Wind turbine generator control method and device, terminal equipment, medium and product

Abstract

The invention discloses a wind turbine generator control method and device, terminal equipment, a medium and a product. The method comprises the following steps: acquiring real-time working data of the wind driven generator; according to the real-time working data, using the trained residual life prediction model and the fault prediction model to predict the current residual life and the current fault probability; determining a current available score according to the current residual life, the current fault probability and the current power output of the wind driven generator; and reporting the current available score to an energy management system, so that the energy management system selects a target wind driven generator according to the received current available score and the current power grid demand, and issues a control instruction to the target wind driven generator. According to the technical scheme provided by the invention, the flexible scheduling of the fan is realized, the potential of the fan is fully utilized, and meanwhile, the fault tolerance and the calculation efficiency of wind turbine generator management are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of wind power generation, and in particular, to a control method, device, terminal device, medium and product for a wind turbine generator set. Background Art

[0002] Currently, the control system of a wind farm mainly adopts a centralized control strategy, that is, all the wind turbines in the wind farm are uniformly scheduled and controlled through a central energy management system (EMS). Specifically, the central energy management system collects the operation data from each wind turbine, calculates the target power output of each wind turbine according to the collected data and the power grid demand, and then sends the target power instruction to each wind turbine, so that each wind turbine executes the instruction and feeds back the execution result.

[0003] However, this centralized control strategy is difficult to adapt to the individual differences of wind turbines, which may cause some wind turbines to operate at high load for a long time, accelerating equipment aging, while some wind turbines may not be fully utilized. When a certain wind turbine fails, the central energy management system also needs to recalculate and allocate power, resulting in a slow response speed and affecting the overall operation efficiency. Moreover, as the scale of the wind farm expands, the computational burden on the central energy management system will also increase sharply, making it difficult to meet the real-time requirements. Summary of the Invention

[0004] The embodiments of the present invention provide a control method, device, terminal device, medium and product for a wind turbine generator set to achieve flexible scheduling of wind turbines, fully utilize the potential of wind turbines, and at the same time improve the fault tolerance and computational efficiency of wind turbine generator set management.

[0005] In a first aspect, the embodiments of the present invention provide a control method for a wind turbine generator set, the method comprising:

[0006] Obtaining real-time working data of a wind turbine generator;

[0007] Predicting a current remaining life and a current failure probability according to the real-time working data by using a trained remaining life prediction model and a failure prediction model;

[0008] Determining a current available score according to the current remaining life, the current failure probability, and the current power output of the wind turbine generator;

[0009] Reporting the current available score to an energy management system, so that the energy management system selects a target wind turbine generator according to the received current available score and the current power grid demand, and sends a control instruction to the target wind turbine generator.

[0010] Optionally, after reporting the current available score to the energy management system so that the energy management system selects a target wind turbine according to the received current available score and the current power grid demand and issues a control instruction to the target wind turbine, the method further includes:

[0011] If it is selected as the target wind turbine, adjust the working state according to the control instruction, update the current state information, and then feedback it to the energy management system; the current state information includes remaining life, failure probability, and actual power output.

[0012] Optionally, determining the current available score according to the current remaining life, the current failure probability, and the current power output of the wind turbine includes:

[0013] Calculating the current available score using a preset scoring formula; the preset scoring formula is:

[0014] S i = w1·L i + w2·(1 - F i ) + w3·P i ;

[0015] Wherein, S i represents the current available score, L i represents the current remaining life, F i represents the current failure probability, P i represents the current power output, w1 represents the remaining life weight coefficient, w2 represents the failure probability weight coefficient, and w3 represents the power output weight coefficient.

[0016] Optionally, the method further includes:

[0017] If there is a wind turbine with a continuously increasing failure probability in the wind turbine set, increase the failure probability weight coefficient;

[0018] If the power grid demand fluctuation exceeds a preset threshold, increase the power output weight coefficient.

[0019] Optionally, the method further includes:

[0020] If there is abnormal data in the real-time working data, trigger a fault warning based on the abnormal data and update the current failure probability.

[0021] Optionally, before predicting the current remaining life and the current failure probability using the trained remaining life prediction model and the fault prediction model according to the real-time working data, the method further includes:

[0022] Obtain the historical working data of the wind turbine;

[0023] Extract the target key features based on the historical working data;

[0024] Use the target key features to train the remaining life prediction model and the fault prediction model.

