On-line adjusting method, device and equipment for operating parameters of wind generating set
By establishing simulation models and pre-training parameter adjustment models in wind turbines, the performance degradation of wind turbines in complex wind conditions is solved, stable control and efficient adaptation are achieved, and labor costs are reduced.
Patent Information
- Application Number
- CN202510599891.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-09
- Publication Date
- 2025-07-29
AI Technical Summary
In the prior art, wind turbines are difficult to achieve flexible adjustment in complex and variable wind conditions, resulting in reduced performance and low parameter setting costs and low efficiency based on manual experience.
By establishing a simulation model of wind turbine sets, pre-training parameter adjustment models, simulating various wind conditions, improving training accuracy, and deploying the pre-trained model to the actual unit, adjusting the operating parameters in real time to adapt to complex wind conditions.
It realizes stable control of wind turbines under complex wind conditions, improves performance and adaptability, reduces labor costs, and optimizes wind power efficiency.
Smart Images
Figure CN120384840A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of wind power generation, and particularly to an online adjustment method, device and equipment for operating parameters of a wind turbine generator set. Background Art
[0002] In the technical field of wind power generation, how to improve the environmental adaptability of wind turbine generator sets has become a technical problem to be urgently solved.
[0003] In the related art, it mainly relies on manual experience to preset the operating parameters of the wind turbine generator set under fixed working conditions for wind power generation.
[0004] However, this implementation method requires staff to have a high professional level and rich prior knowledge, so that the parameter adjustment results can ensure the actual on-site effects, and thus requires a high labor cost. Moreover, this implementation method cannot cope with complex and changeable wind conditions, lacks a certain degree of flexibility, and thus affects the performance of the wind turbine generator set. Summary of the Invention
[0005] The embodiments of the present application provide an online adjustment method, device and equipment for operating parameters of a wind turbine generator set, so as to achieve the effect of improving the control of the wind turbine generator set, enhancing the adaptability of the wind turbine generator set, and further improving the performance of the wind turbine generator set.
[0006] In a first aspect, the embodiments of the present application provide an online adjustment method for operating parameters of a wind turbine generator set, including:
[0007] Pre-training a parameter adjustment model according to a pre-established simulation model of the wind turbine generator set to obtain a pre-trained parameter adjustment model; wherein, the pre-trained parameter adjustment model is used to output a to-be-executed action; the to-be-executed action is used to indicate a control instruction of the wind turbine generator set;
[0008] After deploying the pre-trained parameter adjustment model to the actual wind turbine generator set, determining the to-be-executed action of the actual wind turbine generator set at the next moment according to the first operating state of the actual wind turbine generator set at the current moment, the estimated wind speed information and the pre-trained parameter adjustment model;
[0009] Controlling the operation of the actual wind turbine generator set according to the control instruction of the wind turbine generator set indicated by the to-be-executed action at the next moment.
[0010] In a possible implementation manner, pre-training a parameter adjustment model according to a pre-established simulation model of the wind turbine generator set to obtain a pre-trained parameter adjustment model includes:
[0011] Obtain wind condition data determined by a wind condition generation tool;
[0012] According to the wind condition data and the wind turbine simulation model, pre-train the parameter adjustment model to obtain the pre-trained parameter adjustment model.
[0013] In a possible implementation manner, pre-training the parameter adjustment model according to the wind condition data and the wind turbine simulation model to obtain the pre-trained parameter adjustment model includes:
[0014] According to the wind condition data read by the wind turbine simulation model and the target execution actions determined by the parameter adjustment model, control the operation of the wind turbine simulation model to obtain the second operating state of the wind turbine simulation model at different times;
[0015] Determine model training data according to the second operating state at different times and the target execution actions corresponding to each of the second operating states;
[0016] Train the parameter adjustment model according to the model training data, and when it is determined that the second operating state at the target time meets the training end requirement, end this training and start the next training until the training is completed to obtain the pre-trained parameter adjustment model.
[0017] In a possible implementation manner, the method further includes:
[0018] Obtain the third operating state of the wind turbine simulation model at the current time and the fourth operating state at the previous time;
[0019] Determine the target reward information according to the difference between the target parameter information in the third operating state and the target parameter information in the fourth operating state, and / or the difference between the target execution actions corresponding to the third operating state and the target execution actions corresponding to the fourth operating state;
[0020] Train the parameter adjustment model according to the target reward information.
[0021] In a possible implementation manner, the to-be-executed action includes parameter adjustment information; after determining the to-be-executed action of the actual wind turbine at the next time, the method further includes:
[0022] Determine whether the parameter adjustment information included in the to-be-executed action at the next time is within the safety threshold range;
[0023] If so, control the operation of the actual wind turbine according to the wind turbine control instruction indicated by the action to be executed at the next moment;
[0024] If not, determine a new action to be executed according to the safety threshold range, and determine the new action to be executed as the action to be executed by the actual wind turbine at the next moment.
[0025] In a possible implementation manner, after determining the action to be executed by the actual wind turbine at the next moment, the method further includes:
[0026] Determine the stable state of the actual wind turbine;
[0027] According to the stable state of the actual wind turbine, determine a control strategy for controlling the operation of the actual wind turbine according to the action to be executed by the actual wind turbine at the next moment; wherein, the control strategy is used to indicate the manner of controlling the operation of the actual wind turbine according to the action to be executed at the next moment;
[0028] According to the control strategy, determine a new action to be executed, and determine the new action to be executed as the action to be executed by the actual wind turbine at the next moment.
[0029] In a possible implementation manner, the method further includes:
[0030] In response to an update operation on the reward function, based on the updated reward function, pre-train the parameter adjustment model to obtain a pre-trained parameter adjustment model; wherein, the updated reward function is determined according to at least part of the parameter information in the second operating state corresponding to the wind turbine simulation model.
