Research and Design of a Fuzzy Neural Expert System

We have developed a fuzzy neural expert system that has the precision and learning ability of a neural network.Knowledge is acquired from domain experts as fuzzy rules and membership functions.Then,they are converted into a neural network which implements fuzzy inference without rule matching.The neural network is applied to problem-solving and learns from the data obtained during operation to enhance the accuracy.The learning ability of the neural network makes it easy to modify the membership functions defined by domain experts.Also,by modifying the weights of neural networks adaptively,the problem of belief propagation in conventional expert systems can be solved easily.Converting the neural network back into fuzzy rules and membership functions helps explain the inner representation and operation of the neural network.     

Integrating artificial neural networks with rule-based expert systems

The Rule-Based (RB) and the Artificial Neural Network (ANN) approaches to expert systems development have each demonstrated some specific advantages and disadvantages. These two approaches can be integrated to exploit the advantages and minimize the disadvantages of each method used alone. An RB/ANN integrated approach is proposed to facilitate the development of an expert system which provides a “high-performance” knowledge-based network, an explanation facility, and an input/output facility. In this case study an expert system designed to assist managers in forecasting the performance of stock prices is developed to demonstrate the advantages of this integrated approach and how it can enhance support for managerial decision making.     

结晶器液位专家系统控制研究

针对传统的结晶器液位控制系统方案存在缺点,根据对象机理,比较了塞棒比例积分（PI）控制和拉速PI控制和对结晶器液位控制效果,分析了塞棒PI控制液位存在的问题,设计了基于径向基函数神经网络（RBFNN）的塞棒控制器和拉速补偿器。使用规则模型,研究了基于专家系统的塞棒和拉速控制液位的协调控制系统,解决了液位控制器切换存在的干扰。系统在MATLAB上进行了仿真,并在实际中进行了测试。实际测试结果表明该系统的液位控制精度、稳定性和动态响应优于传统的单回路控制方案。     

"MDZY型，冲压模方案设计专家系统的设计与实现

作为一种先进的金属零件加工方法,冲压生产在工业领域中已经得到了广泛的应用,但其生产效率的进一步提高却由于冲压模具设计的复杂性和耗时性而受到了较大程度制。CAD技术的引入虽然简化了图形的绘制和处理工作,但对冲模设计当中普遍存在的推理、判断等非数值问题却无能为力。文中利用人工智能技术,在Visual C 5．0环境下实现了一个冲压模方案设计专家系统,用来帮助设计者在模具设计前期完成合理的方案设计。系统在执行设计任务时采用了实例匹配和归纳推理两种不同手段,系统中的设计知识分别采用了数据库记录、链表以及集成了框架、规则和方法的对象等多种形式来表示,各类知识的学习和修改采用了包括神经网络在内的多种机制。     

A knowledge sorting process for a product design expert system

Knowledge sorting is one way to organize the knowledge acquired from the domain expert(s) and various sources. Well-organized knowledge representation will make rule extraction much easier. This paper describes a knowledge sorting process that was developed to facilitate the rule extraction for a product design expert system. The process capitalizes on the relationships between design attributes and factors, dependent and independent variables, and consists of three stages: identification of knowledge sources, generation of taxonomic trees, and organization of acquired knowledge. An example applies the sorting process to the development of an expert system for the design of wood head golf clubs.     

A reasoning method for a ship design expert system

Abstract: The ship design process is a highly data‐oriented, dynamic, iterative and multi‐stage algorithm. It utilizes multiple abstraction levels and concurrent engineering techniques. Specialized techniques for knowledge acquisition, knowledge representation and reasoning must be developed to solve these problems for a ship design expert system. Consequently, very few attempts have been made to model the ship design process using an expert system approach. The current work investigates a knowledge representation–reasoning technique for such a purpose. A knowledge‐based conceptual design was developed by utilizing a prototype approach and hierarchical decompositioning. An expert system program called ALDES (accommodation layout design expert system) was developed by using the CLIPS expert system shell and an object‐oriented user interface. The reasoning and knowledge representation methods of ALDES are explained in the paper. An application of the method is given for the general arrangement design of a containership.     

