Search

Crushing plants are widely used around the world as a pre-processing step in the mineral and mining industries or as standalone processing plants for final products in the aggregates industry. Despite automation and different types of advanced model predictive control, many the processes are still managed by operators. The skill of the operators influences the process performance and thus production yield. Therefore, it is important to train the operators so they know how to behave in different situations and to make them able to operate the process in the best possible way. Different types of models for crushers and other production units have been developed during the years and the latest improvement is the addition of dynamic behavior which gives the crushing plants a time dependent behavior and performance. This can be used as a simulator for operators training. By connecting an Internet based Human Machine Interface (WebHMI) to a dynamic simulator with the models incorporated, an on-line training environment for operators can be achieved. In this paper, a dynamic crushing plant simulator implemented in MATLAB/SIMULINK has been connected to a WebHMI. The WebHMI is accessible via the Internet, thus creating a realistic control room for operators’ training. In the created training environment, the operators can be trained under realistic conditions. Simple training scenarios and how they could be simulated are discussed. Apart from the increased level of knowledge and experience among the operators, the time aspect is an important factor. While a real crushing plant is still being built, the operators to be can already be trained, saving a lot of the commissioning and ramp up time.

This technical study describes the application of Ecological Interface Design (EID) method that may improve operator performance in digitalized system environment such as main control room in next generation nuclear power plant. While cognitive interface design method such as Information Rich Display (IRD) is mainly focused on stereotyped information for the user, EID method helps users' mental resources to be allocated to tasks that process highly abstract information including diagnosis and situation assessment. However, the sophisticated and relationship-embedded EID displays may cease to be intuitive when many of them are presented without consistently designed common elements. This study aims at suggesting consistent elements of EID displays for a digitalized Main Control Room of Nuclear Power Plant taking the example of the control process of Steam Generator (SG) Water Level. The proposed EID displays were reviewed by domain experts and operators to conjecture how such EID displays might benefit the operator's decision making. The experience from this study indicates that situation awareness in digitalized interface design may be improved by appropriately designed visualization of abstract information and that operator may be aided by the represented dynamics in EID display.

A presentation of the project Marine INspection rObotic Assistant System (MI-NOAS) is given. It proposes the re-engineering of the overall vessel inspection methodology, by introducing an innovative system concept that integrates human personnel with high locomotion enabled robots, effectively tele-porting the human inspector from the vessel's hold to a control room with virtual reality properties. Project motivations, requirements and basic architecture are discussed as well as the locomotion techniques which will be evaluated, if necessary, through prototyping of ad-hoc mobile robots.

In modern control rooms operators monitor complex operations which are highly automated and optimized. In normal conditions the automation system keeps the processes at an optimum. In abnormal situations the automation system might generate an avalanche of alarms and messages to which the operator needs to respond in the most appropriate way. Some of these alarm avalanches have lead to serious incidents, with damage and injuries as a consequence. Further to such an incident, guidelines have been published to reduce the workload of operators. This paper discusses the origin of the alarm overload, the measures taken in recent years and the advanced methods which are being developed and applied in industry. More attention should be paid to operator guidance to deal with and to avoid abnormal situations. Model-based reasoning and expert systems offer a good basis for such operator guidance. Recent data mining techniques make it possible to shorten engineering of such decision support systems. Further steps can be taken to automate the workflow between data mining tools and decision support tools for better operator guidance.

Although it is well known that the probability of accidents is very low in NPPs(Nuclear Power Plants), many researches are still being performed on these accidents since their results may be so severe. Furthermore, the possibility of accidents caused by Human Factors is not negligible. For that reason, many HRA(Human Reliability Analysis) techniques have been developed, but most of them have the following problems. First, they have mainly focused on a quantitative risk analysis rather than a qualitative analysis of systems. And, it is difficult to evaluate the possibility of Error of Commissions(EOCs). Also, the results of human error analyses are not adequate for a use in design fields, and the sensitivity of the evaluation procedures is not sufficient. In this study, as a solution to these problems, a new method for the analysis of human error in NPPs was proposed. The basic concept of this method is “to evaluate the human error possibilities related to the MCR(Main Control Room) equipment”. It means a qualitative analysis for the possibility of human error which is affected by the failures or the features and the limitations of the HSI(Human System Interface) equipment introduced in the MCR. The proposed method can be useful to analyze human errors affected by faults and failures of the HSI in the MCR, which were difficult to analyze with the conventional techniques.

