Knowledge Acquisition Advisor (KA): An interactive knowledge-acquisition tool for expert system development

A new interactive knowledge acquisition tool, called Knowledge Acquisition Advisor (KA²), is presented in this paper. The new tool will help knowledge engineers to conduct effective knowledge-elicitation interviews with domain experts through structured knowledge acquisition for both analytic and synthetic problems. A graphic modeling data structure, called Knowledge Graph is proposed, which allows knowledge engineers to model domain problems with their images and understanding. By using Knowledge Graph, knowledge engineers are able to decompose a domain problem into several components, to model the feature of each component, and to explore their relations by linking them with sets of questions. These questions can later be employed to guide the KA interview. Moreover, KA² is particularly useful for interview through computer networks, so the knowledge acquisition can take place remotely.     

多变量PID控制器的在线自整定

本文研究具有ＰＩＤ结构反馈控制器的多变量控制系统的参数自整定，提出了具有校正因子的多变换ＰＩＤ控制算法及在线自动调整校正因子的专家自整定方法，将该方法用于火电单元机组负荷控制系统自整定仿真研究，结果表明系统具有好的完整性和鲁棒性。     

SISO approaches for linear programming based methods for tuning decentralized PID controllers

Two tuning techniques are proposed to design decentralized PID controllers for weakly coupled and general MIMO systems, respectively. Each SISO loop is designed separately, and the controller parameters are obtained as a solution of a linear programming optimization problem with constraints on the process stability margins. Despite the SISO approach, loop interactions are accounted for either by Gershgorin bands (non-iterative method) or an equivalent open-loop process (iterative method). The tuning results and performance from both methods are illustrated in four simulations of linear processes, and a laboratory-scale application in a Peltier process. Four applications contemplate closed-loop performance comparisons between the proposed techniques and techniques from the literature. One application illustrates the feasibility of the proposed iterative method, based on EOPs, in tuning decentralized PIDs for a 5 × 5 system. Moreover, an analysis of the effect of model uncertainty in the phase and gain margins of the closed-loop process is performed.     

Expert knowledge acquisition and the unwilling expert: a knowledge engineering perspective

Expert systems are an evolving technology with the potential to make human expertise widely and cheaply available. The literature describing the development of expert systems generally assumes that experts willingly give up their knowledge. This is unrealistic and may be a reason why most expert system projects fail. This paper explores the problem of unwilling experts from the perspective of a knowledge engineer building an expert system. The link between knowledge and organizational power is established and human motivation theories are discussed. Finally, a new motivational approach is introduced to help the knowledge engineer deal with unwilling experts.     

A Design Of Multiloop Predictive Self‐Tuning Pid Controllers

In this paper, a new design scheme of multiloop predictive self‐tuning PID controllers is proposed for multivariable systems. The proposed scheme firstly uses a static pre‐compensator as an approximately decoupling device, in order to roughly reduced the interaction terms of the controlled object. The static matrix pre‐compensator is adjusted by an on‐line estimator. Furthermore, by regarding the approximately decoupled system as a series of single‐input single‐output subsystems, a single‐input single‐output PID controller is designed for each subsystem. The PID parameters are calculated on‐line based on the relationship between the PID control and the generalized predictive control laws. The proposed scheme is numerically evaluated on a simulation example.     

基于迭代继电反馈的多变量PID控制器自整定

PID控制是工业过程中最常用的控制方法,但在实际生产过程中,被控过程往往是多变量、有耦合的,常规PID控制器参数往往整定不良、性能欠佳,对运行工况的适应性较差。为此,将迭代反馈理论和继电整定方法有机结合起来,提出一种适用于存在耦合的多变量系统PID控制器的参数整定方法。运用该方法整定PID参数,不需要被控对象的数学模型,而且具有速度快、效果好等优点。     

An online expert system for fault diagnosis in hydraulic systems

Fault diagnosis has become increasingly important for industrial automation and many approaches have been investigated for the online diagnostic task. This work demonstrates a new online expert system for dynamic industrial automated processes. The emphasis of this diagnostic system lies in the functions provided for fault detection, prediction, compensation and diagnosis. The system uses experiential knowledge cooperatively with scientific knowledge in a new interactive formation. It is developed in two parts using the DASYLab software for the numerical computations and the KPWin development tool for the symbolic representations. The system was developed in collaboration with the company Automation Systems S.A., which specializes in hydraulic systems, and is used for fault detection in production machines.     