An expert system for design for robotic assembly

In recent years, due to the various advantages associated with automation and robotics, much work has been done in developing robotic systems for assembly operations. Since part design plays a major role in assembly, this paper deals with the design of parts for ease of robotic assembly. Considerable knowledge is available in the form of design for robotic assembly rules. In addition, a large amount of data is required for decisions regarding suitability of parts for robotic assembly. The implementation of design for robotic assembly rules would be much easier with the help of an expert system, which would guide the designer toward choosing the design alternative that can best facilitate ease of assembly from a robotic point of view.To this end, a prototype expert system for design for robotic assembly is developed and presented in this paper. The expert system was implemented as a production system, which consists of rules and Object-Attribute-Value (O-A-V) triplets to represent domain knowledge. In order to best utilize the domain specific knowledge, a state space search-based inference mechanism was employed. The implementation of the prototype system is illustrated with examples.     

基于混合专家网络的智能检测系统

为解决原油含水率测量中存在的问题,提出一种基于多传感器技术的智能检测系统,对原油的水分、温度、水矿化度进行检测,并采用混合专家网络对多传感器信息进行智能融合处理,以提高含水率的检测精度。在提出的混合专家网络中,采用了专家决策、多神经网络集成等先进技术,且专家子网络采用LM算法和最优停止法相结合以确定网络最佳权值,提高模型的泛化能力。仿真结果表明:该混合专家网络模型在原油含水率测量系统中取得很好的应用效果,获得较高的测量精度。     

Development of a hybrid expert system for nurse shift scheduling

In this article, we discuss the effect of nurse shift job on circadian rhythm, work stress, and some important ergonomics criteria. We also review and compare different nurse shift scheduling methodologies via the criteria of flexibility, consideration of nurse preference, and consideration of ergonomics principles. A hybrid expert system, entitled NURSE-HELP, is developed to facilitate the nurse scheduling process with an emphasis on considering ergonomics criteria. Moreover, the combination of a linear zero-one goal programming and an expert system program reduces the program run time while maintaining the quality of the schedule. The evaluation of the system is done by comparing 18 sets of four-week schedules generated by the head nurses manually and by NURSE-HELP. Concerning the amount of time to generate the schedules, NURSE-HELP averages less than 20 minutes while the head nurses spend about two to four hours. The quality of the schedules is measured by the following four criteria; minimum staff level not satisfied, day off request not granted, backward rotation, and maximum consecutive work periods on the night shift. The results show that NURSE-HELP is superior than the head nurses in preparing schedules, both in terms of time and quality.     

搅拌设备化工预设计专家系统的知识处理

详细论述了搅拌设备化工预设计专家系统的知识处理方法。系统采用黑板结构将多种知识源组合在一起来实现问题的求解,即采用基于规则的推理、基于模糊逻辑的模式匹配和传统的数据库查询等技术,来分别处理搅拌设备设计过程中的人类专家经验、设计公式和机械部件标准等不同种类的知识。此专家系统的结构和知识处理的方法具有较强的灵活性,非常适用于复杂的工业过程设计领域。     

基于关系数据库的化工过程模拟辅助设计专家系统的研究

根据专家系统基本原理,充分发挥关系数据库的强大数据处理能力,使用VISUAL BASIC6．0和MICROSOFT ACCESS 2003开发了一个用于化工过程模拟的专家系统。将知识和案例用数据库中表的形式表示出来,采用数据库语言进行推理,指导用户通过以往的相似案例来进行新工艺过程的模拟,从而使全流程模拟得到了简化。同时简洁友好的界面也便于领域专家操作。系统实现了与化工模拟软件ASPEN PLUS之间的接口,使二者可以进行数据传输。     

An embedded fuzzy expert system for adaptive WFQ scheduling of IEEE 802.16 networks

WiMAX is a futuristic technology which provides simultaneous support for web, video, and voice applications. WiMAX networks are best suitable to real time traffic however the quantity of non real time and best effort traffic cannot be neglected. Distribution of resources in such heterogeneous applications is therefore a challenging task. There are many schedulers available for WiMAX but adaptive and adequate schedulers are still in growing stage of development. This paper introduces a novel method using which a system is developed based on concepts of fuzzy logic to schedule traffic in WiMAX networks. The proposed fuzzy expert system simplifies fair allocation of resources to real as well as non real time traffic. The implementation is based on changing the weights of the queues serving real and non- real time traffic adaptively. New weights will be calculated for each bandwidth request made to base station and these weights will in turn decide amount of bandwidth allocated to different traffic classes. The weights are calculated based on three parameters that are amount of real time and non real time traffic in queues, change in throughput requirement for non real time flows and latency requirement of real time input data. Results obtained by virtue of simulations justify the significance of the proposed method.     

