Knowledge-based Control Systems Research Articles

Development of knowledge­based control systems with built­in functions of rules verification and correction

Two improved models of control rules were proposed. A model in a form of AND/OR graph; in contrast to the known graphical model of general rules, is based on dividing the rules into groups based on the controlled object state. The graph contains special markup that allows to convert the graph paths corresponding to the rules into Boolean expressions including formulas for both and inverse rule sets. The basic formulas of the rules model in a form of Boolean expressions cannot be constructed for general rules and based on these formulas the three methods for verification of the rules were developed: ‒ the method for verifying the control rules premises for inconsistency based on the SAT problem for Boolean formulas; ‒ the method for verifying the control rules for completeness based on visualization of both direct and inverse rules with conclusions in “inverse” rules opposite to the conclusions of the original rules; ‒ the method for verifying reachability of the object state vertices from the control rules. The main advantage of these methods is that they allow to find errors in the rules at early stages when specialists in the field for which the knowledge-based system is used (experts and decision makers work with them). The specificity of the control tasks makes it possible to do this effectively from the point of view of analysis and verification of the rule quality. The developed procedure of the control rules verification and correction assists in to bringing together and placing in a correct order various types of verification and correct errors in an automated mode. Main components were proposed for knowledge-based control systems: the rule editor for knowledge engineers and experts and the control system itself which includes extraction of the controlled object parameters essential for analysis as well as analysis of these parameters and their transfer to a DM for making a decision. A rule editor has been developed and control systems for two domains: safe operation with electric installations and control of computer networks. The presented experimental results on the management of the training process using the developed systems have shown that the number of errors in the created rules was reduced. When verifying for reachability of the object states, errors in an average of 5.4 % control rules were found and removed. When verifying for inconsistency of the rule premises, errors were found and corrected on average in 11.5 % of rules. When verifying for completeness, the rules base was expanded by on average of 12.3 %. In addition, due to consulting, verification and correction of the rules, the time spent by trainees on execution of their work was reduced by an average of 8 %.

Knowledge-based unmanned automation and control systems for the SBR wastewater treatment process

Wastewater treatment processes are usually located in rural seclusion. we designed an unmanned and automated control system for a sequencing batch reactor (SBR) wastewater treatment pilot plant. The pilot plant was constructed in the countryside, a small distance from a large city. Networks and wireless modules were employed for data transmission. A local controller was installed in the SBR pilot plant as a client, and a monitoring system was located in another place as a server. The communication system consisted of an asymmetric digital subscriber line (ADSL) network and a code division multiple access (CDMA) module. A remote control and monitoring system were constructed in a laboratory in the city. We describe a fuzzy inference system which was constructed with the operator's assistance, and acquired sensor data to determine the threshold and influent.

The BIOGENES system for knowledge-based bioprocess control

The application of knowledge-based control systems in the area of biotechnological processes has become increasingly popular over the past decade. This paper will outline the structure of the advanced knowledge-based part of the BIOGENES © control system for the control of bioprocesses such as the fed-batch Saccharomyces cerevisiae cultivation. First, a brief overview of all the tasks implemented in the knowledge-based level including process data classification, qualitative process state identification and supervisory process control is given. The procedures performing the on-line identification of metabolic states and supervisory process control (setpoint calculation and control strategy selection) are described in more detail. Finally, the performance of the system is discussed using results obtained from a number of experimental cultivation runs in a laboratory unit.

The numerical modelling and process simulation for the fault diagnosis of rotary kiln incinerator.

The numerical modelling and process simulation for the fault diagnosis of rotary kiln incinerator were accomplished. In the numerical modelling, two models applied to the modelling within the kiln are the combustion chamber model including the mass and energy balance equations for two combustion chambers and 3D thermal model. The combustion chamber model predicts temperature within the kiln, flue gas composition, flux and heat of combustion. Using the combustion chamber model and 3D thermal model, the production-rules for the process simulation can be obtained through interrelation analysis between control and operation variables. The process simulation of the kiln is operated with the production-rules for automatic operation. The process simulation aims to provide fundamental solutions to the problems in incineration process by introducing an online expert control system to provide an integrity in process control and management. Knowledge-based expert control systems use symbolic logic and heuristic rules to find solutions for various types of problems. It was implemented to be a hybrid intelligent expert control system by mutually connecting with the process control systems which has the capability of process diagnosis, analysis and control.

