Cs Egresits

Design and real-time reconfiguration of robust manufacturing systems by using design of experiments and artificial neural networks

Abstract A methodology for design and real-time reconfiguration of robust manufacturing systems is described which combines design of experimental technology, Taguchi method and knowledge-based simulation techniques. Artificial neural networks are proposed for mapping between design factors and system performance. The applicability of the approach is analysed through experiments for the estimation of the throughput time, and the determination of Automated Guided Vehicle (AGV) speed in a given system. In contrast to the simulation-based approach, the solution using artificial neural networks can also be used in real-time circumstances.

On hybrid learning and its application in intelligent manufacturing

Abstract For most real-world problems, the information concerning design, evaluation, realisation, control, monitoring, etc., can be classified into two groups, e.g. numerical information usually obtained by sensor measurements, and linguistic information obtained from human experts. Trainable systems must also rely on these kinds of information (sampled input-outputs pairs, and human experience). Artificial neural networks (ANNs) and symbolic (expert) systems can be mentioned as characteristic techniques. The paper demonstrates that neuro-fuzzy solutions can combine the above information sources, i.e. they have hybrid learning abilities. Combined use of the neural and fuzzy techniques in cutting tool monitoring is illustrated. The results are compared with ANN and previous neuro-fuzzy (NF) approaches. The paper shows that the NF technique can comply with the above; fundamental requirements of intelligent manufacturing, i.e. real-time nature, uncertainty handling and learning abilities, with the additional benefits of managing both symbolic and numeric information, hybrid learning, and a kind of explanation facility. Finally, the integration of such a hybrid system in an intelligent manufacturing environment is investigated.

Soft Computing and Hybrid AI Approaches to Intelligent Manufacturing

The application of pattern recognition (PR) techniques, artificial neural networks (ANNs), and nowadays hybrid artificial intelligence (Al) techniques in manufacturing can be regarded as consecutive elements of a process started two decades ago. The fundamental aim of the paper is to outline the importance of soft computing and hybrid AI techniques in manufacturing by introducing a genetic algorithm (GA) based dynamic job shop scheduler and the integrated use of neural, fuzzy and GA techniques for modeling, control and monitoring purposes.

Multistrategy learning approaches to generate and tune fuzzy control structures and their application in manufacturing

Intelligence is strongly connected with learning adapting abilities, therefore such capabilities are considered as indispensable features of intelligent manufacturing systems (IMSs). A number of approaches have been described to apply different machine learning (ML) techniques for manufacturing problems, starting with rule induction in symbolic domains and pattern recognition techniques in numerical, subsymbolic domains. In recent years, artificial neural network (ANN) based learning is the dominant ML technique in manufacturing. However, mainly because of the ‘black box’ nature of ANNs, these solutions have limited industrial acceptance. In the paper, the integration of neural and fuzzy techniques is treated and former solutions are analysed. A genetic algorithm (GA) based approach is introduced to overcome problems that are experienced during manufacturing applications with other algorithms.

Modelling and Monitoring of Milling through Neuro-Fuzzy Techniques

Abstract Real-time nature, uncertainty handling and learning ability are essential requirements for knowledge representation and processing techniques to be applied at lower levels of intelligent manufacturing systems. The paper demonstrates and compares the applicability of neural networks and neuro fuzzy techniques for monitoring of milling tools. Learning and classification performances of back propagation (BP) networks and the neuro-fuzzy approach using different learning techniques are compared. The possible role of such a hybrid solution in an intelligent manufacturing environment is investigated.

Approaches to coupling connectionist and expert systems in intelligent manufacturing

Abstract Artificial neural networks are successfully applied in different fields of manufacturing, mostly where multisensor integration, robustness, real-timeness, and learning abilities are needed. Since the higher levels of the control and the monitoring hierarchy require symbolic knowledge representation and processing techniques, the integrated use of the symbolic and subsymbolic approaches is straightforward. The paper describes two hybrid artificial intelligence systems for control and monitoring of manufacturing processes on different hardware and software bases. The first experiences gained by their usage are outlined. Finally, further possible applications of these hybrid solutions in an intelligent manufacturing environment are enumerated.

Knowledge-Based Hybrid Techniques Combined with Simulation: Application to Robust Manufacturing Systems

Publisher Summary Knowledge-based hybrid systems (KBHSs) play an increasing role in industrial applications. New techniques of artificial intelligence (AI) such artificial neural networks (ANNs), genetic algorithms (GEs), machine learning (ML), or the fuzzy set theory have created new ways of solving problems. Most of these new techniques, however, are not appropriate for handling complex tasks alone; therefore, their integration with or connection to other systems is obligatory. The different types of KBHSs are the result of this development trend, as in their case the goal is to solve complex problems of a well-defined limited area. This chapter defines and classifies KBHSs, including the integration possibilities with simulation. Several examples of industrial applications of various types of KBHSs are described. KBHSs will have an increasing role in industry in future, as they are a significant source of support for users in a variety of applications. The representation, design, and control of dynamic, stochastic, and complex systems (like manufacturing systems) can be intensively supported by KBHSs.

Virtual manufacturing using a hieararchically coupled hybrid AI system

Virtual Manufacturing (VM) techniques are used in the paper for development and evaluation of submodules for Intelligent Manufacturing Systems (IMSs). A hierarchically coupled hybrid AI solution for control and monitoring of manufacturing processes is introduced. The combined use of the VM and IMS concepts is illustrated through this hybrid control and monitoring system.

Genetic algorithms for automatic design of fuzzy decision systems for intelligent manufacturing

Abstract Artificial neural networks (ANNs) have been successfully applied in different fields of manufacturing. Monitoring and modelling of manufacturing processes obviously belong to the most promising areas, where real-time nature, uncertainty handling and learning abilities are essential. However, mainly because of the “black box” nature of ANNs, these solutions have limited industrial acceptance. In the paper, a combined use of the neural and fuzzy techniques in cutting tool monitoring is illustrated. We introduce a genetic algorithm based approach to overcome problems, including rule selection and redundancy checking. Finally the results are compared with ANN and previous neurofuzzy (NF) approaches.

Combined structure and parameter learning for intelligent manufacturing applications

Integrated use of the neural and fuzzy techniques in cutting tool monitoring is illustrated in the paper. The results are compared with ANN and previous neuro-fuzzy (NF) approaches. The paper shows that the NF technique can comply with the fundamental requirements of intelligent manufacturing, i.e. real-time nature, uncertainty handling and learning abilities, with the additional benefits of managing both symbolic and numeric information, hybrid learning, and a kind of explanation facility. Finally, the integration of such a hybrid system in an intelligent manufacturing environment is investigated.