[0025] In a second aspect, an embodiment of the present invention further provides a wind turbine control device, which includes:

[0026] A working data acquisition module, configured to acquire the real-time working data of the wind turbine;

[0027] A scoring parameter prediction module, configured to predict the current remaining life and the current fault probability according to the real-time working data by using the trained remaining life prediction model and the fault prediction model;

[0028] An available score determination module, configured to determine the current available score according to the current remaining life, the current fault probability, and the current power output of the wind turbine;

[0029] An available score reporting module, configured to report the current available score to the energy management system, so that the energy management system selects a target wind turbine according to the received current available score and the current grid demand, and sends a control instruction to the target wind turbine.

[0030] In a third aspect, an embodiment of the present invention further provides a terminal device, which includes:

[0031] One or more processors;

[0032] A memory, configured to store one or more programs;

[0033] When the one or more programs are executed by the one or more processors, the one or more processors implement the wind turbine control method provided by any embodiment of the present invention.

[0034] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the wind turbine control method provided by any embodiment of the present invention.

[0035] In a fifth aspect, an embodiment of the present invention further provides a computer program product, which includes a computer program, and when the program is executed by a processor, it implements the wind turbine control method provided by any embodiment of the present invention.

[0036] An embodiment of the present invention provides a wind turbine control method. First, real-time working data of a wind turbine is acquired. Then, based on the real-time working data, a trained remaining life prediction model and a fault prediction model are used to predict the current remaining life and the current fault probability. Next, the current available score is determined according to the current remaining life, the current fault probability, and the current power output of the wind turbine. After that, the current available score is reported to an energy management system, so that the energy management system selects target wind turbines according to the received current available score and the current grid demand, and issues control instructions to each target wind turbine. The wind turbine control method provided by the embodiment of the present invention improves the flexibility of scheduling by enabling the wind turbine to autonomously evaluate its available state and allowing the energy management system to directly schedule according to the available states of each wind turbine, realizes the reasonable distribution of the fan load, avoids the situation where some fans are overused while some fans are not fully utilized, reduces the maintenance cost and improves the power generation efficiency. At the same time, it also avoids the impact of some fan failures on the overall operation efficiency, improves the fault tolerance. Through the distributed decision-making mechanism, the computational pressure on the central energy management system is also reduced, and the real-time performance of scheduling is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flowchart of the wind turbine control method provided by Embodiment 1 of the present invention;

[0038] Figure 2 It is a schematic structural diagram of the wind turbine control device provided by Embodiment 2 of the present invention;

[0039] Figure 3 It is a schematic structural diagram of the terminal device provided by Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The present invention will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present invention, rather than limiting the present invention. Additionally, it should be noted that for the sake of description, only parts related to the present invention are shown in the drawings rather than all structures.

[0041] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of the steps can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the steps can be rearranged. The process can be terminated when its operations are completed, but it can also have additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0042] Embodiment 1

[0043] Figure 1 The flowchart of the wind turbine control method provided in the first embodiment of the present invention. This embodiment is applicable to the situation of scheduling wind turbines in real time based on grid demand. This method can be executed by the wind turbine control device provided in the embodiments of the present invention, and the device can be implemented in a hardware and / or software manner, and is generally integrated in a terminal device. As Figure 1 shown, the specific steps are as follows:

[0044] S11. Obtain the real-time working data of the wind turbine.

[0045] S12. According to the real-time working data, use the trained remaining life prediction model and fault prediction model to predict the current remaining life and the current fault probability.

[0046] S13. Determine the current available score according to the current remaining life, the current fault probability, and the current power output of the wind turbine.

[0047] S14. Report the current available score to the energy management system, so that the energy management system selects a target wind turbine according to the received current available score and the current grid demand, and sends a control instruction to the target wind turbine.

[0048] Specifically, each wind turbine (wind machine) can be modeled as an agent, and the distributed characteristics of the multi-agent system can be used to realize the autonomous management and optimal scheduling of the wind turbine unit. Each wind turbine agent has an independent remaining life prediction model and a fault prediction model, and can independently evaluate its own state. Specifically, the real-time working data of itself can be collected through sensors at intervals, including operation data such as power output, rotation speed, vibration amplitude, and bearing temperature, and can also include environmental data such as wind speed, wind direction, environmental temperature, and environmental humidity. Every time the real-time working data is obtained, the local trained remaining life prediction model and fault prediction model can be used to predict the current remaining life and the current fault probability of itself.

