AI-Guided Reasoning-Based Operator Support System for the Nuclear Power Plant Management

In the Nuclear Power Plant (NPP) control room, the operators’ performance in emergencies is impacted by the need to monitor many indicators on the control room boards, the limited time to interact with dynamic events, and the incompleteness of the operator’s knowledge. Recent research has been directed toward increasing the level of automation in the NPP system by employing modern AI techniques that support the operator’s decisions. In previous work, the authors have employed a novel AI-guided declarative approach (namely, Answer Set Programming (ASP)) to represent and reason with human qualitative knowledge. This represented knowledge is structured to form a reasoning-based operator support system that assists the operator and compensates for any knowledge incompleteness by performing reasoning to diagnose failures and recommend executing actions in real time. A general ASP code structure has been proposed and tested against simple scenarios, e.g., diagnosis of pump failures that result in loss of flow transients and generating the needed plans for resolving the issue of stuck valves in the secondary loop. In this work, we investigate the potential of the previously proposed ASP structure by applying ASP to a realistic case study of the Three Mile Island, Unit 2 (TMI-2) accident event sequence (in particular, the first 142 minutes). The TMI scenario presents many challenges for a reasoning system, including a large number of variables, the complexity of the scenario, and the misleading readings. The capability of the ASP-based reasoning system is tested for diagnosis and recommending actions throughout the scenario. This paper is the first work to test and demonstrate the capability of an automated reasoning system by applying it to a realistic nuclear accident scenario, such as the TMI-2 accident.

The author discusses human factors engineering as applied to main nuclear power plant (NPP) control room design, including the design principles, control room layout, control and information equipment, environment conditions, communications system, etc. The main control room for a new 300 MW NPP will be modified and improved on the basis of the Qinshan 300 MW NPP design. >

Operator situational awareness in Nuclear Power Plant (NPP) control rooms is quintessential in ensuring plant safe operation. Towards improving situational awareness, a novel framework of the EYE - Expert Supervisory System capable of visually monitoring human machine interfaces (HMI) in cyber physical system (CPS) is proposed with target application in NPP control room environment. EYE applied on HMI (EYE-on-HMI), conceptually offers an unique opportunity to reduce human-in-the-loop errors inherent in feedback control systems and improve overall safety. While assessing the feasibility of the proposed framework, we reflect on implementation challenges and review key enabling state-of-the-art technologies for EYE-on-HMI framework. EYE-on-HMI as a framework is envisioned to be portable to various industrial applications with prospective safety improvement to next generation of commercial applications with “driverless” operations.

The design of computerized procedure presentation for nuclear power plants

The control of nuclear power plants covers a wide range of problems and requires understanding of the many interactions and objectives involved in obtaining operational flexibility and safety. Improved solutions must be developed as nuclear plants will increasingly require operation in a load-following mode. The authors discuss applications, problems, trends, and their perspectives. Areas where researchers developing control methods can contribute toward improved control design are outlined.

New methods of information presentation and interface design are changing the working conditions in the modern Nuclear Power Plant (NPP) control room. Symptom-oriented EOPs (SOPs) with their structures and practical application are described. The Computerized Symptom-oriented Operating Procedures (CSOP) is researched, which can help the operator analyze all the symptom signals of steam generator tubes rupture (SGTR) and provide the computerized procedures corresponding to the symptom signals. This paper analyzes the accident of SGTR; the accident management of SGTR is important in reactor safety because SGTR is one of the relatively high-frequency events in pressurized water reactors PWRs. The symptom signals of SGTR and the possible accidents corresponding to the symptom signals are analyzed. The homologous measures of symptoms are summarized. The disposal of SGTR adopts the method of based on symptoms. The programs are developed by VxWorks that is a real-time operating system. The debugging of programs is processed on simulator. The test results indicated that the programs can provide operating procedures according to the symptoms of accidents. After adopting the Computerized Emergency Operating Procedures, the labor intensity and mental burden of operators are lightened. Computerized Emergency Operating Procedures can enhance the reliability, safety and efficiency of Nuclear Power Plant.

Development of an EEG-based workload measurement method in nuclear power plants

We apply the theory of regulatory focus in the field of nuclear power plant (NPP) control. The first pilot study was conducted at the German simulator centre for NPPs. Here, we tested the influence of accident training lessons on the experts' regulatory focus. In a second study, we tested and validated two chronically regulatory focus questionnaires to decide which measurement would be more suitable for future research. The Work Regulatory Focus scale was translated into German, and the Regulatory Concerns Questionnaire was adapted to a work environment. The combined evidence of both studies shows that the regulatory focus of the experts does not change during the training lessons. Study 2 shows that both questionnaires were able to adequately measure the regulatory focus, but that they produce artificially contradictory outcomes that could be misleading. Recommendations and considerations for using the questionnaires are provided. Theoretical and practical implications on the influence of experts' regulatory focus on working and decision-making behaviour using standard procedures in NPP control are given.

