Model-free adaptive control method for permanent magnet synchronous motor based on extended state observer

Abstract

The invention discloses a permanent magnet synchronous motor model-free adaptive control method based on an extended state observer, and the method comprises the steps: firstly building a permanent magnet synchronous motor nonlinear discrete time data model considering disturbance under a synchronous rotating coordinate system through a tight-format dynamic linearization method; then, a control law of the system is derived through a corresponding cost function, an improved projection algorithm is used for estimating a pseudo partial derivative parameter matrix, so that a reference voltage given value is obtained, the permanent magnet synchronous motor is controlled after SVPWM modulation, and model-free adaptive control is achieved; meanwhile, a discrete time expansion state observer is designed to observe and compensate unknown interference caused by external disturbance and system uncertainty, the complexity of a pseudo partial derivative matrix is reduced, and parameter estimation is easier to carry out; the anti-interference capability of the system is enhanced, and the control performance is effectively improved. The method is easy to implement and higher in robustness, and the stability of a motor control system under complex working conditions can be remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of permanent magnet synchronous motor adaptive control, and specifically relates to a model-free adaptive control method for permanent magnet synchronous motors based on an extended state observer. Background Technique

[0002] Due to advantages such as simple structure, high output torque, and wide speed range, permanent magnet synchronous motors are widely used in fields such as aerospace and new energy vehicles. With the rapid development of computer technology and microcontrollers, more and more high-performance model-based permanent magnet synchronous motor control methods have become research hotspots, such as model predictive control in discrete or continuous sets, deadbeat control, etc. These methods require accurate modeling of permanent magnet synchronous motors. However, although the established mathematical models of permanent magnet synchronous motors generally have characteristics such as multi-variable, strong coupling, and non-linearity, there are still a large number of unmodeled dynamics that are not taken into account, and it is difficult to guarantee the accuracy of the model. At the same time, as the operating state of the motor, the external operating environment, etc. change, the parameters of the motor will also change accordingly, making the control effects of these model-based control methods worse and even losing stability.

[0003] Common solutions are: adding a parameter identification module, using active disturbance rejection control technology, or adopting model-free control technology. The first two methods not only significantly increase the complexity and design difficulty of the system, but also it is difficult to guarantee the stability of the system. Therefore, model-free control technology has received more and more attention in recent years. Among them, the model-free adaptive control method is a new data-driven model-free control method, which has been widely used in various fields such as chemical engineering and electrical engineering. Applying it to the field of permanent magnet synchronous motor control can make the control completely independent of the accurate mathematical model of the permanent magnet synchronous motor, avoiding the influence brought by model mismatch and parameter mismatch, and enhancing the robustness of the system. However, since the model-free adaptive control method is a data-driven control method, it is difficult to handle the influence brought by external disturbances and system uncertainties, and the control effect is poor under complex working conditions. Summary of the Invention

[0004] In order to solve the technical problems existing in the background technique, the present invention aims to provide a model-free adaptive control method for permanent magnet synchronous motors based on an extended state observer, which models the disturbance and designs a discrete-time extended state observer to observe and compensate for the disturbance, thereby solving the technical problems of being difficult to handle the influence brought by external disturbances and system uncertainties and having a poor control effect under complex working conditions.

[0005] In order to solve the technical problems, the technical solution of the present invention is:

[0006] A model-free adaptive control method for permanent magnet synchronous motors based on an extended state observer, the method comprising:

[0007] S1: Establish a nonlinear discrete-time data model of a permanent magnet synchronous motor considering disturbances in the synchronous rotating coordinate system by using the compact-form dynamic linearization method. The model is based on the differential mean value theorem, describes the influence of current increment on the input voltage with a pseudo partial derivative matrix, and includes a total disturbance term.

[0008] S2: Construct a control cost function through the data model and the current current measurement value to obtain the control law of the voltage, i.e., the control input. Use an improved projection algorithm to estimate the pseudo partial derivative matrix online to obtain the parameter estimation law, i.e., the pseudo partial derivative estimation value.