The analog typed interface of main control room in nuclear power plant (NPP) is gradually being replaced to the one suitable for the digital environment. SKN 3,4 Nuclear Power Plant is currently developed in such a way to employ digitalized displays and controls such as computerized procedure system(CPS), large display panel(LDP), and Soft control. The main control room (MCR) of the SKN3,4 was designed based on the human factors guidelines, which somehow lack of the confidence for the rapidly changing digital environment. The suitable review guideline for the digitalized information displays was developed with an extension to the total digital environment including task environment, hardware and workstation. For the guideline development, tremendous guidelines and technical papers related to evaluation issues of digital environment has been collected, analyzed and transformed to electric database forms and then built on database management system to retrieve the appropriate issues for the practical usage of evaluators-in-field.

In abnormal or emergency situations, correct communication between main control room (MCR) operators is important to safe operation of safety-critical systems such as nuclear power plants (NPPs) and large-scale chemical plants. To improve the communications between MCR operators in NPPs during emergency situations, we developed a standard communication protocol for the emergency operation procedures (EOPs) of Korean Standard Nuclear Plant (KSNP). A general comment from real NPP operators was that the standard communication protocol is expected to increase the safety and economy of NPPs by reducing communication-related human errors.

This paper is presenting a wearable control room prototype developed for testing technology, usability, wearability and applicability of wearable computers in process environments. Problem areas related to introduction of wearable equipment in the process industry are discussed.

Advanced Power Reactor 1400 (APRI400) Man-Machine Interface Systems (MMIS) designs, which encompass the design of the Main Control Room (MCR) and Instrumentation and Control (I&C) systems, make use of advanced digital technologies which have been proven through industrial operating experience and extensive testing and development program. I&C system has been designed with the network-based distributed control architecture. Diversity between safety I&C systems and non-safety I&C systems along with hardwired switches are provided for the defense in depth against common mode failure of software in the safety I&C systems. There are some design improvement such as adoption of multi-loop controllers, multiplexing devices for both safety and non safety control, universal soft controller in main control room for both safety and nonsafety I &# C equipment controls.

Korea Water Resources Corporation (KOWACO), the only organization specialized in water resources management in Korea, is planning to establish a centralized control system for a related operation, in the central control room, of decision making as to water management and electric power generation. This plan is intended to convert the existing on-site operation system in each multi-purpose dam across the nation into a central operation system in the Corporation's head office, for which Generation Integrated Operating System (GIOS) will be established control center in the Corporation's head office and modernized nine (9) power stations, substations (RDAC). This article discusses realization of GIOS, especially real-time operation of hydropower plants and a specific architectural features for system communication.

The advanced human-machine interface (HMI) in nuclear power plants provides an information environment that supports the operators' burdensome cognitive tasks. Safety function status check (SFSC), one of such tasks, can benefit from well-designed information aid. This paper describes an interface design for supporting the operator's SFSC task as an element of HMI for Advanced Power Reactor 1400 (APR1400) of Korea. A prototype interface scheme for the SFSC task was developed based on the task analysis on APR1400 HMI and an evaluation of existing plants. The interface consists of three hierarchical levels of information: the goal tree including safety functions and their success paths, the flow structure of the success paths, and the system's topological structures related to the success paths. The interface is expected to have advantages in terms of cognitive information processing, which are currently being verified through systematic evaluations under diverse contexts and will be accordingly improved.

The objective of this paper is to describe and discuss, firstly, how cognitive ergonomics principles have been applied for the evaluation of the activities carried out in the main control room of a fire fighting system of a large city. The same principles, coupled with engineering creativity, have then been applied for re-designing the allocation of functions and interfaces in order to improve efficiency of communication, decision making and working context.

This paper reports an observational study of a Computerized Procedure System (CPS) under development for an advanced main control room in Korea. The purpose of this study was to examine possible effects of using the CPS. The crew, a shift supervisor and two operators, admitted that the CPS reduced their mental workload and they thought the CPS was better than or as good as paper-based procedures. One significant effect was that the amount of communication between the shift supervisor and two board operators was reduced considerably.

The rationale for automation is to achieve an optimum solution for a manufacturing process. It is implicitly assumed that the elementary functions are correctly implemented, and it is therefore reasonable to set up a surveillance level to assess the process state at all times and thus to improve process availability and maintainability. Model-based surveillance is now a classic solution, but early fault detection is still an innovative concept in the control room. Plant status diagnostics cannot be separated from process control: the supervisory system aid must benefit the operator, assess the process and take control actions into account. A multiple viewpoint model of the control situation allows behavior models to be selected for diagnostic purposes and, more generally, constitutes an effective decision-making aid. Finally, the issue of operator aids is situated in the broader context of control of complex situations.

