Abstract: The injection molding process has been well-known non-linear complex dynamics and
the approach extensively applied manual control and rely on experienced engineers. An intelligent
optimization controller has been designed with two series neural networks and the multi-losses
function has been proven can automatically adjust the machine setting overcome the complex
dynamics to upgrade part’s quality and reduce experienced engineers. The proposed method has
shown promising future for expediting the on-line process parameter tuning work to other
complicate non-linear system in the future.
884
Authors: Xuan Ling, Xu Dong Wang
Abstract: Radar leveling system is the key equipment for improving radar high mobility and survival capability. In this paper, an integrated hybrid controller, comprising of a sets of fuzzy cylinder controllers and one synchronous neural controller, is proposed for a antenna trailer of radar with four points supporting, which is driven by one hydraulic oil supply, to achieve synchronization motion under he various conditions of system uncertainties and disturbances. A new control strategy for better synchronization performance is proposed and a self-adaptive learning algorithm to adjust the weights of the neural network, is also derived. To verify the effectiveness and potential of the proposed controller, a series of experiments were made. Experimental results have demonstrated its high accuracy synchronization performance and fast dynamic response as well as stability of the leveling servo system. The hybrid controller can drive the radar truck leveling platform accurately, quickly and stably.
935
Authors: Xiao Li, Xiao Hua Yang
Abstract: Considering the nonlinear and time-variable characteristics of the injection cylinder system of hydraulic injection molding machine (HIMM), a self-learning control method based on fuzzy neural network is proposed in this paper. A self-learning controller was designed based on the combination of PID controller, fuzzy neural network controller, and learning mechanism. It was applied to the position control of injection cylinder. The experimental results show that the controller has the property of higher position tracking accuracy under the high speed and variable track movement of injection cylinder, compared with PID controller. The conclusion of this research can provide the beneficial reference for designing high speed and high precision HIMM.
1117
Authors: D.M. Elias, Y. Yusof, Mohammad Minhat
Abstract: The Function Block Development Kit (FBDK) platform proposed for STEP-NC controller architecture . It can be used to create an open CNC architecture on ISO 14649 and Function Block (IEC 61499). This offers interoperability, portability, and adaptability. The proposed framework consists of three main modules: (1) STEP-NC Interface, (2) Simulator, and (3) Communication. The system will be implemented in CNC controller on actual milling machine. It will simplify the design of CNC machine controller with the architecture layers responsible for data processing, data storage, execution and feedback. This paper also discusses towards the requirements for global interoperable manufacturing for real-life machining system with the architecture layers responsible for data storage, data processing, and execution.
2365
Authors: Gong She Shi, Lei Huang, Wei Hu
Abstract: The brushless DC motor (BLDCM) non-linear and the complexity of the working conditions are likely to cause the conventional PID servo control performance is not satisfactory. In order to improve the performance of the BLDCM servo control system and PID parameter tuning efficiency, this paper designs an adaptive fuzzy PID controller. Fuzzy logic PID controller parameters Kp, Ki, Kd are adjusted online real time to achieve the effect of optimal control, the BLDCM speed is as to the control object, and in the Matlab of Simulink toolbox simulation is used to achieve speed closed loop of BLDCM. According to comparative analysis of the conventional PID and adaptive fuzzy PID of Dynamic response curve, adaptive Fuzzy PID quick start for brushless DC motors, anti-disturbance has better control effect.
838