Permanent magnet synchronous motor sliding mode integrated control method based on disturbance observer

Abstract

The invention discloses a permanent magnet synchronous motor sliding mode integrated control method based on a disturbance observer, and the method comprises the steps: constructing a hyperlocal mathematical model of a speed ring of a permanent magnet synchronous motor, combining a state equation with a first-order nonlinear model, and reducing the dependence on a precise system model; designing a model-free adaptive fast integration terminal sliding mode controller; a current loop backstepping controller is designed, electromagnetic parameter disturbance is considered, d-axis and q-axis current control laws are recursively constructed, and the anti-interference capability of a current loop is enhanced; and designing an extended super-spiral disturbance observer, estimating total unknown disturbance of the system and performing feed-forward compensation, thereby improving the control precision. The speed tracking precision of the system is improved, and the steady-state error is reduced.

Description

Technical Field

[0001] This invention relates to the field of motor speed control, and specifically to a sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) have been widely used in many fields, such as rail transportation, aerospace, industrial automation, and new energy vehicles. Due to their high power density and high efficiency, the control performance of PMSMs is often affected by factors such as model uncertainties, parameter disturbances, and external interference in practical applications. Although traditional PI control is widely used, its ability to reject disturbances and maintain stability is insufficient, making it difficult to meet requirements under complex operating conditions and dynamic environments. Therefore, sliding mode control has attracted considerable attention due to its robustness to system parameters.

[0003] In recent years, many scholars have proposed methods such as integral sliding mode control, neural network sliding mode control, and higher-order sliding mode control based on traditional sliding mode control. These methods effectively improve the control performance and robustness of permanent magnet synchronous motor systems. Compared with traditional linear sliding mode control, terminal sliding mode control has the advantage of finite-time convergence. However, it can also lead to singularity problems. Some scholars have proposed a non-singular fast terminal sliding mode control strategy, which effectively avoids singularity problems and improves both response speed and control accuracy. However, its adaptability to complex nonlinear systems and robustness under practical operating conditions still need further improvement. Some scholars have designed a composite control method combining a non-singular terminal sliding mode controller and a disturbance observer to overcome the contradiction between fast response and severe chatter in traditional non-singular fast terminal sliding mode controllers. However, the high gain of the observer can lead to oscillations and overshoot in the system. Some scholars have proposed a non-singular fast terminal sliding mode control strategy based on a higher-order super-torsion observer, which enhances the robustness and accuracy of permanent magnet synchronous motor position tracking. However, it fails to fully consider the impact of disturbances on the integrity of system state estimation and overall control performance.

[0004] Permanent magnet synchronous motor (PMSM) systems are often difficult to model accurately and completely due to factors such as external disturbances, parameter uncertainties, and nonlinear characteristics. Some researchers have proposed a model-free control method by constructing a hyperlocal model based on the system's input and output. This method reduces dependence on specific mathematical models and effectively addresses the impact of parameter uncertainties, unknown disturbances, and unmodeled dynamics on motor control performance. Building on this, some researchers have proposed an improved hyperlocal model that simplifies the design of model-free controllers by separating the known and unknown parts of the system.

[0005] In summary, existing methods have improved the dynamic performance, disturbance rejection and control accuracy of permanent magnet synchronous motors in several aspects, but they do not take into account the effects of parameter disturbances and external interference under complex operating conditions. Summary of the Invention

[0006] The purpose of this invention is to provide a sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer in order to solve the above-mentioned problems, thereby addressing the existing issues of considering the influence of PMSM parameter disturbances and external disturbances under complex operating conditions.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer, comprising:

[0008] S1. Construct a hyperlocal mathematical model of the speed loop of the permanent magnet synchronous motor, combining the state equation with the first-order nonlinear model to reduce the dependence on the accurate system model;

[0009] S2. Design a model-free adaptive fast integral terminal sliding mode controller;

[0010] S3. Design a current loop backstepping controller, considering electromagnetic parameter disturbances, recursively construct d-axis and q-axis current control laws, and enhance the anti-interference capability of the current loop.

[0011] S4. Design an extended superspiral disturbance observer to estimate the total unknown disturbance of the system and perform feedforward compensation to improve control accuracy.

