Specific subharmonic suppression pulse width modulation method and device based on Levenberg-Marquardt method

Abstract

One or more embodiments of this specification provide a method and apparatus for pulse width modulation with specific subharmonic suppression based on the Levenberg-Marquardt method. This method transforms the problem of pulse width modulation with specific subharmonic suppression in a converter into a system of transcendental equations, representing each subharmonic as a function of the switching angle. This eliminates the need for carrier generation and comparison with the modulated wave, simplifying the control process. Furthermore, the Levenberg-Marquardt method overcomes the potential unsolvable nature of existing algorithms, ensuring that the switching angle adjustment direction can be stably determined for the pulse width modulation problem.

Description

Technical Field

[0001] One or more embodiments of the present specification relate to the field of power electronics science and technology, and in particular, to a method and device for pulse width modulation with specific subharmonic suppression based on the Levenberg-Marquardt method. Background Art

[0002] With the advancement of power electronics technology and the cross-disciplinary integration of interdisciplinary studies, the problem of converter harmonic suppression has gradually shifted from traditional solutions such as improving converter topology and installing power quality compensators to improving converter harmonic suppression algorithms.

[0003] Traditional carrier comparison modulation methods, such as the SPWM (Sinusoidal Pulse Width Modulation) algorithm, are commonly used to suppress converter harmonics. This requires generating a carrier wave and comparing it with the modulating wave. This makes real-time converter control difficult and has low reliability.

[0004] Compared to carrier comparison modulation, PWM with subharmonic suppression offers advantages such as low switching frequency, low switching losses, good output waveform quality, high inverter efficiency, and a compact output filter. However, solving the nonlinear equations associated with the switching angles in PWM with subharmonic suppression remains a technical bottleneck, and existing methods often lack solutions, thus failing to address practical engineering problems. Summary of the Invention

[0005] In view of this, the purpose of one or more embodiments of this specification is to propose a pulse width modulation method and device for suppressing specific subharmonics based on the Levenberg-Marquardt method to solve the problem that existing pulse width modulation methods and devices have no solution and therefore cannot solve practical engineering problems.

[0006] Based on the above objectives, one or more embodiments of this specification provide a pulse width modulation method for suppressing specific subharmonics based on the Levenberg-Marquardt method, including:

[0007] Obtaining the output phase voltage of the three-level converter;

[0008] Representing each harmonic of the output phase voltage by a switching angle to obtain a group of transcendental equations related to the switching angle;

[0009] Solving the transcendental equations about the switching angles using the Levenberg-Marquardt method to obtain a set of switching angles;

[0010] The pulse width is modulated according to the set of switching angles.

[0011] Optionally, before obtaining the output phase voltage of the three-level converter, the method further includes:

[0012] The output phase voltage of the three-level converter is filtered to remove high-order harmonics.

[0013] Optionally, the step of expressing each harmonic of the output phase voltage by a switching angle to obtain a group of transcendental equations related to the switching angle includes:

[0014] Perform Fourier transform on the output phase voltage:

[0015]

[0016] Among them, V n (t) represents the output phase voltage;

[0017] t represents the time variable;

[0018] ω represents the fundamental angular frequency;

[0019] n represents the harmonic order;

[0020] a0 represents a constant term;

[0021] a n represents the sine term coefficient, a n =0;

[0022] b n represents the cosine term coefficient,

[0023]

[0024] V dc Indicates the output phase DC voltage;

[0025] α k represents the kth switching angle, k = 1, 2, 3, ... N;

[0026] Let the fundamental amplitude b1 satisfy the preset modulation ratio M, that is:

[0027] Let the amplitude of the remaining harmonics to be eliminated be b N =0.

