Matrix motor inverse system decoupling control method, system, equipment and medium

Abstract

The invention discloses a decoupling control method, system and device for a matrix motor inverse system and a medium. The decoupling control method comprises the following steps: acquiring a state parameter v of the current matrix motor inverse system; inputting the state parameter v of the current matrix motor inverse system into a trained inverse system model based on a deep neural network to obtain an action parameter u output by the inverse system model; according to the action parameter u output by the inverse system model, closed-loop control is conducted on the control matrix motor, decoupling control of the inverse system of the matrix motor can be achieved through the method, the system, the equipment and the medium, and reliability and accuracy are high.

Description

Technical Field

[0001] The present invention belongs to the technical field of motor control, and relates to a matrix motor inverse system decoupling control method, system, equipment and medium. Background Art

[0002] Electric drive systems are increasingly being used in aerospace, ship propulsion, military equipment and many other important fields, and the requirements for drive motors are becoming increasingly higher. Many applications have complex and harsh working environments, are inconvenient to maintain, and have strict restrictions on installation space and weight. Therefore, in addition to extremely high requirements for power density, extremely stringent requirements are placed on the continuous output capacity of torque. For example, helicopter power systems and aircraft fly-by-wire control systems require a failure rate of less than 10%. -9 / flight hours. The matrix motor is a magnetic field modulation motor with multiple redundant magnetic sources configured inside. Multi-phase armature windings and permanent magnets are arranged on both the stator and the rotor. The matrix motor generates multiple torque components distributed in a matrix through multiple magnetic field modulation effects, which significantly improves the torque density of the motor. At the same time, the multi-magnetic source configuration also enhances its ability to continuously output torque after a fault.

[0003] There is cross-coupling between the AC and DC axis circuit equations in the mathematical model of the motor, and the magnitude of the coupling is related to the speed, AC and DC axis current and inductance. When the motor is running under heavy load and high speed conditions, the cross-coupling problem has strong disturbance characteristics; when the motor parameters change due to nonlinear factors, the value of the coupling term will fluctuate, with strong uncertainty. Since the matrix motor has two sets of multi-phase armature windings, the stator and rotor, the structure of the coupling term in its circuit equation is more complex. At present, the decoupling control methods of the motor can be mainly summarized into three categories: diagonal decoupling, anti-interference decoupling and inverse system decoupling. Among them, the construction of diagonal decoupling and anti-interference decoupling controllers directly requires the precise mathematical model parameters of the motor, and the controller setting is relatively complex, which is difficult to adapt to the time-varying parameters caused by nonlinear factors. Inverse system decoupling requires the inverse model of the system, and its calculation process also requires the precise mathematical model of the motor, and the calculation of the inverse model is relatively complex. With the increase of the order of the motor mathematical model, the complexity of solving the inverse model increases significantly. Although some researchers have combined machine learning algorithms such as neural networks and support vector machines with inverse system decoupling, so that the inverse system of the motor can be solved and applied to the decoupling control of traditional three-phase permanent magnet synchronous motors, the reliability and accuracy of motor systems such as matrix motors with higher-order circuit equations and more complex coupling terms still need to be improved. Summary of the invention

[0004] The purpose of the present invention is to overcome the disadvantages of the above-mentioned prior art, and provides a decoupling control method, system, device and medium for a matrix motor inverse system, which can achieve the decoupling control of the matrix motor and has high reliability and accuracy.

[0005] To achieve the above object, the present invention discloses a decoupling control method for a matrix motor inverse system, including:

[0006] Obtain the state parameter v of the current matrix motor inverse system, and the state parameter where, i ds1 is the fundamental wave space current amplitude of the stator d-axis, i ds3 is the third harmonic space current amplitude of the stator d-axis, i dr is the fundamental wave current amplitude of the rotor d-axis, ω es is the stator electrical angular velocity, ω er is the rotor electrical angular velocity, T L is the load torque, i dr ', i ds1 ', i ds3 ', ω es ', ω er ' and T L ' are the first derivatives of i dr , i ds1 , i ds3 , ω es , ω er and T L respectively; ω es ” and ω er ” are the second derivatives of ω es and ω er respectively;

[0007] Input the state parameter v of the current matrix motor inverse system into the trained inverse system model based on a deep neural network to obtain the action parameter u output by the inverse system model, where u = (u ds1 , u qs1 , u ds3 , u qs3 , u dr , u qr ), T where, u ds1 , u qs1 are the fundamental wave space voltage amplitudes of the stator d and q axes respectively; u ds3 , u qs3 are the third harmonic space voltage amplitudes of the stator d and q axes respectively, u dr , u qr are the fundamental wave voltage amplitudes of the rotor d and q axes respectively;

[0008] Perform closed-loop control on the matrix motor according to the action parameter u output by the inverse system model.