[0049] After obtaining the current remaining life, the current fault probability, and the current power output, the adjustable ability of the corresponding wind turbine agent can be evaluated based on a preset scoring rule, and the current available score can be calculated. Each wind turbine agent in the wind turbine unit can calculate the corresponding current available score, and report the current available score to the energy management system through its own communication module. The communication module initialization can be completed in advance, and the communication connection between each wind turbine agent and the energy management system can be established to support real-time communication and ensure the real-time and reliability of data transmission.

[0050] Optionally, determining the current available score according to the current remaining life, the current failure probability, and the current power output of the wind turbine includes: calculating the current available score using a preset score formula; the preset score formula is:

[0051] S i = w1·L i + w2·(1 - F i ) + w3·P i ;

[0052] where S i represents the current available score, L i represents the current remaining life, F i represents the current failure probability, P i represents the current power output, w1 represents the remaining life weight coefficient, w2 represents the failure probability weight coefficient, and w3 represents the power output weight coefficient. Each of the weight coefficients can be initialized according to the wind farm operation strategy and satisfies w1 + w2 + w3 = 1. Exemplarily, w1 = 0.4, w2 = 0.3, and w3 = 0.3. Additionally, the current remaining life, the current failure probability, and the current power output can all be pre-normalized to obtain corresponding values between 0 and 1 to ensure the accuracy of the score. For the current remaining life, the ratio of the remaining life duration T remaining to the designed life duration T design can be used, that is, L i = T remaining / T design . When the value is 1, it means the maximum remaining life of the fan (almost brand new), and when the value is 0, it means the fan is about to fail. Exemplarily, the designed life duration is 20 years. For the current failure probability, it represents the failure probability of the fan in the current operating state, that is, the possibility of the fan failing in a future period of time, and the probability value is between 0 and 1. For the current power output, the normalized value can be calculated based on the collected power output value in combination with the wind speed and the current operating state.

[0053] The energy management system can receive the current available scores reported by each fan agent through the score receiving module and can quickly select the fan agent that needs to be scheduled as the target wind turbine based on the score level through the decision-making module, so as to enable each fan agent to compete for positions through scoring, thereby realizing efficient and adaptive collaborative control and energy management of the wind farm. Specifically, the energy management system can sort all the received current available scores from high to low, and then select several fans with the highest scores as the target wind turbines to participate in power regulation according to the current grid demand (such as the demand power P grid ). Exemplarily, if P gridis relatively high, select the top N wind turbines with the highest scores. If P grid is relatively low, select the top M wind turbines with the highest scores, where M < N. After determining the target wind turbines, corresponding control commands can be sent to each target wind turbine. The control command can specifically include adjusting the target power P target and the adjustment time T, etc., to meet the current grid demand.

[0054] Based on the above technical solution, optionally, after reporting the current available score to the energy management system so that the energy management system selects target wind turbines according to the received current available score and the current grid demand and sends control commands to the target wind turbines, it further includes: If selected as the target wind turbine, adjust the working state according to the control command, and update the current state information and then feedback it to the energy management system; the current state information includes remaining life, failure probability, and actual power output.

[0055] Specifically, after the energy management system sends a control command, correspondingly, the wind turbine agent selected as the target wind turbine can receive the corresponding control command through its communication module, and thus perform autonomous adjustment according to the control command. Specifically, its power output can be adjusted to the target power P target . If the original power output is less than the target power, increase the power output. If the original power output is greater than the target power, decrease the power output. After the adjustment is completed, since the operation data changes, it may affect the remaining life and failure probability. Therefore, the remaining life and failure probability can be predicted and updated based on the updated working data, and the actual power output is also updated. Then, the updated remaining life, failure probability, actual power output and other current state information can be fed back to the energy management system. The energy management system can also monitor the operation state of the wind farm in real time to ensure the stability of the system. Further, the corresponding target wind turbine can also re-rate based on the updated current state information and report it to the energy management system for the next scheduling, thus forming a dynamic closed-loop control. The above process can also be cycled according to a preset period, that is, each wind turbine agent re-acquires real-time working data for scoring and reporting, and then the energy management system performs scheduling, so as to realize the continuous and efficient energy management of the wind farm. Each wind turbine agent can adjust its own behavior according to the latest state in each round of cycle, ensuring the maximization of the overall operation efficiency of the system.