We have developed an expert system for continuous supervision of part. of a pulp process. This is part of a prototype of KE 2000, a workstation for process industry developed by the Computing Science Department of Uppsala University and SCA Research AB. The system ran continuously in an operator control room at the SCA Ostrand pulp plant from early 1991 to mid 1993. In this paper we evaluate the experiences gained so far.

Research, development and semi-industrial implementation of Model-Incorporating Multi-Mode Controller MIMC on a pilot distillation column is described as a case study for future industrial implementation in water treatment processes of hydrolyzing stripping column to be controlled by integrated Honeywell TDC 3000 process control system, OMRON Fuzzy Interface Technology and Hewlett Packard 9000 Series computer installed at the control room of urea production.

Automation that is appropriate for application in realistically complex sociotechnical domains should be based on an integrated understanding of the technical, human, organizational, ecologic, economic and even cultural attributes of the application. Moreover, it is the engineer's responsibility to consider the direct and indirect effects of the automation technology being used. In order to integrate comprehensive sets of requirements into system planning and design, one must understand them in terms of causes and effects, dependencies and contradicitons, and short-und long-term effects. Integration is, therefore, a question of balancing different values A design philosophy is advocated which aims at developing the means whereby people in complex systems can achieve the operational objectives for which they are responsible: people are in charge The key questions for man-machine interface, control room and team design is, then, how to support operators so that they can be in charge successfully Innovative design concepts, principles and methods will be summarized. Encouraging experiences show how such approaches lead to better overall system performance. Much research on open questions remains to be done, though

This paper presents the strategies adopted in the development and implementation of a small scale low-cost distributed control system and its more important characteristics. The communications way adopted was the serial field bus, which confers great economy to the system. The structure used was the star type, where a master unit handles communication with slave units. The system was developed under personal computer's environment and includes several utilities, in local mode from the process control stations and in remote mode from the central control room of the plant. As a result, we have achieved a reliable and modular control system, capable of powerful additional possibilities with only small efforts.

The traditional Control Room evolution through Computer Aided solutions is growing together with the design/construction of new kind of plants, like the Combined Cycles and the Fluid Bed. The Constructions and Research & Development ENEL Directions have realized a Combined Cycles/Gas Turbine simulator that allows the full design of the main automation systems to validate the command/control functionalities by a computer aided MMI, one for all the plant systems, despite of the present Control Rooms. To verify such operator interface unification, the control system and subsystems of the real-time simulator is implemented by a real kind of industrial control system, with an advanced operator interface, based on process keyboards and graphic touch-screens.

This paper presents the initial experience gained from operating a VDU-based control room with consistently networked VDU-based operating units. System-related weaknesses of serial process control are indicated, requirements for VDU-based control rooms stated, supplemented by a treatment of the beneficial effects of networking on displays, operation control concepts, plant availability and working procedures.

Recent industrial accidents have proved the need for considering also human acts influencing the safety of industrial installations. An understanding of human errors relies on an understanding of the cognitive mechanisms of human decision making. Different attempts have been made to build models of the human operator, but the models have mostly concentrated on single individuals and isolated tasks. The models developed are also descriptive and are not very useful in guiding the design of new systems. At present there is no model, by which accurate predictions of the performance of a socio-technical system can be given. It is necessary to develop new modeling techniques, before it is possible to model the human parts of industrial installations to such a degree of accuracy that it is possible to predict their safety. The chapter discusses different modeling approaches, their advantages and disadvantages in providing guidance for designing future systems. The chapter discusses also some general requirements on the modeling methods of human-machine systems.

The paper deals with the selection and the design of process control systems based on distributed microcomputers for a large petrochemical complex. Taking into account the experiences gained in numerous process computer applications at PCK Schwedt a realistic control conception is presented using the potential advantages of the modern digital control hardware to the possible extent. Emphasis is given to such aspects as man-plant interface, control room design, local distribution and functional assignment of hardware, coexistence of conventional and modern technologies, reliability.

At the Zolder mine a S/7 computer is used for two purposes: - to monitor production at the coal face - to monitor the distribution of mine cars. By using sensors attached to equipment at the coal face and at the loading points,the computer provides realtime information and also summarized reports, which help management to control production better. The computer will display this information to the control room, where two operators monitor all production events and the availability of the right numbers of mine cars at every production face. All benefits obtained from the system are described.

During 1974-77 Store Kopparberg expanded its pulp production capacity at the Skutskar pulp mill and installed a large computer system for process supervision. Recently, production control and pollution control was implemented. The common basis for the three systems is a computer-based information system which itself is an integral part of control room instrumentation. The paper deals with the considerations made before the information system was decided on, especially with regard to mill organization. The main functions of the information system are described and the special features are discussed. The findings after a three-year operating period are summarized which may serve as a guidance for those who plan to install a similar system.