Uncertain variables and learning algorithms in knowledge-based control systems

This paper concerns a class of knowledge-based control systems with unknown parameters in the knowledge representation describing a plant. For the deterministic case, the logic-algebraic method of determination of the control decisions has been developed. The purpose of this paper is to present an extension of this method for the case with unknown parameters. In the first part, an approach based on uncertain variables is presented. In the second part, the method and algorithms of current knowledge updating in the learning system are proposed. The main idea of this approach consists of a step-by-step estimation of the unknown parameters in the knowledge representation using the successive values of the control decisions and their results. This concept may be considered as an extension of the known ideas of identification and adaptation for the traditional case.

KNOWLEDGE-BASED CONTROL SYSTEMS FOR SINGLE STEM ROSE PRODUCTION — PART II: IMPLEMENTATION AND FIELD EVALUATION

Field evaluations of knowledge-based control systems (KBS) for single stem rose production were performed. Two fuzzy inference systems (FIS) for automated selection of daytime and nighttime temperature were incorporated into the design of the integrated environmental control systems. Assessments of systems’ operational characteristics for daytime and nighttime temperature settings, single stem rose performance, and potential supplemental heating energy savings were performed over three consecutive crops. It was found that the KBS is capable of making a critical decision for daytime temperature settings based on the principle of economic optimization. Single stem roses produced by the KBS were graded by stem length. Overall crop quality was in the medium to high grades for the first two tests and poor for the last test. From the investigations, it was learned that undesirable low daytime temperature and light level during plant establishment promotes flower bud abortion. Potential benefits of heating energy savings using the KBS were estimated compared to a blueprint set point schedule (21°C). A moderate reduction in heating energy costs could be realized by using the KBS. The new method demonstrated the success of incorporating a FIS into decision support for daily environment control decisions, where conflicting economic goals must be dynamically re-evaluated as crop conditions and weather vary.

KNOWLEDGE-BASED CONTROL SYSTEMS FOR SINGLE STEM ROSE PRODUCTION — PART I: SYSTEMS ANALYSIS AND DESIGN

Systems analysis was performed to develop knowledge-based control systems (KBS) for single stem rose production. System components and control strategies for the KBS were identified from the analysis. The control strategy for daytime temperatures was derived by balancing the conflicting goals of rose production and costs of environmental modification. Nighttime temperature control strategy was determined through qualitative reasoning of the relationship between the plant’s photosynthetic efficiency and respiratory energy metabolism. A fuzzy inference paradigm was applied to facilitate system design. Two fuzzy inference systems (FIS) were designed to determine the daytime and nighttime greenhouse set point temperatures. A set of experiments was conducted to obtain the information on dimensional measurements of single stem rose (Rosa hybrida L., cv. Lady Diana) and the corresponding stem dry weight. An adaptive neuro-fuzzy inference system (ANFIS) was devised that predicted the dry weight for single stem roses from simple nondestructive measurements. An on-line computational algorithm for greenhouse heating and ventilation costs was developed according to the daily energy consumption of each component of heating and ventilation equipment. Two sets of rule bases were derived for daytime and nighttime temperature settings.

Characteristics recognition for the supervision of knowledge-based control systems

This paper presents a pattern-based information-processing approach, called characteristics recognition (CR), for on-line supervision of knowledge-based control systems. This approach provides a method for the analysis and automatic extraction of different types of process behaviour features during operation without a precise description of the process model. By combining the use of symbolic manipulation, graded fuzzy membership and numeric computation, the characteristics patterns that can capture system behaviour features are deduced from on-line data, and subsequently used for process monitoring and control. A multi-mode information-extraction and representation mechanism is provided to meet the needs of different tasks of system supervision. This leads to an effective and practical way of taking full advantage of on-line information. The basic methodology and a general application of CR to process monitoring and supervisory control are given.

Towards Design of Complete Rule-Based Control Systems

In this paper the problems of design of complete knowledge-based control systems are considered. The main goal of this paper is to present logical bases for a methodology for systematic design of rule-based control systems capable of control for a prespecified set of input situations. Such systems are complete with respect to the prespecified control area. A definition of generalized backward dual resolvent - a basic logical tool for completeness verification - is given and a definition of completeness is presented. An idea of a special semantic tree for representing various branching possibilities for formulae being components of rule preconditions is proposed. Such a tree constitute a basic scheme guiding the design of rule preconditions. A definition and criteria for such a tree to be complete are proposed. The criteria assure completeness of the designed set of rules. Further, an approach to design complete rule-based systems with use of such a tree as a basic design supporting tool is outlined.

