Intelligent tracking control of a PMLSM using self‐evolving probabilistic fuzzy neural network

A hybrid supervisory control system using a recurrent fuzzy neural network (RFNN) is proposed to control the mover of a permanent magnet linear synchronous motor (PMLSM) servo drive for the tracking of periodic reference inputs. First, the field-oriented mechanism is applied to formulate the dynamic equation of the PMLSM. Then, a hybrid supervisory control system, which combines a supervisory control system and an intelligent control system, is proposed to control the mover of the PMLSM for periodic motion. The supervisory control law is designed based on the uncertainty bounds of the controlled system to stabilize the system states around a predefined bound region. Since the supervisory control law will induce excessive and chattering control effort, the intelligent control system is introduced to smooth and reduce the control effort when the system states are inside the predefined bound region. In the intelligent control system, the RFNN control is the main tracking controller which is used to mimic a idea control law and a compensated control is proposed to compensate the difference between the idea control law and the RFNN control. The RFNN has the merits of fuzzy inference, dynamic mapping and fast convergence speed, In addition, an online parameter training methodology, which is derived using the Lyapunov stability theorem and the gradient descent method, is proposed to increase the learning capability of the RFNN. The proposed hybrid supervisory control system using RFNN can track various periodic reference inputs effectively with robust control performance.

In this paper, the maximum thrust force per ampere (MTPA) strategy for saturated permanent magnet linear synchronous motors over the entire field weakening region is presented. In other words, influence of magnetic saturation on the MTPA and field weakening strategies as well as motor-inverter limitations is considered. Using the modified MTPA strategy leads to minimized copper losses, and so maximum capacity of motor is utilized. The proposed scheme incorporates all optimal strategies and operates the drive within the voltage and current limits of the motor/inverter. Simulation results for a prototype interior permanent magnet linear synchronous motor demonstrate that the MTPA and field weakening strategies can be successfully implemented for vector controlled saturated permanent magnet linear synchronous motor drives.

A field-programmable gate array (FPGA) based functional link radial basis function network (FLRBFN) control is proposed in this study to control the mover of a permanent magnet linear synchronous motor (PMLSM) servo drive system to track periodic reference trajectories. First, the dynamics of the field-oriented control PMLSM servo drive with a lumped uncertainty, which contains parameter variations, external disturbances and nonlinear friction force, is derived. Then, to achieve accurate trajectory tracking performance with robustness, an intelligent control approach using FLRBFN is proposed for the field-oriented control PMLSM servo drive system. The proposed FLRBFN is a radial basis function network (RBFN) embedded with a functional link neural network (FLNN). Moreover, the on-line learning algorithm of the FLRBFN, including the connective weights, the centers and the centers' width of the receptive field functions, are derived using back-propagation (BP) method. Furthermore, an FPGA chip is adopted to implement the developed control and on-line learning algorithms for possible low-cost and high-performance industrial applications using PMLSM. Finally, some experimental results are illustrated to show the validity of the proposed control approach.

Sensorless control of permanent magnet linear synchronous motors (PMLSMs) offers advantages such as reduced cost, stronger robustness, and easier maintenance. Existing back electromotive force (BEMF)-based sensorless control methods often assume three-phase circulant symmetry, which is not valid due to end effects, thereby losing accuracy. To reduce the reliance on symmetry assumptions and improve accuracy, this article first proposes a jumping curve jumping amplitude (JCJA) algorithm to reconstruct asymmetric three-phase flux linkages. Next, instead of using the traditional Clarke transform, this work proposes a new cotangent transform (CotT), which can precisely extract orthogonal and equal-amplitude <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\boldsymbol{\alpha}$</tex-math></inline-formula>-<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\boldsymbol{\beta}$</tex-math></inline-formula> components without requiring three-phase symmetry. Experimental results on a PMLSM show a significant improvement in position and speed estimation accuracy compared with existing methods assuming symmetry. Finally, sensorless trajectory-tracking control is realized with accuracy and robustness, providing new insights into the field of precision sensorless control of linear motors.