[0012] Preferably, the hyperlocal mathematical model is described in the dq synchronous rotating reference frame as follows:

[0013]

[0014] Among them, u d and u q These represent the stator voltages along the d-axis and q-axis, respectively; i d and i q L represents the stator current along the d-axis and q-axis, respectively; d and L q These represent the stator inductance along the d-axis and q-axis, respectively; ω e and ω m T represents the electric angular velocity and the mechanical angular velocity, respectively. e and T L These represent electromagnetic torque and load torque, respectively; ψ r Indicates the magnetic flux linkage of permanent magnets, n p R represents the number of magnetic pole pairs. s J represents the stator resistance, B represents the moment of inertia, and B represents the damping coefficient.

[0015] Preferably, the hyperlocal mathematical model can also be described in the dq synchronous rotating reference frame in the following form:

[0016]

[0017] Where Δu d and Δu q Representing the parameters R respectively s 、L d 、L q and ψ r Voltage deviation caused by disturbances in the circuit;

[0018] The equation of motion for the machine is:

[0019]

[0020] The electromagnetic torque equation is:

[0021]

[0022] Where, ΔT e ΔP represents the deviation of the electromagnetic torque. n This represents the deviation caused by the disturbances in parameters J and B;

[0023] The state equation for the velocity loop is:

[0024]

[0025] Preferably, the design of the model-free adaptive fast integral terminal sliding mode controller includes:

[0026] The error between a given velocity value and the actual value is defined as:

[0027]

[0028] in, and ω e These represent the given and actual electric angular velocities of the electric motor, respectively.

[0029] Speed ​​error e ω Defined as a state variable, we get:

[0030]

[0031] The adaptive fast reaching law is:

[0032]

[0033] Among them, c1(e ω ) = c 10 +α|e ω | / (1+|eω |), c2(s)=c 20 +β|s| is the adaptive control gain, 0 < a < ln2, 0 < b < 1.

[0034] Preferably, the current loop backstepping controller design considers electromagnetic parameter disturbances and recursively constructs d-axis and q-axis current control laws to enhance the anti-interference capability of the current loop; including:

[0035] The current equation for a permanent magnet synchronous motor can be rewritten as follows:

[0036]

[0037] f1=-ΔR s i d +ΔL q ω e i q -ΔL d di d / dt and f2=-ΔR s i q -Δψ r ω e -ΔL d ω e i d -ΔL q di d / dt is the interference caused by changes in electromagnetic parameters, where ΔR s ΔL q ΔL d and Δψ r It is the deviation of electromagnetic parameters generated during the operation of the motor;

[0038] set up This is the current error. For the perturbation error, the Lyapunov function is defined as:

[0039]

[0040] Calculating the derivative of the above equation, we get:

[0041]

[0042] Where p1 and p2 are the positive gains to be designed;

[0043] By combining equations, The d-axis current loop control law can be obtained by separation:

[0044]

[0045] Similarly, the q-axis current loop control law is given in the same way as the d-axis current loop control law;

[0046] The Lyapunov function is defined as:

[0047]

[0048] Calculating the derivative of the above equation, we get:

[0049]

[0050] Where p3 and p4 are the positive gain to be designed, e iq It is the q-axis current error;

[0051] q-axis current loop control law:

[0052]

[0053] Preferably, the design of the extended superspiral disturbance observer estimates the total unknown disturbance of the system and performs feedforward compensation to improve control accuracy; including:

[0054] design:

[0055]

[0056] Among them, electric angular velocity and ω e The unknown part F is the state variable, and the q-axis current i q It is the control input, ω e This is system output;

[0057] The dynamic error equation for ESTDO is given by the following equation:

[0058]

[0059] in It is a speed observation error, it is The observation error of the unknown total disturbance;

[0060] Prove that the observer's estimation error converges to zero in a finite time by choosing a positive definite quadratic Lyapunov function:

[0061]

[0062] Simplifying, we get:

[0063]

[0064] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0065] This method combines model-free adaptive fast integral terminal sliding mode control with backstepping control based on an extended superspiral disturbance observer. First, a novel hyperlocal mathematical model is constructed to reduce the control method's dependence on an exact model. Second, by combining an adaptive fast reaching law and a non-singular fast integral terminal sliding surface, a model-free adaptive fast integral terminal sliding mode controller is designed for the velocity loop to achieve high-precision velocity control. Next, a backstepping controller is designed for the current loop to suppress current oscillations. Finally, an extended superspiral disturbance observer is designed to estimate the total unknown disturbance of the system, and the estimated value is feedforward compensated to improve control accuracy. Comparative experiments verify the effectiveness of the proposed method. Attached Figure Description