DAI-DEPUR: an integrated and distributed architecture for wastewater treatment plants supervision

The activated sludge process — the main biological technology usually applied to wastewater treatment plants (WWTP) — directly depends on live beings (microorganisms), and therefore on unforeseen changes produced by them. It could be possible to get a good plant operation if the supervisory control system is able to react to the changes and deviations in the system and can take the necessary actions to restore the system's performance. These decisions are often based both on physical, chemical, microbiological principles (suitable to be modelled by conventional control algorithms) and on some knowledge (suitable to be modelled by knowledge-based systems). But one of the key problems in knowledge-based control systems design is the development of an architecture able to manage efficiently the different elements of the process ( integrated architecture), to learn from previous cases ( specific experimental knowledge) and to acquire the domain knowledge ( general expert knowledge). These problems increase when the process belongs to an ill-structured domain and is composed of several complex operational units. Therefore, an integrated and distributed AI architecture seems to be a good choice. This paper proposes an integrated and distributed supervisory multi-level architecture for the supervision of WWTP, that overcomes some of the main troubles of classical control techniques and those of knowledge-based systems applied to real world systems.

An Investigation into the Control of Brushed Yarn Properties: The Application of Machine Vision and Knowledge-based Systems. Part I: The Yarn Brushing Machine and Experimental Work

This work is concerned with the design and construction of a yarn brushing machine to produce and monitor such yarns by means of the application of knowledge-based, computer vision and on-line control systems. Part I describes an investigation into the effects that input yarn properties and changes in machine parameters have on the characteristics of the brushed yarn.

Real‐time qualitative analysis of the temporal shapes of (bio) process variables

AbstractOne of the limitations of today's knowledge‐based (KB) systems for diagnostics and supervision is a lack of adequate temporal reasoning mechanisms. Most of these systems are designed primarily to operate with the current values of the process variables and, sometimes, with their derivatives. Such simple capabilities, however, are not always sufficient to identify some complex dynamic phenomena, which in many cases leave their own unique “stamp” on the process behavior, expressed in the form of characteristic temporal shapes of the related variables. To detect and diagnose adequately the events of interest, the KB system should be able to reason about the temporal shapes of the process variables. Although during manual supervision process operators rely heavily on such characteristic shapes as reliable symptoms of underlying phenomena, their exploitation has not been considered seriously by the designers of KB control systems. We propose a generic methodology for qualitative analysis of the temporal shapes of continuous process variables designed to be embedded into a real‐time KB environment. It is applicable to bioprocesses, as well as to other complex dynamic systems.

Control Under Uncertainty Through Zone Logic

The manufacturing plant represents a complex environment, rife with uncertainty. The complexity arises from the multitude of interactions that must be considered when attempting to model most manufacturing processes. Important process variables can remain unidentified; or even if they are identified, their interactions may remain uncertain. This complexity and the uncertainties that are often its derivatives cause various inefficiencies when conventional control methods are employed. In an attempt to remedy this situation, an intelligent control methodology termed zone logic has been advanced. Various extensions to it have been proposed which are designed to increase its domain of applicability. This paper further extends zone logic into the area of stochastic controls by using concepts from Bayesian belief networks. An information theoretic analysis of an initial application of stochastic zone logic is performed. This analysis indicates that an object oriented computational scheme best matches the real-time performance requirements for knowledge-based control systems.

Low-Cost Environment for Analysis and Design of Knowledge-Based Control Algorithms

A dedicated low-cost environment for analysis and design of knowledge-based control systems of continuous-time processes REXCON is described. It is implemented on IBM PC/AT compatible computers under MS-DOS and includes three main parts - a special real-time operating system shell (RTOSS), a medium scale expert system shell (ESS) and a simulation program named RSCOPE. The so-called "progressive reasoning" mechanism is in the base of the ESS in order to improve the speed of reasoning and ensure a proper control signal at each sample time for various control system structures. REXCON supplies the user with facilities to analize a system working in two modes - direct control and simulation. Thus it is very useful for education of students and training of control engineers in the field of intelligent control.

