Anthonie Cilliers

Temperature sensing is one of the most common measurements of a nuclear reactor monitoring system. The coolant fluid flow in a reactor core depends on the reactor power state. We investigated the monitoring and estimation of the thermocouple time series using machine learning for a range of flow regimes. Measurement data were obtained, in two separate experiments, in a flow loop filled with water and with liquid metal Galinstan. We developed long short-term memory (LSTM) recurrent neural… Show more

Temperature sensing is one of the most common measurements of a nuclear reactor monitoring system. The coolant fluid flow in a reactor core depends on the reactor power state. We investigated the monitoring and estimation of the thermocouple time series using machine learning for a range of flow regimes. Measurement data were obtained, in two separate experiments, in a flow loop filled with water and with liquid metal Galinstan. We developed long short-term memory (LSTM) recurrent neural networks (RNNs) for sensor predictions by training on the sensor’s own prior history, and transfer learning LSTM (TL-LSTM) by training on a correlated sensor’s prior history. Sensor cross-correlations were identified by calculating the Pearson correlation coefficient of the time series. The accuracy of LSTM and TL-LSTM predictions of temperature was studied as a function of Reynolds number (Re). The root-mean-square error (RMSE) for the test segment of time series of each sensor was shown to linearly increase with Re for both water and Galinstan fluids. Using linear correlations, we estimated the range of values of Re for which RMSE is smaller than the thermocouple measurement uncertainty. For both water and Galinstan fluids, we showed that both LSTM and TL-LSTM provide reliable estimations of temperature for typical flow regimes in a nuclear reactor. The LSTM runtime was shown to be substantially smaller than the data acquisition rate, which allows for performing estimation and validation of sensor measurements in real time. Show less

This paper introduces the stochastic Monte Carlo (MC) modelling of a nuclear reactor core using the Geant4 framework. The simulation is exercised in the context of a High Temperature Gas Cooled Reactor (HTGCR) that uses helium (a noble gas) as a coolant and graphite as a neutron moderator. The study presents the results from the implementation of basic neutronics, scalability, geometrical discretisation for studying the spatial variation of physical parameters, time slicing and adaptation of… Show more

This paper introduces the stochastic Monte Carlo (MC) modelling of a nuclear reactor core using the Geant4 framework. The simulation is exercised in the context of a High Temperature Gas Cooled Reactor (HTGCR) that uses helium (a noble gas) as a coolant and graphite as a neutron moderator. The study presents the results from the implementation of basic neutronics, scalability, geometrical discretisation for studying the spatial variation of physical parameters, time slicing and adaptation of Geant4 for correct intra-slice persistence, a scheme of integration with thermal hydraulics by workflow scheduling, validation of the thermal macroscopic cross-section behaviour, the process of fission, burn, decay, and differential energy depositions for the various physics processes, validation of the Xenon effects on the neutronics, criticality and core follow over multiple time steps. Show less

We are developing a microwave cavity-based transducer for applications in high-temperature fluid flow sensing. This sensor is a hollow metallic cylindrical cavity, which can be fabricated from stainless steel, and expected to be resilient to ionizing radiation, high temperature (above 500C) and corrosive environment of molten salt cooled and liquid sodium cooled nuclear reactors. Currently, limited options exist for flow sensing in such fluids, particularly for salts such as FLiBe (mixture of… Show more

We are developing a microwave cavity-based transducer for applications in high-temperature fluid flow sensing. This sensor is a hollow metallic cylindrical cavity, which can be fabricated from stainless steel, and expected to be resilient to ionizing radiation, high temperature (above 500C) and corrosive environment of molten salt cooled and liquid sodium cooled nuclear reactors. Currently, limited options exist for flow sensing in such fluids, particularly for salts such as FLiBe (mixture of lithium fluoride and beryllium fluoride), with melting point at ambient pressure above 450C. The principle of sensing consists of making one flat wall of the cylindrical cavity flexible enough so that dynamic pressure, which is proportional to fluid velocity, will cause microscopic membrane deflection. Cavity volume change due to membrane deflection leads to a measurable shift in the resonant frequency. Membrane thickness is … Show less