An investigation of expert systems for die design

This article introduces fuzzy set theory to process the design details of the uncertain portion in die design, and assist the designer to transform those design items with fuzziness into those with definite and reasonable design attributes. For the design parameters of die block thickness, die clearance angle and die sets choice in die design, which possess intermediate features, fuzzy cluster analysis is used to obtain the design attributes. As for single-sided die clearance, stripper pressure and guide bushing-type die design, whose theory or empirical formulas possess uncertainty coefficients or preference design parameters, the fuzzy weighted average method is adopted to obtain the feature parameters that conform with the die design requirement. This study established an expert system prototype to combine the aforementioned uncertainty problems into two kinds of die design, and help the designer obtain a definite design strategy when faced with uncertain design items.     

Expert system for selecting expert system shells

A prototype expert system is described that advises end-users in the selection of a suitable development environment for small- to medium-scale expert system applications. The system, running on a microcomputer, assesses the suitability of an application as a whole for expert systems techniques and recommends the five ‘best’ products from its knowledge base of 42 products. The paper takes a case-study approach, using the system itself as a recursive discussion example, applying some of the concepts involved in building an expert system. It describes the operation of the system and suggests improvements; it includes a feature list for classifying expert system shells.     

An expert system for the selection of software design patterns

Abstract: This paper describes the development of a prototype expert system for the selection of design patterns that are used in object-oriented software. Design patterns provide one method of software reuse, which supports the goal of improved software development productivity. The prototype system represents an initial step towards providing an automated solution regarding the design pattern application problem, i.e. leading a designer to a suitable design pattern which is applicable to the problem at hand. The feasibility of using expert system technology to aid in the selection problem is demonstrated.     

Analysis of assembly operations difficulty: a fuzzy expert system approach

Typical mechanical products can be assembled in various sequences of assembly operations. These sequences have high impact on the assembly time, machine utilization and even on the product quality. In order to select the best sequence of assembly operations, proper evaluation of the various sequences of assembly operations is required. This, in turn, requires the consistent evaluation of each assembly operation in the sequence. The assembly operations can be evaluated for various criteria, of which the operation difficulty is the most meaningful. This paper describes a methodology to analyse the assembly operations and calculates an operation's degree of difficulty using an expert system. This analysis consists of two steps: the first one identifies the main parameters that affect the assembly difficulty and assigns fuzzy triangular values to these parameters. The second step assigns weights to the parameters in order to maximize the agreement with a domain expert. The expert system analyses the difficulty of the assembly operation performed in two orientations: horizontal and vertical. The expert system then assigns a triangular fuzzy number as the aggregate measure of the operation's difficulty.     

An expert system for the dyeing recipes determination

This article deals with the modelization of the reasoning of an expert in dyeing. The dyeing operation is one of the latest stage in the whole textile process which alters the intrinsic properties of the articles. At present, there are expert systems for dyeing industries. Most of them have been developed by dyes and chemical auxiliary suppliers, and these tools are not widely used in dyeing factories. This paper describes the dyeing operation, and the rules of the expert in dyeing. The objects that he reasons about, his reasoning, and the critical points that he considers are also presented.     

基于神经网络专家系统的钻井事故诊断

结合石油钻井工程的实际情况,依据钻井过程的监测参数,设计了利用神经网络进行知识获取、专家系统进行事故诊断的钻井工程事故智能诊断系统。通过神经网络对钻井复杂问题实例的不断学习训练,获得用于智能诊断的知识,完成对事故发生可能性的初步诊断。经过专家系统的进一步启发式反向推理验证事故是否存在,给出最后确诊,以此监控钻井参数,指导钻井参数调整的实施。应用实例结果表明,该智能诊断系统应用于钻井事故诊断是有效的,对减少钻井事故的发生与发展具有重大的实际应用价值。     

一种改进的神经网络机械故障诊断专家系统

针对传统BP神经网络训练中收敛速度较慢的缺点,提出一种基于L-M算法的神经网络应用于机械设备故障诊断的专家系统。论述了神经网络的专家系统结构,并以7216圆锥轴承试验研究为例,建立了基于该算法的故障诊断模型。仿真结果表明:该模型显著缩短了训练时间,具有较高的准确性。运用该神经网络专家系统进行机械故障诊断是有效的。