A data-driven and experiment-enhanced modulation method and device for power converter

Abstract

The application discloses a kind of based on data driving and experimental enhancement's electric energy converter modulation method, comprising the following steps: obtaining simulation performance index and corresponding simulation modulation data, and actual experimentally determined real performance index and corresponding experimental modulation data;Based on extreme gradient boosting algorithm, a prediction model is constructed, and the prediction model is trained using simulation performance index and corresponding simulation modulation data;Experimental modulation data is input into the trained prediction model to obtain predicted performance index;The difference between the predicted performance index and the real performance index is analyzed using the extreme gradient boosting algorithm to obtain the corresponding error correction parameter;The error correction parameter is introduced into the output end of the trained prediction model to obtain the best prediction model.The application also provides an electric energy converter modulation device.The method provided by the application can accurately reveal the behavior law and characteristics of the electric energy converter, providing better guidance for the selection of subsequent modulation parameters.

Description

Technical Field

[0001] This invention belongs to the technical field of power converters, and particularly relates to a power converter modulation method and apparatus based on data-driven and experimental enhancement. Background Technology

[0002] In recent years, with the continuous development and application of renewable energy, power converters, as key energy conversion and transmission devices, have played a crucial role. The instability and uncontrollability of renewable energy sources such as solar and wind power pose significant challenges to the modulation of power converters. Traditional power converter modulation modeling is typically based on empirical rules and theoretical models, which have limitations in fully considering the nonlinearity and complexity of the system. Furthermore, power converters face challenges in practical applications such as load variations and environmental disturbances, and traditional theoretical models often fail to adequately account for the impact of these practical factors.

[0003] Patent document CN116187540 A discloses an ultra-short-term power prediction method for wind farms based on spatiotemporal deviation correction. The steps are as follows: import numerical weather forecast data of the wind farm and its four adjacent grids, and simultaneously import real-time weather data obtained from the wind measurement tower of the wind farm; construct time-series features, and construct meteorological features after preprocessing the weather data; use the weather data as training data and optimize the sample set to establish an error correction model and obtain the weather data correction result; obtain the optimal model hyperparameters through Bayesian optimization method, establish a power prediction model, and predict the power generation of the wind farm.

[0004] Patent document CN 114237087A discloses a monitoring system early warning method, device, and computer-readable storage medium. The method includes: collecting operating data of at least one monitoring device; obtaining target feature data corresponding to the operating data according to a preset rule, and inputting the target feature data as input into a fault prediction model; obtaining a fault coefficient calculated by the fault prediction model based on the target feature data; and outputting early warning information based on the fault coefficient. Summary of the Invention

[0005] The purpose of this invention is to provide a modulation method and apparatus for a power converter. This method can more accurately reveal the behavior and characteristics of the power converter, and provide better guidance for the selection of subsequent modulation parameters.

[0006] To achieve the first objective of this invention, a data-driven and experimentally enhanced power converter modulation method is provided, comprising the following steps: The acquisition includes simulation performance indicators and corresponding simulation modulation data obtained from simulation, as well as real performance indicators and corresponding experimental modulation data measured in actual experiments.

[0007] A prediction model is constructed based on the extreme gradient boosting algorithm, and the prediction model is trained using simulation performance metrics and corresponding simulation modulation data.

[0008] The experimental modulation data is input into the trained prediction model to obtain the corresponding prediction performance index.

[0009] The extreme gradient boosting algorithm is used to analyze the difference between the predicted performance index and the actual performance index in order to obtain the corresponding error correction parameters.

[0010] Error correction parameters are introduced into the output of the trained prediction model to obtain the optimal prediction model for predicting the best power converter modulation parameters.

[0011] The extreme gradient boosting algorithm used in this invention consists of multiple decision trees, the number of which is consistent with the number of input modulation data. The objective function of the first decision tree is the first performance index corresponding to the first modulation data, and the output of the first decision tree is the first prediction result of the first modulation data. The objective function of the second decision tree is the difference between the first performance index and the first prediction result. The above process is repeated until the decision tree calculation for all simulated modulation data is completed, and the outputs of each decision tree are accumulated.

[0012] Specifically, the simulation performance indicators are obtained by using PLECS or MATLAB to simulate the modulation parameters and corresponding operating power and operating voltage across the entire range, in order to obtain the performance indicators under different modulation parameters in the simulation.

[0013] Specifically, the actual performance indicators are obtained by analyzing and calculating the modulation parameters and corresponding operating power and operating voltage collected in the physical experiment, so as to obtain the performance indicators under different modulation parameters in the experiment.

[0014] Specifically, the expression for the optimal prediction model is as follows: In the formula, This represents the prediction model obtained through training. This indicates the error correction parameter.

[0015] Specifically, the expression of the prediction model is as follows: In the formula, This represents the total number of decision trees in the extreme gradient boosting algorithm. Indicates the first The output is the number of decision trees, where the total number of decision trees is consistent with the amount of input data.