[0066] In order to more clearly illustrate the embodiments of the present invention 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0067] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0068] Figure 2 This is a comparison diagram of the phase plane trajectories of the present invention;

[0069] Figure 3 This is a comparison chart of the convergence time of the sliding surface in this invention;

[0070] Figure 4 This is a block diagram of the MFAFITSMC of the present invention;

[0071] Figure 5 This is a block diagram of the backstepping control of the present invention;

[0072] Figure 6 This is a block diagram of the ESTDO of the present invention;

[0073] Figure 7 This is a block diagram of the PMSM control system of the present invention;

[0074] Figure 8 This is a comparison graph of the speed and torque response of the present invention;

[0075] Figure 9 This is a comparison diagram of the current response waveforms of the present invention;

[0076] Figure 10 This is a comparison chart of the observation results of the present invention;

[0077] Figure 11 This is a comparison chart of the experimental results of this invention. Detailed Implementation

[0078] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0079] like Figure 1 As shown in this embodiment, the sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer includes the following steps:

[0080] S1. Construct a hyperlocal mathematical model of the speed loop of the permanent magnet synchronous motor, combining the state equation with the first-order nonlinear model to reduce the dependence on the accurate system model;

[0081] S2. Design a model-free adaptive fast integral terminal sliding mode controller;

[0082] S3. Design a current loop backstepping controller, considering electromagnetic parameter disturbances, recursively construct d-axis and q-axis current control laws, and enhance the anti-interference capability of the current loop.

[0083] S4. Design an extended superspiral disturbance observer to estimate the total unknown disturbance of the system and perform feedforward compensation to improve control accuracy.

[0084] First, a hyperlocal mathematical model of the speed loop of the permanent magnet synchronous motor is constructed. The mathematical model of the permanent magnet synchronous motor in the dq synchronous rotating reference frame is given by the following equation:

[0085]

[0086] Among them, u d and u q These represent the stator voltages along the d-axis and q-axis, respectively; i d and i q L represents the stator current along the d-axis and q-axis, respectively; d and L q These represent the stator inductance along the d-axis and q-axis, respectively; ω e and ω m T represents the electric angular velocity and the mechanical angular velocity, respectively. e and T L These represent electromagnetic torque and load torque, respectively; ψ r It is a permanent magnet flux linkage, n p It is the number of magnetic pole pairs, R s J is the stator resistance, J is the moment of inertia, and B is the damping coefficient.

[0087] The parameters of a permanent magnet synchronous motor (PMSM) (such as stator resistance, moment of inertia, and damping coefficient) are subject to disturbances under complex operating conditions such as high temperature and magnetic saturation. Considering the influence of electromagnetic parameter disturbances, the stator voltage equation of the PMSM is given by the following formula:

[0088]

[0089] Where Δu d and Δu q Representing the parameters R respectively s 、L d 、L q and ψ r Voltage deviation caused by disturbance.

[0090] The electromagnetic torque equation can be rewritten as follows:

[0091]

[0092] The mechanical motion equations of a permanent magnet synchronous motor can be rewritten as follows:

[0093]

[0094] Where, ΔT e ΔP represents the deviation of the electromagnetic torque. n This represents the deviation caused by the disturbances in parameters J and B.

[0095] Considering parameter disturbances and unknown interferences, the state equation for the speed loop of the permanent magnet synchronous motor is:

[0096]

[0097] This embodiment discloses a novel hyperlocal model for the speed loop of a permanent magnet synchronous motor:

[0098] The first-order nonlinear novel hyperlocal model is:

[0099]

[0100] in y is the state variable, γ and δ are the non-physical constants to be designed, u is the control input, y is the system output, and F represents the total disturbance of the system and satisfies the Lipschitz continuity condition and boundedness.

[0101] To ensure robust and accurate control performance under complex operating conditions, a novel hyperlocal velocity loop model is constructed by combining some equations:

[0102]

[0103] in δ = -B / J, where ζ(t) represents the rate of change F.

