Operator Control Actions Research Articles

Investigation of the source term and radiological consequences of small break LOCA with failure of the emergency core cooling system in Bushehr nuclear power plant

In the present study, the in-containment source term derivation, environmental dispersion and deposition as well as radiological consequences of Bushehr Nuclear Power Plant (BNPP) due to Small Break Loss of Coolant Accident (SB-LOCA) have been studied. First, the calculation of in-containment release and distribution of the source terms due to the severe accident of SB-LOCA with failure of the active part of the Emergency Core Cooling System (ECCS) has been carried out. In this calculation, two scenarios, including with and without operator control actions, have been considered. Then, the environmental leakage of radionuclides as well as their dispersion and deposition considering two meteorological scenarios have been investigated. Consequently, public exposure has been calculated for each meteorological scenario and the obtained results have been reported and discussed. For two accidents with the same meteorological conditions, at a distance of 400 m from the release point, the received dose in the first 24 h is 18 and 342 mSv without and with operator action, respectively. The results show that due to the intervention of the operator and the use of emergency cooling supply systems, more water enters the reactor. Therefore, if the operator's actions do not prevent the core from melting, more source terms have entered the containment after the core melts and worse conditions than the baseline scenario without operator action occurred.

CPV of the Future: AI-Powered Continued Process Verification for Bioreactor Processes.

According to the standard guidelines by the FDA, process validation in biopharma manufacturing encompasses a life cycle consisting of three stages: process design (PD), process qualification (PQ), and continued process verification (CPV). The validity and efficiency of the analytics methods employed during the CPV require extensive knowledge of the process. However, for new processes and new drugs, such knowledge is often not available from Process performance qualification and Validation (PPQV). In this work, the suitability of methods based on machine learning/artificial intelligence (ML/AI) for the CPV applied in bioprocess monitoring and cell physiological control of the yeast Pichia pastoris (Komagataella phaffii) was studied with limited historical data. In particular, the production of recombinant Candida rugosa lipase 1 (Crl1) under hypoxic conditions in fed-batch cultures was considered as a case study. Supervised and unsupervised machine learning models using data from fed-batch bioprocesses with different gene dosage clones under normoxic and hypoxic conditions were evaluated. Firstly, a multivariate anomaly detection (isolation forest) model was applied to the batch phase of the bioprocess. Secondly, a supervised random forest model for prediction of required operator's control actions during the semiautomated fed-batch phase under hypoxic conditions was assessed to maintain the respiratory quotient (RQ) within the desired range for maximizing the specific production rate (qP ). The performance of these models was tested on historical data using independent evaluation of the process by the process control engineer (subject matter expert-SME), and on real-time data in the case of manual action prediction, where the model was implemented to guide the control of the bioprocess. The work presented here constitutes a proof-of-concept that multivariate analytics methods, based on machine learning, can be a valuable tool for real-time monitoring and control of biopharma manufacturing bioprocesses to improve its efficiency and to assure product quality.

HMM-based models of control room operator's cognition during process abnormalities. 1. Formalism and model identification

Operators' mental models play a central role in safety-critical domains like the chemical process industries. Accurate mental models, i.e., a correct understanding of the process and its causal linkages, are prerequisites for safe operation. Mental models are often defined and explained in abstract terms that make their interpretation subjective and prone to bias. In this work, we propose a Hidden Markov Model (HMM) based formalism to characterize control room operators' mental models while handling abnormal situations. We show that a suitable HMM representing the operator's mental model – including the states, state transition probabilities, and emission probability distributions – can be identified experimentally using data of the operator's control actions, eye gaze, and process variable values. This HMM can be used for the quantitative assessment of operators' mental models as illustrated using various case studies. We discuss the potential applications of the model in identifying various cognitive errors and human reliability assessment. In Part 2 of this paper, we use the proposed approach to assess operators' learning during training.