[0009] S3: Based on the current current measurement value, the calculated pseudo partial derivative estimation value, and the control input, design an extended state observer to observe the total disturbance, and feedback the total disturbance estimated by the extended state observer to the control law in step S2 for disturbance compensation. Finally, calculate the control input to be applied, i.e., the voltage reference value.

[0010] Furthermore, the step S1 includes:

[0011] Use unknown nonlinear functions f1(·), f2(·) to describe the discretized voltage equations of the permanent magnet synchronous motor in the dq axes of the synchronous rotating coordinate system:

[0012] y(k + 1) = f(y(k), …, y(k - n y ), u(k), …, u(k - n u ), d(k), …, d(k - n d ))

[0013] where f(·) = [f1(·), f2(·)] T is the unknown nonlinear function vector; n y , n u , n d is the system order greater than zero;

[0014] In the permanent magnet synchronous motor system, y(k) = i dq (k) = [i d (k), i q (k)] T is the output current vector, u(k) = u dq (k) = [u d (k), u q (k)] T , d(k) = d dq (k) = [d d (k), d q (k)] TThey are the input voltage vector and the unknown external disturbance vector respectively; to reduce the computational burden, a compact-form dynamic linearization data model is adopted, that is, the pseudo-orders of the input, output, and disturbance are 0, 1, and 1 respectively; according to the differential mean value theorem, the nonlinear discrete-time data model of the permanent magnet synchronous motor considering the disturbance can be established as follows:

[0015] Δi dq (k + 1) = Φ c (k)Δu dq (k) + ψ(k)

[0016] where Φ c (k) is the pseudo-partial derivative matrix and ψ(k) is the total disturbance term.

[0017] Furthermore, the step S2 includes:

[0018] Design the control cost function as follows:

[0019]

[0020] where λ > 0 is the weight factor;

[0021] Substitute the data model into it to obtain the control law of the voltage;

[0022] Design the parameter estimation cost function as follows:

[0023]

[0024] where μ > 0 is the weight factor;

[0025] Substitute the data model into it to obtain the corresponding parameter estimation law.

[0026] Furthermore, the step S3 includes:

[0027] Construct the state space model:

[0028] Select the state variable as x(k) = [i dq (k), ψ(k)] T , the output vector is y(k), the control input u(k) = Δu dq (k), w(k) = ψ(k + 1) - ψ(k); write out the discrete-time state space model of the data model:

[0029]

[0030] where I = I 2×2 , 0 = 0 2×2 are the second-order identity matrix and zero matrix respectively;

[0031] Design the extended state observer ESO:

[0032]

[0033] Where L = [l1I, l2I] T is the gain vector;

[0034] Disturbance compensation:

[0035] Each time control starts, the current measurement value and estimated value at the previous moment are first Observe ψ(k), and then compensate for the disturbance in the control law and parameter estimation law to obtain the control input u at the current moment dq (k) and

[0036] Further, in step S1, Φ c (k) is the pseudo partial derivative matrix:

[0037]

[0038] They are f1(·), f2(·) for u d (k),u q (k) In [u d (k),u d (k-1)] and [u q (k),u q The partial derivative at (k-1)];

[0039] Φ c (k) is a diagonally dominant matrix, satisfying: |φ ij (k)|≤b1,b2≤|φ ij (k)|≤αb2,α≥1,b2≥b1(2α+1)(m+1),i=1,…,m,j=1,…,m,i≠j, and Φ c The sign of any element of (k) remains unchanged; in this permanent magnet synchronous motor control system, m=2.