[0009] A further improvement of the matrix motor inverse system decoupling control method of the present invention lies in:

[0010] Further, before inputting the state parameter v of the current matrix motor inverse system into the trained inverse system model based on a deep neural network, it further includes:

[0011] Obtain historical experimental data;

[0012] Construct a data set according to the historical experimental data;

[0013] Use the data set to train the deep neural network to obtain a trained inverse system model based on the deep neural network.

[0014]

[0015] Further, use the data set and adopt the gradient descent method to train the deep neural network.

[0016] The present invention discloses a matrix motor inverse system decoupling control system, including:

[0017] An acquisition module, configured to acquire the state parameter v of the current matrix motor inverse system, and the state parameter where, i ds1 is the fundamental wave space current amplitude of the stator d-axis, i ds3 is the third harmonic space current amplitude of the stator d-axis, i dr is the fundamental wave current amplitude of the rotor d-axis, ω es is the stator electrical angular velocity, ω er is the rotor electrical angular velocity, T L is the load torque, i dr ', i ds1 ', i ds3 ', ω es ', ω er ' and T L ' are the first derivatives of i dr , i ds1 , i ds3 , ω es , ω er and T L respectively; ω es ” and ω er ” are the second derivatives of ω es and ω er respectively;

[0018] A prediction module, configured to input the state parameter v of the current matrix motor inverse system into a trained inverse system model based on a deep neural network, and obtain an action parameter u output by the inverse system model, where u = (u ds1 , u qs1 , u ds3 , u qs3 , u dr , u qr ) T , where u ds1 , u qs1 are respectively the fundamental wave space voltage amplitudes of the stator d and q axes; u ds3 , u qs3 are respectively the third harmonic space voltage amplitudes of the stator d and q axes, and u dr , u qr are respectively the fundamental wave voltage amplitudes of the rotor d and q axes;

[0019] A control module, configured to perform closed-loop control on the matrix motor according to the action parameter u output by the inverse system model.

[0020] A further improvement of the matrix motor inverse system decoupling control system according to the present invention lies in:

[0021] Further, it further includes:

[0022] An acquisition module, configured to acquire historical experimental data;

[0023] A construction module, configured to construct a data set according to the historical experimental data;

[0024] A training module, configured to train a deep neural network by using the data set to obtain a trained inverse system model based on the deep neural network.

[0025]

[0026] Further, the deep neural network is trained by using the data set and adopting the gradient descent method.

[0027] The present invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the matrix motor inverse system decoupling control method are implemented.

[0028] The present invention discloses a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the matrix motor inverse system decoupling control method are implemented.

[0029] The present invention has the following beneficial effects:

[0030] When the matrix motor inverse system decoupling control method, system, device and medium of the present invention are specifically operated, the deep neural network has powerful feature learning ability, big data processing ability, scalability and flexibility. Therefore, the inverse system model based on the deep neural network can process large-scale data sets. Compared with the analytical inverse system solution, it takes into account the variability of stator inductance, resistance and other parameters under different conditions. Compared with the traditional neural network with only one hidden layer, more hidden layers are added, thereby improving the reliability and accuracy of decoupling control. In addition, the observables and action quantities in the present invention are more than those of the conventional permanent magnet synchronous motor, and the data to be processed is more complex. If the number of phases of the matrix motor increases significantly, the accuracy of the conventional decoupling method is insufficient. Compared with the conventional decoupling method, the accuracy of the present invention is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings forming a part of this specification are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0032] Figure 1 Schematic diagram of a deep neural network with m inputs and n outputs for a matrix motor;

[0033] Figure 2 Topological diagram of matrix motor decoupling control.

[0034] Figure 3 Flowchart of steps. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0036] In the description of the present invention, it should be understood that the terms "including" and "comprising" indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0037] It should also be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0038] It should be further understood that the term "and / or" used in the specification and appended claims of the present invention refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. For example, A and / or B can represent three cases: A exists alone, A and B exist simultaneously, and B exists alone. Additionally, in the present invention, the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0039] It should be understood that although terms such as first, second, third, etc. may be used to describe preset ranges in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0040] Depending on the context, the word "if" as used herein can be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detected (stated condition or event)" can be interpreted as "when determined" or "in response to determining" or "when detected (stated condition or event)" or "in response to detecting (stated condition or event)".