신형원전(APR+)을 위한 범용소프트제어기의 내고장성 설계

최근 범용소프트제어기 설계는 원자력발전소의 첨단주제어실에 적용되고 있다. 범용소프트제어기는 고집적 주제어실에서 비안전 기기뿐만이 아니라 안전기기를 제어할 수 있는 소프트웨어 기반의 수동제어 수단이다. 따라서 범용소프트제어기는 신형주제어실의 단일 워크스테이션 구현을 위한 필수적인 설계특성을 갖고 있다. 전통적인 주제어실은 컴퓨터 기반으로 하는 통합운전원 인터페이스 체계로 대체되고 있다. 범용소프트제어기의 오작동신호 발생 가능성을 줄이기 위해 어떠한 기기의 조작을 위해서는 2단계의 구분된 운전원 조작을 요구하는 설계를 고려하였다. 범용소프트제어기 오작동 가능성은 매우 낮기 때문에 범용소프트제어기 그자체로 발전소의 트립 가능성을 증가시키지는 않는다. 범용소프트제어기는 원자력발전소의 계측제어분야/인간연계 분야의 혁신을 대표한다. 범용소프트제어기는 인간연계를 기반으로 하는 단일 표시장치에 다양한 디비젼의 제어와 표시기를 통합하고 있다. 범용소프트제어기의 고장으로부터 안전기능 수행의 영향을 막기 위해 안전기기 및 기능에는 공학적 안전설비 신호가 적용된다. 또한 안전등급 수동스위치는 범용소프트제어기의 신호보다 우선한다. 그러므로 범용소프트제어기의 오작동 신호는 안전관련 스위치로부터의 제어신호에 의해 차단되어질 수 있다.

Predictive human operator model to be utilized as a controller using linear, neuro-fuzzy and fuzzy-ARX modeling techniques

Modeling human operator's behavior as a controller in a closed-loop control system recently finds applications in areas such as training of inexperienced operators by expert operator's model or developing warning systems for drivers by observing the driver model parameter variations. In this research, first, an experimental setup has been developed for collecting data from human operators as they controlled a nonlinear system. Appropriate reference signals and scenarios were designed according to the system identification and human operator modeling theory, to collect data from subjects. Different modeling schemes, namely ARX models as linear approach, and adaptive-network-based fuzzy inference system (ANFIS) as intelligent modeling approach have been evaluated. A hybrid modeling method, fuzzy-ARX (F-ARX) model, has been developed and its performance was found to be better in terms of predicting human operator's control actions as well as replacing the operator as a stand-alone controller. It has been concluded that F-ARX models can be a good alternative for modeling the human operator.

Hybrid predictive control for real-time optimization of public transport systems’ operations based on evolutionary multi-objective optimization

A hybrid predictive control formulation based on evolutionary multi-objective optimization to optimize real-time operations of public transport systems is presented. The state space model includes bus position, expected load and arrival time at stops. The system is based on discrete events, and the possible operator control actions are: holding vehicles at stations and skipping some stations. The controller (operator) pursues the minimization of a dynamic objective function to generate better operational decisions under uncertain demand at bus stops. In this work, a multi-objective approach is conducted to include different goals in the optimization process that could be opposite. In this case, the optimization was defined in terms of two objectives: waiting time minimization on one side, and impact of the strategies on the other. A genetic algorithm method is proposed to solve the multi-objective dynamic problem. From the conducted experiments considering a single bus line corridor, we found that the two objectives are opposite but with a certain degree of overlapping, in the sense that in all cases both objectives significantly improve the level of service with respect to the open-loop scenario by regularizing the headways. On average, the observed trade-off validates the proposed multi-objective methodology for the studied system, allowing dynamically finding the pseudo-optimal Pareto front and making real-time decisions based on different optimization criteria reflected in the proposed objective function compounds.