A Method for On-Line Reasoning about the Time-Profiles of Process Variables

Abstract Most of today’s knowledge-based control systems lack an adequate temporal reasoning mechanisms. These systems are designed primarily to operate with the current values of the process variables and, sometimes, with their derivatives. However, such simple capabilities are not always sufficient to identify some complex dynamic phenomena which result in intricate temporal profiles of the related variables. To detect and diagnose adequately the events of interest, the KB system should be able to reason about the temporal shapes of the process variables. The present paper proposes a generic procedure for qualitative analysis of the temporal shapes of continuous process variables, designed to be embedded into a real-time knowledge-based environments. Examples from its application to the supervision of recombinant amino acid production are provided.

Contemporary Systems and Cybernetics

Information Processing Model of Knowledge Systems Two Massachusetts Institute of Technology (MIT) scientists, Kaan Egilmez and Steven H. Kim, writing in Applied Artificial Intelligence, Vol. 5 No. 3, 1991, pp. 353–92, believe that to date, knowledge‐based systems (KBSs) have been developed on an ad‐hoc basis, without the benefit of quantitative tools for analysis and synthesis. They claim that to approach the field of knowledge engineering systematically, we must develop performance measures to characterize the effectiveness and efficiency of knowledge systems, and thereby serve to highlight directions for improvement. They say that, when KBSs are applied to control tasks, real‐time performance becomes critical and in consequence the internal processes of knowledge‐based control systems (KBCS) must be optimized to fulfil real‐time requirements. In their contribution they derive relationships between information attributes and system performance.

Low-cost environment for analysis and design of knowledge-based control algorithms

A dedicated low-cost environment for analysis and design of knowledge-based control systems of continuous-time processes HEXOON is described. It is implemented on IBM PC/AT compatible computers under MS-DOS and includes three main parts — a special real-time operating system shell (RTOSS), a medium scale expert system shell (ESS) and a simulation program named RSOOPE. The so-called “progressive reasoning” mechanism is in the base of the ESS in order to improve the speed of reasoning and ensure a proper control signal at each sample time for various control system structures. HEXOON supplies the user with facilities to analize a system working in two modes — direct control and simulation. Thus it is very useful for education of students and training of control engineers in the field of intelligent control.

Real-time knowledge-based control systems

This special section is the outgrowth of a workshop on real-time knowledge-based control systems, held during the 1990 national conference of the American Association for Artificial Intelligence (AAAI). The workshop was motivated by the recognition that constructing robotic systems capable of autonomous, flexible, intelligent behavior is an inherently interdisciplinary task. In particular, professionals from the fields of artificial intelligence (AI), control theory, process control and real-time scheduling were invited. To facilitate communication among these fields, the workshop organizers decided to focus on software design principles for real-time systems, and so sought out those who had built working systems and could express their experiences in terms of techniques they had tested and/or refuted.

AN INFORMATION PROCESSING MODEL OF KNOWLEDGE SYSTEMS

To date, knowledge-based systems ( KBS's) have been developed on an ad-hoc basis, without the benefit of quantitative tools for analysis and synthesis. If we are to approach the field of knowledge engineering systematically, we must develop performance measures to characterize the effectiveness and efficiency of knowledge systems, and thereby serve to highlight directions for improvement. When KBS's are applied to control tasks, real-time performance becomes critical. Consequently, the internal processes of knowledge-based control systems ( KBCS) must be optimized to fulfill real-time requirements. This article derives relationships between information attributes and system performance. The performance measures are applied to a knowledge-based design system ( KBDS) for the synthesis of KBCS. The performance measures reflect intuitively satisfying properties of the KBDS and provide insights into the system not possible through a straight orward inspection of the program code. This type of system can be use...

Knowledge-based systems for industrial control

Artificial intelligence can assist in industrial control in many ways, but this paper concentrates on the role of knowledge-based systems in the areas of control, detection and diagnosis of faults, scheduling and planning. The paper argues that the application of knowledge-based techniques can provide a unification of these and other activities, and outlines the requirements of knowledge representation and inference techniques that are needed to make such systems a reality. The advances in hardware, including sensors as well as computers, that will be driven by and in turn stimulate the introduction of the knowledge-based industrial control systems are also considered.