Xiaoqi Tang

Stable adaptive PI control for permanent magnet synchronous motor drive based on improved JITL technique

In this paper, a stable adaptive PI control strategy based on the improved just-in-time learning (IJITL) technique is proposed for permanent magnet synchronous motor (PMSM) drive. Firstly, the traditional JITL technique is improved. The new IJITL technique has less computational burden and is more suitable for online identification of the PMSM drive system which is highly real-time compared to traditional JITL. In this way, the PMSM drive system is identified by IJITL technique, which provides information to an adaptive PI controller. Secondly, the adaptive PI controller is designed in discrete time domain which is composed of a PI controller and a supervisory controller. The PI controller is capable of automatically online tuning the control gains based on the gradient descent method and the supervisory controller is developed to eliminate the effect of the approximation error introduced by the PI controller upon the system stability in the Lyapunov sense. Finally, experimental results on the PMSM drive system show accurate identification and favorable tracking performance.

Adaptive two-degree-of-freedom PI for speed control of permanent magnet synchronous motor based on fractional order GPC.

In this paper, an adaptive two-degree-of-freedom (2Dof) proportional-integral (PI) controller is proposed for the speed control of permanent magnet synchronous motor (PMSM). Firstly, an enhanced just-in-time learning technique consisting of two novel searching engines is presented to identify the model of the speed control system in a real-time manner. Secondly, a general formula is given to predict the future speed reference which is unavailable at the interval of two bus-communication cycles. Thirdly, the fractional order generalized predictive control (FOGPC) is introduced to improve the control performance of the servo drive system. Based on the identified model parameters and predicted speed reference, the optimal control law of FOGPC is derived. Finally, the designed 2Dof PI controller is auto-tuned by matching with the optimal control law. Simulations and real-time experimental results on the servo drive system of PMSM are provided to illustrate the effectiveness of the proposed strategy.

Adaptive PIF Control for Permanent Magnet Synchronous Motors Based on GPC

To enhance the control performance of permanent magnet synchronous motors (PMSMs), a generalized predictive control (GPC)-based proportional integral feedforward (PIF) controller is proposed for the speed control system. In this new approach, firstly, based on the online identification of controlled model parameters, a simplified GPC law supplies the PIF controller with suitable control parameters according to the uncertainties in the operating conditions. Secondly, the speed reference curve for PMSMs is usually required to be continuous and continuously differentiable according to the general servo system design requirements, so the adaptation of the speed reference is discussed in details in this paper. Hence, the performance of the speed control system using a GPC-based PIF controller is improved for tracking some specified signals. The main motivation of this paper is the extension of GPC law to replace the traditional PI or PIF controllers in industrial applications. The efficacy and usefulness of the proposed controller are verified through experimental results.

Implementation of CL points preprocessing methodology with NURBS curve fitting technique for high-speed machining

CL points can be grouped into three problematic cases: short distance, unequal adjacent distance and sharp points.Using the bisection method, a NURBS curve is efficiently fitted using CL points.The simulation performed on CL points for a wave surface demonstrates that the proposed approach reduces the NC blocks to about 3.3%, within the tolerance of deviation.The machining experiment shows that the proposed approach can increase the machining quality while reducing the machining time by about 19%. Due to the limitations of linear/circular interpolation, parametric interpolation typically represented by NURBS curves, has played a key role in the computer control of machine tools. To achieve the highest quality parts, generated trajectories must describe the desired toolpath accurately. However, a NURBS toolpath cannot be generated directly from a CAD model, due to certain limitations. CL points preprocessing methodology with NURBS curve fitting technique for high-speed machining is proposed in this paper. CL points are initially preprocessed, and unsuitable CL points can be grouped into three problematic cases: short distance, unequal adjacent distance and sharp points. Each case is identified and corrected, which can improve the fitted NURBS curve quality. CL points corner and distance rules are then used to extract CL points for each segmentation. Using the bisection method, a NURBS curve is efficiently fitted using CL points. If there is no advantage to using a curve in this case, CL points will be output directly. The simulation performed on CL points for a wave surface demonstrates that the proposed approach reduces the NC blocks to about 3.3%, while staying within the tolerance of deviation. The machining experiment shows that the proposed approach can increase the machining quality while reducing the machining time by about 19%.

Fractional order controllers tuning strategy for permanent magnet synchronous motor servo drive system based on genetic algorithm–wavelet neural network hybrid method

In this paper, a novel tuning strategy for the fractional order proportional integral and fractional order [proportional integral] controllers is proposed for the permanent magnet synchronous motor servo drive system. The tuning strategy is based on a genetic algorithm–wavelet neural network hybrid method. Firstly, the initial values of the control parameters of the fractional order controllers are selected according to a new global tuning rule, which is based on the genetic algorithm and considers both the time- and frequency-domain specifications. Secondly, the wavelet neural network is utilized to update the control parameters based on the selected initial values in an online manner which improves the capability of handling parameter variations and time-varying operating conditions. Furthermore, to improve the computational efficiency, a recursive least squares algorithm, which provides information to the wavelet neural network, is used to identify the permanent magnet synchronous motor drive system. Finally, experimental results on the permanent magnet synchronous motor drive system show both of the two proposed fractional order controllers work efficiently, with improved performance comparing with their traditional counterpart.