[0040] Furthermore, in step S1, the total disturbance model is as follows:

[0041] The change in output is expressed as:

[0042] Δy(k+1)=f(y′(k),u′(k),d′(k))-f(y′(k-1),u′(k-1),d′(k-1))-f(y′(k),[u(k-1),u(k-1),…,u(kn u )],d′(k))+f(y′(k),[u(k-1),u(k-1),…,u(kn u )],d′(k))

[0043] where \(y^{\prime}(k)=[y(k),\ldots,y(k - n y )] T , \(u^{\prime}(k)=[u(k),\ldots,u(k - n u )] T , \(d^{\prime}(k)=[d(k),\ldots,d(k - n d )] T

[0044] The total system disturbance is:

[0045] \(\psi(k)=f(y^{\prime}(k),[u(k - 1),u(k - 1),\ldots,u(k - n u )],d^{\prime}(k)) - f(y^{\prime}(k - 1),u^{\prime}(k - 1),d^{\prime}(k - 1))\).

[0046] Furthermore, in the step S2, the control cost function is given by the speed loop; substituting the data model into the control cost function and letting obtain the control law of the voltage:

[0047]

[0048] where \(\lambda>0\) is the weight factor; \(\rho\in[0,1]\) is the step factor; used in the stability analysis.

[0049] Furthermore, in the step S2, an improved projection algorithm is used to estimate the pseudo - partial derivative matrix online, including:

[0050] Substituting the data model and letting obtain the following parameter estimation law:

[0051]

[0052] where \(\mu>0\) is the weight factor; \(\eta\in[0,2]\) is the step factor; used in the stability analysis;

[0053] To ensure the convergence of the parameter estimation algorithm, a reset mechanism is designed as follows:

[0054] If or or has a different sign from the initial value , then let for reset;

[0055] If or has a different sign from the initial value , then let Perform reset.

[0056] Furthermore, in the step S3, the discrete-space transfer function of the extended state observer is:

[0057]

[0058] Its characteristic equation is:

[0059] z 2 +(l1 - 2)z - l1 + l2 + 1 = 0

[0060] The two poles are respectively:

[0061]

[0062] Configure both poles to z0, |z0| < 1;

[0063] The gain of the extended state observer is set as:

[0064]

[0065] When |z0| approaches 0, the dynamic performance of the observer deteriorates. When |z0| approaches 1, the stability of the system is affected; for the current loop of the permanent magnet synchronous motor, as the inner loop of control, set a smaller |z0| to meet the bandwidth requirements.

[0066] Furthermore, for the output voltage reference value, control pulses are obtained through SVPWM modulation to control the conduction state of the switching tubes, thereby realizing the control of the permanent magnet synchronous motor.

[0067] Compared with the prior art, the advantages of the present invention are as follows:

[0068] (1) The present invention does not rely on any accurate mathematical model of the permanent magnet synchronous motor, and only completes the model-free adaptive current control through the input and output data of the control system. It is insensitive to model and parameter mismatches and can significantly improve the robustness of the motor control system;

[0069] (2) The present invention designs a discrete-time extended state observer to observe the unknown disturbances caused by external disturbances and system uncertainties, and compensates for them in the control law and parameter estimation law. On the one hand, it reduces the complexity of the pseudo partial derivative matrix, making parameter estimation easier; on the other hand, it enhances the anti-disturbance ability of the system and effectively improves the control performance. It can significantly improve the stability of the motor control system under complex working conditions. Description of the Drawings

[0070] Figure 1 The flowchart of the method provided by the present invention;

[0071] Figure 2 1. Block diagram of the method provided by the present invention;

[0072] Figure 3 2. Waveform diagrams of the measured d - axis and q - axis currents and current reference values of a permanent - magnet synchronous motor driven based on the present invention;

[0073] Figure 4 3. Waveform diagrams of the measured d - axis and q - axis currents and current reference values of a permanent - magnet synchronous motor driven based on the present invention under various parameter mismatch conditions;

[0074] Figure 5 4. Harmonic distribution diagrams of the three - phase currents of a permanent - magnet synchronous motor driven based on the present invention;

[0075] Figure 6 5. Observed total disturbance waveform diagrams of the d - axis and q - axis of a permanent - magnet synchronous motor driven based on the present invention;

[0076] Figure 7 6. Comparative waveform diagrams of the measured d - axis and q - axis currents and current reference values of a permanent - magnet synchronous motor driven based on the present invention and a traditional model - free adaptive control method;