Considering internal parameters variation and external disturbance will greatly affect the performance of permanent magnet linear synchronous motor (PMLSM) control system, the active disturbance rejection control (ADRC) strategy is studied in this paper to enhance the anti-disturbance ability and dynamic performance of the system. Compared with the traditional PI control, ADRC has smaller overshoot, better adaptability, greater robustness to parameters variation and stronger anti-interference ability. The corresponding Simulink model is set up and its Simulink results are analyzed to prove the above advantages.

The high-performance feature of the Permanent Magnet Linear Synchronous Motor (PMLSM) makes it a reliable and valuable motor for use in the automotive industry, especially for electric vehicle (EVs) applications. This research proposes a bond graph approach in modeling the PMLSM as a multi-domain dynamical system. However, A time-based simulation was performed using 20-sim software to simulate the dynamical behavior of the motor. An equivalent model of the motor was first obtained and then modeled and simulated using 20-sim software. The model of the PMLSM drive system was modeled separately and incorporated with PMLSM Motor equivalent model to form a global model. Moreover, the motor drive system response was studied based on the sensor resolutions and the inverter switching frequency. The block diagram and the transfer function methods validated the bond graph model obtained. Two classical PIs such as continuous and discrete were implemented on the motor response to control the velocity of the motor.

In permanent magnet linear synchronous motor (PMLSM) direct drive system, disturbance caused by load disturbance and perturbation of motor parameters would reduce the control precision in a certain extent. In order to restrict the disturbance in PMLSM direct drive system, a vector control (VC) strategy based on the disturbance observer (DOB) was proposed in this investigation. By using the observed disturbance of DOB to adjust the given value of current in speed loop, the compensation for a sudden disturbance is optimized and a high precision of PMLSM drive system is obtained. For the sake of improving the dynamic performance under unknown disturbances, the traditional PID controller in speed loop was replaced by a single neuron PID controller, which leads to a good adaptive capability and a better starting performance in a given speed. Finally, a higher steady precision and robustness would be realized.

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> A recurrent functional link (FL)-based fuzzy neural network (FNN) controller is proposed in this study to control the mover of a permanent-magnet linear synchronous motor (PMLSM) servo drive to track periodic reference trajectories. First, the dynamic model of the PMLSM drive system is derived. Next, a recurrent FL-based FNN controller is proposed in this study to control the PMLSM. Moreover, the online learning algorithms of the connective weights, means, and standard deviations of the recurrent FL-based FNN are derived using the back-propagation (BP) method. However, divergence or degenerated responses will result from the inappropriate selection of large or small learning rates. Therefore, an improved particle swarm optimization (IPSO) is adopted to adapt the learning rates of the recurrent FL-based FNN online. Finally, the control performance of the proposed recurrent FL-based FNN controller with IPSO is verified by some simulated and experimental results. </para>

Aiming at the problems of poor stability and low system accuracy in the vector control of permanent magnet synchronous linear motor (PMSLM), a vector control strategy for PMSLM based on model reference adaptive system (MRAS) is proposed. Improve the accuracy of speed and mover position estimation by selecting the appropriate adaptive law, and introduce a sliding mode control (SMC) strategy in the speed loop to optimize the vector control system. The simulation results shows that this method can effectively reduce the overshoot and chattering, and improve the stability of the system.

Based on the analysis of power angle characteristics of the low-speed permanent magnet linear synchronous motor (PMLSM). An approach is made to determining motor steady operation sections and the basic methods to prevent out of step for the low-speed PMLSM. And the out-of-step prevention strategy is set up by using the fuzzy control theory, which adopts dual-mode control structure. With the variable voltage and frequency, the motor's out-of-step preventive control is realized. Experimental results show that the method is simple, high quality. It could effectively avoid out-of-step occurrence, and the research result is valuable to engineering application.