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the present invention described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0042] Various structural schematic diagrams according to the disclosed embodiments of the present invention are shown in the drawings. These figures are not drawn to scale, where for the purpose of clear expression, some details are enlarged and some details may be omitted. The shapes of various regions and layers shown in the figures and their relative sizes and positional relationships are only exemplary. In practice, there may be deviations due to manufacturing tolerances or technical limitations, and those skilled in the art can design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0043] Embodiment 1

[0044] Reference Figure 1 and Figure 2 , the decoupling control method of the matrix motor inverse system according to the present invention includes the following steps, and the steps are as Figure 3 shown.

[0045] step1 Construct the original system model of the matrix motor

[0046] In the d-q rotating coordinate system, the accurate voltage equation of the matrix motor considering nonlinear factors, parameter mismatch, and unmodeled dynamics can be derived:

[0047]

[0048] Among them, u ds1 , u qs1 are the amplitudes of the fundamental space voltages of the stator d and q axes of the five-phase respectively; i ds1 , i qs1 are the amplitudes of the fundamental space currents of the stator d and q axes of the five-phase respectively; L ds1 , L qs1 are the inductances corresponding to the fundamental space of the stator d and q axes of the five-phase respectively; u ds3 , u qs3 are the amplitudes of the third harmonic space voltages of the stator d and q axes of the five-phase respectively; i ds3 , i qs3 are the amplitudes of the third harmonic space currents of the stator d and q axes of the five-phase respectively; L ds3 , L qs3 are the inductances corresponding to the third harmonic space of the stator d and q axes of the five-phase respectively; u dr , u qr are the amplitudes of the fundamental voltages of the rotor d and q axes respectively; i dr , i qr are the amplitudes of the fundamental currents of the rotor d and q axes respectively; S is the differential operator; L dr , L qr are the inductances corresponding to the fundamental of the rotor d and q axes respectively; R s is the stator phase resistance value, R r is the rotor phase resistance value; M sr is the fundamental amplitude of the mutual inductance between the stator winding and the rotor winding; ω es is the stator electrical angular velocity, ω er is the rotor electrical angular velocity; ψ ms1 is the magnetic flux value of the stator fundamental permanent magnet flux linkage, ψ ms3 is the magnetic flux value of the stator third harmonic permanent magnet flux linkage, ψ mr is the magnetic flux value of the rotor permanent magnet flux linkage; ΔR s represents the parameter error value of the stator resistance, ΔR r represents the parameter error value of the rotor resistance, ΔL ds1, ΔL qs1 are the inductance parameter error values corresponding to the fundamental wave spaces of the stator d and q axes, respectively, and ΔL ds3 , ΔL qs3 are the inductance parameter error values corresponding to the third harmonic spaces of the stator d and q axes, respectively, and ΔL dr , ΔL qr are the inductance parameter error values corresponding to the fundamental waves of the rotor d and q axes, respectively; ΔM sr is the fundamental wave amplitude error value of the mutual inductance between the stator winding and the rotor winding; D ds1 , D qs1 represent the unmodeled dynamics of the fundamental waves of the stator d and q axes, respectively, and D ds3 , D qs3 represent the unmodeled dynamics of the third harmonics of the stator d and q axes, respectively, and D dr , D qr represent the unmodeled dynamics of the fundamental waves of the rotor d and q axes, respectively, and i ds1 ', i ds3 ', i dr ', ω es ', ω er ' and T L ' are the first-order differentials of i ds1 , i ds3 , i dr , ω es , ω er and T L ; ω es ” and ω er ” are the second-order differentials of ω es and ω er respectively, and p ns is the equivalent conversion pole pair number of the stator, and p nr is the equivalent conversion pole pair number of the rotor;

[0049] The electromagnetic torque equation is:

[0050]

[0051] where, T e 5-3 is the electromagnetic torque, T es is the stator electromagnetic torque, T er is the rotor electromagnetic torque, and p ns is the equivalent conversion pole pair number of the stator, and p nr is the equivalent conversion pole pair number of the rotor;

[0052] Assume that the input variable of the original system of the matrix motor is u and the output variable is y, i.e.:

[0053] u = (u1, u2, u3, u4, u5, u6) T = (u ds1 , uqs1 , u ds3 , u qs3 , u dr , u qr ) T

