The Research Of Intelligent Control And Optimization About Stacker

With the development of industrial automation, Automated Storage and Retrieval System (AS/RS) is used more and more widely. As the core equipment of AS/RS, stacker plays an important role in AS/RS system about safety, reliability and efficiency. The requirement of stacker is from to loading or unloading goods simply turn to its speed and positioning accurate. Therefore, using of advanced control technology is an inevitable trend, especially in improve and enhance stackers operation speed, precision and be able to adapt to the different work environment. In recent years, the development of intelligent control technology, such as fuzzy control and neural network, made a lot of progress in the field of engineering application. So it's better for stacker to achieve precise control of its performance when intelligent control technology is applied for its movement control. This paper mainly content and results are as follows:1. Stacker movement control use FCMAC network which has rapidly and self-learning ability. It is to expound the structure and work principle of fuzzy control and Cerebellar Model Articulation Controller (CMAC), and according to the performance characteristics of both, the fuzzy control technology is introduced into the CMAC which make up a FCMAC network. It introduced in detail the structure, the working mechanism and the learning algorithm of the newly FCMAC network. The FCMAC controller is analyzed on its performance by Matlab7.0, and the simulation results show that FCMAC controller can complete the control requirements of stacker.2.According to analysis the characteristics of the stacker movement, the stackers intelligent control system based on FCMAC network is designed. The design of the structure of FCMAC controller was discussed in detail, and to explain the process of fuzzy, the selection of system parameter and the selection principle and step of fuzzy membership functions.3.According to the shortage of the FCMAC controller, this paper propose to use the improved particle swarm algorithm to adapt the quantification factors, scaling factor and study factors of the control system. The improvement of PSO which is introduced the crossover and mutation of Genetic Algorithm in order to solute the shortage of easy to fall into the local optimum, and tested its performance with test function.4.In Matlab7.0environment, adjustment results of parameters and the curve of speed response in no-load and load conditions has got. The results show that the improved PSO can better optimized parameters and adaptive to the adjustment, improve the performance of the control system.