[0028] Optionally, the step of expressing each harmonic of the output phase voltage by a switching angle to obtain a group of transcendental equations related to the switching angle includes:

[0029]

[0030] The transcendental equations about the switching angle are sorted out to obtain the sorted transcendental equations about the switching angle:

[0031]

[0032] Optionally, the use of the Levenberg-Marquardt method to solve the transcendental equations related to the switching angle to obtain the switching angle set includes:

[0033] The above-arranged transcendental equations about the switching angle are converted into a nonlinear equation system to obtain:

[0034] F(x)=0

[0035] Wherein, x represents the switching angle set;

[0036] Perform a first-order Taylor expansion on the nonlinear equations at the current iteration point to obtain:

[0037] F(x k+1 )=F(x k )+J(x k )d k

[0038] Among them, d k represents the correction amount of the switch angle set; and d k =x k+1 -x k ;

[0039] J(x k ) is F(x k ) at the current calculation point x k The first-order partial derivative at ;

[0040] Set the number of iterations k = 1, the convergence accuracy ε>0, the adaptive factor β, the constant m, and β>m>0,0 <P0<P1<P2<1;

[0041] When || J(x k ) T F(x k )||<ε holds true, output the switching angle set;

[0042] When || J(x k ) T F(x k )||<ε does not hold,

[0043] Calculates the correction for the current iteration of the Levenberg-Marquardt method

[0044]

[0045] Among them, μ krepresents the damping factor of the Levenberg-Marquardt method;

[0046] Indicates the correction value of the current iteration step of the Levenberg-Marquardt method;

[0047] Continue and calculate the approximate iteration step of the Levenberg-Marquardt method

[0048]

[0049] in, represents the approximate iteration step of the Levenberg-Marquardt method;

[0050] Define a new iteration step as s k :

[0051]

[0052] Define a model function and define the actual reduction value Ared of the model function k :

[0053]

[0054] Define the estimated reduction value Pred of the model function k :

[0055]

[0056] Define the actual reduction value Ared k and the estimated reduction value Pred k The ratio is the iterative step size cut-off index ρ k :

[0057]

[0058] Calculate the iterative step size cut-off index ρ k , determine whether to accept the new iteration step s k :

[0059]

[0060] Update the adaptive factor β k :

[0061]

[0062] The number of iterations increases by one until ||J(x k ) T F(xk )||<ε, output the switching angle set.

[0063] Optionally, also include:

[0064] The obtained switching angle set is stored in a database, and in an offline state, the corresponding switching angle set is selected from the database according to the output phase voltage and a preset modulation ratio.

[0065] Optionally, also include:

[0066] Constructing a sample set including a plurality of samples; wherein the samples include: sample data and label data; the sample data includes historical output phase voltages and historical preset modulation ratios of the converter; the label data includes a training set of switching angles corresponding to the historical output phase voltages and historical preset modulation ratios of the converter;

[0067] According to the sample set, a pulse width modulation model is constructed and trained by a predetermined machine learning algorithm;

[0068] According to the output phase voltage of the converter and the preset modulation ratio, the corresponding switching angle set is obtained through the pulse width modulation model.

[0069] Based on the same inventive concept, one or more embodiments of this specification provide a pulse width modulation device for suppressing specific subharmonics based on the Levenberg-Marquardt method, including:

[0070] A voltage acquisition module, used to obtain the output phase voltage of the three-level converter;

[0071] An information representation module, configured to represent each harmonic of the output phase voltage by a switching angle to obtain a group of transcendental equations related to the switching angle;

[0072] An information operation module, configured to solve the transcendental equations related to the switching angles using a Levenberg-Marquardt method to obtain a set of switching angles;

[0073] A pulse width modulation module is used to modulate the pulse width according to the switching angle set.

[0074] Based on the same inventive concept, one or more embodiments of this specification provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the program.

[0075] Based on the same inventive concept, one or more embodiments of this specification provide a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the above method.

[0076] As can be seen from the foregoing, the Levenberg-Marquardt method-based pulse width modulation method and apparatus for suppressing specific subharmonics, provided in one or more embodiments of this specification, transform the problem of suppressing specific subharmonics of a converter into a problem of solving a system of transcendental equations in which each harmonic is represented by a switching angle. This eliminates the need to generate a carrier wave and the need to compare with the modulated wave, thereby simplifying the control process. Furthermore, the Levenberg-Marquardt method can overcome the unsolvable situations that may exist in existing algorithms, ensuring that the switching angle adjustment direction can be stably obtained for the pulse width modulation problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only one or more embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0078] Figure 1 A schematic flow chart of a pulse width modulation method for suppressing specific subharmonics based on the Levenberg-Marquardt method provided in one or more embodiments of this specification;