[0031] In a second aspect, an embodiment of the present application provides an online adjustment device for operating parameters of a wind turbine, including:
[0032] A training unit, configured to pre-train a parameter adjustment model according to a pre-established wind turbine simulation model to obtain a pre-trained parameter adjustment model; wherein, the pre-trained parameter adjustment model is used to output an action to be executed; the action to be executed is used to indicate a wind turbine control instruction;
[0033] A determination unit, configured to, after deploying the pre-trained parameter adjustment model to an actual wind turbine, determine an action to be executed by the actual wind turbine at the next moment according to the first operating state of the actual wind turbine at the current moment, the estimated wind speed information, and the pre-trained parameter adjustment model;
[0034] A control unit for controlling the operation of the actual wind turbine according to the wind turbine control instruction indicated by the action to be executed at the next moment.
[0035] In a possible implementation manner, a training unit is configured to:
[0036] Obtain wind condition data determined by a wind condition generation tool;
[0037] Pre-train the parameter adjustment model according to the wind condition data and the wind turbine simulation model to obtain the pre-trained parameter adjustment model.
[0038] In a possible implementation manner, a training unit is configured to:
[0039] Control the operation of the wind turbine simulation model according to the wind condition data read by the wind turbine simulation model and the target execution action determined by the parameter adjustment model, to obtain the second operating state of the wind turbine simulation model at different moments;
[0040] Determine model training data according to the second operating state at different moments and the target execution action corresponding to each second operating state;
[0041] Train the parameter adjustment model according to the model training data, and end this training and start the next training when it is determined that the second operating state at the target moment meets the training end requirement, until the training is completed, to obtain the pre-trained parameter adjustment model.
[0042] In a possible implementation manner, the training unit is further configured to:
[0043] Obtain the third operating state of the wind turbine simulation model at the current moment and the fourth operating state at the previous moment;
[0044] Determine target reward information according to the difference between the target parameter information in the third operating state and the target parameter information in the fourth operating state, and / or the difference between the target execution action corresponding to the third operating state and the target execution action corresponding to the fourth operating state;
[0045] Train the parameter adjustment model according to the target reward information.
[0046] In a possible implementation manner, the action to be executed includes parameter adjustment information; the device is further configured to:
[0047] After determining the action to be executed by the actual wind turbine at the next moment, determine whether the parameter adjustment information included in the action to be executed at the next moment is within the safety threshold range;
[0048] If so, control the operation of the actual wind turbine according to the wind turbine control instruction indicated by the action to be executed at the next moment;
[0049] If not, determine a new action to be executed according to the safety threshold range, and determine the new action to be executed as the action to be executed by the actual wind turbine at the next moment.
[0050] In a possible implementation, the device is further configured to:
[0051] Determine the stable state of the actual wind turbine;
[0052] According to the stable state of the actual wind turbine, determine a control strategy for controlling the operation of the actual wind turbine according to the action to be executed by the actual wind turbine at the next moment; wherein, the control strategy is used to indicate the manner of controlling the operation of the actual wind turbine according to the action to be executed at the next moment;
[0053] According to the control strategy, determine a new action to be executed, and determine the new action to be executed as the action to be executed by the actual wind turbine at the next moment.
[0054] In a possible implementation, the device is further configured to:
[0055] In response to an update operation on the reward function, pre-train the parameter adjustment model based on the updated reward function to obtain a pre-trained parameter adjustment model; wherein, the updated reward function is determined according to at least part of the parameter information in the second operating state corresponding to the wind turbine simulation model.
[0056] In a third aspect, an embodiment of the present application provides a computer device, including: a memory, a processor;
[0057] The memory stores computer execution instructions;
[0058] The processor executes the computer execution instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementation manners of the first aspect.
[0059] Fourthly, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the first aspect and / or various possible implementation manners of the first aspect as described above.
[0060] Fifthly, an embodiment of the present application provides a computer program product, including a computer program, which when executed by a processor, implements the first aspect and / or various possible implementation manners of the first aspect as described above.
[0061] The online adjustment method, device and equipment for the operating parameters of the wind turbine generator set provided by the embodiments of the present application can pre-train the parameter adjustment model according to the pre-established wind turbine generator set simulation model to obtain the pre-trained parameter adjustment model, so that the parameter adjustment model can be trained according to the wind turbine generator set simulation model, and thus the training accuracy of the parameter adjustment model can be improved by simulating various wind conditions, and further the safety of deploying the parameter adjustment model to the actual wind turbine generator set can be improved, and further the problem of unstable control of the wind turbine generator set caused by directly deploying the parameter adjustment model to the real wind turbine generator set can be avoided. After that, after the pre-trained parameter adjustment model is deployed to the actual wind turbine generator set, according to the first operating state of the actual wind turbine generator set at the current moment, the estimated wind speed information and the pre-trained parameter adjustment model, the action to be executed by the actual wind turbine generator set at the next moment is determined. Finally, according to the wind turbine generator set control instruction indicated by the action to be executed at the next moment, the operation of the actual wind turbine generator set is controlled. This implementation manner can determine the action to be executed by the actual wind turbine generator set at the next moment in real time according to the deployed parameter adjustment model, so as to realize the real-time adjustment of the operating parameters of the wind turbine generator set, optimize the control of the wind turbine generator set, and enable the wind turbine generator set to adapt to the complex and changeable wind condition environment, thereby improving the performance of the wind turbine generator set. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] The drawings here are incorporated into the description and form a part of this description, showing the embodiments consistent with the present application, and are used together with the description to explain the principles of the present application.