Typical distributed computer control systems for thermal power plants recently developed in Western Europe are presented. The complete control and supervision system for a modern power plant has a typical hiararchical structure, which is well suited for gradual introduction of micro computer modules or subsystems on the different levels. The expensive cabling between the functional groups, the central control room, and the process, is substituted by serial busses for digital data transmission. The structures of control loops and sequential programs are defined interactively by the design engineer by means of a user-oriented machine-independant programming language via display and keyboard. The reliability of central power plant computers proved all ready in the seventies to be very satisfactory, and by distributing the computer power the overall reliability and availability is expected to fullfill the most strict requirements.

A control project for a complex batch process producing organic fine chemicals is discussed. Because of process characteristics special attention was devoted to the safety of the control system. In the design of the control room general principles of man-manchine interaction were utilized. New systematic review methods were developed as Finnish contribution to the European Co-operation in the Field of Scientific and Technical Research (COST-A1 project). In alarm handling a new concept with dynamic alarm limits was introduced. Experiences are discussed. Based on these, some ideas for the development of new safety oriented design and review methods are presented.

In the process industries, control (in the wide sense) and organization show interactions on different levels. Process operators supervise and correct the automation system, and take care of off-normal events. For these purposes, they have to observe and to assess the process state. In some types of processes, state observation includes direct access to the process equipment; in other types indirect access (e.g. by means of VDU's) is usually sufficient. This difference has strong implications for the organization of the operator team and for the design of man/machine-interfaces. There is a trend to integrate process units, in order to save energy and raw materials. In this way, communication between control room operators becomes more and more important, leading to aggregation and centralization. However, this imposes requirements on control room lay-out, personnel selection and training, and job rotation schemes. Multi-level systems are often advocated for controlling complex systems, such as oil refineries or petrochemical plants. In practice, these systems encompass computers and people, with their various communication requirements and limitations. Hence, careful mutual adaptation of computer networks and organizational structures is necessary. These problems become even more complicated if other enterprise functions are included, such as marketing, buying, and maintenance. The paper ana1yzes and evaluates these interactions between process control and organization on the basis of industrial case studies. The results lead to modified relationships between production and instrumentation technology, jobs and work organization, control room and operator console lay-out, and operator/automation system interaction.

This survey paper indicates various aspects of traffic control of subways: The general concepts of the carrier (transportation supply, disturbance, vulnerability, transportation costs and human concepts) and some ideas to organize the operation. For illustration, many examples are given in the following fields: — automatic driving and driver monitoring devices; — central control room; — ticket issuing and checking; — public address systems . The conclusion points out that the effectiveness of any system is conditional on a balanced consideration of all the various viewpoints: technical, economic and human.

The paper describes the use of programmable controllers, micro-and minicomputers and Supervising computer for blast furnace control. The control system is divided into a number of control areas in order to achieve high availability, fast commissioning, etc. Each control area and its control are described. Details are given of the different control levels and their use. Finally, the paper presents a supervisory computer system, which replaces most of the conventional instrumentation and gives a new look to the central control room of the blast furnace plant.

One simple form of combined cycle system consists of a gas turbine, a heat recovery steam generator, a steam turbine and a generator. One set ot these combined cycle cumpuneiits may be called a “stage”. By combining several Stages, a large capacity power plant can be realized with the following features; (1) high efficiency (2) high efficiency at partial load and (3) rapid start-up and shut-down. To achieve control room operations efficiency similar to plant operation efficiency, it is desirable to implement comprehensive automation of each stage, overall, integrated control of the set of stages, and to provide well designed man-machine communications. This simplifies control room operations and enables ease ot load control over the complete plant output range. This paper describes with respect to this type of power plant; (1) a desirable configuration of the overall control system, (2) the concept of automatic control (3) how the total load is controlled and (4) the simulation result of total load control.

Over the last eight years an extensive program on the modelling of human supervisory control behavior and performance has been carried out at the Delft University of Technology. This paper will summarize the positive and negative results as they have been experienced over this period. Modelling human supervisory control behavior is not just an interesting scientific activity, it is meant to result into engineering tools on the basis of which a control room can be designed optimally in terms of adapting it to man's capabilities and avoiding his limitations. In fact, one aims for a quantitative description or model where the influence of task variables, such as display configuration or dynamics of the system to be supervised, can be predicted on the human supervisory control behavior. Such models exist for manual control, but for supervisory control most of the modelling attempts have failed. This fact is not only caused by the variability in supervision tasks. Certainly, there is a great difference in supervisory tasks during normal and abnormal operation of a plant. In normal operation the major task is one of optimizing the plant e.g. by controlling the setpoints, whereas in abnormal operation, tasks like fault management and intervention have to be executed. Moreover, during start-stop procedures a mixture of these tasks occurs. But, the fact that in a particular process control condition, tasks are poorly defined in a quantitative way, is probably the most important reason for the many disappointing results. In fact, the following statement can be made: Only for those tasks which are well defined, the modelling of human supervisory behavior and performance may be successful. In this paper our experience on human supervisory control behavior modelling in different experimental setups will be discussed. It will be emphasized that the approach as followed in manual control studies cannot be followed here; directions for further research will be given