[0079] Figure 2 A schematic diagram of a flow chart for solving a system of transcendental equations related to switching angles using the Levenberg-Marquardt method provided in one or more embodiments of this specification;

[0080] Figure 3 A schematic structural diagram of a pulse width modulation device for suppressing specific subharmonics based on the Levenberg-Marquardt method provided in one or more embodiments of this specification;

[0081] Figure 4 A more specific schematic diagram of the hardware structure of an electronic device is provided for one or more embodiments of this specification. DETAILED DESCRIPTION

[0082] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0083] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of this specification should have the usual meanings understood by people with ordinary skills in the field to which this disclosure belongs. The "first", "second" and similar words used in one or more embodiments of this specification do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0084] The specific subharmonic suppression pulse width modulation control algorithm can be regarded as the problem of solving a set of nonlinear equations with the switching angle as the variable. For solving the nonlinear equations, common algorithms include: Newton's method, homology method, least squares method, Walsh method, etc.

[0085] However, these algorithms are sensitive to the choice of initial values ​​and have range constraints on the modulation ratio. This can lead to non-convergence or even a lack of a solution. Furthermore, if the algorithm still fails to converge under specific initial values, a feasible solution cannot be obtained, and thus the direction for adjusting the switching angle under the current state cannot be determined, making it impossible to solve practical engineering problems.

[0086] In order to solve the above problems, one or more embodiments of this specification provide a specific subharmonic suppression pulse width modulation method and device based on the Levenberg-Marquardt method. The method and device can be applied to various electronic devices, including memory, processor, and computer program stored in the memory and executable on the processor, as well as non-transitory computer-readable storage media, which are not specifically limited in this disclosure.

[0087] Figure 1 A flowchart of a pulse width modulation method for suppressing specific subharmonics based on the Levenberg-Marquardt method is provided for one or more embodiments of this specification. The pulse width modulation method for suppressing specific subharmonics based on the Levenberg-Marquardt method includes:

[0088] S101 : Obtain output phase voltage of a three-level converter.

[0089] In some implementations, before obtaining the output phase voltage of the three-level converter, the method further includes:

[0090] The output phase voltage of the three-level converter is filtered to remove high-order harmonics.

[0091] Filtering the output phase voltage of a three-level converter can be achieved using a filter. Using pulse width modulation to suppress low- and high-order harmonics requires significant computational memory resources. Therefore, in situations where stringent grid-side harmonic requirements are required, combining a specific harmonic suppression algorithm with a filter can be considered to effectively suppress low- and high-order harmonics, as well as specific harmonics, to meet practical engineering needs.

[0092] S102 : Expressing each harmonic of the output phase voltage by a switching angle to obtain a group of transcendental equations related to the switching angle.

[0093] In some implementations, the set of transcendental equations for the switching angle includes:

[0094]

[0095] The transcendental equations about the switching angle are sorted out to obtain the sorted transcendental equations about the switching angle:

[0096]

[0097] In some implementations, expressing each harmonic of the output phase voltage by a switching angle to obtain a set of transcendental equations related to the switching angle includes:

[0098] Performing Fourier transform on the output phase voltage.

[0099] To ensure the symmetry of the output waveform, the specific subharmonic suppression pulse width modulation method only needs to solve the switching angle of 1 / 4 cycle, and the remaining part can be obtained accordingly based on the symmetry. Assume that the switching angles of the first 1 / 4 cycle are α1, α2, ... α N , the output phase voltage is Fourier transformed as follows:

[0100]

[0101] Among them, V n (t) represents the output phase voltage;

[0102] t represents the time variable;

[0103] ω represents the fundamental angular frequency;

[0104] n represents the harmonic order;

[0105] a0 represents a constant term;

[0106] a nrepresents the sine term coefficient,

[0107]

[0108] b n represents the cosine term coefficient,

[0109]

[0110] V dc Indicates the output phase DC voltage;

[0111] α k represents the kth switching angle, k = 1, 2, 3, ... N;

[0112] Since the voltage waveform satisfies the 1 / 4 cycle symmetry, the sine term coefficient a n =0;

[0113] Since the voltage waveform satisfies the 1 / 2 cycle before and after 180 degrees symmetry, the cosine term coefficient b n Satisfy the following formula:

[0114]

[0115] Let the fundamental amplitude b1 satisfy the preset modulation ratio M, that is:

[0116]

[0117] Let the amplitude of the remaining harmonics to be eliminated be b N =0.