[0077] Figure 8 7. Partially enlarged comparative waveform diagrams of the measured d - axis and q - axis currents and current reference values of a permanent - magnet synchronous motor driven based on the present invention and a traditional model - free adaptive control method. Detailed implementation manners

[0078] The following describes the specific implementation manners of the present invention in conjunction with embodiments:

[0079] It should be noted that the structures, ratios, sizes, etc. shown in this specification are only used to cooperate with the content disclosed in the specification for those skilled in this technology to understand and read, and are not used to limit the implementation conditions of the present invention. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope covered by the technical content disclosed in the present invention.

[0080] At the same time, the terms such as "upper", "lower", "left", "right", "middle", and "one" cited in this specification are only for the convenience of clear narration and are not used to limit the scope of implementation of the present invention. The change or adjustment of their relative relationships, without substantial change in the technical content, should also be regarded as the scope of implementation of the present invention.

[0081] Embodiment 1:

[0082] This embodiment discloses a model - free adaptive control method for a permanent - magnet synchronous motor based on an extended state observer, as shown in Figure 1 and Figure 2 The specific description is as follows:

[0083] The motor obtains the actual rotational speed ω through an encoder m and compares it with the desired rotational speed to obtain the current reference value through the speed regulator for the error value, which is then input into the model-free adaptive current controller of the permanent magnet synchronous motor in this embodiment. First is the discrete-time extended state observer part. Based on the current i dq (k - 1) and the estimated value of the pseudo partial derivative matrix the disturbance ψ(k) at the current moment is estimated; then it enters the model-free adaptive control part. After obtaining the current moment by the parameter estimation law and determining whether resetting should be performed, the dq-axis voltage reference values are calculated The output voltage reference value is used to obtain control pulses through SVPWM modulation to control the conduction state of the switching tubes, thereby achieving the control of the permanent magnet synchronous motor.

[0084] Specifically, in combination with Figure 2 , the specific steps of the model-free adaptive control method for the permanent magnet synchronous motor based on the extended state observer in this embodiment are as follows:

[0085] Step 1: Use the unknown nonlinear functions f1(·) and f2(·) to describe the discretized voltage equations of the permanent magnet synchronous motor in the dq-axis of the synchronous rotating coordinate system:

[0086] y(k + 1) = f(y(k), …, y(k - n y ), u(k), …, u(k - n u ), d(k), …, d(k - n d ))

[0087] where f(·) = [f1(·), f2(·)] T is the unknown nonlinear function vector; n y , n u , n d are positive system orders.

[0088] In the permanent magnet synchronous motor system, y(k) = i dq (k) = [i d (k), i q (k)] T is the output current vector, u(k) = u dq (k) = [u d (k), u q (k)] T , d(k) = d dq (k) = [d d (k), d q (k)] TThey are the input voltage vector and the unknown external disturbance vector respectively. To reduce the computational load, a compact-form dynamic linearization data model is adopted, that is, the pseudo-orders of the input, output, and disturbance are 0, 1, and 1 respectively. According to the differential mean value theorem, the nonlinear discrete-time data model of the permanent magnet synchronous motor considering the disturbance can be established as follows:

[0089] Δi dq (k + 1) = Φ c (k)Δu dq (k) + ψ(k)

[0090] where Φ c (k) is the pseudo-partial derivative matrix and ψ(k) is the total disturbance term.

[0091] Step 2: Design the control cost function as follows:

[0092]

[0093] where λ > 0 is the weight factor.

[0094] Substituting the data model, the control law of the voltage can be obtained.

[0095] Design the parameter estimation cost function as follows:

[0096]

[0097] where μ > 0 is the weight factor.

[0098] Substituting the data model, the corresponding parameter estimation law can be obtained.