[0104] Model-free adaptive fast integral terminal sliding mode control includes:

[0105] The error between a given velocity value and the actual value is defined as:

[0106]

[0107] in and ω e These represent the given and actual electric angular velocities of the electric motor, respectively.

[0108] Speed ​​error e ω Define it as a state variable, then:

[0109]

[0110] To suppress surface jitter while ensuring arrival time, an adaptive fast approach law is designed as follows:

[0111]

[0112] Among them, c1(e ω ) = c 10 +α|e ω | / (1+|e ω |), c2(s)=c 20 +β|s| is the adaptive control gain, 0 < a < ln2, 0 < b < 1, c 10 >0, c 20 >0.

[0113] The adaptive gain and saturation features in AFRL ensure finite-time convergence and suppress chatter, and c1(e ω c2(s) adapts to the system state and the sliding surface. This enhances robustness and improves control accuracy.

[0114] NFITSMS is designed as follows:

[0115]

[0116] Where l1, l2, l3p and q are constants, and 1 < p / q < 2. By combining dynamic error, nonlinear integral and proportional control, the sliding surface exhibits fast convergence and nonsingularity.

[0117] Calculating the derivative of (11), we get:

[0118]

[0119] Using AFRL, a model-free adaptive fast integral terminal sliding mode control law is designed for:

[0120]

[0121] To verify the improved performance of the proposed sliding mode control strategy, a phase plane comparison was performed using a constant velocity reaching law and a non-singular terminal sliding mode control (NTSMC) method. Both methods simulated [x1,x2]=[1,0] under the same initial conditions. A saturation function was introduced in both controllers to suppress chatter.

[0122] Figure 2 and Figure 3 The performance improvements of the proposed sliding mode control strategy are demonstrated, characterized by better convergence and chatter suppression compared to NTSMC. It produces a smoother trajectory and reduces the sliding surface time to approximately 0.17 seconds, below zero, while NTSMC requires 0.66 seconds.

[0123] To prove the stability of the designed MFAFITSMC, a Lyapunov function is defined. Its derivative V is:

[0124]

[0125] Therefore, the designed controller satisfies the Lyapunov stability condition, and the system asymptotically converges to stability.

[0126] The block diagram of the designed speed loop controller is as follows: Figure 4 The diagram illustrates the internal structure of the Model-Free Adaptive Fast Integral Terminal Sliding Mode Controller (MFAFITSMC), highlighting the design of the sliding surface and arrival law, enabling the control system to achieve fast and stable control without relying on a model.

[0127] Under complex operating conditions, traditional current-loop PI control is sensitive to changes in motor parameters and has poor disturbance suppression capabilities, making it difficult to meet high-performance control requirements. Backstepping control, as an efficient nonlinear control method, has been widely used in motor drive systems. It ensures accurate current tracking of the reference value through recursive controller design. Furthermore, recursive Lyapunov stability analysis is used to systematically handle complex nonlinearities. Therefore, stability and robustness of the current loop are achieved.

[0128] Considering the influence of electromagnetic parameter disturbances, the PMSM current equation is rewritten as:

[0129]

[0130] Where ΔR s ΔL q ΔLd and Δψ r It is the deviation of electromagnetic parameters generated during the operation of the motor; f1=-ΔR s i d +ΔL q ω e i q -ΔL d di d / dt and f2=-ΔR s i q -Δψ r ω e -ΔL d ω e i d -ΔL q di d / dt is interference caused by changes in electromagnetic parameters.

[0131] set up This is the current error. This represents the perturbation error. The Lyapunov function is defined as follows:

[0132]

[0133] Calculate the derivative of (16) to obtain:

[0134]

[0135] p1 and p2 are the positive gains to be designed.

[0136] By combining equations (15) and (17), The d-axis current loop control law can be obtained by separation:

[0137]

[0138] Similarly, the q-axis current loop control law is given in the same manner as the d-axis current loop control law. The Lyapunov function is defined as:

[0139]

[0140] Differentiating(19) yields:

[0141] p3 and p4 are the positive gains to be designed. iq It is the Q-axis current error.

[0142] q-axis current loop control law:

[0143]

[0144] The block diagram of the designed current loop backstepping controller is as follows: Figure 5 As shown, it clearly illustrates the control law of the dq-axis current loop.