Nuclear fission power for 21st century needs: Enabling technologies for large-scale, low-risk, affordable nuclear electricity

We examine the principal concerns regarding provision of a large fraction of human energy needs with nuclear fission reactor-derived electricity, and offer robust physics and engineering responses to each of them. We then propose a representative system-level integration of these solutions to the longstanding problems that have confronted nuclear fission-based power. This integration obviates all fuel supply issues, including the entire set of isotopic enrichment ones, while rendering comparably useful as nuclear fuels all of the actinide elements and isotopes. It entirely avoids transport and reprocessing and the full set of ad hoc waste disposal issues, and completely precludes all those involving proliferation/diversion of fissile isotopes into weapons' programs. It offers high-grade heat in pressurized helium gas for thermodynamically efficient, economically appealing, environmentally attractive combined-cycle conversion to electricity while robustly avoiding prospects of internal overheating of any portion of the reactor's core or fuel. It provides highly redundant means of any desired statistical reliability for prevention of core meltdown in LOCA circumstances. It provides zero biospheric hazard in event of either natural or man-made catastrophe. It requires – indeed, admits of – no operator control actions, other than initial start-up and final shutdown commands, so that operator errors are entirely precluded; during the half-century of potentially full-power operational life in between these two commands, it thermostatically regulates in an entirely automatic manner its own nuclear power generation to match the heat removed from its core in a time-varying fashion. The thorium-burning variant of this new class of reactors involves no long-lived actinide isotopes, thereby obviating a present-day keystone issue of long-term reactor waste storage and disposal. Each of these novel features is technologically separable, so that these new reactor design concepts may be applied piecewise to enhance prospects of nuclear reactor-centered power generation in many different utilization circumstances. However, synergisms arising from their full integration seem likely to be compellingly attractive in most situations, for a constellation of economic and safety reasons. We therefore project a bright future for cheap electricity safely obtained in >10 TWe quantities from nuclear power reactors of this new type, moreover over multi-century time frames. We observe that pertinent aspects of neutron physics and modern technology together offer a far richer spectrum of possibilities for nuclear power reactors than has been significantly explored through the present; the present architecture is merely exemplary.

Understanding Perceived Complexity in Human Supervisory Control

Industrial processes are becoming more complex owing to technological developments and new opportunities. Technological developments, hardware and software, have become more reliable and system configurations more robust. However, the reliability of operator control actions has not improved at the same pace. Consequently human reliability has become the relatively weakest aspect of automated, operator-supervised systems. Hence, understanding how the human operator experiences increasing complexity may play an important role in task allocation and human–machine system design. In this paper the perceived complexity is studied within four typical operational environments in supervisory control. Mathematical formulations for these four operational environments are proposed, and their properties are analysed. A laboratory system is used to investigate the perceived complexity under various operational environments. The experimental results show a significantly different perceived complexity for the coupled and uncoupled operation environments. Extrapolation of the results revealed that the operator would have perceived the system as extremely complex if he/she would have to operate more than eight strongly interconnected subsystems extensively in 30 minutes. Implications of this study are also addressed.

The Software Simulator as an Effective Tool for Testing Control Algorithms

This paper describes a software simulator used in the development of control algorithms for the pulp cooking process. The laboratory-type simulator has been built up, based on the same control system hardware and software already installed in an industrial plant. The simulator itself emulates some operator control actions, the behavior of typical pulp cooking process variables and important signals from the process equipment. The simulation can be performed in real time or in two different accelerated modes. In this paper, the background of the simulator is also introduced and a discussion of advantages, based on experience, is attached at the end.

Queuing Network Theory Applied to Manned System Performance Prediction

This paper is a condensation of material presented in greater detail elsewhere (Kornfeld, et. al, 1985). An attempt is made to show the development of a queuing network theory model and how that model might be used for quantitative human factors analysis of a hypothetical message handling system. The system is postulated to consist of a menu-driven report preparation process with a pressure/touch sensitive screen for detecting operator control actions based upon human information processing and decision-making that occur when messages arrive at an operator's console.

Derivation of Fuzzy Control Rules from Human Operator's Control Actions

So called fuzzy control has been developed for the purpose of realization of a man-like controller with the aid of computer. However, most of fuzzy controllers have two very important problems. First is a defect of the reasoning algorithm. Second is a way to acquire control rules, particulaly the precise parameters in the rules. In this Daper considering the above problems, we propose a realistic fuzzy reasoning algorithm and a method to identify control rules from human operators actual control actions. Further we examine the performance of the proposed algorithm by applying it to water cleaning process control.