[0016] Specifically, the expression for the error correction parameter is as follows: In the formula, Indicates the first The number of group decision trees is based on the output of predicted performance metrics and actual performance metrics.

[0017] Specifically, the performance indicators include the efficiency of the power converter and / or current stress.

[0018] Specifically, the modulation parameters include the inner phase shift angle of the primary side bridge arm, the inner phase shift angle of the secondary side bridge arm, and the outer phase shift angle of the primary and secondary bridge arms.

[0019] To achieve the second objective of the present invention, a power converter modulation apparatus is provided, comprising a computer memory, a computer processor, and a computer program stored in the computer memory and executable on the computer processor, wherein the computer memory stores the data-driven and experimentally enhanced power converter modulation method as described above.

[0020] When the computer processor executes the computer program, it performs the following steps: Multiple preset modulation parameters are input into the optimal prediction model to output the modulation parameters that enable the power converter to achieve its optimal performance.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows: Based on the simulation data-driven modeling method, this paper improves upon it by integrating real experimental data into the dataset to train the extreme gradient boosting algorithm. This method can more accurately reveal the behavior and characteristics of power converters under complex operating conditions and reduce the total amount of data required. Attached Figure Description

[0022] Figure 1 This is a flowchart of the power converter modulation method provided in this embodiment; Figure 2 This is a flowchart of the prediction model provided in this embodiment; Figure 3 This is a flowchart illustrating the prediction process based on predicted and actual performance metrics, provided in this embodiment. Detailed Implementation

[0023] Exemplary embodiments of the invention will now be described with reference to the accompanying drawings. However, the invention may be embodied in many different forms and is not limited to the embodiments described herein. These embodiments are provided to fully and completely disclose the invention and to fully convey its scope to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the drawings is not intended to limit the invention. In the drawings, the same units / elements are referred to by the same reference numerals.

[0024] Unless otherwise stated, the terms used herein (including technical terms) have their common meaning as understood by one of ordinary skill in the art. Furthermore, it is understood that terms defined in commonly used dictionaries should be understood to have a meaning consistent with the context of their relevant field, and not to be interpreted as having an idealized or overly formal meaning.

[0025] like Figure 1 As shown, this embodiment provides a power converter modulation method based on data-driven and experimental enhancement, which includes the following steps: The acquisition includes simulation performance indicators and corresponding simulation modulation data obtained from simulation, as well as real performance indicators and corresponding experimental modulation data measured in actual experiments.

[0026] A prediction model is constructed based on the extreme gradient boosting algorithm, and the prediction model is trained using simulation performance metrics and corresponding simulation modulation data.

[0027] The experimental modulation data is input into the trained prediction model to obtain the corresponding prediction performance index.

[0028] The extreme gradient boosting algorithm is used to analyze the difference between the predicted performance index and the actual performance index in order to obtain the corresponding error correction parameters.

[0029] Error correction parameters are introduced into the output of the trained prediction model to obtain the optimal prediction model for predicting the best power converter modulation parameters.

[0030] More specifically, the simulation proposed in this embodiment uses PLECS simulation software to collect performance index values ​​under different modulation parameters. In the dual phase-shift modulation method of the bidirectional active bridge converter, each set of modulation parameter combinations includes the inner phase shift angle of the primary bridge arm, the inner phase shift angle of the secondary bridge arm, and the outer phase shift angle of the primary and secondary bridge arms. The corresponding performance indexes include efficiency and current stress, and a certain performance index can be selected for collection according to the actual application scenario.

[0031] Experimental measurements are performed by setting up a physical hardware platform, maintaining the same circuit parameters and operating conditions, and collecting performance index values ​​corresponding to different modulation parameters. Similarly, in the dual-phase-shift modulation method of the bidirectional active bridge converter, each set of modulation parameters includes the inner phase shift angle of the primary bridge arm, the inner phase shift angle of the secondary bridge arm, and the outer phase shift angle of both the primary and secondary bridge arms. The corresponding performance indicators include efficiency and current stress, and a specific performance indicator can be selected for collection based on the actual application scenario.

[0032] Extreme gradient boosting is an additive model consisting of multiple decision trees, requiring initial predictions. Residuals are calculated based on the predicted and observed values, and a decision tree with these residuals is created using the similarity scores of the residuals. The output of each decision tree becomes a new residual for the dataset, used to build another decision tree. This process is repeated until the residuals stop decreasing or a specified number of iterations are reached.

[0033] A prediction model is constructed based on the extreme gradient boosting algorithm. The simulation performance metrics and corresponding simulation modulation data are used as the training set to train the prediction model. The process is as follows: Figure 2 As shown, the objective function of the first decision tree is the first performance index corresponding to the first modulation data, and the output of the first decision tree is the first prediction result of the first modulation data. The objective function of the second decision tree is the difference between the first performance index and the first prediction result. The above process is repeated until the decision tree calculation of all simulated modulation data is completed, and the outputs of each decision tree are accumulated.