Why spend a lot of money on simplified training simulators modelling the power system, when the real system with no simplifications can be used for operator training? Procuring a simulator for training of operating personnel is often too expensive. This holds true especially for a small power system with limited resources. Nevertheless training is very important to ensure a proper quality of supply by reducing brown-outs and black-outs. In the ELKRAFT power pool the operator training is done by introducing planned alert states during daily operation, and a very realistic situation is thereby achieved. As it often happens during disturbances, unforeseen and unexpected situations are likely to occur. The achievement of a general understanding of the need for everybody doing, what is expected, is an additional benefit compared with the use of a training simulator. The cost of a training programme of this type is estimated as USD 10-15,000 each time, which is very reasonable considering that the training is done in the actual control room instead of in simplified training rooms. As a benefit, additionally to the training of the control centre personnel, the whole power system (power stations, grid, and personnel) is also tested and trained.

Fault-tolerant automation subsystems have to be added to a modern distributed process control system in order to fulfill the most stringent requirements in view of reliability and safety of process control. The paper describes design features and experience with both new 1-out-of-2 hot standby system and 2-out-of-3 system, each one operating in a fully synchronized mode. The subsystems have the same scope of functions as a basic non-redundant system and are suitable for non-stop process control. Operator communication, observation and configuration can be carried out either from a central control room via a bus system and special operating subsystems, or from a direct coupled local operating station.

A control room completely computerized will operate the new French N4 series of PWR's nuclear plants (1, 400 MW); this paper, after having described the main goals of the system, presents its architecture (and specially the control room and the operator console), gives some key-figures of this system, and finally shows the main choices done for its development.

With the advanced technological possibilities for man-machine interfaces in control rooms of complex dynamic systems, a new quality of questions and problems for their design need to be addressed and investigated. Expert system advances to support the designer of control room interfaces, and later, the control room operator, play a crucial role in the ongoing discussion within the cognitive engineering community, and most of all between all organizational groups involved in the process of system design. In this paper, general design issues for modern man-machine interfaces are discussed in relation to problem solving and engineering tasks. A knowledge representation for a new graphical interface design is described that accounts for cognitive and ergonomic factors, that guarantees for higher consistency and improved ease of the design process, and that can be applied both during the design phase and the operation phase of a control system by a set of expert systems. In conjunction with data and knowledge structures for the application of expert systems to the design phase, and later, the operation phase of real-time systems, some of the requirements for the reasoning techniques are discussed. Finally, the question of applicable knowledge acquisition methods to provide proper contents for the suggested representations is addressed.

This paper will describe a program for the planning and implementation of advanced process control in an existing refinery. Such a program is a typical example of new concepts and methodologies utilized by companies seeking improved operating efficiency through increased yields of valuable products and decreased consumption of required energy. Advanced process control applications vary in complexity and hierarchy due to the variations of control objectives among the refinery units. For the refinery discussed here, advanced process control was applied to the following primary units:. - Two topping units, with downstream vacuum units. - Visbreaker. - Fluid catalytic cracking unit. - Storage and blending. These advanced control strategies were specifically designed for this refinery to address a complex pattern of changes in crude oil type and product demand mix. For example, the atmospheric and vacuum distillation units require strategies for the dynamics of crude changes every three days and up to 18 different product demand patterns. The implementation of advanced process control is part of an overall refinery control modernization project. The project entails replacement of the current main control room pneumatic control system with an up-to-date digital control system which specifically provides for implementing advanced control strategies. Implementation of the control system changeover will be an online operation and carried out on a unit-by-unit basis.