High-temperature advanced reactors under development, such as sodium fast reactors (SFR) and molten salt cooled reactors (MSCR), are expected to offer lower levelized cost of energy (LCOE) compared to existing light water reactor (LWR’s). In the existing light water reactors (LWR’s), operation and maintenance (O&M) expenses constitute the largest fraction of the total operating cost. Some of the O&M costs are related maintenance of sensors which can fail due to exposure to harsh environment in… Show more

High-temperature advanced reactors under development, such as sodium fast reactors (SFR) and molten salt cooled reactors (MSCR), are expected to offer lower levelized cost of energy (LCOE) compared to existing light water reactor (LWR’s). In the existing light water reactors (LWR’s), operation and maintenance (O&M) expenses constitute the largest fraction of the total operating cost. Some of the O&M costs are related maintenance of sensors which can fail due to exposure to harsh environment in a reactor. The O&M costs of Advanced Reactor (AR)’s are expected to constitute a significant fraction of the total cost as well, because of high temperature and radiation level in AR are likely to cause material fatigue and premature failure of sensors and components. The O&M costs in AR’s could be reduced through integration of advanced informatics of performance-related sensors into a digital twin designed for reactor monitoring. For example, machine learning (ML) could be employed for real-time validation and correction of performance-related sensors, and reducing the number of performance-related physical sensor units through virtual sensing. As part of the effort, we investigate real-time validation of thermal hydraulic sensors through one-step ahead forecasting of sensor values using long short-term memory (LSTM) recurrent neural networks (RNN). The sensors are installed in a flow loop containing a thermal mixing tee, which is a common experimental model to study thermal fatigue in a thermal hydraulic loop. In addition, nonlinear transients generated in a thermal mixing tee constitute a good challenge data set for training and validation … Show less

A condition-based maintenance (CBM) regime in a nuclear plant will result in eliminating unnecessary maintenance cost without jeopardising the safety of the plant. The foundation of a good CBM regime is an accurate and timely fault detection. A method has been developed to identify transients and detect fault in a Nuclear power plant in transients. This is to aid condition-based maintenance in a nuclear power plant. This method was achieved by using the nuclear plant simulator as a dynamic… Show more

A condition-based maintenance (CBM) regime in a nuclear plant will result in eliminating unnecessary maintenance cost without jeopardising the safety of the plant. The foundation of a good CBM regime is an accurate and timely fault detection. A method has been developed to identify transients and detect fault in a Nuclear power plant in transients. This is to aid condition-based maintenance in a nuclear power plant. This method was achieved by using the nuclear plant simulator as a dynamic reference. At steady state, a fault is easily detected but in transients, it is difficult. This gives rise to the introduction of a machine-learning tool like artificial neural networks (ANN) to train both the simulator and plant parameters. The neural network outputs of the plant and simulator are then compared and this results in a better identification of faults in transients. Show less

Abstract Condition-based maintenance (CBM) involves undertaking maintenance activities based on the health of the system. CBM has found useful applications in many industries. This paper presents a survey on the state of condition-based maintenance in the nuclear industry. This is achieved by systematically looking at the major phases of CBM, which are monitoring, diagnostics and prognostics. A methodical review has been done on these aspects of CBM. This includes the current practices in the… Show more

Abstract Condition-based maintenance (CBM) involves undertaking maintenance activities based on the health of the system. CBM has found useful applications in many industries. This paper presents a survey on the state of condition-based maintenance in the nuclear industry. This is achieved by systematically looking at the major phases of CBM, which are monitoring, diagnostics and prognostics. A methodical review has been done on these aspects of CBM. This includes the current practices in the nuclear industry and the ... Show less

With nuclear plant full scope simulator technology reaching maturity, the possibility of using its capabilities in expert systems such as online plant diagnostics has become a reality. The effectiveness of plant diagnostics using real-time simulated measurements as a plant reference has been shown in previous papers. However, in order to implement these systems, the full scope plant simulator needs to be designed specifically with this application in mind. This will help in maximising the… Show more