[0034] Its expression is: In the formula, This represents the total number of decision trees in the extreme gradient boosting algorithm. Indicates the first The output is the number of decision trees, where the total number of decision trees is consistent with the amount of input data.

[0035] like Figure 3 As shown, the error correction parameter takes the predicted performance metric and the actual performance metric as input, allowing the prediction model to learn the difference and thus obtain the corresponding error correction parameter. Its expression is as follows: In the formula, Indicates the first The number of group decision trees is based on the output of predicted performance metrics and actual performance metrics.

[0036] Finally, the prediction model and error correction parameters proposed in the above embodiments are combined to obtain the optimal prediction model, the expression of which is as follows: In the formula, This represents the prediction model obtained through training. This indicates the error correction parameter.

[0037] This embodiment also provides a power converter modulation device, which is based on the optimal prediction model proposed in the above embodiment. It includes a computer memory, a computer processor, and a computer program stored in the computer memory and executable on the computer processor. The computer memory uses the above-mentioned optimal prediction model.

[0038] When the computer processor executes the computer program, it performs the following steps: Multiple preset modulation parameters are input into the optimal prediction model to output the modulation parameters that enable the power converter to achieve its optimal performance.

[0039] This invention combines simulation data and experimental data, and uses a data-driven method to model the modulation mode of power converters. This helps to accelerate the engineering modeling process, reduce the difficulty of analyzing complex systems, and help solve the problem of errors between the model and actual operating conditions. It has great engineering application value and promotion prospects.

[0040] Those skilled in the art will understand that the embodiments of this application can be provided as methods, systems, or computer program products. The solutions in the embodiments of this application can be implemented using various computer languages, such as MATLAB, Python, C++, Java, etc. The simulation software used in the embodiments of this application... Various commonly used power electronics simulation software can be used, such as MATLAB and PLECS.

[0041] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0042] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0043] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application. Clearly, those skilled in the art can make various alterations and variations to this application without departing from its spirit and scope. Thus, if such alterations and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such alterations and variations.

Claims

1. A power converter modulation method based on data-driven and experimental enhancement, characterized in that, Includes the following steps: Acquire simulation performance indicators and corresponding simulation modulation data obtained from simulation, as well as real performance indicators and corresponding experimental modulation data measured in actual experiments; A prediction model is constructed based on the extreme gradient boosting algorithm, and the prediction model is trained using simulation performance metrics and corresponding simulation modulation data. The experimental modulation data is input into the trained prediction model to obtain the corresponding prediction performance index; The extreme gradient boosting algorithm is used to analyze the difference between the predicted performance index and the actual performance index in order to obtain the corresponding error correction parameters; Error correction parameters are introduced into the output of the trained prediction model to obtain the optimal prediction model for predicting the best power converter modulation parameters.

2. The power converter modulation method based on data-driven and experimental enhancement according to claim 1, characterized in that, The simulation performance indicators are obtained by using PLECS or MATLAB to simulate the modulation parameters and corresponding operating power and operating voltage across the entire range, in order to obtain the performance indicators under different modulation parameters in the simulation.

3. The power converter modulation method based on data-driven and experimental enhancement according to claim 1, characterized in that, The actual performance indicators are obtained by analyzing and calculating the modulation parameters and corresponding operating power and operating voltage collected in the physical experiment, so as to obtain the performance indicators under different modulation parameters in the experiment.

4. The power converter modulation method based on data-driven and experimental enhancement according to claim 1, characterized in that, The expression for the optimal prediction model is as follows: In the formula, This represents the prediction model obtained through training. This indicates the error correction parameter.

5. The data-driven and experimentally enhanced power converter modulation method according to claim 1 or 4, characterized in that, The expression for the prediction model is as follows: In the formula, This represents the total number of decision trees in the extreme gradient boosting algorithm. Indicates the first Output the number of group decision trees.

6. The data-driven and experimentally enhanced power converter modulation method according to claim 1 or 4, characterized in that, The expression for the error correction parameter is as follows: In the formula, Indicates the first The number of group decision trees is based on the output of predicted performance metrics and actual performance metrics.

7. The power converter modulation method based on data-driven and experimental enhancement according to claim 1, characterized in that, The performance indicators include the efficiency of the power converter and / or current stress.

8. A power converter modulation device, comprising a computer memory, a computer processor, and a computer program stored in the computer memory and executable on the computer processor, characterized in that, The computer memory stores the power converter modulation method based on data-driven and experimental enhancement as described in claim 1; When the computer processor executes the computer program, it performs the following steps: Multiple preset modulation parameters are input into the optimal prediction model to output the modulation parameters that enable the power converter to achieve its optimal performance.

Images