The 1980's will see the profitable integration of low cost digital computers, powerful process control software, and advanced control and optimization technology in petroleum refineries to reduce plant energy consumption and maximize valuable product yields. This paper surveys the experience of the last two decades for the third and most critical segment: on-line closed-loop optimization technology. The requirements for successful application emphasize practical guidelines, economic scope and justification, model requirements, mathematical algorithms, objective functions, project costs, and maintenance requirements. Matching appropriate models and algorithms to particular process units and business objectives will be outlined. The proper role of steady-state and dynamic optimization will be explained for petroleum refineries. The ingredients to expand technical successes to lasting economic successes are listed. The successful systems have a fundamental impact on the gap between control room operations, higher level management objectives, planning, and environmental restrictions. Beyond the mathematical techniques for solving the problem, much attention is focusing on the information interface between the optimizing computer, the operators charged with running the process, and engineering managers charged with setting operating objectives in response to business and market changes. People need to know much more than what the optimum computer answer is, to be assured the computer optimum is true and to utilize the tool for business control. The systems methodology for successful implementation on crude units, vacuum units, fluid catalytic crackers, catalytic reformers, gas plants, olefins plants, aromatics plants, hydrocrackers, and utilities will be developed. Some optimizations worth $100/day, and some worth $10, 000/day will be described. Future trends for integrated units and refinery wide optimization are given. These examples will clearly illustrate the role on-line process optimization is playing to reduce gasoline shortages and conserve energy.

The use of real-time simulotors for control system development is a well established application area. With the growing number of large-scale training simulators their use for other purposes than training such as human factors studies and control room improvements is becoming a common practice. For the development and evaluation of highly sophisticated operator support systems the use of large-scale plant simuI a tors as test facilities is becoming a practical necessity. These development trends are leading to the introduction of a new type of real-time simulators. the dedicated large-scale research simulators. The NORS ( NO kia R esearch S imulator) is of this new type. It was implemented during 1983 Halden Reactor Project's Man-Machine Systems Laboratory (HAMMLAB). The HAMM-LAB research programme on advanced computer and colour display based control rooms covers the establishment of experimental techniques and methods for the new facility basic Man-Machine Systems research with development of concepts and theories experiments on information presentation and exercise of control and experimental validation of advanced computerized operator support systems. The NORS simulator was designed to effectively support the needs of the versatile research programme. The design goal was to combine the simulation scope and quality of large-scale training simulators with a flexible software and hardware structure readily adaptable to various experimental conditions. Novel features of the simulator software include extensive use of computer aided modelling and generic modelling packages for simulation of plant systems and instrumentation, highly modular structure and comprehensive use of a data base and software development system.

The CERN Large Electron Positron (LEP) collider ring under construction and the existing Super Proton Synchrotron (SPS) accelerator will both be operated from the same main control room over a new control network implementing the token-ring protocol in conformity to the IEEE 802.2 and 802.5 standards. To cover the circumference of the LEP and SPS accelerator rings and the long distances (up to 4.6 kms) between geographically distributed control areas an unconventional transmission method has been implemented and successfully tested with the contribution of the IBM Zurich Research Laboratories. The LEP/SPS control network, consisting of several token-rings, inter-connected by bridges, and of gateways to the present SPS multi-star network and to the CERN computer centre, will be described as well as the network reconfiguration and diagnostic facilities.

The application of a distributed microcomputer based automation system to a new board mill is described. The board mill is quite complex with various production possibilities and is using different qualities of raw materials with a large percent of recycled fiber. The automation system used is based on several independent data processing units with dedicated functions. The data processing units are interconnected with serial buses which are using the HDLC communication protocol. The operating station are based on a colour visual display unit which has made a new type of control room concept possible. To ensure a sufficient high availability to make it possible to have maintenance staff available only during normal working hours the substations and the buses are backed up by standby substations and buses. The paper also gives a short account on the history and experience of the project.

This paper describes the controls and instrumentation for the Susquehanna Steam Electric Station (SSES), which is under construction for Pennsylvania Power & Light Co. (PP&L), Allentown, Pennsylvania, USA. The SSES is a two-unit boiling water reactor (BWR) station — each unit has a rated output of 1100 MWe. Of particular interest is the advanced control room featuring computer- driven displays with resultant reduction in the number and size of conventional controls and instrumentation. The principal design goal is for improved operability through the application of human engineering factors developed over the past several years. Successful execution of the advanced control room design and manufacture is centered in a team effort among the owner, the architect-engineer, and the NSSS manufacturer.

In the case of a traditional control room the operators get too many particular information, especially in an alarm situation, which they simply cannot properly evaluate in a short time. They make measures usually by intuition and this can lead to serious failures. An expert system based on special knowledge of technology organized by help of artificial intelligence methods can be used to assist the operators’ work. By help of expert system it is possible to select the primer alarms from the secondary ones and to give the operator the necessary compressed information only. The alarm situation analysis based on cause-consequences trees can be combined with state-analysis by help of mathematical models of the whole process or their subsystems. In the paper we describe an expert system for diagnosis purposes in large power plants. In the case of several thousand process signals not only the system functions, principles and structure are worth of interest, but the collecting and selecting the primer information – the technological knowledge – and the building up and proving the correctness of the expert system are very difficult problems, too.

