of the multivariable controller is as follows: u (u) min ≤u(t)≤u max
Wherein u is min And u max Minimum and maximum values for the control quantity; the state constraints of the multivariable controller are: p (P) min ≤P(t)≤P max ；
Wherein P is min And P max Is the minimum and maximum of the power output.
Specifically, the method for adjusting the control quantity by the multivariable controller according to the prediction error comprises the following steps:
the control amount adjusted according to the error is calculated using the following formula:
wherein u (t) represents the control amount at the current time, γ represents the learning rate, P (t-k) and P ref (t-k) represents the predicted power and the reference power within the kth window, respectively.
The cost function is an indicator used to measure system performance and control effectiveness. The cost function is typically made up of multiple parts, where each part corresponds to a particular objective or constraint. In the cost function, a number of parameters may be involved, including the k parameter in the cost function.
K in the cost function is typically a non-negative integer representing the time index in the control time domain window. In particular, terms about the prediction horizon k may be included in the cost function, which may be used to constrain the behavior of the control quantity and state variable over k time steps in the future. In practical applications, k in the cost function is generally determined according to specific problems and systems, for example, may be determined according to factors such as response time, complexity of a model, computing resources, and the like of the system.
Specifically, the method for predicting the wind speed by the multivariable controller through the sliding window comprises the following steps:
the predicted wind speed is calculated using the following formula:
wherein V is forecast (t + k) represents the predicted wind speed at time k,for the current wind speed, p is the autoregressive order, phi i For the autoregressive coefficients, q is the moving average order, θ j For the moving average coefficient +.>For the residual in the ith window, +.>Is the noise in the j-th window.
The purpose of predicting the wind speed is to timely adjust the control quantity of the wind driven generator when the wind speed changes so as to maintain the stability and consistency of the power output. The change of wind speed can affect the power output of the wind driven generator, so that when the wind speed changes, the control quantity needs to be adjusted to ensure the stability and controllability of the power output. Without a prediction of wind speed, it is difficult to accurately predict the power output of the wind turbine and the response time of the controller may be relatively long, which may affect the power output of the generator and the stability and performance of the overall wind power generation system.
By predicting the wind speed, the change trend of the wind speed in a period of time in the future can be known in advance, so that the control quantity is better adjusted, and the wind driven generator can stably output power under different wind speed conditions. Specifically, if the predicted outcome indicates that wind speed will drop, the controller will increase the control amount accordingly to maintain the stability of the power output; if the predicted outcome indicates that wind speed will rise, the controller will reduce the control accordingly to avoid overload of the power output or exceeding the maximum capacity.
Therefore, predicting wind speed is a very important ring in wind power generation systems, and can improve the efficiency and reliability of the system, thereby better meeting the power demand.
Specifically, the historical wind speed data is divided into a plurality of windows according to a time sequence, wherein each window comprises p+q data points, wherein p is an autoregressive order, and q is a moving average order. Within each window, a wind speed prediction model may be employed to fit and predict wind speed. The wind speed prediction model is a three-parameter model, comprising an autoregressive term, a differential term and a moving average term, and the wind speed can be predicted by adjusting the parameters.
The prediction formula of the wind speed prediction model is as follows:
wherein V is forecast (t + k) represents the predicted wind speed at time k,for the current wind speed phi i As autoregressive coefficient, θ j For the moving average coefficient +.>For the residual in the ith window, +.>Is the noise in the j-th window.
In the prediction process, the wind speed at the current moment is used as a prediction starting point, the autoregressive term and the moving average term in the historical data are used for prediction, the predicted value is used as a prediction starting point at the next moment, and the predicted value at the k moment is obtained through sequential recursion. Meanwhile, the autoregressive order and the moving average order are monitored and adjusted to adapt to different data and prediction requirements.
While specific embodiments of the present invention have been described above, it will be understood by those skilled in the art that these specific embodiments are by way of example only, and that various omissions, substitutions, and changes in the form and details of the methods and systems described above may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, it is within the scope of the present invention to combine the above-described method steps to perform substantially the same function in substantially the same way to achieve substantially the same result. Accordingly, the scope of the invention is limited only by the following claims.