[0099] Step 3: Design an extended state observer to observe and compensate for the total disturbance ψ(k):

[0100] Select the state variables as x(k) = [i dq (k), ψ(k)] T , the output vector is y(k), the control input u(k) = Δu dq (k), and w(k) = ψ(k + 1) - ψ(k). The discrete-time state space model of the data model can be written as:

[0101]

[0102] where I = I 2×2 , 0 = 0 2×2 are the second-order identity matrix and zero matrix respectively.

[0103] Design the extended state observer as follows:

[0104]

[0105] where \(L = [l_1I, l_2I]\) T is the gain vector.

[0106] At the beginning of each control, first, the current measurement value and the estimated value at the previous moment are used to observe \(\psi(k)\), and then the disturbance is compensated in the control law and the parameter estimation law to obtain the control input \(u\) dq (k) and

[0107] In a preferred embodiment of the present invention, \(\varPhi\) c (k) described in step one is a pseudo partial derivative matrix:

[0108]

[0109] are the partial derivative values of \(f_1(\cdot)\) and \(f_2(\cdot)\) with respect to \(u\) d (k) and \(u\) q (k) at \([u\) d (k), \(u\) d (k - 1)] and \([u\) q (k), \(u\) q (k - 1)].

[0110] \(\varPhi\) c (k) is a diagonally dominant matrix, satisfying: \(|\varphi\) ij (k)| \(\leq b_1\), \(b_2 \leq |\varphi\) ij (k)| \(\leq \alpha b_2\), \(\alpha \geq 1\), \(b_2 \geq b_1(2\alpha + 1)(m + 1)\), \(i = 1, \ldots, m\), \(j = 1, \ldots, m\), \(i \neq j\). At the same time, any element sign of \(\varPhi\) c (k) remains unchanged. In this permanent magnet synchronous motor control system, \(m = 2\).

[0111] In a preferred embodiment of the present invention, the total disturbance considered in step one is modeled as follows:

[0112] The change in the output can be expressed as:,

[0113] \(\Delta y(k + 1)=f(y'(k),u'(k),d'(k)) - f(y'(k - 1),u'(k - 1),d'(k - 1)) - f(y'(k),[u(k - 1),u(k - 1),\ldots,u(k - n u )],d'(k))+f(y'(k),[u(k - 1),u(k - 1),\ldots,u(k - n u )],d'(k))

[0114] where \(y'(k)=[y(k),\ldots,y(k - n y )]T , u′(k) = [u(k), …, y(k - n u )] T , d′(k) = [d(k), …, d(k - n d )] T

[0115] The total system disturbance is:

[0116] ψ(k) = f(y′(k), [u(k - 1), u(k - 1), …, u(k - n u )], d′(k)) - f(y′(k - 1), u′(k - 1), d′(k - 1))

[0117] In a preferred embodiment of the present invention, the in the control cost function in step two is given by the speed loop. Substitute the data model into the control cost function and let The control law of the voltage can be obtained:

[0118]

[0119] where λ > 0 is the weight factor; ρ ∈ 0, 1 is the step size factor. Used in stability analysis.

[0120] In a preferred embodiment of the present invention, the parameter estimation algorithm in step two is the improved projection algorithm. Substitute the data model and let The following parameter estimation law can be obtained:

[0121]

[0122] where μ > 0 is the weight factor; η ∈ 0, 2 is the step size factor. Used in stability analysis.

[0123] To ensure the convergence of the parameter estimation algorithm, a reset mechanism is designed as follows:

[0124] If or or has a different sign from the initial value , then let be reset;

[0125] If or has a different sign from the initial value , then let be reset.

[0126] In a preferred embodiment of the present invention, the discrete space transfer function of the extended state observer in step three is as follows:

[0127]

[0128] Its characteristic equation is:

[0129] z 2 +(l1 - 2)z - l1 + l2 + 1 = 0

[0130] The two poles are respectively:

[0131]

[0132] Both poles are configured to z0, |z0| < 1.