Linear motors have many advantages in compare with rotary motors due to directly creating linear motion without gears or belts,… It is hard to apply the online model predictive controller for electro-mechanism systems with input, output and state constraints because it requires the good ability in calculations of microcontroller and the nonlinear property of these models. In this paper, the exact linearization is applied to obtain the corresponding linear model and we propose the offline model predictive control law based on multi-parametric programming (MPP) for permanent magnet linear synchronous motors (PMLSM) to remarkably reduce amount of calculation. Thus, it is convenient for embedding the control law into microcontroller. Several simulation results show the ability of the proposed method in this paper.

The long primary permanent magnet linear synchronous motor (PMLSM) used for high speed maglev train needs to adopt flux weakening (FW) control to increase its speed due to the limitation of the traction inverter output voltage. In this paper, a subsection FW scheme in the linear modulation region based on the direct traction control (DTC) is proposed. By controlling the reference primary flux directly in different working states, the target speed is achieved faster compared with SFOC-FW control, and the demagnetization current is saved. The primary flux and traction fluctuation are reduced in the low speed working condition by leading the space vector modulation (SVM) algorithm into the proposed DTC-FW scheme. The simulation analysis is carried out and the effectiveness of the proposed scheme is verified.

Active disturbance rejection control (ADRC) has been widely used in the permanent magnet linear synchronous motor (PMLSM) drive system for its strong anti‐disturbance ability and fast response speed. However, a large number of adjustment parameters add difficulties to its application. To solve this problem, a novel sliding mode‐based active disturbance rejection control (SM‐ADRC) with autotuned parameters is proposed in this paper. In this control strategy, the SM is incorporated with extended state observer (ESO) and nonlinear state error feedback (NLSEF) to reduce the number of adjustment parameters. The simulated annealing particle swarm optimization (SAPSO) algorithm is carried out to achieve the optimal configuration and realize parameter self‐tuning. The experimental results of the PMLSM speed servo system demonstrate the proposed control method has better suppression of chattering effect, disturbance rejection ability and fast dynamic response. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC.

The paper presents a sensorless direct force (DFC) control method for a surface permanent magnet linear synchronous motor (SPMLSM) drive that provides high-performance speed control without position and speed sensors. To solve the practical problem of sensorless propulsion, DFC method is selected to control the thrust force and stator flux linkage of SPMLSM. The sensorless control scheme is based on the measurement and observation of stator current, voltage, and flux vector. Two phase voltages and two stator currents are measured and processed. The rotor speed is estimated through the stator flux and its derivative estimation. Flux and its derivative are calculated in the stationary reference frame and used to estimate the speed. In this paper a DFC based on fuzzy logic is adopted, so that the proper stator voltage vectors can be selected according to the force error. To test the control strategy a simulation using SIMULINK is built. The results show that the proposed control technique gives a good performance for tracking applications.

For any faults of automotive engines, the diagnosis can be performed based on variety of symptoms. Traditionally, the description of the faulty symptom is just existence or not. However, this description cannot lead to a high accuracy because the symptom sometimes appears in different degrees. Therefore, a knowledge representation method which could precisely reflect the degree of the symptom is necessary. In this paper, the fuzzy logic is firstly applied to quantify the degrees of symptoms. A probabilistic classification system is then constructed by using the fuzzified symptoms and a new technique namely Fuzzy Relevance Vector Machine (FRVM). Moreover, both Fuzzy Probabilistic Neural Network (FPNN) and Fuzzy Probabilistic Support Vector Machine (FPSVM) are used to respectively construct similar classification systems for comparison with FRVM. Experimental results show that FRVM produces higher diagnosis performance than FPNN and FPSVM.KeywordsFuzzy Probabilistic Neural NetworkFuzzy Probabilistic Support Vector MachineFuzzy Relevance Vector MachineEngine fault diagnosis