Recently, human error has been highlighted as one of the main causes of accidents in nuclear power plants (NPPs). In order to prevent human errors during the main control room (MCR) operations, which are highly complex and mentally taxing activities, improved interfaces and operator support systems have been developed for advanced MCRs. Although operator support systems have the capability to improve the safety and reliability of an NPP, inappropriate designs can have adverse effects on the system safety. Designs based on systematic development frames and validation/verification of the systems are pivotal strategies to circumvent the negative effects of operator support systems. In this paper, an integrated operator support system designed to aid the cognitive activities of operators as well as theoretical and experimental evaluation methods of operator support systems are reviewed. From this review, it was concluded that not only issues about systems (e.g., the accuracy of the system outputs),but also issues about human operators who use the systems (for instance, information quality, the operator’s trust and dependency on support systems) should be considered in the design of efficient operator support systems.

This paper describes factors that cause the information complexity and how to measure the information complexity, and proposes solutions to design a human-machine interface to reduce the complexity of system information in nuclear power plants. The factors include nuclear power control system itself, an improper automation design, new technologies of fully digital instrumentation, control, and human-machine interface that has created tremendous functionality and the price of increased system complexity, and the ill-designed humancomputer user interface. The NASA-TLX and Task COMplexity (TACOM) were adopted to measure the information complexity of screen design and emergency procedure responses, respectively. An overview of the instrumentation and control system of the EPR (Evolutionary Power Reactor), an advanced pressurized-water reactor (PWR), is presented. Its information complexity is discussed and solutions are proposed according to the general principles of humanmachine interface design. We focus on two principles that direct the interface design for the EPR project. The first is to "provide information in a satisfactory manner" to the human operator. The second is to "provide adequate information" to the human operator. This paper is the preliminary report of part of the large multi-year project addressing information complexity problems and interface design in constructions of new nuclear power plants.

Ergonomic field studies in a nuclear power plant control room

ADVANCED MMIS TOWARD SUBSTANTIAL REDUCTION IN HUMAN ERRORS IN NPPS

The average share of nuclear generation in a power balance of Russia approximates 20%, in some parts of a power supply system of the country the contribution of the nuclear power plants (NPP) significantly above. NPP output is increasing (despite the decommissioning of RBMK units) and is provided by the commissioning of capacities of new advanced generation III+ units operating under the Power Delivery Program (power supply contracts). Power unit No. 4 of Beloyarsk NPP, No. 6 and 7 of Novovoronezh NPP, and No. 5 of Leningrad NPP have been commissioned recently. The power unit №6 on the Leningrad NPP will be put into operation in 2020. The dynamic construction of VVER-TOI at Kursk site is underway. An increase in the share of power generated by a nuclear power plant leads to the need for their regulation and frequency response in accordance with the daily load pattern and management of emergency situations in power systems. This means that operation of NPP power units in a load following mode to ensure the stable operation of the power system is a resent challenge. The paper analyzes the existing possibilities of daily control of the plant load, and also proposes to arrange a system of daily loading control of nuclear power plants based on the experience gained.

Control room modernization is an important part of life extension for the existing light water reactor fleet. None of the 99 currently operating commercial nuclear power plants in the U.S. has completed a full-scale control room modernization to date. Nuclear power plant main control rooms for the existing commercial reactor fleet remain significantly analog, with only limited digital modernizations. Upgrades in the U.S. do not achieve the full potential of newer technologies that might otherwise enhance plant and operator performance. The goal of the control room upgrade benefits research is to identify previously overlooked benefits of modernization, identify candidate technologies that may facilitate such benefits, and demonstrate these technologies through human factors research. This report describes a pilot study to test upgrades to the Human Systems Simulation Laboratory at INL.

Hardware-in-the-loop (HIL) simulation technology, where the part of a system to be verified adopts real objects, is one of the important methods for the research of advanced nuclear power plant (NPP) instrumentation and control (I&C) technology. With the development of advanced NPP I&C technology, especially the multi-module NPP technology, the HIL simulation technology is facing the challenge of communication signals booming and model extension to deal with the requirement of modules increasing and thermal-electricity generation. Driven by the above requirement of research and engineering, it is necessary to develop a novel HIL simulation technology that has well flexible scalability and avoids the high computational burden of the distributed control system (DCS). In this paper, an original distributed simulation method applied to the transformation extension of the NPP I&C HIL simulation verification platform is proposed. The initial opinion of the method is deploying a third-party system utilized for numerical simulation and form a close loop with DCS by network communication. With the support of third-party equipment represented by the real-time target machine, the functions of the system can be flexibly expanded through the MODBUS series protocol, and algorithms with high sampling frequency requirements can be deployed. The method has the characteristics of economical communication consumption, standardized and reliable communication protocol, and flexible downloading models and algorithms mean. Aside from this, due to the relative independence from DCS, the distribute simulation method is promising to be an original platform for verifying the technology advanced control or fault diagnosis in addition to DCS computing servers.