Graphical tuning method for non-linear fractional-order PID-type controllers free of analytical model

The main focus of this paper is on a graphical tuning method of non-linear fractional-order PID (FOPID)-type controllers, i.e. a class of FOPID-type controllers that non-linearly depend on the control parameters, e.g. FO[PI], FO[PD] etc. Firstly, a method is proposed to determine the stabilizing region of non-linear FOPID-type controllers, namely the complete sets of FOPID-type controllers providing stability of the control system. Secondly, two different approaches are proposed to determine the H∞ region of these FOPID-type controllers, namely the complete sets achieving H∞ robust performance specifications. The first approach maps the H∞ constraints into the parameter space by solving a series of non-linear equations. The second approach transforms the original H∞ region problem into simultaneous stabilization of a family of characteristic polynomials. It turns out that these two approaches are both very flexible, and the second approach is more efficient than the former. The main advantage of our proposed graphical tuning method is that the exact mathematical model of the controlled plant is not needed. The stabilizing and H∞ regions can be computed only from the frequency response data of the plant. Finally, numerical and experimental results are presented to demonstrate the proposed graphical tuning method.

Predictive IP controller for robust position control of linear servo system.

Position control is a typical application of linear servo system. In this paper, to reduce the system overshoot, an integral plus proportional (IP) controller is used in the position control implementation. To further improve the control performance, a gain-tuning IP controller based on a generalized predictive control (GPC) law is proposed. Firstly, to represent the dynamics of the position loop, a second-order linear model is used and its model parameters are estimated on-line by using a recursive least squares method. Secondly, based on the GPC law, an optimal control sequence is obtained by using receding horizon, then directly supplies the IP controller with the corresponding control parameters in the real operations. Finally, simulation and experimental results are presented to show the efficiency of proposed scheme.

Reliability Analysis and Accelerated Statistical Model of CNC PCB for Electrochemical Migration

The effect of conductor spacing and applied voltage on the insulation performance of printed circuit board (PCB) has been investigated under temperature-humidity-bias (THB) conditions in this paper. Based on the electrochemical migration failure mechanism, this paper proposes an accelerated statistical model, which includes a life-stress relationship model and a life distribution model. The accelerated statistical model can effectively quantify the effect of concurrent conductor spacing and applied voltage on the characteristic life of the PCB. Because THB test can produce less life data in limited test period, an accelerated degradation theory is applied for the modeling research. In addition, a methodology on how to verify the accelerated statistical model for the CNC PCB is introduced. In the case of small samples, the least square regression method is applied for parameter calculation due to its high precision. Furthermore, the residual analysis is carried out to check the fitting effect. The results show that the accelerated statistical model can describe the life characteristics of CNC PCBs preferably under the comprehensive action of conductor spacing stress and applied voltage stress.

Fractional order modeling and high-frequency repetitive motion control for flexible swing arm system

Flexible swing arm system (FSAS) with flexibility and nonlinearity is the key component of die bonder in LED packaging industry. This paper explores the application of fractional calculus on modeling and high-frequency repetitive motion control for FSAS. Considering of the complex characteristics of high-frequency repetitive motion on different stages, fractional order models are designed to describe FSAS more accurately. The proposed fractional order models are identified using particle swarm optimization (PSO) algorithm. Based on the comprehensive control performance evaluation, proportional integral (PI) controllers are optimized for FSAS according to the identified fractional order models. Simulation and experiments will be conducted to demonstrate the existence and practical viability of the proposed fractional order models for the FSAS.

Humidity Threshold Model of Numerical Control Multilayer Printed Circuit Board for Electrochemical Migration

In this paper, numerical control finished boards are employed as test samples to evaluate the effects of the environment and bias voltage on the surface insulation resistance of printed circuit boards. The results indicate that the surface insulation resistance degradation processes leading to failure exhibit a relative humidity threshold value, which is mostly related to temperature, test patterns, and bias voltage. Then, a humidity threshold model is built on the basis of electrochemical migration to quantify the relationship between the relative humidity threshold and conductor spacing and bias voltage under the controlled environments. Finally, experiments and tests have been carried out to validate the effectiveness and practicability of the model. Furthermore, the residual analysis is applied to check the fitting effect of the proposed model. According to the new model, the humidity threshold value can be obtained to provide the guidance for reliability problems in application. It is crucial to the design and deployment of reliable printed circuit boards. Copyright