[0118] The amplitude of each harmonic of the output voltage can be expressed by the N switching angles in the first 1 / 4 cycle, and the N switching angles can be obtained by inversely solving the transcendental equation.

[0119] In addition to satisfying the fundamental amplitude condition, let the amplitude of the remaining harmonics to be eliminated be b N = 0, solving N transcendental equations can yield N angles. To solve the complete switching table, the transcendental equations must be solved under different modulation indices M, and a set of solutions must be calculated for each modulation indices.

[0120] S103 , using the Levenberg-Marquardt method to solve the group of transcendental equations related to the switching angles to obtain a set of switching angles.

[0121] The present disclosure adopts the Levenberg-Marquardt method to solve the transcendental equations about the switching angle. The Levenberg-Marquardt method has high-order convergence characteristics, can avoid the problem that traditional optimization algorithms are sensitive to initial values, and is not inferior to the conventional Newton method in terms of calculation speed.

[0122] More importantly, under specific computational conditions, when conventional algorithms like the Newton method and the homology method experience divergence or unsolvable problems due to improper initial value selection, the Levenberg-Marquardt method can still provide a least-squares solution, providing reference information for subsequent adjustment of the switching angle. It can also provide reference information for adjusting parameters such as initial values ​​and modulation ratios.

[0123] S104: Modulate the pulse width according to the switching angle set.

[0124] Optionally, the specific subharmonic suppression pulse width modulation method based on the Levenberg-Marquardt method further includes:

[0125] The obtained switching angle set is stored in a database, and in an offline state, the corresponding switching angle set is selected from the database according to the output phase voltage and a preset modulation ratio.

[0126] Among them, when the converter computing memory resources can meet the demand, the present disclosure can realize online calculation of the switching angle value at each moment. However, in the case where the converter computing memory resources are insufficient relative to the demand or relatively economical converter computing memory resources are considered, the switching angle set calculated by the specific subharmonic suppression pulse width modulation method based on the Levenberg-Marquardt method of the present disclosure can be stored in a database. In an offline state, according to the actual operating conditions of the converter, such as the output phase voltage and the preset modulation ratio, the corresponding switching angle set is selected from the database. In this case, the converter computing memory resources are saved, and the scope of application of the present disclosure is expanded, so that the present disclosure can meet the needs of actual engineering projects.

[0127] Optionally, the specific subharmonic suppression pulse width modulation method based on the Levenberg-Marquardt method further includes:

[0128] Constructing a sample set including a plurality of samples; wherein the samples include: sample data and label data; the sample data includes historical output phase voltages and historical preset modulation ratios of the converter; the label data includes a training set of switching angles corresponding to the historical output phase voltages and historical preset modulation ratios of the converter;

[0129] According to the sample set, a pulse width modulation model is constructed and trained by a predetermined machine learning algorithm;

[0130] According to the output phase voltage of the converter and the preset modulation ratio, the corresponding switching angle set is obtained through the pulse width modulation model.

[0131] Among them, the predetermined machine learning algorithm can be selected from one or more of the naive Bayes algorithm, decision tree algorithm, support vector machine algorithm, kNN algorithm, neural network algorithm, deep learning algorithm and logistic regression algorithm.

[0132] Combining the Levenberg-Marquardt method disclosed in this invention with a machine learning algorithm can effectively discover the inherent laws of pulse width modulation for suppressing specific subharmonics, combine the predicted parameters with the actual dynamic parameters, and thus quickly adjust and correct the switching angle.