[0054] y = (y1, y2, y3, y4, y5, y6) T = (i ds1 , i ds3 , i dr , ω es , ω er , T L ) T

[0055] Step 2: Obtain historical experimental data. Since there is coupling between the d - q axis currents in the above voltage equation, current loop decoupling is required. According to the voltage equation and the electromagnetic torque equation, it is chosen to express the q - axis current in the coupling term with the d - axis current, and the analytical expression for eliminating the q - axis current is shown as follows. The purpose of the analytical expression is to clarify the input variables of the inverse system;

[0056]

[0057] Among them, let k1 and k3 be constants, and let

[0058]

[0059] Then the input of the inverse system model is v, and the output is u, that is, the following historical experimental data needs to be obtained:

[0060] v = (v1, v2, v3, v4, v5, v6, v7, v8, v9, v 10 , v 11 , v 12 , v 13 , v 14 ) T = (i ds1 , i ds1 ', i ds3 , i ds3 ', i dr , i dr , ω es , ω es , ω es ”, ω er , ω er , ω er , T L , T L ') T

[0061] u = (u1, u2, u3, u4, u5, u6) T = (u ds1 , u qs1 , u ds3 , u qs3 , u dr , u qr ) T

[0062] where i ds1 ', i ds3 ', i dr ', ω es ', ω er ' and T L ' are the first-order differentials of i ds1 , i ds3 , i dr , ω es , ω er and T L respectively; ω es ” and ω er ” are the second-order differentials of ω es and ω er respectively.

[0063] Step 3: Construct a data set based on historical experimental data

[0064] Step 4: Use the data set to perform regression training on the deep neural network to obtain a trained inverse system model based on the deep neural network;

[0065] Among them, the specific training process of the deep neural network model: As Figure 1 shown, set the input layer, output layer and hidden layer of the deep neural network model. Among them, the number of neurons in the input layer is set to be equal to the number of input features, the number of neurons in the output layer is set to be equal to the number of output features, and the size of the hidden layer can be adjusted according to the actual situation.

[0066] Divide the data set into a training set and a test set, and preprocess all the data. For example, check for outliers, missing values or duplicate values in the data, and perform data normalization, etc. Select the mini-batch gradient descent algorithm (Mini-batch Gradient Descent, MBGD). When updating the parameters each time, randomly select a certain number of samples (i.e., the batch size) from the training set for learning. For large-scale data sets, mini-batch gradient descent can significantly improve the training speed and reduce the computational cost.

[0067] Since there is a gap between the predicted value and the true value, a loss function is selected to calculate the gap, and the gradient is calculated by going back layer by layer; the Adam optimizer is used for adaptive learning rate optimization. And a suitable number of iterations is set to continuously train to make the expected value close to the true value.

[0068] Use the test set data to test the fitting accuracy. If the accuracy is not high, it is necessary to recheck whether there are outliers in the data, adjust the number of hidden layers and the number of nodes in each layer, adjust the batch size of the mini-batch gradient descent algorithm, and adjust the number of iterations to prevent overfitting or underfitting.

[0069] Step5 Obtain the state parameter v of the current matrix motor inverse system, input the state parameter v of the current matrix motor inverse system into the trained inverse system model based on the deep neural network, obtain the action parameter u output by the inverse system model, and perform closed-loop control on the matrix motor according to the action parameter output by the inverse system model. The decoupling control topology diagram of the matrix motor is as Figure 2 shown.

[0070] Embodiment 2

[0071] The decoupling control system of the matrix motor inverse system of the present invention includes:

[0072] An acquisition module for acquiring the state parameter v of the current matrix motor inverse system, where the state parameter where, i ds1 is the amplitude of the fundamental space current of the stator d-axis five-phase, i ds3 is the amplitude of the third harmonic space current of the stator d-axis five-phase, i dr is the amplitude of the fundamental current of the rotor d-axis, ω es is the electrical angular velocity of the stator, ω er is the electrical angular velocity of the rotor, T L is the load torque, i dr ', i ds1 ', i ds3 ', ω es ', ω er ' and T L ' are the derivatives of i dr , i ds1 , i ds3 , ω es , ω er and T L respectively; ω es ” and ω er ” are the second derivatives of ω es and ω er respectively.