With nuclear plant full scope simulator technology reaching maturity, the possibility of using its capabilities in expert systems such as online plant diagnostics has become a reality. The effectiveness of plant diagnostics using real-time simulated measurements as a plant reference has been shown in previous papers. However, in order to implement these systems, the full scope plant simulator needs to be designed specifically with this application in mind. This will help in maximising the effectiveness and scope of use of the system. This paper investigates the various simulator technologies available as well as the development strategies and focus areas to establish the design requirements of a single, full scope engineering and training nuclear plant simulator. This can be implemented to provide a real-time dynamic reference to the plant diagnostic system. Show less

South Africa has long been regarded as an active country in the nuclear industry with two operating power reactors and a research reactor. In recent years’ research and development projects, such as the Pebble Bed Modular Reactor, has established additional expertise in the country situated at various institutions. After the PBMR project was stopped, the expertise became fragmented throughout the country and some experts even left the country. A number of training and research facilities have… Show more

South Africa has long been regarded as an active country in the nuclear industry with two operating power reactors and a research reactor. In recent years’ research and development projects, such as the Pebble Bed Modular Reactor, has established additional expertise in the country situated at various institutions. After the PBMR project was stopped, the expertise became fragmented throughout the country and some experts even left the country. A number of training and research facilities have also completed their research cycle and are in the process of being decommissioned. With the renewed interest in nuclear technology and the states position to complete the procurement of 9600 MW of nuclear power before the end of the year, nuclear knowledge gap has been identified and the need to capture all nuclear education and research in an educational network as well as to establish new nuclear training and research facilities such as small training reactors and research laboratories to support the national new build programme. This expertise and research facilities were combined into SAN-NEST (South African Network for Nuclear Education, Science and Technology) for South Africa and the African continent, with links to AFRA-NEST. The paper reports on the successes and challenges of the establishment and operation of SAN-NEST. Show less

Stress tests required evaluation of the consequences of loss of safety functions from any initiating event (e.g., earthquake or flooding) causing loss of electrical power, including station blackout (SBO). The SBO scenario involves a loss of offsite power, failure of the redundant emergency diesel generators, failure of alternate current (AC) power restoration and the eventual degradation of the reactor coolant pump (RCP) seals resulting in a long term loss of coolant. Using PCTRAN/CPR1000… Show more

Stress tests required evaluation of the consequences of loss of safety functions from any initiating event (e.g., earthquake or flooding) causing loss of electrical power, including station blackout (SBO). The SBO scenario involves a loss of offsite power, failure of the redundant emergency diesel generators, failure of alternate current (AC) power restoration and the eventual degradation of the reactor coolant pump (RCP) seals resulting in a long term loss of coolant. Using PCTRAN/CPR1000, this study analyses the station blackout on a Chinese CPR1000 which is the most representative type reactor in terms of number of reactors, operating period, power capacity and geological distance from Korean Peninsula. Both the physical effects of the accidents as well as the releases of radioisotopes are calculated and discussed. Station blackout simulation was conducted in this study. The resulting effects seen are consistent with other stress test station blackout tests used utilizing licensed simulation codes. An exact comparison is however not possible as the plants on which the simulations was done vary greatly and the limitations of availability to Chinese FSAR. PCTRAN/CPR1000 is an extremely useful simulation package that provides engineers and scientists very accurate feedback to how a nuclear power plant would react as a whole under various plant conditions. It is able to do this extremely fast as well. As a training tool PCTRAN/CPR1000 provides hands-on experience with many of the primary plant operations and develops an intuitive understanding of the plant Show less

Monazite is a light rare earth phosphate which is difficult to process when using conventional chemical processes. It is considered to be one of the most important commercial sources of thorium and lanthanides. Using conventional chemicals to process monazite have caused severe environmental damage as has been demonstrated in the Baotou region, China. Monazite is found in combination with other minerals in nature such as bastnaesite and xenotime. The conventional techniques and chemicals used… Show more