[0133] In a preferred embodiment of the present invention, the gain of the observer is set as:

[0134]

[0135] When |z0| approaches 0, the dynamic performance of the observer deteriorates. When |z0| approaches 1, the stability of the system is affected. For the current loop of the permanent magnet synchronous motor, as the inner loop of control, a smaller |z0| is set to meet the bandwidth requirement.

[0136] In a preferred embodiment of the present invention, the output voltage reference value is used to obtain control pulses through SVPWM modulation to control the conduction state of the switching tubes, thereby realizing the control of the permanent magnet synchronous motor.

[0137] In a specific example of adopting the present invention, in order to verify the effectiveness of the method of the present invention, a corresponding system simulation model is built in the Matlab / Simulink environment. The motor parameters are as follows: rated voltage U = 311V, rated current I = 20A, number of pole pairs of the motor P n = 4, stator resistance R s = 3Ω, stator inductance L s = 8.5mH, permanent magnet flux linkage ψ f = 0.1688Wb. The motor model is given a torque step (from 0N to 2N) at 0.8s and another torque step (from 2N to 5N) at 1.4s, and the reference speed is 1000r / min.

[0138] The parameters are selected as follows: the discrete-time extended state observer pole configuration is z0 = 0.15, the weight factors of the model-free adaptive control part are λ = 2.53, μ = 1.60, the step factors are ρ = 0.84, η = 1.70.

[0139] The result is as Figure 3As shown, the dq-axis stator current has small fluctuations and can quickly follow the given value when the load torque changes. At the same time, as Figure 4 shown, an experiment on motor parameter mismatch was carried out. From left to right are the current waveforms with the inductance, resistance, and permanent magnet flux linkage mismatched by one time. It can be seen from the figure that the change of motor parameters has little impact on the performance of the motor control system and has better robustness. Fourier analysis was performed on the three-phase stator current. As Figure 5 shown, the total harmonic distortion THD = 2.36% was obtained, the sinusoidal degree of current control is high, and the control effect is good. Figure 6 is the waveform diagram of the total disturbance on the dq-axis observed by the extended state observer.

[0140] Figure 7 and Figure 8 compare the dq-axis stator current waveform diagrams of the method provided by the present invention and the traditional model-free adaptive control method. The left is the traditional model-free adaptive control method, and the right is the method provided by the present invention. From the Figure 8 local enlarged view (from 1.35 s to 1.6 s), it can be seen that when a sudden load is applied, the overshoot of the current of the method provided by the present invention is smaller, and at the same time, it has a faster current response speed and stronger anti-disturbance ability.

[0141] Embodiment 2:

[0142] This embodiment provides a terminal device, which includes a processor and a memory. The memory is used to store a computer program, and the computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for operations based on the extended state observer permanent magnet synchronous motor model-free adaptive control method, including the following steps:

[0143] S1: Adopt the compact-form dynamic linearization method to establish a nonlinear discrete-time data model of a permanent magnet synchronous motor considering disturbances in the synchronous rotating coordinate system; the model is based on the differential mean value theorem, describes the influence of current increment on the input voltage with a pseudo partial derivative matrix, and includes a total disturbance term;

[0144] S2: Construct a control cost function through the data model and the current current measurement value to obtain the control law of voltage, that is, the control input. Use an improved projection algorithm to perform online estimation on the pseudo partial derivative matrix to obtain the parameter estimation law, that is, the pseudo partial derivative estimation value;

[0145] S3: Based on the current current measurement value, the calculated pseudo partial derivative estimation value and the control input, design an extended state observer to observe the total disturbance, and feedback the total disturbance estimated by the extended state observer to the control law in step S2 for disturbance compensation, and finally calculate the control input to be applied, that is, the voltage reference value.

[0146] Embodiment 3:

[0147] This embodiment provides a storage medium, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a terminal device for storing programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space that stores the operating system of the terminal. And, one or more instructions suitable for being loaded and executed by a processor are stored in this storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory.