[0133] One or more embodiments of the present specification provide a method and apparatus for pulse width modulation with specific subharmonic suppression based on the Levenberg-Marquardt method, which transforms the problem of pulse width modulation with specific subharmonic suppression of a converter into the problem of solving a system of transcendental equations in which each subharmonic is represented by a switching angle. This eliminates the need to generate a carrier wave and to compare it with a modulated wave, thereby simplifying the control process. Furthermore, the use of the Levenberg-Marquardt method can overcome the unsolvable situations that may exist in existing algorithms, ensuring that the switching angle adjustment direction can be stably obtained for the pulse width modulation problem.

[0134] It can be understood that the method can be executed by any device, equipment, platform, or device cluster with computing and processing capabilities.

[0135] Figure 2 A flowchart of using the Levenberg-Marquardt method to solve a system of transcendental equations related to a switching angle is provided for one or more embodiments of this specification. Using the Levenberg-Marquardt method to solve a system of transcendental equations related to a switching angle includes:

[0136] S201 , converting the transcendental equations related to the switching angle into a nonlinear equations system, and performing a first-order Taylor expansion on the nonlinear equations system at the current iteration point.

[0137] In some implementations, the sorted transcendental equations for the switching angle are converted into a nonlinear equation system to obtain:

[0138] F(x)=0

[0139] Where x represents the set of switching angles, i.e., the switching angles α1, α2, α3, ... α k A collection of components.

[0140] The nonlinear equation group F(x)=0 is subjected to a first-order Taylor expansion at the current iteration point to obtain:

[0141] F(x k+1 )=F(x k )+J(xk )d k

[0142] Among them, d k represents the correction amount of the switch angle set;

[0143] and d k =x k+1 -x k ;

[0144] J(x k ) is F(x k ) at the current iteration point x k The first-order partial derivative at .

[0145] S202. Set the number of iterations, convergence accuracy, adaptive factor and constant.

[0146] Set the number of iterations k = 1, the convergence accuracy ε>0, the adaptive factor β, and the constant m.

[0147] Among them, β>m>0,0 <P0<P1<P2<1。

[0148] S203, judge || J(x k ) T F(x k )||<ε holds true.

[0149] S204, when || J(x k ) T F(x k )||<ε holds true, output the switching angle set.

[0150] S205, when || J(x k ) T F(x k )||<ε does not hold, calculate the correction amount and approximate iteration step of the Levenberg-Marquardt method, and define a new iteration step.

[0151] Among them, the correction amount of the current iteration step of the Levenberg-Marquardt method is calculated:

[0152]

[0153] Among them, μ k represents the damping factor of the Levenberg-Marquardt method;

[0154] Indicates the correction amount of the current iteration step of the Levenberg-Marquardt method.

[0155] Continuing, we calculate the approximate iterations of the Levenberg-Marquardt method:

[0156]

[0157] in, represents the approximate iteration step of the Levenberg-Marquardt method.

[0158] Define a new iteration step:

[0159]

[0160] Among them, s k represents the iteration step of the Levenberg-Marquardt method.

[0161] S206: Calculate an iteration step size rejection index to determine whether to accept the new iteration step.

[0162] Calculate the iterative step size cut-off index ρ k , determine whether to accept the iterative step s k .

[0163] in,

[0164] In some implementations, a model function is defined, and the actual reduction value Ared of the model function is defined. k and the estimated reduction value Pred k .

[0165] Among them, the actual reduction value Ared defined by the model function k :

[0166]

[0167] Define the estimated reduction value Pred of the model function k :

[0168]

[0169] Define the actual reduction value Ared k and the estimated reduction value Pred k The ratio is the iterative step size cut-off index ρ k .

[0170]

[0171] S207: Update the adaptive factor.

[0172]

[0173] S208: The number of iterations is increased by one.

[0174] Let k = k + 1, and repeat the above calculation steps until || J(x k ) T F(x k )||<ε, output the switching angle set.

[0175] When conventional algorithms such as the Newton method and the homology method cause divergence or no solution due to improper initial value selection, the Levenberg-Marquardt method can still provide a least squares solution, providing reference information for subsequent switch angle adjustments and solving practical engineering problems.

[0176] It should be noted that the methods of one or more embodiments of this specification can be performed by a single device, such as a computer or server. The methods of this embodiment can also be applied in a distributed scenario, where multiple devices cooperate to perform the method. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the methods of one or more embodiments of this specification, and the multiple devices will interact with each other to complete the method.