The design process of expert systems development: some concerns

Abstract: A discussion is presented of why some expert systems that organizations have developed are not successful. The concept of design process plays a very significant role at the outset of the expert system development process. This concept has not been the subject of much debate and attention in expert systems development. From the author's point of view, one of the main issues is how the designer (knowledge engineer) thinks about the design process. In general, the designer's process is influenced by the knowledge engineer's conception. This paper endeavors to disclose some of the main factors related to the knowledge engineer's conception of the design process and an attempt is made to put forward a conceptual model of the expert system design process. This conceptual model is an initial step towards a successful implementation of expert system projects.     

ARPO-2: An expert system for legal advice on the breach of building contracts

Although Berman and Hafner [Berman 1989, pp. 928–938] presented the possibility to adapt the model of reasoning of development of an expert system for medical diagnosis to the reasoning of a judge when he/she sentences criminals does not resemble the reasoning found in the decisions of physicians, mathematicians or statisticians.When a lawyer reasons, he/she not only looks for the solution of a case; he/she simultaneously looks for the bases on which his/her reasoning can rest [Galindo 1992, pp. 363–367]. That is to say, he/she not only needs to find the solution but moreover he/she has to find the references (laws, jurisprudence and bibliography) that allow him/her to argue the solution.In many cases, computer solutions to these reasoning processes have been made in a separated way: the solution to the cases using expert systems, and the search of documentation using information retrieval systems.This paper presents the ARPO-2 prototype, a solution integrating the two aspects of legal reasoning: an expert system which is able to simultaneously find the solution to a problem and to give the necessary references so that the lawyer argues the solution. The subject on which the prototype solves problems is the breach of building contracts.In this paper, we describe the process of development of an expert system for solving, justification and documentation of breach of contracts, giving details on the way how the objects that intervene in the case were defined as well as on the reasoning followed.This paper was funded in part by DGICYT, Spanish Civil Law Computerization Project: PB870-632.     

Simulations of adaptive controllers for a paper machine headbox

The paper headbox is a crucial part of a paper machine, largely influencing the paper quality. Since this can be improved by better control, the evaluation of advanced control methods is certainly justified from an economical point of view. Not only do disturbances have to be rejected, but smooth startups and fast setpoint changes must also be possible. To design a good controller, some parameters need to be known, the determination of which can be cumbersome. However, this can be considerably alleviated by adding an on-line parameter estimator to the controller. This allows the controller to adapt itself to slowly changing process characteristics. In this paper different adaptive control methods are used on the simulated process and compared to the classical PI-controls. Particularly, a single-output self-tuning controller with feedforward action and a multivariable adaptive deadbeat controller are promising. The last one provides simultaneous offset-free control of both process outputs.     

An expert system for assembly based on Axiomatic Design principles

Increasing demands for lower cost of goods, due to international competition, has developed a new area of rationalization for increased productivity which can be achieved by applying new radical methods of production principles. Design for Assembly (DFA) is one of such methods that should be considered in the early design stage in order to cope with the increasing demands of lower cost, improved working environment, and higher wages. When Axiomatic Design (AD) is implemented with DFA in an intelligent way, a new area of design concepts would be realized. This study examines some elements of intelligent design systems to assess manufacturability of a product through the development of a knowledge based expert system for assembly (KBESA). The knowledge base has been acquired from DFA along with AD concepts with emphasis on the conceptual design stage where the structure of the product as a whole is considered. These concepts have been implemented in a case study illustrating its applicability.     

An agent-based framework for cooperating expert systems in concurrent engineering

This paper describes an agent-based framework with a tandem integration architecture for cooperating distributed, heterogeneous, and autonomous concurrent engineering systems. The approach has been experimented with using the prototype AGENTS system. It is demonstrated that design tools can be dynamically connected and configured to construct highly complex systems on an incremental basis, designing and proving one system at a time without bothering the rest. A major advantage of the approach is the balance between the interoperability and the autonomy of individual systems.     

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.     