[0145] To eliminate the impact of external load disturbances and system parameter changes on system stability, an ESTDO method is proposed to estimate and compensate for unknown disturbances in the speed control system.

[0146] ESTDO is designed according to (6):

[0147]

[0148] Among them, electric angular velocity and ω e The unknown part F is the state variable, and the Q-axis current i q It is the control input, ω e This is system output.

[0149] According to formulas (6) and (22), the dynamic error equation of ESTDO is given by the following equation:

[0150]

[0151] in It is a speed observation error, it is The observation error of the unknown total disturbance.

[0152] Prove that the observer's estimation error converges to zero in a finite time by choosing a positive definite quadratic Lyapunov function:

[0153]

[0154] Simplify (24) outputs

[0155]

[0156] Where P is a positive definite symmetric matrix. Then, The eigenvalues ​​of the λ matrix are denoted by , and the sum of the Euclidean norms of the ||υ||2 vectors satisfies .

[0157] From the derivative υ of the orientation quantity, we obtain:

[0158]

[0159] Differential (25) yields:

[0160]

[0161] in It is a positive definite symmetric matrix. Therefore, hold

[0162] Since both P are positive definite symmetric matrices of Q, the system error will converge to zero in a finite time when the condition is met.

[0163] Therefore, the speed loop control law is designed as follows:

[0164]

[0165] The design of the ESTDO block diagram is as follows: Figure 6 As shown.

[0166] A model-free nonsingular terminal sliding mode control (MFNTSMC-PI) control system based on ESTDO was constructed in MATLAB / Simulink and compared with the model-free nonsingular terminal sliding mode control (MFNTSMC-PI) method. The initial velocity was set to 500 r / min and increased to 1000 r / min at 0.2 s. The system parameters used in the simulation are shown in Table 1. The control method employed...

[0167] Figure 7 A block diagram of a PMSM control system based on ESTDO and MFAFITSMC-BC is shown. The core components of the control system include the controller, disturbance observer, coordinate transformation, and system feedback. During control execution, ESTDO estimates unknown disturbances acting on the PMSM. These estimates are applied in a feedforward manner in the MFAFITSMC-BC controller to enhance the system's ability to suppress disturbances caused by parameter variations.

[0168] Table 1 Parameters of Permanent Magnet Synchronous Motor

[0169]

[0170] By changing the electromagnetic parameter ψ r R s 、L d and L q The load torque TL was used to verify the control performance of the two control methods to simulate complex operating conditions. The parameter variations are shown in Table 2.

[0171] Table 2 Disturbance parameters and range

[0172]

[0173]

[0174] Figure 8 Speed ​​and torque are displayed. Figure 8As clearly shown in (a), the MFAFITSMC-BC method reaches the target speed approximately 25% faster than the MMFNTSMC-PI method during startup, with a sudden speed change at 0.2 s. When the load torque disturbance occurs at 0.3 s, the fluctuation amplitude is reduced by 41%. When electromagnetic parameter disturbances occur, the MFAFITSMC-BC method quickly recovers to a steady state with smaller fluctuation amplitudes. In other words, the MFAFITSMC-BC method demonstrates a more stable response to sudden disturbances and effectively suppresses severe oscillations.

[0175] Figure 8 (b) shows the torque response. In the initial stage, both methods can quickly track the reference torque value, but instantaneous overshoot and significant ripple occur during startup. When load disturbances occur, the MFAFITSMC-BC converges to the reference value faster than the MNFTSMC-PI. The instantaneous peak value of the MFAFITSMC-BC method is 16.58 N·m, which is 9.26% lower than the 18.27 N·m peak value observed by the MNFTSMC-PI method. When parameter disturbances occur, the response curve of the MNFTSMC-PI exhibits greater fluctuations than that of the MFAFITSMC-BC. The MFAFITSMC-BC exhibits a more stable response, faster tracking, and higher steady-state accuracy.

[0176] In summary, the MFAFITSMC-BC demonstrates significant advantages in control speed, system robustness, and control accuracy, effectively reducing the impact of external disturbances under complex operating conditions.