[0177] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0178] Figure 3 A schematic structural diagram of a pulse width modulation device for suppressing specific subharmonics based on the Levenberg-Marquardt method provided in one or more embodiments of this specification, wherein the pulse width modulation device for suppressing specific subharmonics based on the Levenberg-Marquardt method includes:

[0179] The voltage acquisition module 301 is configured to acquire the output phase voltage of the three-level converter.

[0180] The information representation module 302 is configured to represent each harmonic of the output phase voltage by a switching angle to obtain a set of transcendental equations related to the switching angle.

[0181] The information operation module 303 is used to solve the transcendental equations related to the switching angles using the Levenberg-Marquardt method to obtain a set of switching angles.

[0182] The pulse width modulation module 304 is configured to modulate the pulse width according to the switching angle set.

[0183] For the convenience of description, the above devices are described as being functionally divided into various modules. Of course, when implementing one or more embodiments of this specification, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0184] The apparatus of the above embodiment is used to implement the corresponding method in the above embodiment and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0185] Figure 4 A more specific hardware structure diagram of an electronic device is provided for one or more embodiments of this specification. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other within the device via the bus 1050.

[0186] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0187] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 1020 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0188] The input / output interface 1030 is used to connect input / output modules to implement information input and output. The input / output modules can be configured as components within the device (not shown in the figure) or can be externally connected to the device to provide corresponding functions. Input devices may include a keyboard, mouse, touch screen, microphone, various sensors, etc., and output devices may include a display, speaker, vibrator, indicator light, etc.

[0189] The communication interface 1040 is used to connect to a communication module (not shown) to enable communication between the device and other devices. The communication module can communicate via a wired method (such as USB, network cable, etc.) or a wireless method (such as mobile network, WiFi, Bluetooth, etc.).

[0190] The bus 1050 comprises a path for transmitting information between the various components of the device (eg, the processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 ).

[0191] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, it will be understood by those skilled in the art that the above device may only include the components necessary to implement the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0192] The computer-readable media of this embodiment include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0193] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Based on the concept of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present specification as described above, which are not provided in detail for the sake of simplicity.

[0194] In addition, to simplify the description and discussion, and so as not to obscure one or more embodiments of the present specification, well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, devices may be shown in block diagram form to avoid obscuring one or more embodiments of the present specification, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which one or more embodiments of the present specification will be implemented (i.e., these details should be fully within the purview of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present disclosure, it will be apparent to those skilled in the art that one or more embodiments of the present specification may be implemented without these specific details or with variations in these specific details. Accordingly, these descriptions should be considered illustrative rather than restrictive.

[0195] Although the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0196] The one or more embodiments of this specification are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of this specification shall be included within the scope of protection of this disclosure.

Claims

1. A pulse width modulation method for suppressing specific subharmonics based on the Levenberg-Marquardt method, characterized in that: include: Obtaining the output phase voltage of the three-level converter; Representing each harmonic of the output phase voltage by a switching angle to obtain a group of transcendental equations related to the switching angle; The transcendental equations about the switching angle are transformed into a nonlinear equation system to obtain: in, represents the set of switch angles; Perform a first-order Taylor expansion on the nonlinear equations at the current iteration point to obtain: in, d k represents the correction amount of the switching angle set; and ; J ( x k )for F ( x k ) at the current iteration point x k The first-order partial derivative at ; Set the number of iterations k =1, convergence accuracy ε >0, adaptive factor β ,constant m ,and β>m>0,0 <P 0 <P 1 <P 2 <1 ; when ||J ( x k ) T F ( x k ) ||<ε When it is established, the switching angle set is output; when ||J ( x k ) T F ( x k ) ||<ε When it is not established, Calculates the correction for the current iteration of the Levenberg-Marquardt method : in, represents the damping factor of the Levenberg-Marquardt method; represents the correction amount of the Levenberg-Marquardt method; Continue and calculate the approximate iteration step of the Levenberg-Marquardt method : in, represents the approximate iteration step of the Levenberg-Marquardt method; Define a new iteration step as s k : Define a model function and define the actual reduction value of the model function : Define the estimated reduction value of the model function : Define the actual reduction value The estimated reduction The ratio is the iterative step size selection index : Calculate the iterative step size cut-off index , determine whether to accept the new iteration step s k : Update the adaptive factor β k : The number of iterations increases by one until the ||J ( x k ) T F ( x k ) ||<ε , output the switching angle set; The pulse width is modulated according to the set of switching angles.