Monazite is a light rare earth phosphate which is difficult to process when using conventional chemical processes. It is considered to be one of the most important commercial sources of thorium and lanthanides. Using conventional chemicals to process monazite have caused severe environmental damage as has been demonstrated in the Baotou region, China. Monazite is found in combination with other minerals in nature such as bastnaesite and xenotime. The conventional techniques and chemicals used for the processing of monazite are expensive. In order to address this issue, South Africa would like to beneficiate monazite as part of its mineral beneficiation strategy. Doing so competitively would require a new cheaper and more environmentally friendlier process. A new method for the processing of monazite is currently being investigated. It is proposed to feed the monazite is fed into a plasma reactor to crack it. Potentially the cracking will allow the monazite to be more reactive and susceptible to react with less harsh chemicals. The plasma treated monazite is reacted with a fluorinating compound such as ammonium bifluoride. Ammonium bifluoride is used rather than fluorine or hydrogen fluoride as it is less dangerous to handle. The fluorinated rare earth mixture can now be separated using various methods. Show less

Early fault identification systems enable detecting and diagnosing early onset faults or fault causes which allow maintenance planning on the equipment showing signs of deterioration or failure. This includes valve and leaks and small cracks in steam generator tubes usually detected by means of ultrasonic inspec- tion.
We have shown (Cilliers and Mulder, 2012) that detecting faults early during transient operation in NPPs is possible when coupled with a reliable reference to compare plant… Show more

Early fault identification systems enable detecting and diagnosing early onset faults or fault causes which allow maintenance planning on the equipment showing signs of deterioration or failure. This includes valve and leaks and small cracks in steam generator tubes usually detected by means of ultrasonic inspec- tion.
We have shown (Cilliers and Mulder, 2012) that detecting faults early during transient operation in NPPs is possible when coupled with a reliable reference to compare plant measurements with during transients. We have also shown (Cilliers, 2013) that by correlating the fault detection information as received from distributed systems it is possible to diagnose the faults in terms of location and magnitude.
This paper makes use of the techniques and processes developed in the previous papers and apply it to a case study of the Three Mile Island accident. In this way we can determine how the improved informa- tion available could present the operator with a better idea to the state of the plant during situations where a combination of faults and transients prevents the operator and conventional systems to recog- nise the abnormal behaviour. Show less

Early fault identification systems enable detecting and diagnosing early onset faults or fault causes which allow maintenance planning on the equipment showing signs of deterioration or failure. This includes valve and leaks and small cracks in steam generator tubes usually detected by means of ultrasonic inspection.
We have shown (Cilliers and Mulder, 2012) that detecting faults early during transient operation in NPPs is possible when coupled with a reliable reference to compare plant… Show more

Early fault identification systems enable detecting and diagnosing early onset faults or fault causes which allow maintenance planning on the equipment showing signs of deterioration or failure. This includes valve and leaks and small cracks in steam generator tubes usually detected by means of ultrasonic inspection.
We have shown (Cilliers and Mulder, 2012) that detecting faults early during transient operation in NPPs is possible when coupled with a reliable reference to compare plant measurements with during transients. The problem introduced by the distributed application of control systems operating independently to keep the plant operating within the safe operating boundaries was solved by re-introducing the
fault information it into the measurement data, thereby improving plant diagnostic performance.
This paper introduces the use of improved fault detection information received from all distributed systems in the plant control system and correlating the information to not only detect the fault but also to
diagnose it based on the location and magnitude of the fault cause. Show less

Being able to detect and describe hardware deterioration in nuclear power plants benefits the nuclear industry tremendously as it would enable appropriate outage and maintenance planning. Being able to detect and describe this faulty behaviour also assists in fault analysis of nuclear power plants.

This thesis describes the development of narrow band dynamic operating envelope that makes use of real-time simulated plant measurements and control operations to compare with actual plant… Show more

Being able to detect and describe hardware deterioration in nuclear power plants benefits the nuclear industry tremendously as it would enable appropriate outage and maintenance planning. Being able to detect and describe this faulty behaviour also assists in fault analysis of nuclear power plants.