[0177] Figure 9 The current is displayed. For example... Figure 9 As shown in (c) and 9(d), both methods exhibit current overshoot when a sudden change in load torque is applied at 0.3 s. The peak q-axis current of the MFAFITSMC-BC method is 15.79 A, ​​a 9.3% reduction compared to the 17.41 A peak current of the MMFNTSMC-PI method. When motor parameters are disturbed, the steady-state value of the dq-axis current of the MFAFITSMC-BC method shows a 50% smaller steady-state error than that of the MMFNTSMC-PI method. The MFAFITSMC-BC method effectively suppresses current ripple caused by electromagnetic parameter disturbances.

[0178] Figure 9 (e) shows the current response curve for phase a. At 0–0.2 s, i aThe fluctuation ranges of the MFAFITSMC-BC and MMFNTSMC-PI methods are ±0.2A and ±0.4A, respectively, with a difference of nearly 50%. The MFAFITSMC-BC method exhibits no current overshoot and a smoother current waveform during sudden load changes. This indicates that the control system based on the MFAFITSMC-BC method possesses good dynamic performance and steady-state robustness.

[0179] Figure 10 The hyperlocal model of the velocity loop and estimates of unknown disturbances in the velocity tracking errors of STDO and ESTDO are shown. From Figure 10 As can be clearly seen in (g), ESTDO accurately estimated the unknown disturbance with small flutter. From Figure 10 As clearly shown in (h), when the load change is 0, the velocity estimation error of the ESTDO method is 0.015 r / min lower than that of the STDO method. When the parameter disturbance occurs at 0.6 s, and the load changes, the velocity estimation error of the ESTDO method is 0.01 r / min lower than that of the STDO method. In other words, the ESTDO method has higher accuracy in observing unknown disturbances than the STDO method. This allows the system to maintain a smaller velocity tracking error and effectively improves the dynamic performance and robustness of the system.

[0180] After simulation, Table 3 provides a quantitative comparison of control performance based on ISE, ITSE, IAE, and TV. Clearly, the proposed MFAFITSMC-BC method outperforms MFAFITSMC-PI on all exponents, exhibiting faster error convergence, better steady-state accuracy, and smoother control input, further confirming the effectiveness and robustness of the strategy.

[0181] Table 3 Performance Evaluation Indicators

[0182]

[0183]

[0184] To verify the performance of the designed method, a hardware-in-the-loop (HILS) simulation experiment was conducted on the PMSM drive system using the RT-LAB platform. The required hardware platform consisted of a DSP controller (TMS320F2812) and a motor drive system simulated by the RT-LAB platform. The experimental parameters were consistent with the simulation parameters. Figure 11 Experimental results for the MFNTSMC-PI and MFAFITSMC-BC methods are shown.

[0185] from Figure 11It is evident that the dq-axis current fluctuations are large and the steady-state torque error is high under the MMFNTSMC-PI method. In contrast, the MFAFITSMC-BC method exhibits lower current fluctuations and lower steady-state torque error. When motor parameters are adjusted and unknown disturbances exist, the MFAFITSMC-BC method can quickly track the steady-state values ​​of dq-axis current, torque, and speed. Compared to the MFAFITSMC-BC method, the a-phase current exhibits greater oscillations under the MMFNTSMC-PI method. Figure 11 The results show the observations of two different observers, and it is clear that the ESTDO observer significantly outperforms the STDO observer. These results further demonstrate the strong robustness and excellent anti-interference ability of the MFAFITSMC-BC method.

[0186] To address the stability degradation problem of permanent magnet synchronous motors under complex operating conditions, a MFAFITSMC-BC control method based on ESTDO is proposed. Through simulation and experimental analysis, we draw the following conclusions:

[0187] First, by establishing a novel permanent magnet synchronous hyperlocal mathematical model, the dependence of the control method on the precise mathematical model of the system is effectively reduced, laying the foundation for achieving high-precision control. Second, the MFAFITSMC-BC method exhibits strong robustness to parameter variations and unknown external disturbances. It can converge to steady state quickly while significantly reducing torque ripple and current fluctuations. Third, the ESTDO design and feedforward compensation for estimated disturbances significantly improve the system's speed tracking accuracy and reduce steady-state error. Finally, experimental results verify the effectiveness of the designed method.

[0188] The designed control method demonstrates good adaptability and practicality in real-world applications. Future research could further explore how to enhance the system's adaptability and stability under more complex operating conditions.