Claims (4)
1. Wind farm power control system based on primary frequency modulation, characterized in that it comprises: a sensor module for measuring the wind speed and the power output of the wind generating set; a control module for controlling a power output of a wind turbine, wherein the control module comprises: the primary frequency modulation controller is used for calculating the maximum power output of the wind generating set; a fuzzy logic controller for controlling the power output of the wind generating set according to the calculated maximum power output of the wind generating set to approach the maximum power output; the power output optimization module is used for optimizing the power output to obtain optimized power output; the power classifying and adjusting module is configured to classify the power output by using a classifier, determine the category to which the current optimized power output belongs, and adjust the control output of the wind driven generator according to the category so as to realize the control and stability maintenance of the power output; the power output optimizing module is used for optimizing the power output, and the process of optimizing the power output comprises the following steps: the power output is optimized using the following formula:
wherein P is s (t) represents the optimized power output, N w Representing the size of the optimization window, V being the time within the optimization window;
the power classifying and adjusting module classifies the optimized power output by using a classifier, and the method for determining the category to which the current optimized power output belongs comprises the following steps: extracting the characteristics of the power output of the wind speed and wind generating set to obtain a characteristic vector reflecting the power output change; marking the data according to the classification standard of the feature vector to obtain corresponding category information; training the marked data by using a classifier to obtain model parameters of the classifier; classifying the power output data acquired in real time by adopting a trained classifier, and determining the category to which the current power output belongs; according to the classification result of the power output, adjusting the control output of the wind driven generator to realize the control and stability maintenance of the power output;
the classifier expression is expressed using the following formula:
wherein P is state (t) represents the power state at the current time, W is a weight vector, φ (P) s (t)) is a feature vector, b is a bias term, and argmax represents a state corresponding to the maximum value.
2. A primary frequency modulation based wind farm power control method based on the system of claim 1, comprising the steps of:
step 1: acquiring current wind speed and power output of a wind generating set from a sensor module;
step 2: calculating the maximum power output of the wind generating set by using the primary frequency modulation controller;
step 3: calculating the control quantity of the next moment by using a preset multivariable controller based on the current wind speed and power error;
step 4: taking the calculated control quantity as output, and adjusting the power output of the wind generating set;
step 5: continuously repeating the steps 1-4 to realize the power control of the wind generating set;
the expression of the multivariable controller is:
wherein P (t+1) is the power output at the next moment, u (t) is the control quantity at the current moment, N is the length of the control time domain, and alpha s And beta s Is a weight coefficient; s is the time step;
the optimization objective of the multivariable controller is to minimize the following cost function:
wherein P is ref (t+k) is the reference power output at the next time, P (t+k) is the power output at the predicted time, u (t+k) is the control amount, N is the length of the control time domain, N u The length of the time domain is the control quantity; k is a time index in a control time domain window; the first term of the cost function (P ref (t+k)-P(t+k)) 2 For describing the error between the predicted power output and the reference power output, the smaller the error, the smaller the cost; second term of cost functionFor describing the magnitude of the control quantity if the control quantity is excessiveThe cost is correspondingly increased when the control quantity of the multivariable controller is large or too small, and the control quantity constraint of the multivariable controller is as follows: u (u) min ≤u(t)≤u max
Wherein u is min And u max Minimum and maximum values for the control quantity; the state constraints of the multivariable controller are: p (P) min ≤P(t)≤P max ；
Wherein P is min And P max Minimum and maximum values of power output;
the method for adjusting the control quantity by the multivariable controller according to the prediction error comprises the following steps:
the control amount adjusted according to the error is calculated using the following formula:
wherein u (t) represents the control amount at the current time, γ represents the learning rate, and P (t-k) and P ref (t-k) represents the predicted power and the reference power within the kth window, respectively.
3. The method according to claim 2, wherein the method for calculating the maximum power output of the wind turbine using the primary frequency modulation controller in step 2 comprises: the maximum power output of the wind turbine is calculated using the following formula:
wherein P is max For the maximum power output of the wind generating set, ρ is the air density, A is the blade area, V rated For rated wind speed, C p Is the power coefficient lambda opt Is the optimal torque coefficient.