[0073] A prediction module, configured to input the state parameter v of the current matrix motor inverse system into a trained inverse system model based on a deep neural network, and obtain an action parameter u output by the inverse system model, where u = (u ds1 , u qs1 , u ds3 , u qs3 , u dr , u qr ) T , where u ds1 , u qs1 are respectively the fundamental wave space voltage amplitudes of the stator d-axis and q-axis; u ds3 , u qs3 are respectively the third harmonic space voltage amplitudes of the stator d-axis and q-axis, and u dr , u qr are respectively the fundamental wave voltage amplitudes of the rotor d-axis and q-axis;

[0074] A control module, configured to perform closed-loop control on the control matrix motor according to the action parameter u output by the inverse system model.

[0075] In this embodiment, it further includes:

[0076] An acquisition module, configured to acquire historical experimental data;

[0077] A construction module, configured to construct a data set according to the historical experimental data;

[0078] A training module, configured to train the deep neural network by using the data set to obtain a trained inverse system model based on the deep neural network.

[0079] In this embodiment,

[0080]

[0081] In this embodiment, the deep neural network is trained by using the data set and adopting the gradient descent method.

[0082] Embodiment 3

[0083] A computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the matrix motor inverse system decoupling control method are implemented. For example, it includes: acquiring the state parameter v of the current matrix motor inverse system, and the state parameter where i ds1 is the fundamental wave space current amplitude of the stator d-axis, i ds3 is the third harmonic space current amplitude of the stator d-axis, i dr is the fundamental wave current amplitude of the rotor d-axis, ω es is the stator electrical angular velocity, ωer is the rotor electrical angular velocity, T L is the load torque, i dr ', i ds1 ', i ds3 ', ω es ', ω er ' and T L ' are respectively the first derivatives of i dr , i ds1 , i ds3 , ω es , ω er and T L ; ω es ” and ω er ” are the second derivatives of ω es and ω er ; Input the state parameter v of the current matrix motor inverse system into the trained inverse system model based on a deep neural network to obtain the action parameter u output by the inverse system model, where u = (u ds1 , u qs1 , u ds3 , u qs3 , u dr , u qr ), T where u ds1 , u qs1 are respectively the fundamental wave space voltage amplitudes of the stator d and q axes; u ds3 , u qs3 are respectively the third harmonic space voltage amplitudes of the stator d and q axes, u dr , u qr are respectively the fundamental wave voltage amplitudes of the rotor d and q axes; Perform closed-loop control on the controlled matrix motor according to the action parameter u output by the inverse system model. Among them, the memory may include internal memory, such as high-speed random access memory, and may also include non-volatile memory, such as at least one disk memory, etc.; The processor, network interface, and memory are interconnected through an internal bus, and this internal bus can be an Industry Standard Architecture bus, a Peripheral Component Interconnect standard bus, an Extended Industry Standard Architecture bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs. Specifically, the program can include program code, and the program code includes computer operation instructions. The memory can include internal memory and non-volatile memory and provide instructions and data to the processor.

[0084] Example 4

[0085] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the decoupling control method of the matrix motor inverse system. For example, it includes: obtaining the state parameter v of the current matrix motor inverse system, and the state parameter where i ds1 is the fundamental wave space current amplitude of the stator d-axis, i ds3 is the third harmonic space current amplitude of the stator d-axis, i dr is the fundamental wave current amplitude of the rotor d-axis, ω es is the stator electrical angular velocity, ω er is the rotor electrical angular velocity, T L is the load torque, i dr ', i ds1 ', i ds3 ', ω es ', ω er ' and T L ' are the first derivatives of i dr , i ds1 , i ds3 , ω es , ω er and T L respectively; ω es ” and ω er ” are the second derivatives of ω es and ω er respectively; input the state parameter v of the current matrix motor inverse system into the trained inverse system model based on a deep neural network to obtain the action parameter u output by the inverse system model, where u = (u ds1 , u qs1 , u ds3 , u qs3 , u dr , u qr ), T where u ds1 , u qs1 are the fundamental wave space voltage amplitudes of the stator d and q axes respectively; u ds3 , u qs3 are the third harmonic space voltage amplitudes of the stator d and q axes respectively, u dr , u qr are the fundamental wave voltage amplitudes of the rotor d and q axes respectively; perform closed-loop control on the control matrix motor according to the action parameter u output by the inverse system model. Specifically, the computer-readable storage medium includes but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disc, magnetic disk, etc.