This thesis describes the development of narrow band dynamic operating envelope that makes use of real-time simulated plant measurements and control operations to compare with actual plant measurements and control operations. By simulating the plant behaviour in real-time whilst comparing it with the real-time transient the plant is following, a second set of plant measurements is generated. The newly generated plant measurements represent plant measurements if the control system did not introduce control operations to nullify the effect of the fault. Show less

With the fairly recent adoption of digital control and instrumentation systems in the nuclear industry a lot of research now focus on the development expert fault identification systems. The fault identification
systems enable detecting early onset faults of fault causes which allows maintenance planning on the equipment showing signs of deterioration or failure. This includes valve and leaks and small cracks in steam generator tubes usually detected by means of ultrasonic… Show more

With the fairly recent adoption of digital control and instrumentation systems in the nuclear industry a lot of research now focus on the development expert fault identification systems. The fault identification
systems enable detecting early onset faults of fault causes which allows maintenance planning on the equipment showing signs of deterioration or failure. This includes valve and leaks and small cracks in steam generator tubes usually detected by means of ultrasonic inspection.
Detecting faults early during transient operation in NPPs is problematic due to the absence of a reliable reference to compare plant measurements with during transients. The distributed application of control systems operating independently to keep the plant operating within the safe operating boundaries complicates the problem since the control systems would not only operate to reduce the effect of transient disturbances but fault disturbances as well.
This paper provides a method to adapt the plant measurements that isolates the control actions on the fault and re-introduces it into the measurement data, thereby improving plant diagnostic performance. Show less

The use of real-time plant simulators running in parallel with a nuclear plant to predict the control system behaviour and highlighting unexpected plant behaviour is presented. The study is performed on a
910 MW Generation II Framatome Pressurised Water Reactor model and simulator. By simulating the plant behaviour in real-time whilst comparing it with the real-time transient the plant is following, a
complete second set of expected control operations and simulated plant measurements is… Show more

The use of real-time plant simulators running in parallel with a nuclear plant to predict the control system behaviour and highlighting unexpected plant behaviour is presented. The study is performed on a
910 MW Generation II Framatome Pressurised Water Reactor model and simulator. By simulating the plant behaviour in real-time whilst comparing it with the real-time transient the plant is following, a
complete second set of expected control operations and simulated plant measurements is generated. This enables the calculation of the unknown set of variables introduced into the plant as a fault condition. The benefit of such a system is that plant faults that are too small to detect (especially during transients when the plant operating point is moving around) can be identified as unexpected or faulty plant behaviour.
The behaviour of the control system is also continually predicted so the effect of the control system compensating for fault symptoms (which in most cases hides the fault condition) is used to characterise the fault as a control variable acting on the plant. The approach is illustrated by simulating a specific fault, small enough to go undetected for an extended period of time, during a typical transient. This is continually compared with a plant simulation, simulating the same transient without the fault. Using the described methodology, the fault is detected and characterised long before the plant safety is jeopardised or the fault is detected by the conventional protection system. Show less

Eskom, South Africa's largest supplier of electricity, uses a vast telecommunication network to control and monitor all operations in remote locations such as substations. As Eskom expanded the electricity network across South Africa, the telecommunications network expanded as well. With changing technology in the telecommunications industry, the telecontrol network of Eskom uses different protocols and communication mediums.
This paper covers the study of these different protocols and… Show more

Eskom, South Africa's largest supplier of electricity, uses a vast telecommunication network to control and monitor all operations in remote locations such as substations. As Eskom expanded the electricity network across South Africa, the telecommunications network expanded as well. With changing technology in the telecommunications industry, the telecontrol network of Eskom uses different protocols and communication mediums.
This paper covers the study of these different protocols and mediums and the interconnectivity between them. The purpose of the study is to enable the network administrators of Eskom to easily time-synchronise all nodes on the network. Even more importantly, the study is done to better understand the setup of the Eskom telecontrol network and the delays that occur between different protocols and using different communication mediums. The study quantifies all delays that occur between nodes, considering distance between nodes, switching between mediums and processing time within systems. A network simulation tool is established that enables the network administrator to simulate the network and find all delays that occur in the network before the network is actually implemented. Show less