[0063] Figure 1 Schematic flowchart of an online adjustment method for the operating parameters of a wind turbine generator set provided by an embodiment of the present application Figure 1 ;
[0064] Figure 2 Schematic flowchart of an online adjustment method for the operating parameters of a wind turbine generator set provided by an embodiment of the present application Figure 2 ;
[0065] Figure 3Schematic diagram of the training process of a parameter adjustment model provided by an embodiment of the present application;
[0066] Figure 4 Schematic diagram of the implementation process of an online adjustment method for operating parameters of a wind turbine provided by an embodiment of the present application;
[0067] Figure 5 Schematic diagram of the structure of an online adjustment device for operating parameters of a wind turbine provided by an embodiment of the present application;
[0068] Figure 6 Schematic diagram of the structure of a computer device provided by an embodiment of the present application.
[0069] Through the above drawings, specific embodiments of the present application have been shown, and more detailed descriptions will be given later. These drawings and written descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed implementation manners
[0070] Here, exemplary embodiments will be described in detail, and their examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0071] The term "and / or" in this article only describes an association relationship and indicates that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" in this article represents any one of multiple or any combination of at least two of multiple. For example, including at least one of A, B, and C can represent any one or more elements selected from the set composed of A, B, and C.
[0072] First, the nouns involved in the present application are explained:
[0073] DRL: Deep Reinforcement Learning, deep reinforcement learning;
[0074] DDPG: Deep Deterministic Policy Gradient, deep deterministic policy gradient algorithm;
[0075] PPO: Proximal Policy Optimization, proximal policy optimization.
[0076] In the field of wind power generation technology, how to improve the performance of wind turbines has become a technical problem that urgently needs to be solved.
[0077] In one implementation, the operating parameters of the wind turbine can be preset according to manual experience under fixed operating conditions for wind power generation. However, this method requires staff to have a high professional level and rich prior knowledge so that the parameter adjustment results can ensure the actual on-site effects, and thus requires a high labor cost. Moreover, due to reasons such as the continuous increase in the number of wind turbines, the change of environmental parameters, and the degradation of the wind turbines themselves, the control effect of the wind turbines will gradually deteriorate.
[0078] In another implementation, a model can be deployed on the wind turbine, and through training the model, the optimization of the operating parameters of the wind turbine can be achieved.
[0079] However, the method of directly training the model, on the one hand, requires a large amount of training data to ensure the accuracy of the model. In the face of complex and changeable operating conditions, it is difficult to obtain training data, and thus a model with high accuracy cannot be trained. On the other hand, it also requires high computing power requirements and high resource consumption. In addition, in this implementation, at the beginning of training, it is unable to adapt to the complex and changeable environment, resulting in a poor control effect of the wind turbine and affecting the control accuracy of the wind turbine.
[0080] Based on this, the online adjustment method for the operating parameters of the wind turbine provided in this application pre-trains the parameter adjustment model by establishing a wind turbine simulation model and simulating various wind conditions according to the wind turbine simulation model, so as to improve the model training efficiency and training accuracy at the same time. Then, the pre-trained parameter adjustment model is deployed to the real wind turbine to adjust the operating parameters of the wind turbine and improve the power generation performance of the wind turbine.
[0081] The technical solutions of this application and how the technical solutions of this application solve the above technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.
[0082] Figure 1 Schematic diagram of the process of an online adjustment method for the operating parameters of a wind turbine provided in an embodiment of this application Figure 1 , such as Figure 1 shown, this method includes:
[0083] S101. According to the pre-established wind turbine simulation model, pre-train the parameter adjustment model to obtain the pre-trained parameter adjustment model.
[0084] Among them, the pre-trained parameter adjustment model is used to output the action to be executed; the action to be executed is used to indicate the wind turbine control instruction.
[0085] In one example, the action to be executed may include parameter adjustment information of one or more wind turbine operating parameters, and thus the wind turbine control instruction can be determined according to the parameter adjustment information of one or more wind turbine operating parameters.
[0086] Optionally, the wind turbine operating parameters associated with the action to be executed may include, but are not limited to, at least one of pitch angle, generator torque, and yaw angle.
[0087] In one example, the pre-established wind turbine simulation model can be compiled in any type of simulation software. For example, the simulation software can be the open-source Openfast software.
[0088] In one example, the parameter adjustment model can be deployed in a simulation training environment. For example, the simulation training environment can be a deep learning framework environment or other environments capable of running the parameter adjustment model. There is no limitation here, as long as it can be implemented.
[0089] In one example, the parameter adjustment model can indicate a DRL algorithm. For example, the parameter adjustment model can be a DDPG model or a PPO model, etc. There is no limitation on the specific structure of the parameter adjustment model here, as long as it can be implemented.
[0090] In one example, after deploying the parameter adjustment model in the simulation training environment, a wind turbine controller can also be deployed in the simulation training environment, so that after receiving the action to be executed output by the parameter adjustment model through the wind turbine controller, a wind turbine control instruction can be generated to adjust the operating parameters of the wind turbine, thereby controlling the operation of the wind turbine.
[0091] In one example, after pre-training the parameter training model in the simulation training environment according to the pre-established wind turbine simulation model, the pre-trained parameter adjustment model is obtained.
[0092] S102. After deploying the pre-trained parameter adjustment model to the actual wind turbine, determine the action to be executed by the actual wind turbine at the next moment according to the first operating state of the actual wind turbine at the current moment, the estimated wind speed information, and the pre-trained parameter adjustment model.
[0093] In one example, the first operating state of the actual wind turbine at the current moment can indicate but is not limited to the following parameter information: wind speed information, rotational speed information, rotational speed error information, blade angle information, front-back vibration value, left-right vibration value, power information, time information (e.g., current month or current date), meteorological parameters, etc.
[0094] Optionally, the first operating state may further include parameters obtained by calculating and processing the above parameter information. For example, based on the wind speed information and power information, the wind speed-power matching information can be determined, thereby obtaining a first operating state of the actual wind turbine.
[0095] In one example, the predicted wind speed information can be understood as the wind speed information within a preset time period in the future. For example, the preset time period in the future can indicate the next moment or multiple moments after the current moment. Here, the length of the preset time period is not limited and should be based on what can be achieved.
[0096] S103. Control the operation of the actual wind turbine according to the wind turbine control instruction indicated by the action to be executed at the next moment.
[0097] In specific implementation, the operation parameters of the wind turbine can be adjusted according to the wind turbine control instruction indicated by the action to be executed at the next moment, so as to control the operation of the actual wind turbine.
[0098] As can be seen from the above description, in the embodiment of the present application, based on the pre-established wind turbine simulation model, the parameter adjustment model can be pre-trained to obtain the pre-trained parameter adjustment model. Thus, the parameter adjustment model can be trained according to the wind turbine simulation model, and by simulating various wind conditions, the training accuracy of the parameter adjustment model can be improved. Furthermore, the safety of deploying the parameter adjustment model to the actual wind turbine can be enhanced, and the problem of unstable and inaccurate control of the wind turbine caused by directly deploying the parameter adjustment model to the real wind turbine can be avoided. After that, after deploying the pre-trained parameter adjustment model to the actual wind turbine, according to the first operating state of the actual wind turbine at the current moment, the predicted wind speed information, and the pre-trained parameter adjustment model, the action to be executed by the actual wind turbine at the next moment can be determined. Finally, the operation of the actual wind turbine is controlled according to the wind turbine control instruction indicated by the action to be executed at the next moment. This implementation method can determine the action to be executed by the actual wind turbine at the next moment in real time based on the deployed parameter adjustment model, thereby realizing the real-time adjustment of the operation parameters of the wind turbine, optimizing the control of the wind turbine, and enabling the wind turbine to adapt to complex and changeable wind conditions, thus improving the performance of the wind turbine.
[0099] Figure 2 Flow schematic of an online adjustment method for operating parameters of a wind turbine provided by an embodiment of the present application Figure 2 , such as Figure 2 shown, based on the Figure 1 embodiment, the online adjustment method for the operating parameters of the wind turbine is described in detail. The method includes:
[0100] S201. Obtain the wind condition data determined by the wind condition generation tool according to the wind conditions.
[0101] In one example, the wind condition generation tool can be used to generate wind condition files for different regions and / or different seasons (or different months), so as to determine the wind condition data according to the wind condition files. At this time, the wind condition files can include data such as wind speed, wind direction, turbulence intensity, and wind shear.
[0102] In one example, the wind condition generation tool can be provided according to a pre-established wind turbine simulation model, or can be a separate tool, which is not limited here, as long as it can be implemented.
[0103] In one example, an input wind file covering the actual working conditions of the actual wind farm, that is, a wind condition file, is generated according to the wind condition generation tool. Then, after determining the training data and test data according to the input wind file, the wind condition data for training is determined.
[0104] This implementation method can generate input wind files under various wind conditions according to the wind condition generation tool, so that the determined wind condition data is more comprehensive, and it can also avoid collecting a large amount of real data to determine the wind condition data, thereby improving the efficiency and reducing the cost.
[0105] S202. Pre-train the parameter adjustment model according to the wind condition data and the wind turbine simulation model to obtain the pre-trained parameter adjustment model.
[0106] Among them, the pre-trained parameter adjustment model is used to output the action to be executed; the action to be executed is used to indicate the wind turbine control instruction.
[0107] In one example, the operation of the wind turbine simulation model can be controlled according to the wind condition data read by the wind turbine simulation model and the target execution actions determined by the parameter adjustment model, so as to obtain the second operating state of the wind turbine simulation model at different times. Then, according to the second operating states at different times and the target execution actions corresponding to each second operating state, the model training data is determined. Finally, according to the model training data, the parameter adjustment model is trained, and when it is determined that the second operating state at the target time meets the requirements for the end of training, this training is ended and the next training is started until the training is completed, and a pre-trained parameter adjustment model is obtained.
[0108] In one example, the wind speed information can be estimated according to the second operating states at different times, so that the model training data can be determined according to the second operating states, the estimated wind speed information, and the target execution actions corresponding to each second operating state.
[0109] In specific implementation, the model parameters in the parameter adjustment model can be initialized first to obtain an initialized parameter adjustment model, and the initialized target execution actions can be obtained according to the target execution actions corresponding to the initialized parameter adjustment model. Then, according to the initialized parameter adjustment model and the initialized target execution actions, the operation of the wind turbine simulation model can be controlled to obtain the second operating state of the wind turbine simulation model at different times. At this time, the wind speed information can be estimated according to the second operating states of the wind turbine simulation model at different times, so that the initial training data included in the experience replay buffer corresponding to the parameter adjustment model can be determined according to the second operating states and the estimated wind speed information. Then, according to the initial training data, the model parameters of the parameter adjustment model can be adjusted, so as to adjust the target execution actions output by the parameter adjustment model, and then different second operating states can be obtained, so that new training data can be obtained, thereby constituting the model training data corresponding to the parameter adjustment model. At this time, the process of training the parameter adjustment model while collecting training data can be realized according to the model training data.
[0110] In one example, the training end requirement can be used to indicate the degree of deviation from the second operating state. At this time, for at least some of the parameter information included in the second operating state, corresponding preset thresholds can be determined. Thus, when the parameter information in the second operating state at the target moment exceeds the corresponding preset threshold, it is determined that the training end requirement is met. Exemplarily, if the parameter information included in the second operating state is: wind speed information, rotational speed information, rotational speed error information, blade angle information, front-back vibration value, left-right vibration value, power information, then a corresponding preset threshold can be set for the rotational speed error information. And when it is determined that the rotational speed error information in the second operating state at the target moment exceeds the corresponding preset threshold, it is determined that the training end requirement is met, this training is ended, and the next training is started until the training is completed.
[0111] In one example, during the training of the parameter adjustment model, an optimization goal can be preset in advance, and thus, based on the optimization goal, the parameter adjustment model is trained. Among them, the optimization goal can be determined according to one or more parameter information included in the operating state. At this time, during the training of the parameter adjustment model, the gap between the parameter adjustment model and the optimization goal can be determined through the target reward information, so that the training process of the parameter adjustment model is consistent with the optimization goal.
[0112] Specifically, the third operating state of the wind turbine simulation model at the current moment and the fourth operating state at the previous moment can be obtained; then, based on the difference between the target parameter information in the third operating state and the target parameter information in the fourth operating state, and / or, the difference between the target execution action corresponding to the third operating state and the target execution action corresponding to the fourth operating state, the target reward information is determined; finally, the parameter adjustment model is trained according to the target reward information.
[0113] In the embodiments of the present application, the parameter information included in the third operating state (or, the fourth operating state) at least includes: wind speed information, rotational speed information, rotational speed error information, blade angle information, front-back vibration value, left-right vibration value, power information, time information, wind speed-power matching information. At this time, the target parameter information in the third operating state can indicate: rotational speed information, rotational speed error information, front-back vibration value, left-right vibration value, and power information.
[0114] Based on this, the optimization goal can be: small rotational speed error and fluctuation, small vibration, small power deviation, and small action adjustment amplitude. At this time, the determined target reward information can be seen as shown in the following formula (1).
[0115] R = C1Δw g 2 +C2(w g -w g′ ) 2 + C3a x 2 + C4(P - P * ) 2 + C5(a - a ′ ) 2 (1)
[0117] Among them, C1 - C5 represent weight coefficients, and Δw g represents the rotational speed error information at the current moment, and w g represents the rotational speed information at the current moment, and w g ′ represents the rotational speed information at the previous moment, and a x represents the vibration value at the current moment, P represents the power information at the current moment, and P * represents the rated power information at the current moment, a represents the parameter adjustment information corresponding to the target execution action at the current moment (for example, the pitch angle increases by 5 degrees), and a ′ represents the parameter adjustment information corresponding to the target execution action at the previous moment.
[0118] It should be noted here that this application does not limit this optimization goal, and it is subject to meeting actual needs.
[0119] In the above - mentioned embodiment, the target parameter information and / or the action adjustment amplitude of the target execution action can be used as the target reward information to train the parameter adjustment model, so that the parameter adjustment model can adapt to complex and changeable environmental conditions according to this optimization direction, thereby improving the performance of the online adjustment method for the operating parameters of the wind turbine generator set.
[0120] In a possible implementation manner, the embodiments of this application can also combine the wind speed - power matching information to train the parameter adjustment model. At this time, the parameter adjustment model can be trained according to the deviation degree between the corresponding relationship between the power information at the current moment and the estimated wind speed information (i.e., the wind speed - power matching information) and the standard wind speed - power curve. At this time, the wind speed - power matching information can be used as the input of the parameter adjustment model to train the parameter adjustment model.
[0121] Regarding the above - mentioned training process of the parameter adjustment model, the following is explained in combination with a specific flow diagram. Refer to Figure 3 , Figure 3 which is a training flow diagram of a parameter adjustment model provided by the embodiments of this application. As shown in Figure 3As shown in the figure, a simulation model of a wind turbine generator can be established in advance in simulation software, and a controller and a parameter adjustment model of the wind turbine generator can be built in a simulation training environment. At this time, the parameter adjustment model is used to output target execution actions, and the controller of the wind turbine generator is used to interface with the simulation model and the parameter adjustment model of the wind turbine generator, and control the operation of the simulation model of the wind turbine generator according to the received target execution actions combined with wind condition data.
[0122] Based on this, when training the parameter adjustment model, the wind condition data determined by the wind condition generation tool can be input into the simulation model of the wind turbine generator, and the wind condition data can be sent to the controller of the wind turbine generator, so that the controller of the wind turbine generator controls the operation of the simulation model of the wind turbine generator according to the wind condition data and the target execution actions determined by the parameter adjustment model, and obtains the second operating state corresponding to the simulation model of the wind turbine generator. After that, the wind speed information can be estimated according to the second operating state by the wind speed estimation module, so that after determining the model training data according to the second operating state and the estimated wind speed information, the target reward information can be determined according to the third operating state at the current moment included in the second operating state / model training data / experience replay buffer and the fourth operating state at the previous moment, and the deviation degree of the wind speed-power matching information can be determined according to the third operating state at the current moment. After that, the parameter adjustment model can be trained according to the third operating state, the estimated wind speed information, the deviation degree of the wind speed-power matching information, and the target reward information, combined with the training end requirement, to obtain a pre-trained parameter adjustment model.
[0123] S203. After deploying the pre-trained parameter adjustment model to an actual wind turbine generator, determine the action to be executed by the actual wind turbine generator at the next moment according to the first operating state, the estimated wind speed information, and the pre-trained parameter adjustment model of the actual wind turbine generator at the current moment.
[0124] Optionally, the action to be executed may include one or more parameter adjustment information. At this time, the parameter adjustment information can be used to adjust the magnitude of the operating parameters of the corresponding wind turbine generator.
[0125] Based on this, in order to avoid the operating parameters of the wind turbine generator being adjusted too much, which in turn affects the stability of the operation of the wind turbine generator, after determining the action to be executed by the actual wind turbine generator at the next moment, the steps described in S204 to S206 below can be executed.
[0126] S204. Determine whether the parameter adjustment information included in the action to be executed at the next moment is within the safety threshold range.