[0056] Based on the above technical solution, optionally, the method further includes: If there is a wind turbine with a continuously increasing failure probability in the wind turbine group, increase the failure probability weight coefficient; if the grid demand fluctuation exceeds a preset threshold, increase the power output weight coefficient.

[0057] Specifically, as the energy management system continuously receives the status information fed back by each wind turbine agent, it can also dynamically adjust the weight coefficients in the preset scoring formula according to the changes in the status information to improve the accuracy of scheduling. Specifically, when the failure probability of a certain wind turbine agent continues to increase, the failure probability weight coefficient can be increased to increase the attention to possible failures and avoid overuse. It can also dynamically adjust the weight coefficients in the preset scoring formula according to the fluctuations in grid demand. Specifically, when the grid demand fluctuates greatly, the power output weight coefficient can be increased to increase the attention to power demand, so as to better meet the grid demand. After the energy management system updates the preset scoring formula, it can send the new preset scoring formula to each wind turbine agent so that each wind turbine agent can start using the new preset scoring formula for scoring.

[0058] Based on the above technical solution, optionally, the method further includes: if there is abnormal data in the real-time working data, triggering a fault warning based on the abnormal data and updating the current failure probability.

[0059] Specifically, if a certain wind turbine agent detects abnormal data, such as the vibration amplitude exceeding the threshold or the temperature being abnormal, etc., a fault warning can be triggered based on the abnormal data to indicate the possible fault state, and the current failure probability can be updated to a higher value for reporting the fault state and avoiding the scheduling of the energy management system. Through the independent decision-making ability of each wind turbine agent, it can be ensured that the system can still operate normally when some agents fail. In addition, the energy management system can also send start-stop control instructions to the wind turbine agent, etc., such as controlling the faulty wind turbine to stop working to improve the scheduling efficiency and save energy consumption.

[0060] Based on the above technical solution, optionally, before predicting the current remaining life and the current failure probability using the trained remaining life prediction model and failure prediction model according to the real-time working data, it further includes: obtaining the historical working data of the wind turbine; extracting target key features according to the historical working data; and training the remaining life prediction model and the failure prediction model using the target key features.

[0061] Specifically, the model training can be independently completed by the wind turbine agent. For each wind turbine agent, its own historical working data can be loaded for training, and preprocessing such as normalization and denoising can be performed on the historical working data in advance. For the remaining life prediction model (Remaining Life Model, RLM), it can specifically be used to predict the remaining life duration T remaining, the required historical working data may include power output history, running time (accumulative running hours), maintenance records (such as the time of replacing components), fault history, mechanical state data history (such as rotational speed, vibration amplitude, bearing temperature, etc.), and environmental data history, etc. Then, target key features can be extracted from the historical working data. The target key features required for the remaining useful life prediction model may include, but are not limited to, average power output, standard deviation of vibration amplitude, rising trend of bearing temperature, and accumulative running time, etc. The remaining useful life prediction model can be modeled using machine learning algorithms, such as support vector machine (SVM) or deep neural network (DNN). The support vector machine uses a regression model for prediction, maps the data to a high-dimensional space using a kernel function, finds the optimal hyperplane to minimize the prediction error, is suitable for small-sample data, and can handle non-linear relationships. The deep neural network constructs a multi-layer neural network and optimizes the model weights through the backpropagation algorithm, is suitable for large-scale data, and can capture complex non-linear features. After modeling, the extracted target key features can be used for training. After completion of training, the model performance can also be evaluated by calculating the mean square error (MSE) or mean absolute error (MAE), and further cross-validation can be used to optimize the generalization ability of the model. Subsequently, the remaining useful life prediction model can be applied for prediction, and the formula can be expressed as T remaining = f(Feature1,…,FeatureN), where f represents the trained remaining useful life prediction model, and Feature1,…,FeatureN represent the corresponding target key features.

[0062] For the fault prediction model (Fault Probability Model, FPM), it can be used to predict the fault probability F of the wind turbine under the current operating state i, the required historical working data may include power output history, mechanical state data history (such as rotational speed, vibration amplitude, bearing temperature, etc.), fault history (such as fault type (such as bearing fault, blade damage, etc.), operating state at the time of fault, fault occurrence frequency, etc.), and environmental data history (such as wind speed, temperature, etc.). Then, target key features can be extracted from the historical working data. The target key features required for the fault prediction model may include, but are not limited to, the peak value of the vibration amplitude, the rising rate of the bearing temperature, the standard deviation of the power fluctuation, and the abnormal change of the rotational speed, etc. The fault prediction model can be modeled using statistical methods or machine learning algorithms, such as a logistic regression model or a random forest model. The logistic regression model can calculate the fault probability of each wind turbine using the logistic regression algorithm based on the labeled fault and normal operation data. The random forest model uses the ensemble learning method of multiple decision trees, combines various fault prediction factors (such as vibration, temperature, power fluctuation, etc.), and judges whether there is a fault risk through a voting mechanism. After modeling, the extracted target key features can be used for training. After the training is completed, the effectiveness of the model can also be evaluated through indicators such as accuracy, recall rate, and F1 score. In particular, attention can be paid to whether it can accurately predict the fault risk at an early stage, so as to avoid the wind turbine from being overloaded before a fault occurs. Subsequently, the fault prediction model can be applied for prediction, and the formula can be expressed as F i = g(Vibration, Temperature, Power fluctuation, …), where g represents the trained fault prediction model, and Vibration, Temperature, Power fluctuation, … represent the corresponding target key features, such as vibration amplitude, bearing temperature, power fluctuation, etc.

[0063] The technical solution provided by the embodiments of the present invention first obtains the real-time working data of a wind turbine, then, based on the real-time working data, uses the trained remaining life prediction model and fault prediction model to predict the current remaining life and the current fault probability, and then determines the current available score according to the current remaining life, the current fault probability, and the current power output of the wind turbine. After that, the current available score is reported to the energy management system, so that the energy management system selects the target wind turbines according to the received current available score and the current grid demand, and issues control instructions to each target wind turbine. By enabling the wind turbine to autonomously evaluate its available state and directly scheduling by the energy management system according to the available states of each wind turbine, the flexibility of scheduling is improved, the reasonable distribution of the fan load is realized, the situation where some fans are overused while some fans are not fully utilized is avoided, the maintenance cost is reduced and the power generation efficiency is improved. At the same time, the impact of the failure of some fans on the overall operation efficiency is also avoided, the fault tolerance is improved, and the computing pressure on the central energy management system is reduced through the distributed decision-making mechanism, and the real-time performance of scheduling is improved.

[0064] Embodiment 2

[0065] Figure 2 FIG. is a schematic structural diagram of a wind turbine control device provided by Embodiment 2 of the present invention. This device can be implemented in a hardware and / or software manner and is generally integrated into a terminal device for executing the wind turbine control method provided by any embodiment of the present invention. As Figure 2 shown, the device includes:

[0066] A working data acquisition module 21, configured to acquire the real-time working data of a wind turbine;

[0067] A scoring parameter prediction module 22, configured to predict the current remaining life and the current fault probability according to the real-time working data by using the trained remaining life prediction model and fault prediction model;

[0068] An available score determination module 23, configured to determine the current available score according to the current remaining life, the current fault probability, and the current power output of the wind turbine;

[0069] An available score reporting module 24, configured to report the current available score to the energy management system, so that the energy management system selects the target wind turbines according to the received current available score and the current grid demand, and issues control instructions to the target wind turbines.

[0070] The technical solution provided by the embodiments of the present invention first obtains the real-time working data of a wind turbine, then, based on the real-time working data, uses a trained remaining life prediction model and a fault prediction model to predict the current remaining life and the current fault probability, and then determines the current available score according to the current remaining life, the current fault probability, and the current power output of the wind turbine. After that, the current available score is reported to the energy management system, so that the energy management system selects target wind turbines according to the received current available score and the current grid demand, and issues control commands to each target wind turbine. By having the wind turbine independently evaluate its own available state and having the energy management system directly schedule according to the available states of each wind turbine, the flexibility of scheduling is improved, the reasonable distribution of the fan load is achieved, the situation where some fans are overused while some fans are not fully utilized is avoided, the maintenance cost is reduced and the power generation efficiency is improved. At the same time, the impact of the failure of some fans on the overall operation efficiency is also avoided, the fault tolerance is enhanced, and through the distributed decision-making mechanism, the computational pressure on the central energy management system is also reduced, and the real-time performance of scheduling is improved.

[0071] Based on the above technical solution, optionally, the device further includes:

[0072] A status information feedback module, configured to, after reporting the current available score to the energy management system, so that the energy management system selects target wind turbines according to the received current available score and the current grid demand, and issues control commands to the target wind turbines, if it is selected as the target wind turbine, adjust the working state according to the control command, and after updating the current status information, feedback it to the energy management system; the current status information includes the remaining life, the fault probability, and the actual power output.

[0073] Based on the above technical solution, optionally, the available score determination module 23 is specifically configured to:

[0074] Calculate the current available score using a preset score formula; the preset score formula is:

[0075] S i = w1·L i + w2·(1 - F i ) + w3·P i ;

[0076] Wherein, S i represents the current available score, L i represents the current remaining life, F i represents the current fault probability, and P iP represents the current power output, w1 represents the remaining life weight coefficient, w2 represents the failure probability weight coefficient, and w3 represents the power output weight coefficient.

[0077] Based on the above technical solution, optionally, the device further includes:

[0078] A weight coefficient adjustment module, configured to:

[0079] If there is a wind turbine with a continuously increasing failure probability in the wind farm, increase the failure probability weight coefficient;

[0080] If the grid demand fluctuation exceeds a preset threshold, increase the power output weight coefficient.

[0081] Based on the above technical solution, optionally, the device further includes:

[0082] A fault detection module, configured to trigger a fault warning based on the abnormal data if there is abnormal data in the real-time working data, and update the current failure probability.

[0083] Based on the above technical solution, optionally, the device further includes:

[0084] A model training module, configured to:

[0085] Obtain the historical working data of the wind turbine;

[0086] Extract target key features according to the historical working data;

[0087] Use the target key features to train the remaining life prediction model and the fault prediction model.

[0088] The wind turbine control device provided by the embodiments of the present invention can execute the wind turbine control method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0089] It should be noted that in the embodiments of the above wind turbine control device, the included units and modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0090] Embodiment III

[0091] Figure 3 FIG. is a schematic structural diagram of a terminal device provided by Embodiment III of the present invention, showing a block diagram of an exemplary terminal device suitable for implementing the embodiments of the present invention. Figure 3The displayed terminal device is merely an example and shall not impose any limitation on the functions and scope of use of the embodiments of the present invention. As Figure 3 shown, the terminal device includes a processor 31, a memory 32, an input device 33, and an output device 34; the number of processors 31 in the terminal device may be one or more. Figure 3 Taking one processor 31 as an example, the processor 31, the memory 32, the input device 33, and the output device 34 in the terminal device may be connected through a bus or other means. Figure 3 Taking the connection through the bus as an example.

[0092] The memory 32, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the wind turbine control method in the embodiments of the present invention (for example, the working data acquisition module 21, the scoring parameter prediction module 22, the available score determination module 23, and the available score reporting module 24 in the wind turbine control device). The processor 31 executes various functional applications and data processing of the terminal device by running the software programs, instructions, and modules stored in the memory 32, that is, to implement the above-mentioned wind turbine control method.

[0093] The memory 32 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the terminal device, etc. In addition, the memory 32 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some instances, the memory 32 may further include a memory remotely set relative to the processor 31, and these remote memories can be connected to the terminal device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0094] The input device 33 can be used to obtain the real-time working data of the wind turbine, and generate key signal inputs related to the user settings and function controls of the terminal device, etc. The output device 34 can be used to report the available score to the energy management system, etc.

[0095] Embodiment Four

[0096] Embodiment Four of the present invention further provides a storage medium containing computer-executable instructions, and the computer-executable instructions are used to execute a wind turbine control method when executed by a computer processor. The method includes:

[0097] Obtain the real-time working data of the wind turbine;

[0098] Based on the real-time operating data, predict the current remaining life and the current failure probability using the trained remaining life prediction model and failure prediction model;

[0099] Determine the current available score according to the current remaining life, the current failure probability, and the current power output of the wind turbine;

[0100] Report the current available score to the energy management system, so that the energy management system selects a target wind turbine according to the received current available score and the current grid demand, and sends a control instruction to the target wind turbine.