The background and motivation of a Nordic cooperation project studying man-machine problems in the nuclear power field is described. The work has been done in the areas of control room layout, human reliability and operator training. The methods and results of a system description and a task analysis are considered in more detail. The system description was used to structure the information of the plant and to guide the interviews with the plant operators. The task analysis was intended for the planning of the operator training using a full scope simulator. The training of operators is discussed and the layout of a training session is given.

This paper presents the experimental validation of a computerized operator support system (COSS) for BWR (Boiling Water Reactor) power plants. To assess the effects of the COSS on operator performance, a realistic experimental control room and a full scope BWR plant simulator were developed. The control room includes a main and two auxiliary control panels equipped with color cathode-ray tubes (CRT's), control switches, annunciators and paging devices. A supervisory console with CRT's is also installed in the control room. Test transients were selected by taking account of their severity, complexity, generality, duration time and relevance to the functions of the COSS. Experimental data of the operating crew's performance and the plant status during simulated transients were collected on-line by using computers and audio/video tape recorders. The data were processed and transformed into sunmarized forms to facilitate analysis. Questionnaires and interviews were performed to collect subjective data for analyzing and decision making process of the operators. The results of the experiment show the usefulness of the COSS as an operator aid, especially in diagnosing the causes of disturbances and in establishing counteractive strategies

Supervisory control and the rehabilitation of bodily handicapped seem to be rather different topics: Supervisory control is task oriented, whereas the rehabilitation of handicapped is individual oriented. However, research has shown that both the fields can fertilize each other to a certain extent. Studies on Supervisory control deal with the problem of understanding human behavior during the execution of supervisory tasks in stationary and non-stationary process control modes. As a result, it should lead to a better design of Man-Machine Interfaces. According to the World Health Organisation, the rehabilitation of bodily handicapped persons is the coordinated and integrated set of activities which brings the handicapped back into society. It yields that medical, paramedical, psycho-social and technological aspects all play a certain role. Note, that in supervisory control studies one aims to investigate the capabilities and limitations of healthy men, whereas in rehabilitation one looks for individuals who, for some reason, lost particular capabilities resulting in additional limitations. The study of the effects of these additional limitations on human behavior may just lead to a better understanding of the capabilities which are lost. Until now, the design of control rooms and Man-Machine Interfaces is still more an art than a science. Well-defined relations between task variables and accurate descriptions of human supervisory control behavior are still a dream. Only in those cases where the behavior is based on skills and rules, successful design criteria could be developed. Until now, no descriptions of knowledge based behavior could be achieved. Hence, the fact that just the human supervisor is the essential element in the control room due to his creativity and intelligence, makes that a fundamental basefor the design of Man Machine Interfaces is more or less lacking. A similar situation can be recognized in rehabilitation medicine. The treatment of a patient is still an art. Studies have resulted in pieces of information, but the present state of the art is not that far that general rules applicable to a particular patient have been developed. Many skills and rules are used by the staff of the treatment team, but for an efficient decision making process much information is lacking. Again, for a correct diagnosis and prognosis one needs more than just a number of skills and rules; one asks for knowledge. The two research area's mentioned have many aspects in common. 30th the fields deal with highly multidisciplinary, dynamic, complex processes, where mostly incomplete and subjective data, values and criteria are only available. Moreover, in both the cases, just one trial to design a particular control room or to treat a patient is offered, albeit totally different reasons areinvolved. As a consequence of this similarity, the methodology to approach the problems has much in common. Finally one conspicuous remark should be made. Supervisory control systems are built to serve industrial processes, thus to make money. Rehabilitation is ment to help handicapped individuals, progress in this field leads in general that more money has to be spent, with hopefully is consequence a better rehabilitation result. It is therefore, that the pressure for research is supervisory control, and thus the money available, is much higher than that in rehabilitation. Hence, the possible transfer of methodologies and results obtained in supervisory control studies to rehabilitation research might be much more needed than in the opposite direction. Nevertheless, the author strongly believes that both the fields can help and fertilize each other.