4. A method according to claim 3, wherein the method of predicting wind speed by the multivariable controller using a sliding window is:
the predicted wind speed is calculated using the following formula:
wherein V is forecast (t + k) represents the predicted wind speed at time k,for the current wind speed, p is the autoregressive order, phi i For the autoregressive coefficients, q is the moving average order, θ j For the moving average coefficient +.>As a residual within the i-th window,is the noise in the j-th window.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310603777.6A CN116505556B (en) | 2023-05-26 | 2023-05-26 | Wind farm power control system and method based on primary frequency modulation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310603777.6A CN116505556B (en) | 2023-05-26 | 2023-05-26 | Wind farm power control system and method based on primary frequency modulation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116505556A CN116505556A (en) | 2023-07-28 |
| CN116505556B true CN116505556B (en) | 2024-01-26 |
Family
ID=87326761
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310603777.6A Active CN116505556B (en) | 2023-05-26 | 2023-05-26 | Wind farm power control system and method based on primary frequency modulation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116505556B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
| WO2013174090A1 (en) * | 2012-05-22 | 2013-11-28 | 中国电力科学研究院 | Wind farm active power controlling method for improving the generating efficiency of the wind farm |
| CN103441529A (en) * | 2013-08-22 | 2013-12-11 | 国家电网公司 | Variable-speed wind turbine generator inertia response simulating control method |
| CN110768307A (en) * | 2019-11-11 | 2020-02-07 | 国网山东省电力公司日照供电公司 | Wind power plant primary frequency modulation control method and system |
| CN111946548A (en) * | 2020-08-18 | 2020-11-17 | 国网天津市电力公司电力科学研究院 | Control method for variable-pitch wind power generation system with primary frequency modulation function |
| CN112117768A (en) * | 2020-03-26 | 2020-12-22 | 广西大学 | Wind generating set subsection frequency modulation control method based on power tracking curve switching |
| CN113315166A (en) * | 2021-05-27 | 2021-08-27 | 国网河北省电力有限公司电力科学研究院 | Multi-virtual synchronous machine inertia configuration method and device and terminal equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3901712A1 (en) * | 2020-04-23 | 2021-10-27 | Siemens Gamesa Renewable Energy A/S | Method of operating a wind turbine |
-
2023
- 2023-05-26 CN CN202310603777.6A patent/CN116505556B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013174090A1 (en) * | 2012-05-22 | 2013-11-28 | 中国电力科学研究院 | Wind farm active power controlling method for improving the generating efficiency of the wind farm |
| CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
| CN103441529A (en) * | 2013-08-22 | 2013-12-11 | 国家电网公司 | Variable-speed wind turbine generator inertia response simulating control method |
| CN110768307A (en) * | 2019-11-11 | 2020-02-07 | 国网山东省电力公司日照供电公司 | Wind power plant primary frequency modulation control method and system |
| CN112117768A (en) * | 2020-03-26 | 2020-12-22 | 广西大学 | Wind generating set subsection frequency modulation control method based on power tracking curve switching |
| CN111946548A (en) * | 2020-08-18 | 2020-11-17 | 国网天津市电力公司电力科学研究院 | Control method for variable-pitch wind power generation system with primary frequency modulation function |
| CN113315166A (en) * | 2021-05-27 | 2021-08-27 | 国网河北省电力有限公司电力科学研究院 | Multi-virtual synchronous machine inertia configuration method and device and terminal equipment |
Non-Patent Citations (1)
| Title |
|---|
| 参与电网调频的风电场有功功率控制方法研究;刘淼;《中国优秀硕士论文电子期刊网》(第9期);第1-78页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116505556A (en) | 2023-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lin et al. | An improved moth-flame optimization algorithm for support vector machine prediction of photovoltaic power generation | |
| Chen et al. | Multi-fault diagnosis study on roller bearing based on multi-kernel support vector machine with chaotic particle swarm optimization | |
| CN107392368B (en) | Meteorological forecast-based office building dynamic heat load combined prediction method | |
| CN110414788B (en) | Electric energy quality prediction method based on similar days and improved LSTM | |
| CN110023850B (en) | Method and control device for controlling a technical system | |
| Li et al. | Neuroadaptive variable speed control of wind turbine with wind speed estimation | |
| EP3613982A1 (en) | Method for controlling operation of a wind turbine | |
| CN111222674A (en) | Short-term photovoltaic power generation capacity prediction method based on long-term and short-term memory neural network | |
| CN110685868A (en) | Wind turbine generator fault detection method and device based on improved gradient elevator | |
| US20220186709A1 (en) | Reinforcement learning-based real time robust variable pitch control of wind turbine systems | |
| Chuentawat et al. | The comparison of PM2. 5 forecasting methods in the form of multivariate and univariate time series based on support vector machine and genetic algorithm | |
| CN116307652A (en) | Artificial intelligent resource allocation method for intelligent power grid | |
| CN111178621A (en) | Parameter optimization method of electric heating load prediction support vector regression model | |
| CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
| CN115689353A (en) | Hydropower station speed regulation system health assessment method based on GS-SVM algorithm | |
| Tan et al. | Support vector machine algorithm for artificial intelligence optimization | |
| CN110365053B (en) | Short-term wind power prediction method based on delay optimization strategy | |
| CN111815039A (en) | Weekly scale wind power probability prediction method and system based on weather classification | |
| CN115425680A (en) | Power prediction model construction and prediction method of multi-energy combined power generation system | |
| CN113449847B (en) | Offshore wind power rolling prediction method considering second-level time series wind speed change | |
| CN116865343B (en) | Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network | |
| CN116505556B (en) | Wind farm power control system and method based on primary frequency modulation | |
| US11687772B2 (en) | Optimization of cyber-physical systems | |
| CN116167508A (en) | Short-term photovoltaic output rapid prediction method and system based on meteorological factor decomposition | |
| CN112036601A (en) | Method and system for predicting oil temperature of synchronous phase modifier oil system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |