Method and system for evaluating a speed regulation model of a doubly-fed motor system

Abstract

This invention discloses a method and system for evaluating speed regulation models of doubly-fed induction generator (DFIG) systems, belonging to the field of DFIG technology. The invention controls the DFIG to operate a first speed regulation model within a target cycle and obtains a first score; it acquires the corresponding grid data and environmental data of the DFIG within the target cycle and generates an external feature vector based on the grid data and environmental data; it determines a corresponding second score based on the external feature vector and multiple second speed regulation models, obtaining multiple second scores. The first speed regulation model is the model that actually performs control among multiple preset speed regulation models, and the second speed regulation model is a model other than the model that actually performs control among multiple preset speed regulation models. The evaluation result of each preset speed regulation model is determined based on the first score and the multiple second scores, achieving the beneficial effect of improving the stability of power generation.

Description

Technical Field

[0001] This invention relates to the field of doubly-fed motor technology, and in particular to a method and system for evaluating the speed regulation model of a doubly-fed motor system. Background Technology

[0002] Doubly fed induction generator (DFIG) systems, as core components of wind turbine generators, play a crucial role in the renewable energy sector. With the acceleration of the global energy transition, wind power installed capacity continues to grow. DFIG systems achieve independent control of active and reactive power through rotor-side converters, enabling them to adapt to wind speed fluctuations and changes in grid demand. However, in actual operation, the speed regulation model of the DFIG directly affects the system's dynamic response, power quality, and overall efficiency. Common speed regulation models include scalar control models, vector control models, direct torque control models, and model predictive control models. However, these technologies often have different shortcomings in practical applications, making it difficult for a single speed regulation model to maintain optimal performance under all operating conditions. This results in the inability to select a good speed regulation model during the speed regulation process of the DFIG system, leading to a decrease in power generation stability. Summary of the Invention

[0003] The present invention aims to solve at least one of the problems existing in the prior art, and provides a method and system for evaluating the speed regulation model of a doubly fed motor system.

[0004] One aspect of the present invention provides a method for evaluating a speed regulation model of a doubly-fed induction generator system, comprising: Control the doubly-fed motor to run the first speed regulation model in the target cycle and obtain the first score; Obtain the grid data and environmental data corresponding to the doubly fed motor in the target cycle, and generate an external feature vector based on the grid data and the environmental data; The corresponding second score is determined based on the external feature vector and multiple second speed regulation models, and multiple second scores are obtained. The first speed regulation model is the model that actually performs control among multiple preset speed regulation models, and the first speed regulation model is a model other than the model that actually performs control among multiple preset speed regulation models. The evaluation result of each preset speed regulation model is determined based on the first score and the plurality of second scores.

[0005] Optionally, the step of determining the corresponding second score based on the external feature vector and multiple second speed regulation models to obtain multiple second scores includes: The external feature vector is input into the digital twin model to update the state of the digital twin model; Parallel simulations are performed on each second speed regulation model based on the updated digital twin model to obtain virtual response data for each second speed regulation model; The second score is calculated based on the virtual response data of each second speed regulation model.

[0006] Optionally, the step of performing parallel simulations on each second speed control model based on the updated digital twin model to obtain virtual response data for each second speed control model includes: Calculate the corresponding virtual control quantity for each second speed regulation model; The virtual speed deviation, virtual power tracking error, and virtual power device temperature rise are simulated based on the virtual control quantity and the digital twin model. The virtual response data is determined based on the virtual speed deviation, the virtual power tracking error, and the temperature rise of the virtual power device.

[0007] Optionally, before determining the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores, the method further includes: The tendency score of the first speed regulation model is determined based on the output of the CMDP policy network. Calculate the inverse probability weight based on the propensity score; The first score is weighted and corrected according to the inverse probability weights to obtain the corrected score.

[0008] Optionally, calculating the inverse probability weight based on the propensity score includes: Calculate the initial inverse probability weights based on the propensity scores; The smoothing inverse probability weights are determined based on the initial inverse probability weights and the smoothing algorithm; The smoothed inverse probability weights are normalized to obtain the inverse probability weights.

[0009] Optionally, after determining the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores, the method further includes: Update the constrained Markov decision process policy network based on the evaluation results of all preset speed regulation models; The model selection probability distribution for the next cycle is determined based on the updated policy network; The speed regulation model to be actually executed in the next cycle is selected based on the probability distribution of the model.

[0010] Optionally, obtaining the grid data and environmental data corresponding to the doubly-fed motor in the target cycle includes: Collect grid voltage, grid frequency, and grid harmonics as grid data; Collect wind speed, ambient temperature, and equipment temperature as environmental data; The external feature vector is generated based on the power grid data and the environmental data.

[0011] Another aspect of the present invention provides a speed regulation model evaluation system for a doubly-fed motor system, comprising: The control module is used to control the doubly fed motor to run the first speed regulation model in the target cycle and obtain the first score; The monitoring module is used to acquire the grid data and environmental data corresponding to the doubly fed motor in the target cycle, and generate an external feature vector based on the grid data and the environmental data. The simulation module is used to determine the corresponding second score based on the external feature vector and multiple second speed regulation models, and obtain multiple second scores, wherein the first speed regulation model is the model that actually performs control among multiple preset speed regulation models, and the first speed regulation model is a model other than the model that actually performs control among multiple preset speed regulation models. The evaluation module is used to determine the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores.

[0012] In another aspect, the present invention provides a speed regulation model evaluation device for a doubly-fed motor system, the doubly-fed motor system speed regulation model evaluation device comprising: a memory, a processor, and a doubly-fed motor system speed regulation model evaluation program stored in the memory and executable on the processor, the doubly-fed motor system speed regulation model evaluation program being configured to implement the steps of the doubly-fed motor system speed regulation model evaluation method described above.

[0013] In another aspect, the present invention provides a storage medium storing a doubly fed motor system speed regulation model evaluation program, wherein when the doubly fed motor system speed regulation model evaluation program is executed by a processor, the program implements the steps of the doubly fed motor system speed regulation model evaluation method described above.

[0014] The present invention provides a method and system for evaluating speed control models of doubly-fed induction generator (DFIG) systems. This method controls the DFIG to run a first speed control model within a target cycle and obtains a first score. It also acquires the corresponding grid data and environmental data for the DFIG during the target cycle. Simultaneously, it generates an external feature vector based on the grid data and environmental data. Based on the external feature vector and multiple second speed control models, it determines corresponding second scores, resulting in multiple second scores. Finally, it determines the evaluation result of each preset speed control model based on the first score and the multiple second scores. This allows for accurate evaluation of the speed control models, facilitating the subsequent selection of better-performing models for controlling the DFIG, thereby improving stability. Attached Figure Description

[0015] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0016] Figure 1 This is a schematic diagram of the structure of the speed regulation model evaluation device for the doubly fed motor system in the hardware operating environment involved in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the first embodiment of the speed regulation model evaluation method for doubly-fed motor systems of the present invention. Figure 3 This is a flowchart illustrating the second embodiment of the speed regulation model evaluation method for doubly fed motor systems of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details are presented in the embodiments of the present invention to facilitate a better understanding of the invention. However, the technical solutions claimed in the present invention can be implemented even without these technical details and various variations and modifications based on the following embodiments. The division of the following embodiments is for ease of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined with and referenced by each other without contradiction.

[0018] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of the speed regulation model evaluation device for the doubly fed motor system in the hardware operating environment involved in the embodiments of the present invention.

[0019] like Figure 1As shown, the speed regulation model evaluation device for the doubly-fed motor system may include: a processor 1001, a communication bus 1002, an interactive device 1003, a network interface 1004, and a memory 1005. The processor 1001 may be a central processing unit (CPU). The communication bus 1002 is used to enable communication between these components. The interactive device 1003 may include a display screen or an input unit such as a keyboard. Optionally, the interactive device 1003 may also be connected to the communication bus via standard wired or wireless interfaces. The network interface 1004 may optionally include standard wired or wireless interfaces (such as a Wi-Fi interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0020] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the evaluation equipment for the speed regulation model of the doubly fed motor system. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0021] The memory 1005, which serves as a storage medium, may include an operating system, a data storage module, a network communication module, a user interface module, and a speed regulation model evaluation program for a doubly-fed motor system.

[0022] exist Figure 1 In the doubly-fed motor system speed regulation model evaluation device shown, the network interface 1004 is mainly used for data communication with other devices; the interactive device 1003 is mainly used for data interaction with the user; the processor 1001 and memory 1005 in the doubly-fed motor system speed regulation model evaluation device of the present invention can be set in the doubly-fed motor system speed regulation model evaluation device. The doubly-fed motor system speed regulation model evaluation device calls the doubly-fed motor system speed regulation model evaluation program stored in the memory 1005 through the processor 1001 and executes the doubly-fed motor system speed regulation model evaluation method provided in the embodiment of the present invention.

[0023] This invention provides a method for evaluating the speed regulation model of a doubly-fed induction generator system, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the speed regulation model evaluation method for a doubly fed motor system according to the present invention.

[0024] In this embodiment, the speed regulation model evaluation method for the doubly fed motor system includes steps S1 to S4.

[0025] Step S1: Control the doubly fed motor to run the first speed regulation model in the target cycle and obtain the first score.

[0026] It should be noted that there are various types of speed regulation models. Currently, common speed regulation models generally include scalar control models, vector control models, direct torque control models, and model predictive control models. Different speed regulation models generally have different technical advantages and disadvantages. For example, the principle of the direct torque control model is to directly control the stator flux linkage and electromagnetic torque through switching, without the need for an inner current loop or rotating coordinate transformation. Its advantage is extremely fast dynamic response, generally less than 1 millisecond, but it has large torque ripple and the switching frequency is not fixed. In this embodiment, the first speed regulation model is not limited to one type of control model and can be any common speed regulation model. Specifically, the first speed regulation model here is a model that directly regulates the doubly-fed motor.

[0027] Step S2: Obtain the grid data and environmental data corresponding to the doubly fed motor in the target cycle, and generate an external feature vector based on the grid data and the environmental data.

[0028] It should be noted that the target period here can be set by the user, and corresponding power grid data is collected within the target period. Commonly, power grid voltage, power grid frequency, and harmonics included in the power grid can be collected. Simultaneously, wind speed, ambient temperature, and equipment temperature are typically collected as environmental data. Further, the external feature vector is generated based on the power grid data and the environmental data.

[0029] Step S3: Determine the corresponding second score based on the external feature vector and multiple second speed regulation models to obtain multiple second scores, wherein the first speed regulation model is the model that actually performs control among multiple preset speed regulation models, and the first speed regulation model is a model other than the model that actually performs control among multiple preset speed regulation models.

[0030] In this embodiment, a simulated virtual environment is generated based on the external feature vector. Simulated data corresponding to each second speed control model is calculated within this virtual environment. The simulated data is then scored to obtain the second score. Here, the first speed control model is a model that actually controls the doubly-fed motor among multiple preset speed control models, with the control limited to the target cycle time. Different models can be selected as the first speed control model in other time intervals outside the target cycle, thereby making the overall scoring process more accurate.

[0031] Step S4: Determine the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores.

[0032] In this embodiment, it should be noted that the first score is determined based on the control results obtained from controlling the doubly-fed induction generator (DFIG). Commonly, the first score is determined based on the stator current total harmonic distortion (THD), active power tracking error integral, reactive power fluctuation amplitude, and dynamic response time. Optionally, the first score can be determined according to a preset scoring table, for example, the scoring table includes: the weight corresponding to each data item, and the score value corresponding to each data item in a data interval. The first score is calculated by comparing the obtained stator current THD, active power tracking error integral, reactive power fluctuation amplitude, and dynamic response time with the preset scoring table.

[0033] In this embodiment, the doubly-fed induction generator (DFIG) is controlled to run a first speed regulation model in a target cycle and a first score is obtained. The corresponding grid data and environmental data of the DFIG in the target cycle are also acquired. Simultaneously, an external feature vector is generated based on the grid data and the environmental data. A second score is determined based on the external feature vector and multiple second speed regulation models, resulting in multiple second scores. The evaluation result of each preset speed regulation model is determined based on the first score and the multiple second scores. This allows for accurate evaluation of the speed regulation model, facilitating the subsequent selection of a better-performing model to control the DFIG, thereby improving stability.

[0034] Furthermore, based on the first embodiment, a second embodiment of the speed regulation model evaluation method for doubly-fed motor systems of the present invention is proposed. In this embodiment, referring to... Figure 3 The step of determining the corresponding second score based on the external feature vector and multiple second speed regulation models to obtain multiple second scores includes steps S31 to S33.

[0035] Step S31: Input the external feature vector into the digital twin model and update the state of the digital twin model.

[0036] The collected external feature vectors, such as grid voltage, load torque, and speed, are injected into the digital twin model. The model's internal parameters are updated through online parameter identification and state estimation, ensuring its dynamic characteristics are synchronized with the actual doubly-fed induction generator in real time. This allows the digital twin model to closely approximate the actual situation.

[0037] Step S32: Perform parallel simulations on each second speed regulation model based on the updated digital twin model to obtain virtual response data for each second speed regulation model.

[0038] Multiple candidate second speed regulation models were simulated with the updated digital twin model under the same operating conditions. Parallel simulations were performed, with each model independently outputting control commands and driving the digital twin model to run. The virtual dynamic response was recorded. Data such as speed tracking accuracy and power regulation quality demonstrated by each model during the simulation process were also analyzed.

[0039] Step S33: Calculate the corresponding second score based on the virtual response data of each second speed regulation model.

[0040] Optionally, multiple second scores can be calculated by comparing the simulated data with the preset score table in the first embodiment.

[0041] In this embodiment, the state of the digital twin model is updated by inputting the external feature vector into the digital twin model; parallel simulation is performed on each second speed regulation model according to the updated digital twin model to obtain virtual response data for each second speed regulation model; and the corresponding second score is calculated based on the virtual response data of each second speed regulation model, thereby effectively improving the accuracy of the second score.

[0042] Furthermore, the parallel simulation of each second speed control model based on the updated digital twin model to obtain virtual response data for each second speed control model includes: Calculate the corresponding virtual control quantity for each second speed regulation model; The virtual speed deviation, virtual power tracking error, and virtual power device temperature rise are simulated based on the virtual control quantity and the digital twin model. The virtual response data is determined based on the virtual speed deviation, the virtual power tracking error, and the temperature rise of the virtual power device.

[0043] It should be noted that in this embodiment, the virtual speed deviation, virtual power tracking error, and virtual power device temperature rise are included. In other embodiments, other common data types can be used, such as one or more of the stator current total harmonic distortion rate, active power tracking error integral, reactive power fluctuation amplitude, and dynamic response time. The virtual response data of the above types are obtained through simulation calculation.

[0044] In this embodiment, by calculating the corresponding virtual control quantity according to each second speed regulation model, virtual speed deviation, virtual power tracking error, and virtual power device temperature rise are simulated based on the virtual control quantity and the digital twin model, and the virtual response data is determined based on the virtual speed deviation, the virtual power tracking error, and the virtual power device temperature rise, thereby improving the accuracy of the virtual response data.

[0045] Furthermore, based on the first or second embodiment, a third embodiment of the doubly-fed motor system speed regulation model evaluation method of the present invention is proposed. In this embodiment, before determining the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores, the method further includes: The tendency score of the first speed regulation model is determined based on the output of the CMDP policy network. Calculate the inverse probability weight based on the propensity score; The first score is weighted and corrected according to the inverse probability weights to obtain the corrected score.

[0046] In this embodiment, it should be explained that the CMDP policy network refers to a deep neural network trained based on a constrained Markov decision process. The CMDP policy network outputs a propensity score corresponding to the first speed regulation model. It should be noted that this propensity score refers to the conditional probability by which the CMDP policy network selects the current speed regulation model as the first speed regulation model, given the observed operating conditions. The CMDP policy network needs to receive the external feature vector of the current state.

[0047] It is important to explain that during the actual operation of a doubly-fed induction generator (DFIG), not every speed control model achieves equal performance. Currently, users often choose suitable speed control models for different time periods. For models with good performance and high scores, the system prioritizes high-scoring models and models suitable for the scenario during automatic adjustment. Therefore, models are not selected equally. The purpose of calculating inverse probability weights based on the propensity score is to discount frequently selected models and compensate for rarely selected models, thereby simulating a fair experimental environment of "random allocation." In a specific embodiment, the preset CMDP policy network always performs the following steps: selecting the first type of speed control model during periods of good operating conditions and selecting the second type of speed control model during periods of poor operating conditions. This leads to the first type of speed control model being artificially overestimated, while the score of the second type of speed control model is suppressed. The first and second types of speed control models here are not the same and are not used to limit specific models.

[0048] Furthermore, the step of calculating the inverse probability weight based on the propensity score includes: Calculate the initial inverse probability weights based on the propensity scores; The smoothing inverse probability weights are determined based on the initial inverse probability weights and the smoothing algorithm; The smoothed inverse probability weights are normalized to obtain the inverse probability weights.

[0049] In this embodiment, an initial inverse probability weight is calculated based on the propensity score; a smoothed inverse probability weight is determined based on the initial inverse probability weight and a smoothing algorithm; and the smoothed inverse probability weight is normalized to obtain the inverse probability weight. In this embodiment, the calculation of the initial inverse probability weight follows the reciprocal relationship of the propensity score, i.e., the initial weight value is obtained by dividing the value of 1 by the propensity score. This numerical relationship reflects a statistical compensation mechanism for low selection probability scenarios. The smaller the propensity score, the lower the probability that the model will be selected by the policy network in the corresponding scenario, and thus a higher initial weight is assigned to offset the selection bias. The smoothing algorithm employs either additive smoothing or truncated smoothing strategies. Specifically, additive smoothing calculates the smoothing weight by adding the stabilization coefficient to the propensity score as the denominator, or truncated smoothing sets an upper limit for the weight and takes the smaller of the initial inverse probability weight and the upper limit as the smoothing weight, thereby effectively suppressing the weight explosion problem caused by the propensity score approaching zero. The normalization process calculates the arithmetic mean of all smoothing weights in the current evaluation batch and divides the smoothing weight of a single period by the arithmetic mean to obtain the final inverse probability weight. This normalizes the statistical mean of the inverse probability weight in the evaluation batch to a unit value, forming a Hájek-type weighted estimator, which ensures the consistency of weight scale and numerical stability in subsequent performance evaluations of the speed regulation model.

[0050] Furthermore, based on any of the above embodiments, a fourth embodiment of the doubly-fed motor system speed regulation model evaluation method of the present invention is proposed. In this embodiment, after determining the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores, the method further includes: Update the constrained Markov decision process policy network based on the evaluation results of all preset speed regulation models; The model selection probability distribution for the next cycle is determined based on the updated policy network; The speed regulation model to be actually executed in the next cycle is selected based on the probability distribution of the model.

[0051] In this embodiment, optionally, the model for actual speed adjustment in the next cycle is determined according to the constrained Markov decision process policy network. Preferably, the second external feature vector corresponding to the next cycle is input to the constrained Markov decision process policy network, and the first speed adjustment model corresponding to the rain cycle is determined according to the output result of the constrained Markov decision process policy network.

[0052] In this embodiment, the constrained Markov decision process policy network is updated based on the evaluation results of all preset speed regulation models. The model selection probability distribution for the next cycle is determined based on the updated policy network. The speed regulation model to be actually executed in the next cycle is selected based on the model selection probability distribution, thereby improving the stability of the doubly fed motor operation.

[0053] Furthermore, based on any of the above embodiments, a fifth embodiment of the speed regulation model evaluation method for doubly-fed motor systems of the present invention is proposed. In this embodiment, obtaining the grid data and environmental data corresponding to the doubly-fed motor in the target cycle includes: Collect grid voltage, grid frequency, and grid harmonics as grid data; Collect wind speed, ambient temperature, and equipment temperature as environmental data; The external feature vector is generated based on the power grid data and the environmental data.

[0054] In this embodiment, generally, the voltage amplitude and phase of the power grid, the power grid frequency, and the content of each harmonic are simultaneously collected as power grid data through voltage transformers, frequency detection modules, and harmonic analyzers arranged at the power grid connection points. Additionally, optionally, wind speed, ambient temperature, and equipment temperature are collected as environmental data through wind speed sensors on the top of the nacelle, ambient temperature sensors, and temperature sensors mounted on the stator windings and power device surfaces. Wind speed is used to determine the magnitude of the mechanical torque input to the wind turbine and is the main disturbance source on the load side of the doubly-fed induction generator (DFIG), while ambient temperature affects the motor winding resistance and magnetic circuit characteristics, thereby changing the electromagnetic parameters.

[0055] Furthermore, this invention also proposes a speed regulation model evaluation system for a doubly fed motor system, comprising a control module, a monitoring module, a simulation module, and an evaluation module.

[0056] The control module is used to control the doubly fed motor to run the first speed regulation model in the target cycle and obtain the first score.

[0057] The monitoring module is used to acquire the grid data and environmental data corresponding to the doubly fed motor in the target cycle, and generate an external feature vector based on the grid data and the environmental data.

[0058] The simulation module is used to determine the corresponding second score based on the external feature vector and multiple second speed regulation models, and obtain multiple second scores, wherein the first speed regulation model is the model that actually performs control among multiple preset speed regulation models, and the first speed regulation model is a model other than the model that actually performs control among multiple preset speed regulation models.

[0059] The evaluation module is used to determine the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores.

[0060] Furthermore, this invention also proposes a speed regulation model evaluation device for a doubly-fed motor system. The doubly-fed motor system speed regulation model evaluation device includes: a memory, a processor, and a doubly-fed motor system speed regulation model evaluation program stored in the memory and executable on the processor. The doubly-fed motor system speed regulation model evaluation program is configured to implement the steps of the doubly-fed motor system speed regulation model evaluation method described above.

[0061] Furthermore, this embodiment of the invention also proposes a storage medium storing a speed regulation model evaluation program for a doubly fed motor system. When the doubly fed motor system speed regulation model evaluation program is executed by a processor, it implements the steps of the doubly fed motor system speed regulation model evaluation method described above.

[0062] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0063] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0064] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0065] Those skilled in the art will understand that the above embodiments are specific implementations of the present invention, and in practical applications, various changes can be made in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A method for evaluating the speed regulation model of a doubly-fed induction generator system, characterized in that, include: Control the doubly-fed motor to run the first speed regulation model in the target cycle and obtain the first score; Obtain the grid data and environmental data corresponding to the doubly fed motor in the target cycle, and generate an external feature vector based on the grid data and the environmental data; The corresponding second score is determined based on the external feature vector and multiple second speed regulation models, and multiple second scores are obtained. The first speed regulation model is the model that actually performs control among multiple preset speed regulation models, and the first speed regulation model is a model other than the model that actually performs control among multiple preset speed regulation models. The evaluation result of each preset speed regulation model is determined based on the first score and the plurality of second scores.

2. The evaluation method for speed regulation model of doubly-fed motor system according to claim 1, characterized in that, The step of determining the corresponding second score based on the external feature vector and multiple second speed regulation models, resulting in multiple second scores, includes: The external feature vector is input into the digital twin model to update the state of the digital twin model; Parallel simulations are performed on each second speed regulation model based on the updated digital twin model to obtain virtual response data for each second speed regulation model; The second score is calculated based on the virtual response data of each second speed regulation model.

3. The evaluation method for speed regulation model of doubly-fed motor system according to claim 2, characterized in that, The process of performing parallel simulations on each second speed regulation model based on the updated digital twin model to obtain virtual response data for each second speed regulation model includes: Calculate the corresponding virtual control quantity for each second speed regulation model; The virtual speed deviation, virtual power tracking error, and virtual power device temperature rise are simulated based on the virtual control quantity and the digital twin model. The virtual response data is determined based on the virtual speed deviation, the virtual power tracking error, and the temperature rise of the virtual power device.

4. The evaluation method for the speed regulation model of a doubly-fed motor system according to claim 1, characterized in that, Before determining the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores, the method further includes: The tendency score of the first speed regulation model is determined based on the output of the CMDP policy network. Calculate the inverse probability weight based on the propensity score; The first score is weighted and corrected according to the inverse probability weights to obtain the corrected score.

5. The evaluation method for the speed regulation model of a doubly-fed motor system according to claim 4, characterized in that, The step of calculating the inverse probability weight based on the propensity score includes: Calculate the initial inverse probability weights based on the propensity scores; The smoothing inverse probability weights are determined based on the initial inverse probability weights and the smoothing algorithm; The smoothed inverse probability weights are normalized to obtain the inverse probability weights.

6. The evaluation method for speed regulation model of doubly-fed motor system according to claim 1, characterized in that, After determining the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores, the method further includes: Update the constrained Markov decision process policy network based on the evaluation results of all preset speed regulation models; The model selection probability distribution for the next cycle is determined based on the updated policy network; The speed regulation model to be actually executed in the next cycle is selected based on the probability distribution of the model.

7. The evaluation method for speed regulation model of a doubly-fed motor system according to any one of claims 1 to 6, characterized in that, The step of acquiring the grid data and environmental data corresponding to the doubly-fed motor in the target cycle includes: Collect grid voltage, grid frequency, and grid harmonics as grid data; Collect wind speed, ambient temperature, and equipment temperature as environmental data; The external feature vector is generated based on the power grid data and the environmental data.

8. A speed regulation model evaluation system for a doubly-fed induction generator system, characterized in that, include: The control module is used to control the doubly fed motor to run the first speed regulation model in the target cycle and obtain the first score; The monitoring module is used to acquire the grid data and environmental data corresponding to the doubly fed motor in the target cycle, and generate an external feature vector based on the grid data and the environmental data. The simulation module is used to determine the corresponding second score based on the external feature vector and multiple second speed regulation models, and obtain multiple second scores, wherein the first speed regulation model is the model that actually performs control among multiple preset speed regulation models, and the first speed regulation model is a model other than the model that actually performs control among multiple preset speed regulation models. The evaluation module is used to determine the evaluation result of each preset speed regulation model based on the first score and the plurality of second scores.

9. An evaluation device for a speed regulation model of a doubly-fed induction generator system, characterized in that, The doubly-fed motor system speed regulation model evaluation device includes: a memory, a processor, and a doubly-fed motor system speed regulation model evaluation program stored in the memory and executable on the processor. The doubly-fed motor system speed regulation model evaluation program is configured to implement the steps of the doubly-fed motor system speed regulation model evaluation method according to any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a speed regulation model evaluation program for a doubly fed motor system. When the doubly fed motor system speed regulation model evaluation program is executed by the processor, it implements the steps of the speed regulation model evaluation method for a doubly fed motor system according to any one of claims 1 to 7.

Images