[0127] S205. If so, control the operation of the actual wind turbine according to the wind turbine control instruction indicated by the to-be-executed action at the next moment.
[0128] S206. If not, determine a new to-be-executed action according to the safety threshold range, and determine the new to-be-executed action as the to-be-executed action of the actual wind turbine at the next moment.
[0129] In the above embodiments, according to the safety threshold range, it can be determined whether the to-be-executed action of the determined actual wind turbine at the next moment is within the safe adjustment range, so as to ensure the safety and stability of controlling the wind turbine according to the to-be-executed action at the next moment.
[0130] Optionally, in addition to ensuring the safety and stability of the controlled wind turbine by adjusting the size of the parameter information included in the to-be-executed action, it can also be determined whether it is necessary to control the operation of the wind turbine according to the to-be-executed action at the next moment according to the operating state of the wind turbine, so as to further ensure the safety and stability of the operation of the wind turbine. For details, refer to the steps described in S207 to S208 below.
[0131] S207. Determine the stable state of the actual wind turbine.
[0132] Optionally, the stable state of the wind turbine can be determined according to the operating state determined by the wind turbine based on the actual wind conditions. Exemplarily, the stable state of the actual wind turbine can be determined according to the deviation degree of the parameter information included in the operating state, or the stable state of the actual wind turbine can also be determined according to the target reward information calculated from the operating state. The determination method of the stable state of the actual wind turbine is not limited here, as long as it can be achieved.
[0133] Optionally, the stable state of the actual wind turbine can also be determined according to the action determination frequency of the to-be-executed action at the next moment determined by the parameter adjustment model.
[0134] Optionally, the stable state of the actual wind turbine generator set may include two cases: stable and unstable. At this time, if the stable state of the actual wind turbine generator set is determined according to the deviation degree of the parameter information included in the operating state, then when the parameter information exceeds the corresponding preset threshold, it is determined that the actual wind turbine generator set is unstable; otherwise, it is determined that the actual wind turbine generator set is stable. If the stable state of the actual wind turbine generator set is determined according to the target reward information, then when the calculation result of the target reward information is greater than the preset reward threshold, it is determined that the actual wind turbine generator set is unstable; otherwise, it is determined that the actual wind turbine generator set is stable. If the stable state of the actual wind turbine generator set is determined according to the action determination frequency, then when the action determination frequency is greater than the preset frequency threshold, it is determined that the actual wind turbine generator set is unstable; otherwise, it is determined that the actual wind turbine generator set is stable.
[0135] S208. Determine the control strategy for controlling the operation of the actual wind turbine generator set according to the action to be executed by the actual wind turbine generator set at the next moment based on the stable state of the actual wind turbine generator set.
[0136] Optionally, the control strategy may indicate any one of the following: continue control strategy, start control strategy, stop control strategy, stop this control strategy, adjust control cycle strategy, etc.
[0137] At this time, if the actual wind turbine generator set is stable, it can be determined that the control strategy is: start control strategy, or continue control strategy, or adjust control cycle strategy. For example, if it is determined that the function of controlling the operation of the wind turbine generator set according to the pre-trained parameter adjustment model is not enabled, then the control strategy can be determined as the start control strategy; if it is determined that the function of controlling the operation of the wind turbine generator set according to the pre-trained parameter adjustment model is enabled, then the control strategy can be determined as the continue control strategy, or adjust control cycle strategy (for example, increase or decrease the control frequency).
[0138] If the actual wind turbine generator set is unstable, it can be determined that the control strategy is: stop control strategy, stop this control strategy, adjust control cycle strategy (for example, increase the control frequency).
[0139] S209. Determine a new action to be executed according to the control strategy, and determine the new action to be executed as the action to be executed by the actual wind turbine generator set at the next moment.
[0140] For example, if the control strategy is stop control strategy / stop this control strategy / decrease control frequency, then the new action to be executed is none; if the control strategy is increase control frequency, then the new action to be executed remains unchanged.
[0141] In the above embodiments, the online adjustment method of the operating parameters of the wind turbine can be determined according to the stable state of the actual wind turbine, and the control strategy for controlling the operation of the actual wind turbine can be determined, so as to improve the adaptability of the wind turbine control. It can not only prevent the problems of unstable wind turbine control and insufficient computing power caused by too high adjustment frequency, but also close the control in case of emergency, so as to avoid damage to the wind turbine, and further ensure the safety and stability of the wind turbine control.
[0142] S210. Control the operation of the actual wind turbine according to the wind turbine control instruction indicated by the action to be executed at the next moment.
[0143] Exemplarily, refer to Figure 4 , Figure 4 which is a schematic flowchart of the implementation process of an online adjustment method for the operating parameters of a wind turbine provided by an embodiment of the present application. As Figure 4 shown, after the parameter adjustment model is deployed to the actual wind turbine, the model parameters of the parameter adjustment model can be adjusted in real time according to the actual wind conditions, so as to realize the real-time adjustment of the operating parameters of the wind turbine. At this time, the online adjustment method of the operating parameters of the wind turbine can estimate the wind speed information according to the first operating state of the actual wind turbine, and then determine the data in the experience replay buffer according to the first operating state and the estimated wind speed information, and realize the online adjustment of the operating parameters of the wind turbine according to the data in the experience replay buffer. At this time, the process of online adjusting the operating parameters of the wind turbine is similar to the training process of the above parameter adjustment model, which will not be elaborated here.