2. The pulse width modulation method according to claim 1, wherein: Before obtaining the output phase voltage of the three-level converter, the method further includes: The output phase voltage of the three-level converter is filtered to remove high-order harmonics.

3. The pulse width modulation method according to claim 1, wherein: The harmonics of the output phase voltage are expressed by the switching angle to obtain a set of transcendental equations about the switching angle, including: Perform Fourier transform on the output phase voltage: in, Indicates the output phase voltage; t represents the time variable; ω Indicates the fundamental angular frequency; n Indicates the harmonic order; a 0 represents a constant term; a n represents the sine term coefficient, ; b n represents the cosine term coefficient, ； V dc Indicates the output phase DC voltage; α k Indicates the k Switch angle, k =1,2,3,… N ; Let the fundamental amplitude Meet the preset modulation ratio ,Right now: Let the amplitudes of the remaining harmonics to be eliminated be .

4. The pulse width modulation method according to claim 3, characterized in that: The harmonics of the output phase voltage are expressed by the switching angle to obtain a set of transcendental equations about the switching angle, including: The transcendental equations about the switching angle are sorted out to obtain the sorted transcendental equations about the switching angle: 。 5. The pulse width modulation method according to claim 1, wherein: Also includes: The obtained switching angle sets under different modulation ratios are stored in a database, and in an offline state, the corresponding switching angle set is selected from the database according to the output phase voltage and the preset modulation ratio.

6. The pulse width modulation method according to claim 1, characterized in that: Also includes: Constructing a sample set including a plurality of samples; wherein the samples include: sample data and label data; the sample data includes historical output phase voltages and historical preset modulation ratios of the converter; the label data includes a training set of switching angles corresponding to the historical output phase voltages and historical preset modulation ratios of the converter; According to the sample set, a pulse width modulation model is constructed and trained by a predetermined machine learning algorithm; According to the output phase voltage of the converter and the preset modulation ratio, the corresponding switching angle set is obtained through the pulse width modulation model.

7. A pulse width modulation device for suppressing specific subharmonics based on the Levenberg-Marquardt method, characterized in that: include: A voltage acquisition module, used to obtain the output phase voltage of the three-level converter; An information representation module, configured to represent each harmonic of the output phase voltage by a switching angle to obtain a group of transcendental equations related to the switching angle; The information operation module is used to convert the sorted transcendental equations about the switching angle into a nonlinear equation system to obtain: in, represents the set of switch angles; Perform a first-order Taylor expansion on the nonlinear equations at the current iteration point to obtain: in, d k represents the correction amount of the switching angle set; and ; J ( x k )for F ( x k ) at the current iteration point x k The first-order partial derivative at ; Set the number of iterations k =1, convergence accuracy ε >0, adaptive factor β ,constant m ,and β>m>0,0 <P 0 <P 1 <P 2 <1 ; when ||J ( x k ) T F ( x k ) ||<ε When it is established, the switching angle set is output; when ||J ( x k ) T F ( x k ) ||<ε When it is not established, Calculates the correction for the current iteration of the Levenberg-Marquardt method : in, represents the damping factor of the Levenberg-Marquardt method; represents the correction amount of the Levenberg-Marquardt method; Continue and calculate the approximate iteration step of the Levenberg-Marquardt method : in, represents the approximate iteration step of the Levenberg-Marquardt method; Define a new iteration step as s k : Define a model function and define the actual reduction value of the model function : Define the estimated reduction value of the model function : Define the actual reduction value The estimated reduction The ratio is the iterative step size selection index : Calculate the iterative step size cut-off index , determine whether to accept the new iteration step s k : Update the adaptive factor β k : The number of iterations increases by one until the ||J ( x k ) T F ( x k ) ||<ε , output the switching angle set; A pulse width modulation module is used to modulate the pulse width according to the switching angle set.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

Images