[0086] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0087] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one or more of the processes or multiple processes and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0088] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one or more of the processes or multiple processes and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0089] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the processes or multiple processes and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0090] After considering the specification and the disclosure of the invention, those skilled in the art will readily think of other embodiments of the present invention. The present application is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include common general knowledge or conventional technical means in the technical field not disclosed by the present invention. The specification and the embodiments are only regarded as exemplary, and the true scope and spirit of the present invention are pointed out by the following claims.

[0091] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

[0092] As described above, these are only the preferred embodiments of the present invention and do not impose any limitations on the present invention. Any simple modifications, changes, and equivalent structural changes made to the above embodiments based on the technical essence of the present invention still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A matrix motor inverse system decoupling control method, characterized in that: include: Get the state parameter v of the current matrix motor inverse system, the state parameter Among them, i ds1 is the stator d-axis fundamental space current amplitude, i ds3 is the third harmonic space current amplitude of the stator d axis, i dr is the fundamental current amplitude of the rotor d-axis, ω es is the stator electrical angular velocity, ω er is the rotor electrical angular velocity, T L is the load torque, i dr ',i ds1 ',i ds3 ',ω es ',ω er ' and T L 'respectively i dr ,i ds1 ,i ds3 ,ω es ,ω er and T L The first derivative of es " and er " es and er The second derivative of The state parameter v of the current matrix motor inverse system is input into the trained inverse system model based on deep neural network to obtain the action parameter u output by the inverse system model, where u=(u ds1 ,u qs1 ,u ds3 ,u qs3 ,u dr ,u qr ) T , where u ds1 ,u qs1 are the fundamental spatial voltage amplitudes of the stator d and q axes respectively; u ds3 ,u qs3 are the third harmonic spatial voltage amplitudes of the stator d and q axes, u dr ,u qr are the fundamental voltage amplitudes of the rotor d and q axes respectively; The matrix motor is controlled in a closed loop according to the action parameter u output by the inverse system model.

2. The matrix motor inverse system decoupling control method according to claim 1, characterized in that: Before inputting the state parameter v of the current matrix motor inverse system into the trained inverse system model based on a deep neural network, the method further includes: Obtain historical experimental data; Construct a data set based on historical experimental data; The data set is used to perform regression training on the deep neural network to obtain a trained inverse system model based on the deep neural network.

3. The matrix motor inverse system decoupling control method according to claim 2, characterized in that:

4. The matrix motor inverse system decoupling control method according to claim 2, characterized in that: Using the data set, the deep neural network is trained using the gradient descent method.

5. A matrix motor inverse system decoupling control system, characterized in that: include: The acquisition module is used to obtain the state parameter v of the current matrix motor inverse system. Among them, i ds1 is the stator d-axis fundamental space current amplitude, i ds3 is the third harmonic space current amplitude of the stator d axis, i dr is the fundamental current amplitude of the rotor d-axis, ω es is the stator electrical angular velocity, ω er is the rotor electrical angular velocity, T L is the load torque, i dr ',i ds1 ',i ds3 ',ω es ',ω er ' and T L 'respectively i dr ,i ds1 ,i ds3 ,ω es ,ω er and T L The first derivative of es " and er " es and er The second derivative of The prediction module is used to input the state parameter v of the current matrix motor inverse system into the trained inverse system model based on deep neural network to obtain the action parameter u output by the inverse system model, where u=(u ds1 ,u qs1 ,u ds3 ,u qs3 ,u dr ,u qr ) T , where u ds1 ,u qs1 are the fundamental spatial voltage amplitudes of the stator d and q axes respectively; u ds3 ,u qs3 are the third harmonic spatial voltage amplitudes of the stator d and q axes, u dr ,u qr are the fundamental voltage amplitudes of the rotor d and q axes respectively; The control module is used to perform closed-loop control on the control matrix motor according to the action parameter u output by the inverse system model.

6. The matrix motor inverse system decoupling control system according to claim 5, characterized in that: Also includes: Acquisition module, used to obtain historical experimental data; A construction module, used to construct a data set based on historical experimental data; The training module is used to train the deep neural network using the data set to obtain a trained inverse system model based on the deep neural network.

7. The matrix motor inverse system decoupling control system according to claim 6, characterized in that:

8. The matrix motor inverse system decoupling control system according to claim 6, characterized in that: Using the data set, the deep neural network is trained using the gradient descent method.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the matrix motor inverse system decoupling control method as described in any one of claims 1-4 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the matrix motor inverse system decoupling control method as described in any one of claims 1 to 4 are implemented.

Images