[0144] As Figure 4 shown, when the operating parameters of the wind turbine are adjusted in real time, a control strategy module and a safety monitoring module can be added on the basis of the training process of the parameter adjustment model. At this time, the control strategy module determines the control strategy corresponding to the actual wind turbine according to the stable state of the actual wind turbine. At this time, it can also be determined according to the safety monitoring module whether the parameter adjustment information included in the action to be executed at the next moment determined by the parameter adjustment model is within the safety threshold range, so as to determine whether to control the operation of the actual wind turbine according to the action to be executed at the next moment. The execution order of the control strategy module and the safety monitoring module is not limited here, as long as it can be realized.
[0145] In a possible implementation manner, when controlling the operation of the wind turbine in the above embodiment, the optimization target of the wind turbine can also be adjusted according to actual needs, and then the parameter adjustment model can be correspondingly adjusted to meet the actual control needs of the wind turbine.
[0146] Based on this, in the embodiments of the present application, in response to an update operation on the reward function, the parameter adjustment model can be pre-trained based on the updated reward function to obtain a pre-trained parameter adjustment model.
[0147] Among them, the updated reward function is determined according to at least part of the parameter information in the second operating state corresponding to the wind turbine simulation model.
[0148] In one example, the reward function can be determined by the above formula (1), that is, the reward function can be used to determine the above target reward information.
[0149] In the above implementation manner, by updating the reward function, the optimization direction corresponding to the parameter adjustment model can be flexibly adjusted, so that the online adjustment method for the operating parameters of the wind turbine can be applied to multiple application scenarios, thereby improving the robustness of the online adjustment method for the operating parameters of the wind turbine.
[0150] Figure 5 The following is a schematic structural diagram of an online adjustment device for the operating parameters of a wind turbine provided in the embodiments of the present application. As Figure 5 shown, the online adjustment device 50 for the operating parameters of the wind turbine provided in this embodiment includes:
[0151] A training unit 501, configured to pre-train a parameter adjustment model according to a pre-established wind turbine simulation model to obtain a pre-trained parameter adjustment model; wherein, the pre-trained parameter adjustment model is used to output a to-be-executed action; the to-be-executed action is used to indicate a wind turbine control instruction.
[0152] A determination unit 502, configured to, after deploying the pre-trained parameter adjustment model to an actual wind turbine, determine a to-be-executed action of the actual wind turbine at the next moment according to the first operating state of the actual wind turbine at the current moment, the estimated wind speed information, and the pre-trained parameter adjustment model.
[0153] A control unit 503, configured to control the operation of the actual wind turbine according to the wind turbine control instruction indicated by the to-be-executed action at the next moment.
[0154] In a possible implementation manner, the training unit 501 is configured to:
[0155] Obtain wind condition data determined according to a wind condition generation tool;
[0156] Pre-train the parameter adjustment model according to the wind condition data and the wind turbine simulation model to obtain a pre-trained parameter adjustment model.
[0157] In a possible implementation manner, the training unit 501 is configured to:
[0158] According to the wind condition data read by the wind turbine simulation model and the target execution actions determined by the parameter adjustment model, control the operation of the wind turbine simulation model to obtain the second operating state of the wind turbine simulation model at different times;
[0159] Determine model training data according to the second operating state at different times and the target execution actions corresponding to each second operating state;
[0160] Train the parameter adjustment model according to the model training data, and when it is determined that the second operating state at the target time meets the training end requirement, end the current training and start the next training until the training is completed to obtain the pre-trained parameter adjustment model.
[0161] In a possible implementation manner, the training unit 501 is further configured to:
[0162] Obtain the third operating state of the wind turbine simulation model at the current time and the fourth operating state at the previous time;
[0163] Determine the target reward information according to the difference between the target parameter information in the third operating state and the target parameter information in the fourth operating state, and / or the difference between the target execution actions corresponding to the third operating state and the target execution actions corresponding to the fourth operating state;
[0164] Train the parameter adjustment model according to the target reward information.
[0165] In a possible implementation manner, the to-be-executed action includes parameter adjustment information; the device is further configured to:
[0166] After determining the to-be-executed action of the actual wind turbine at the next time, determine whether the parameter adjustment information included in the to-be-executed action at the next time is within the safety threshold range;
[0167] If so, control the operation of the actual wind turbine according to the wind turbine control instruction indicated by the to-be-executed action at the next time;
[0168] If not, determine a new to-be-executed action according to the safety threshold range and determine the new to-be-executed action as the to-be-executed action of the actual wind turbine at the next time.
[0169] In a possible implementation manner, the device is further configured to:
[0170] Determine the stable state of the actual wind turbine;
[0171] Determine a control strategy for controlling the operation of an actual wind turbine according to the to-be-executed action of the actual wind turbine at the next moment based on the stable state of the actual wind turbine; wherein, the control strategy is used to indicate the manner of controlling the operation of the actual wind turbine according to the to-be-executed action at the next moment.
[0172] Determine a new to-be-executed action according to the control strategy, and determine the new to-be-executed action as the to-be-executed action of the actual wind turbine at the next moment.
[0173] In a possible implementation manner, the device is further configured to:
[0174] In response to an update operation on the reward function, pre-train a parameter adjustment model based on the updated reward function to obtain a pre-trained parameter adjustment model; wherein, the updated reward function is determined according to at least partial parameter information in the second operating state corresponding to the wind turbine simulation model.
[0175] The online adjustment device for the operating parameters of the wind turbine provided in the embodiments of the present application can execute the method provided in the above method embodiments, and its implementation principle and technical effects are similar, which will not be elaborated here in this embodiment.
[0176] Figure 6 This is a schematic structural diagram of a computer device provided in the embodiments of the present application. As Figure 6 shown, the computer device 60 provided in this embodiment includes: at least one processor 601 and a memory 602. Optionally, the device 60 further includes a communication component 603. Among them, the processor 601, the memory 602, and the communication component 603 are connected through a bus 604.
[0177] In a specific implementation process, at least one processor 601 executes the computer execution instructions stored in the memory 602, so that at least one processor 601 executes the above method.