[0189] Within the technical scope disclosed in this invention, any variations or substitutions that can be easily conceived should be included within the protection scope of this invention. Therefore, the protection scope of this invention should be determined by the scope of the claims.

Claims

1. A sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer, characterized in that, include: A hyperlocal mathematical model of the speed loop of a permanent magnet synchronous motor is constructed, which combines the state equation with a first-order nonlinear model to reduce the dependence on the accurate system model. Design a model-free adaptive fast integral terminal sliding mode controller; Design a current loop backstepping controller, consider electromagnetic parameter disturbances, recursively construct d-axis and q-axis current control laws, and enhance the anti-interference capability of the current loop. Design an extended superspiral disturbance observer to estimate the total unknown disturbance of the system and perform feedforward compensation to improve control accuracy.

2. The sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer according to claim 1, characterized in that: The hyperlocal mathematical model is described in the dq synchronous rotating reference frame as follows: Among them, u d and u q These represent the stator voltages along the d-axis and q-axis, respectively; i d and i q L represents the stator current along the d-axis and q-axis, respectively; d and L q These represent the stator inductance along the d-axis and q-axis, respectively; ω e and ω m T represents the electric angular velocity and the mechanical angular velocity, respectively. e and T L These represent electromagnetic torque and load torque, respectively; ψ r Indicates the magnetic flux linkage of permanent magnets, n p R represents the number of magnetic pole pairs. s J represents the stator resistance, B represents the moment of inertia, and B represents the damping coefficient.

3. The sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer according to claim 1, characterized in that: The hyperlocal mathematical model can also be described in the dq synchronous rotating reference frame in the following form: Where Δu d and Δu q Representing the parameters R respectively s L d L q and ψ r Voltage deviation caused by disturbances in the circuit; The equation of motion for the machine is: The electromagnetic torque equation is: Where, ΔT e ΔP represents the deviation of the electromagnetic torque. n This represents the deviation caused by the disturbances in parameters J and B; The state equation for the velocity loop is:

4. The sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer according to claim 2, characterized in that: The design of the model-free adaptive fast integral terminal sliding mode controller includes: The error between a given velocity value and the actual value is defined as: in, and ω e These represent the given and actual electric angular velocities of the electric motor, respectively. Speed ​​error e ω Defined as a state variable, we get: The adaptive fast reaching law is: Among them, c1(e ω ) = c 10 +α|e ω | / (1+|e ω |), c2(s)=c 20 +β|s| is the adaptive control gain, 0 < a < ln2, 0 < b < 1.

5. The sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer according to claim 3, characterized in that: The design of the current loop backstepping controller takes into account electromagnetic parameter disturbances and recursively constructs d-axis and q-axis current control laws to enhance the anti-interference capability of the current loop. include: The current equation for a permanent magnet synchronous motor can be rewritten as follows: f1=-ΔR s i d +ΔL q ω e i q -ΔL d di d / dt and f2=-ΔR s i q -Δψ r ω e -ΔL d ω e i d -ΔL q di d / dt is the interference caused by changes in electromagnetic parameters, where ΔR s ΔL q ΔL d and Δψ r It is the deviation of electromagnetic parameters generated during the operation of the motor; set up This is the current error. For the perturbation error, the Lyapunov function is defined as: Calculating the derivative of the above equation, we get: Where p1 and p2 are the positive gains to be designed; By combining equations, The d-axis current loop control law can be obtained by separation: Similarly, the q-axis current loop control law is given in the same way as the d-axis current loop control law; The Lyapunov function is defined as: Calculating the derivative of the above equation, we get: Where p3 and p4 are the positive gain to be designed, e iq It is the q-axis current error; q-axis current loop control law:

6. The sliding mode integrated control method for permanent magnet synchronous motors based on a disturbance observer according to claim 4, characterized in that: The design extends the superspiral disturbance observer to estimate the total unknown disturbance of the system and perform feedforward compensation, thereby improving control accuracy. include: design: Among them, electric angular velocity and ω e The unknown part F is the state variable, and the q-axis current i q It is the control input, ω e This is the system output: The dynamic error equation for ESTDO is given by the following equation: in It is a speed observation error, it is The observation error of the unknown total disturbance; Prove that the observer's estimation error converges to zero in a finite time by choosing a positive definite quadratic Lyapunov function: Simplifying, we get:

Images