Simulation for training and decision-making in large-scale control systems

Simulators for training nuclear power-plant operators contain a duplicate of the actual plant-control-room instrumentation, which provides proper indications in response to operator control actions. In addition to the control-room instrumentation, the simulator includes an instructor's console, a computer complex, input/output interface equipment, and the attendant computer programs. Over the years simulators have grown in scope of plant system coverage and in complexity of the simulation techniques required. The sophistication of simulator implementation is still rapidly evolving in the areas of both simulation models and instrumentation. 9 references, 1 figure, 1 table.

Manned System Design Using Operator Measures and Criteria

In existing man-machine systems, the human operator is typically presented a predesigned display and he must adapt his control techniques in an attempt to optimize overall system performance. Ultimate performance achieved may not rise to the designer's expectation or to the operator's capability because the design process does not account for the operator's adaptability. In this research program, ship control performance of the Officer of the Deck (OOD) was observed and analyzed in a series of simulator experiments involving ship transit and obstacle avoidance. Three types of displays were included in the analysis. OOD control rules, and measures and criteria which describe the control technique used, as well as the criteria employed in controlling the ship, were derived from the individual performance data. The effect of display type on the measures and criteria which were apparently optimized by the subjects was determined. CONCLUSIONS: Analysis results show that performance differences between OLD and PACS, and RVV and PACS displays are significant at the α = 0.10 level. Summary measures which reveal this performance difference are time-to-CPA and distance travelled. Results confirm that time-to-CPA is an important measure since with it significant differences in performance can be shown. Analysis results also reveal the nature of the performance differences. From these results it is concluded not only that there are significant differences in performance with different displays, but also that performance measures which detect responses leading to a critical condition are more sensitive than summary measures of the critical conditions themselves. Analysis results reveal that models of human operators controlling ships must include a purview factor. The purview factor used here was a radius within which a contact is considered. The model ignores contacts outside that radius. The purview function is required to represent all performances, i.e., without regard to display type. The effect of OMAC parameter values on the comparison of optimal control to the subject control data obtained in the experiment. Results show that performance is at least a function of transit time and CPA/TCPA weighting as well as purview. Results suggests that for the contact density of the experiment problem, PACS and RVV can be rated as at least 12.5 mile displays and that OLD is a 10 mile display. Consider now the primary purpose of the research as identified by the four questions presented in the introduction which concern methods of modeling the human operator. A method of computing the apparent OMAC from a set of candidate OMAC, has been demonstrated. The apparent OMAC has the parameter values of those analyzed that provide the best comparison to the experimental data. Also demonstrated is a method for using the OMAC to predict operator control actions in specific problem situations. The method uses an optimal control synthesis employing the representative OMAC as the criteria. It is important to know whether or not a display configuration can affect the apparent OMAC. The results described here show that performance with different displays is significantly different and also, the results support the hypothesis that purview changes with display type.

Operator-Centered Adaptive Compensation of Continuous Manual Control Systems

The subject of this paper is an adaptive compensator, which has the function of minimizing the compensation required of the human operator in a continuous single-dimensional manual control system. A random input and a compensatory display are assumed. Use of this compensator does not require detailed knowledge of system dynamics since its adaptive action is based solely upon the operator's control actions. A quasi-linear operator model and the parameter-tracking model method of measurement are used to provide the required on-line mathematical description of the operator's control actions. Design of the compensator is based upon past results, which show the human operator to be an adaptive controller who introduces compensation, via his control actions, similar to that which would be chosen by a servo-engineer in minimizing the square of the system error. The compensator's function results from the established fact that an operator can control a system most accurately when the least compensation is required of him. Experimental results demonstrate that the compensator reduces system error while simultaneously simplifying the operator's task. They also establish some of the limitations of the compensation scheme as currently implemented.