The human operator in the control room of a complicated process plant plays a crucial role influencing the safety, availability and economy of the plant. An understanding of human behaviour and limitations is therefore essential both for the assessment of the plant safety and for the design of an optimal man-machine interface. In the paper different models of the human operator and their applicability to process supervision are discussed. A qualitative framework for operator models of a complicated process such as a nuclear power plant is discussed in more detail. The implications of the model in the control room design, the design process itself and the training of operators are indicated. Finally some general needs for new research are discussed

Recent events in U.S. operating nuclear power stations have raised questions as to whether control room operators might be unnecessarily burdened in their response to unusual events. To respond to these questions a team of plant control room operators, system design experts, and a human reliability analyst were assembled by the plant owners to investigate the issue of operator burden. The approach taken focused upon the review of all significant transients which had occurred in one plant type since 1980. The system design organization had previously analyzed each of these events as part of their Transient Assessment Program (TAP). These TAP reports represent a significant data set related to actual transient conditions and operator responses. The reports constituted the primary data base upon which the study was based. The thrust of the effort was the utilization of a multi-disciplinary team to become familiar with the transients and specifically with the operator and equipment responses. This information was then used to indicate generic areas where operator overburden might have occurred. Next information was elicited from the operations staff at each participating facility. In this way plant specific information was obtained which attempted to identify when and in which stations each of the previously identified generic overburdened conditions might apply. The study first required the development of a clear and concise definition of what operator burden was, and the development of generic examples of when it was likely to be observed in actuality. The definition chosen relates the extent of burden to the probability of operator error using an information processing paradigm of operator decision making. The developed definition was then applied to the review of 6 significant transients. This application allowed for the development of 5 generic qualitative categories of operator burden. In a parallel effort, the systems design team members developed detailed scenarios of each transient in a format designed to highlight the key operator decision points. The developed scenarios included a combined presentation of an operator action event tree (von Herrmann, 1983), and an operator concern event timeline. For each of the key decision points identified, a set of questions was developed and correlated to actual plant conditions at that point. These decision point packages were then circulated for response to participating operators and the responses led to a qualitative evaluation of the level of operator burden associated with specific aspects of transient integration.

Based on the recent trends of the international energy policy, the most significant purpose of newly built power generating plants is to reduce the fuel cost as much as possible. Gas/steam combined cycle power generation has been adopted in order to realize this important purpose. The major design philosophy of this control is to harmonize the central control room equipment for one-man operation and the control equipment for each component. A distributed control system is adopted for this combined cycle power plant using the data highway which connects a host computer and several remote stations to realize the following features: (1) High speed updating of CRT display; (2) Nonstop overall control system (high-grade maintainability and reliability); (3) Most advanced man-machine communication system. In this paper, particular emphasis is put on the software frame and function of the data highway control.

Designing control rooms for the process industries is usually a complex matter. Introducing human factors into a design project is even more complicated. Often the methodological approaches found in literature are to some degree theoretical and inflexible. Moreover, in practice one has to compromise in view of the delay in recognition and acceptance of human factors, limited budgets and deadlines. This heavily influences the time schedule for ergonomic contributions. Hence a necessary condition for an effective approach is that the methods are adaptable to a variety of different (local) situations. Such an approach, which was applied in several projects, will be presented, with reference to the design of a new control room for a highly integrated oil refinery. Particular attention will be paid to contributions by process operators to different design phases, and to the role of ergonomists. From this, and other case studies, it can be concluded that the essential design steps could be succesfully applied in a diverse set of industries and design situations.

The shortening of the weekly hours for workers in continuous three-shift work from 40 hours to 36 hours, starting in 1980, forced many process industries to find solutions for decreasing the number of personnel per shift. Automation and concentration of the control to a few or only one control room was chosen also by Partek as a method for reaching this goal. The demand for high availability could be met by a microprocessor based, distributed control system, DAMATIC, which had been developed by Valmet, a Finnish company. In this system even the control room CRT displays are generated by microprocessors, thus eliminating the need for minicomputer or discbased data systems. The same data highways which carry information from the analogic measurement points are also used for motor start-stop control. The process computer, which has turned out easy to program and to operate, is a Hewlett-Packard 1000 with a 20 M byte disc memory. It is connected to DAMATIC through a serial digital channel. The progammes are written in FORTRAN. Besides control tasks going through the DAMATIC system, the computer also controls “intelligent” sensors like X-ray analysers and particle size analysers. Computer programs developed by Partek now control raw material mixing, cement rotary kiln operation, and cement finish grinding. A typical feature of these programs is that several measurements, both continuous and sampled, have to be combined to a model of the process in question to make control actions correct. Through computer control, 2-4 % savings in fuel consumption and 5-10 % savings in energy needed for finish grinding have been achieved. The hierarchical solution, where the process computer has been relieved of primitive routines, like analog data collecting, digital control of simple loops, interlocking, generating of control room displays etc., which all can be taken over by microcomputers, makes it possible for the operating personnel to take advantage of the smooth programming and operating properties of modern minicomputers without hazarding the reliability of the primary supervisory functions in the control room.