[0178] The specific implementation process of the processor 601 can refer to the above method embodiments, and its implementation principle and technical effects are similar, which will not be elaborated here in this embodiment.
[0179] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU for short), or other general-purpose processors, digital signal processors (DSP for short), application specific integrated circuits (ASIC for short), etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly implemented by the execution of the hardware processor, or by the combination of the hardware and software modules in the processor.
[0180] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.
[0181] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, the buses in the drawings of this application are not limited to only one bus or one type of bus.
[0182] This application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.
[0183] This application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the processor executes the computer-executable instructions, the above method is implemented.
[0184] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk or an optical disk. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.
[0185] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be part of the processor. The processor and the readable storage medium can be located in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.
[0186] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed between each other can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.
[0187] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0188] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.
[0189] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or part of this technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks or optical discs that can store program codes.
[0190] Those of ordinary skill in the art will understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including those of the above method embodiments; and the aforementioned storage medium includes: various media such as ROM, RAM, magnetic disks, or optical discs that can store program codes.
[0191] Finally, it should be noted that: after considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include known common knowledge or conventional technical means in the technical field not disclosed by the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.
Claims
1. An on-line adjustment method for operating parameters of a wind turbine generator, characterized in that, Including: Pre-training the parameter adjustment model according to a pre-established wind turbine simulation model to obtain a pre-trained parameter adjustment model; wherein, the pre-trained parameter adjustment model is used to output a to-be-executed action; the to-be-executed action is used to indicate a wind turbine control instruction. After deploying the pre-trained parameter adjustment model to an actual wind turbine, determining the to-be-executed action of the actual wind turbine at the next moment according to the first operating state of the actual wind turbine at the current moment, the estimated wind speed information, and the pre-trained parameter adjustment model. Controlling the operation of the actual wind turbine according to the wind turbine control instruction indicated by the to-be-executed action at the next moment.
2. The method according to claim 1, wherein Pre-training the parameter adjustment model according to a pre-established wind turbine simulation model to obtain a pre-trained parameter adjustment model, including: Obtaining wind condition data determined by a wind condition generation tool. Pre-training the parameter adjustment model according to the wind condition data and the wind turbine simulation model to obtain the pre-trained parameter adjustment model.
3. The method according to claim 2, wherein Pre-training the parameter adjustment model according to the wind condition data and the wind turbine simulation model to obtain the pre-trained parameter adjustment model, including: Controlling the operation of the wind turbine simulation model according to the wind condition data read by the wind turbine simulation model and the target execution action determined by the parameter adjustment model, to obtain the second operating state of the wind turbine simulation model at different moments. Determining model training data according to the second operating state at different moments and the target execution action corresponding to each of the second operating states. Training the parameter adjustment model according to the model training data, and ending the current training and starting the next training when it is determined that the second operating state at the target moment meets the training end requirement, until the training is completed to obtain the pre-trained parameter adjustment model.
4. The method according to claim 3, wherein The method further includes: Obtaining the third operating state of the wind turbine simulation model at the current moment and the fourth operating state at the previous moment. Determining target reward information according to the difference between the target parameter information in the third operating state and the target parameter information in the fourth operating state, and / or the difference between the target execution action corresponding to the third operating state and the target execution action corresponding to the fourth operating state. Training the parameter adjustment model according to the target reward information.
5. The method according to claim 1, wherein The to-be-executed action includes parameter adjustment information; after determining the to-be-executed action of the actual wind turbine at the next moment, the method further includes: Determining whether the parameter adjustment information included in the to-be-executed action at the next moment is within the safety threshold range. If so, controlling the operation of the actual wind turbine according to the wind turbine control instruction indicated by the to-be-executed action at the next moment. If not, determine a new action to be executed according to the safety threshold range, and determine the new action to be executed as the action to be executed by the actual wind turbine at the next moment.
6. The method according to any one of claims 1 to 5, characterized in that, After determining the action to be executed by the actual wind turbine at the next moment, the method further includes: Determine the stable state of the actual wind turbine; According to the stable state of the actual wind turbine, determine a control strategy for controlling the operation of the actual wind turbine according to the action to be executed by the actual wind turbine at the next moment; wherein, the control strategy is used to indicate the manner of controlling the operation of the actual wind turbine according to the action to be executed at the next moment; According to the control strategy, determine a new action to be executed, and determine the new action to be executed as the action to be executed by the actual wind turbine at the next moment.
7. The method according to any one of claims 1 to 5, characterized in that, The method further includes: In response to an update operation on the reward function, based on the updated reward function, pre-train a parameter adjustment model to obtain a pre-trained parameter adjustment model; wherein, the updated reward function is determined according to at least part of the parameter information in the second operating state corresponding to the wind turbine simulation model.
8. An on-line adjustment device for operating parameters of a wind power generating set, characterized in that, Includes: A training unit for pre-training a parameter adjustment model according to a pre-established wind turbine simulation model to obtain a pre-trained parameter adjustment model; wherein, the pre-trained parameter adjustment model is used to output an action to be executed; the action to be executed is used to indicate a wind turbine control instruction; A determination unit for, after deploying the pre-trained parameter adjustment model to the actual wind turbine, determining the action to be executed by the actual wind turbine at the next moment according to the first operating state of the actual wind turbine at the current moment, the estimated wind speed information, and the pre-trained parameter adjustment model; A control unit for controlling the operation of the actual wind turbine according to the wind turbine control instruction indicated by the action to be executed at the next moment.
9. A computer device, characterized in that, Includes: A memory, a processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory, so that the processor executes the method according to any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, Computer execution instructions are stored in the computer-readable storage medium, and when the computer execution instructions are executed by a processor, they are used to implement the method according to any one of claims 1-7.