[0148] One or more instructions stored in the computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer in the above embodiment; one or more instructions in the computer-readable storage medium are loaded and executed by a processor to perform the following steps:

[0149] S1: Adopt the compact-form dynamic linearization method to establish a nonlinear discrete-time data model of a permanent magnet synchronous motor considering disturbances in the synchronous rotating coordinate system; the model is based on the differential mean value theorem, describes the influence of current increment on the input voltage with a pseudo partial derivative matrix, and includes a total disturbance term;

[0150] S2: Construct a control cost function based on the data model and the current current measurement value to obtain the control law of the voltage, i.e., the control input. Use an improved projection algorithm to perform an online estimation of the pseudo-partial derivative matrix to obtain the parameter estimation law, i.e., the pseudo-partial derivative estimation value.

[0151] S3: Based on the current current measurement value, the calculated pseudo-partial derivative estimation value, and the control input, design an extended state observer to observe the total disturbance, and feedback the total disturbance estimated by the extended state observer to the control law in step S2 for disturbance compensation. Finally, calculate the control input to be applied, i.e., the voltage reference value.

[0152] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention 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.) containing computer-usable program code.

[0153] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes 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 process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0154] 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 process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0155] 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. Thus, the instructions executed on the computer or other programmable device provide means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocksFigure 1 Steps of the functions specified in one or more boxes.

[0156] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above embodiments, and various changes can be made without departing from the spirit of the present invention within the knowledge of those of ordinary skill in the art.

[0157] Many other changes and modifications can be made without departing from the concept and scope of the present invention. It should be understood that the present invention is not limited to the specific embodiments, and the scope of the present invention is defined by the appended claims.

Claims

1. A model-free adaptive control method for permanent magnet synchronous motors based on an extended state observer, characterized in that, The method includes the following steps: S1: Adopt the compact-form dynamic linearization method to establish a nonlinear discrete-time data model of a permanent magnet synchronous motor considering disturbances in the synchronous rotating coordinate system. The model is based on the differential mean value theorem, describes the influence of current increment on the input voltage with a pseudo partial derivative matrix, and includes a total disturbance term. S2: Construct a control cost function through the data model and the current measurement value to obtain the control law of voltage, i.e., the control input. Use an improved projection algorithm to perform online estimation on the pseudo partial derivative matrix to obtain the parameter estimation law, i.e., the pseudo partial derivative estimation value. S3: Based on the current measurement value, the calculated pseudo partial derivative estimation value, and the control input, design an extended state observer to observe the total disturbance, and feedback the total disturbance estimated by the extended state observer to the control law in step S2 for disturbance compensation. Finally, calculate the control input to be applied, i.e., the voltage reference value.

2. The model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, wherein The step S1 includes: Use unknown nonlinear functions f1(·) and f2(·) to describe the discretized voltage equations of the permanent magnet synchronous motor in the dq axes of the synchronous rotating coordinate system: y(k + 1) = f(y(k), …, y(k - n y ), u(k), …, u(k - n u ), d(k), …, d(k - n d )) where \(f(\cdot)=[f_1(\cdot), f_2(\cdot)]\) T is an unknown non - linear function vector; \(n\) y , \(n\) u , \(n\) d is the system order greater than zero; In a permanent magnet synchronous motor system, y(k) = i dq (k) = [i d (k), i q (k)] T is the output current vector, u(k) = u dq (k) = [u d (k), u q (k)] T , d(k) = d dq (k) = [d d (k), d q (k)] T are the input voltage vector and the unknown external disturbance vector respectively; To reduce the computational burden, a compact-form dynamic linearization data model is adopted, that is, the pseudo-orders of the input, output, and disturbance are 0, 1, and 1 respectively; According to the differential mean value theorem, a nonlinear discrete-time data model of the permanent magnet synchronous motor considering the disturbance can be established as follows: Δi dq (k + 1) = Φ c (k)Δu dq (k) + ψ(k) where, Φ c (k) is the pseudo partial derivative matrix and ψ(k) is the total disturbance term.

3. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, characterized in that, The step S2 includes: Design the control cost function as follows: where λ > 0 is a weight factor; Substitute the data model to obtain the control law of voltage. Design the parameter estimation cost function as follows: where μ > 0 is a weight factor; Substitute the data model to obtain the corresponding parameter estimation law.

4. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, characterized in that The step S3 includes: Construct a state space model: Select the state variables as \(x(k)=[i dq (k),\psi(k)] T , the output vector is \(y(k)\), the control input \(u(k)=\Delta u dq (k)\), \(w(k)=\psi(k + 1)-\psi(k)\); write the discrete-time state-space model of the data model: where I = I 2×2 , 0 = 0 2×2 are the second-order identity matrix and zero matrix respectively; Design an extended state observer ESO: where L = [l1I, l2I] T is the gain vector; Disturbance compensation: At the beginning of each control, first, based on the current measured current value and the estimated value at the previous moment ψ(k) is observed, and then the disturbance is compensated in the control law and the parameter estimation law to obtain the control input u dq (k) and 5. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, characterized in that In the step S1, Φ c (k) is a pseudo partial derivative matrix: They are the partial derivative values of f1(·) and f2(·) with respect to u d (k) and u q (k) at [u d (k) and u d (k - 1)] and [u q (k) and u q (j - 1)] respectively; Φ c (k) is a diagonally dominant matrix and satisfies: |φ ij (k)| ≤ b1, b2 ≤ |φ ij (k)| ≤ ab2, α ≥ 1, b2 ≥ b1(2α + 1)(m + 1), i = 1, …, m, j = 1, …, m, i ≠ j, and at the same time any element symbol of Φ c (k) remains unchanged; in this permanent magnet synchronous motor control system, m = 2.

6. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, characterized in that, In the step S1, the total disturbance is modeled as follows: The change in output is expressed as: Δy(k + 1) = f(y′(k), u′(k), d′(k)) - f(y′(k - 1), u′(k - 1), d′(k - 1)) - f(y′(k), [u(k - 1), u(k - 1), …, u(k - n u )], d′(k)) + f(y′(k), [u(k - 1), u(k - 1), …, u(k - n u )], d′(k)) where y′(k) = [y(k), …, y(k - n y )] T , u′(k) = [u(k), …, u(k - n u )] T , d′(k) = [d(k), …, d(k - n d )] T The total system disturbance is: ψ(k) = f(y′(k), [u(k - 1), u(k - 1), …, u(k - n u )], d′(k)) - f(y′(k - 1), u′(k - 1), d′(k - 1)).

7. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, characterized in that, In the step S2, the is given by the speed loop; substitute the data model into the control cost function, and let to obtain the control law of the voltage: where λ > 0 is a weight factor; ρ ∈ (0, 1) is a step factor, which is used in the stability analysis.

8. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, characterized in that In the step S2, an improved projection algorithm is used to perform online estimation on the pseudo partial derivative matrix, including: Substitute the data model and let The following parameter estimation law is obtained: where μ > 0 is a weight factor; η ∈ (0, 2) is a step factor, which is used in the stability analysis; To ensure the convergence of the parameter estimation algorithm, a reset mechanism is designed as follows: If or or differs from the initial value in sign, then reset it; If or differs in sign from the initial value then reset to perform reset.

9. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 4, wherein In the step S3, the discrete space transfer function of the extended state observer is: Its characteristic equation is: z 2 +(l1 - 2)z - l1 + l2 + 1 = 0 The two poles are respectively: Configure both poles to z0, |z0| < 1; The gain of the extended state observer is set as: When |z0| is close to 0, the dynamic performance of the observer will deteriorate. When |z0| is close to 1, the stability of the system will be affected. For the current loop of the permanent magnet synchronous motor, as the inner control loop, set a smaller |z0| to meet the bandwidth requirements.

10. A model-free adaptive control method for a permanent magnet synchronous motor based on an extended state observer according to claim 1, characterized in that For the output voltage reference value, control pulses are obtained through SVPWM modulation to control the conduction state of the switching tubes, thereby realizing the control of the permanent magnet synchronous motor.

Images