Adaptive hierarchical tuning of fuzzy controllers

Fuzzy controller design includes both linear and non-linear dynamic analysis. The knowledge base parameters associated within the fuzzy rule base influence the non-linear control dynamics while the linear parameters associated within the fuzzy output signal influence the overall control dynamics. For distinct identification of tuning levels, an equivalent linear controller output and a normalized non-linear controller output are defined. A linear proportional-integral-derivative (PID) controller analogy is used for determining the linear tuning parameters. Non-linear tuning is derived from the locally defined control properties in the non-linear fuzzy output. The non-linearity in the fuzzy output is then represented in a graphical form for achieving the necessary non-linear tuning. Three different tuning strategies are evaluated. The first strategy uses a genetic algorithm to simultaneously tune both linear and non-linear parameters. In the second strategy the non-linear parameters are initially selected on the basis of some desired non-linear control characteristics and the linear tuning is then performed using a trial and error approach. In the third method the linear tuning is initially performed off-line using an existing linear PID law and an adaptive non-linear tuning is then performed online in a hierarchical fashion. The control performance of each design is compared against its corresponding linear PID system. The controllers based on the first two design methods show superior performance when they are implemented on the estimated process system. However, in the presence of process uncertainties and external disturbances these controllers fail to perform any better than linear controllers. In the hierarchical control architecture, the non-linear fuzzy control method adapts to process uncertainties and disturbances to produce superior performance.     

An alternative structure for next generation regulatory controllers: Part I: Basic theory for design, development and implementation

Even though employed widely in industrial practice, the popular PID controller has weaknesses that limit its achievable performance, and an intrinsic structure that makes tuning not only more complex than necessary, but also less transparent with respect to the key attributes of the overall controller performance, namely: robustness, set-point tracking, and disturbance rejection. In this paper, we propose an alternative control scheme that combines the simplicity of the PID controller with the versatility of model predictive control (MPC) while avoiding the tuning problems associated with both. The tuning parameters of the proposed control scheme are related directly to the controller performance attributes; they are normalized to lie between 0 and 1; and they arise naturally from the formulation in a manner that makes it possible to tune the controller directly for each performance attribute independently. The result is a controller that can be designed and implemented much more directly and transparently, and one that outperforms the classical PID controller both in set-point tracking and disturbance rejection while using precisely the same process reaction curve information required to tune PID controllers. The design, implementation and performance of the controller are demonstrated via simulation on a nonlinear polymerization process.     

An application of expert systems techniques to the on-line control and fault diagnosis of a mixing process

An application of expert hierarchical control is described in this paper. The control is implemented in a two-level configuration, where the lower layer performs direct regulation control and the upper layer performs supervisory functions. In the regulation layer, a rule-based controller performs the regulation task, where the controller is constructed upon causal relations between subsystems. The control action is inferred from the measurement of both controlled and noncontrolled variables. In the supervisory layer, the main function is a fault diagnosis system which diagnoses faults on-line. The diagnosis is based upon reasoning from the structure of the system and the functions of its components, and efficient diagnosis is achieved by dividing the system into several subsystems. The overall technique has been successfully implemented on a pilot scale mixing process under on-line computer control.     

Virtual reference feedback tuning: a direct method for the design of feedback controllers

This paper considers the problem of designing a controller for an unknown plant based on input/output measurements. The new design method we propose is direct (no model identification of the plant is needed) and can be applied using a single set of data generated by the plant, with no need for specific experiments nor iterations. It is shown that the method searches for the global optimum of the design criterion and that, in the case of restricted complexity controller design, the achieved controller is a good approximation of the restricted complexity global optimal controller. A simulation example shows the effectiveness of the method.     

Quality control expert systems: a review of pertinent literature

Statistical quality control (SQC) is an effective tool that ensures quality products and services by means of control charts, the essence of SQC, and sampling plans. While the computation of sample statistics and the development of control charts are routine exercises, the interpretation of chart patterns, trends and the associated diagnosis of assignable causes requires expert knowledge. The present trend is to develop a quality control system and apply it throughout the company (company-wide quality control CWQC or total quality control - TQC). This frequently means involvement of non-quality personnel in QC teams. Additionally, many companies are faced with a shortage of experienced quality controllers and individuals who can train and educate others on statistical quality control techniques. Quality control expert systems (QCESs) are considered as one way to alleviate these difficulties. In recent years, quality control expert systems have attracted the attention of both quality researchers and practitioners. This paper reviews existing quality control expert systems and recommends a set of quality engineering techniques that should be used to form a knowledge base, the heart of an expert system.