[0101] The storage medium can be any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media such as CD-ROMs, floppy disks or magnetic tape devices; computer system memories or random access memories such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memories such as flash memories, magnetic media (such as hard disks or optical storage); registers or other similar types of memory elements, etc. The storage medium can also include other types of memories or combinations thereof. Additionally, the storage medium can be located in the computer system in which the program is executed, or can be located in a different second computer system that is connected to the computer system via a network (such as the Internet). The second computer system can provide program instructions to the computer for execution. The term "storage medium" can include two or more storage media that can reside in different locations (such as in different computer systems connected via a network). The storage medium can store program instructions (such as specifically implemented as a computer program) executable by one or more processors.

[0102] Of course, for a storage medium containing computer-executable instructions provided by an embodiment of the present invention, the computer-executable instructions are not limited to the method operations described above, and can also execute related operations in the wind turbine control method provided by any embodiment of the present invention.

[0103] A computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0104] The program code contained on a computer-readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination of the above.

[0105] From the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general-purpose hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disc of a computer, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention.

[0106] Embodiment Five

[0107] Embodiment Five of the present invention further provides a computer program product. The computer program product includes a computer program (which can also be referred to as code, instruction). The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it is used to execute the wind turbine control method provided in any of the above embodiments, and has the corresponding beneficial effects of the execution method.

[0108] Note that the above is only the preferred embodiment of the present invention and the applied technical principle. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described here. Various obvious changes, re-adjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A wind turbine control method, characterized in that: include: Obtain real-time working data of wind turbines; According to the real-time working data, the current remaining life and the current failure probability are predicted using the trained remaining life prediction model and the failure prediction model; determining a current availability score based on the current remaining life, the current probability of failure, and the current power output of the wind turbine; The current availability score is reported to the energy management system, so that the energy management system selects a target wind turbine generator according to the received current availability score and current grid demand, and issues a control instruction to the target wind turbine generator.

2. The wind turbine control method according to claim 1, characterized in that: After reporting the current available score to the energy management system so that the energy management system selects a target wind turbine generator according to the received current available score and the current grid demand, and issues a control instruction to the target wind turbine generator, the method further includes: If it is selected as the target wind turbine, the working state is adjusted according to the control instruction, and the current state information is updated and fed back to the energy management system; the current state information includes the remaining life, failure probability and actual power output.

3. The wind turbine control method according to claim 1, characterized in that: The determining of the current availability score according to the current remaining life, the current failure probability and the current power output of the wind turbine includes: The currently available score is calculated using a preset score formula; the preset score formula is: S i =w1·L i +w2·(1-F i )+w3·P i ; Among them, S i Represents the currently available score, L i represents the current remaining life, F i represents the current failure probability, P i represents the current power output, w1 represents the remaining life weight coefficient, w2 represents the failure probability weight coefficient, and w3 represents the power output weight coefficient.

4. The wind turbine control method according to claim 3, characterized in that: The method further comprises: If there is a wind turbine in the wind turbine group whose failure probability continues to increase, then the failure probability weight coefficient is increased; If the grid demand fluctuation exceeds a preset threshold, the power output weight coefficient is increased.

5. The wind turbine control method according to claim 1, characterized in that: The method further comprises: If there is abnormal data in the real-time working data, a fault warning is triggered based on the abnormal data, and the current fault probability is updated.

6. The wind turbine control method according to claim 1, characterized in that: Before predicting the current remaining life and the current failure probability using the trained remaining life prediction model and the failure prediction model according to the real-time working data, the method further includes: Acquiring historical operating data of the wind turbine; Extracting target key features based on the historical work data; The remaining life prediction model and the fault prediction model are trained using the target key features.

7. A wind turbine control device, characterized in that: include: A working data acquisition module is used to acquire real-time working data of the wind turbine; A scoring parameter prediction module is used to predict the current remaining life and the current failure probability using the trained remaining life prediction model and the failure prediction model according to the real-time working data; an availability score determination module, configured to determine a current availability score according to the current remaining life, the current failure probability, and the current power output of the wind turbine; The available score reporting module is used to report the current available score to the energy management system, so that the energy management system selects a target wind turbine generator according to the received current available score and current grid demand, and issues a control instruction to the target wind turbine generator.

8. A terminal device, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the wind turbine control method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the wind turbine control method as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the wind turbine control method according to any one of claims 1 to 6.

Images