Massimo Sepielli

Using a Hybrid FPGA-CPU Platform for a User-Friendly Environment in Control Systems Design: A Possible Application to Nuclear Plants

In this work, in-depth considerations about FPGA-CPU architectures design methods are presented together with some possible plant applications. FPGAs are used because of their potentials in guaranteeing parallelism, synchronism and high reliability tasks; on the other hand CPUs are crucial to higher precision computations and to reproduce data and results on a graphical interface, thus resulting in an improvement in the human-machine-interface (HMI). Design choices are here reported to justify the benefits from the use of both CPU-FPGA approaches. Taking advantage of these features, the LabVIEW environment and its digital platform is expanded in order to develop a novel tool able to support designers during the fundamental phases of a supervision and control tool realization: from the model development to the digital hardware implementation and testing. LabVIEW is able to communicate with several languages like Matlab or Scilab for a theoretical model study and simulation, and VHDL for a digital implementation, taking the advantages from both development environments.Main results have been reported concerning studies of control systems in nuclear plant facilities, showing how a full-digital platform can be very useful in developing monitoring and control instrumentation. This new tool is thought to improve the so-called hardware in the loop (HIL) simulations and to accelerate the process of prototypes implementation and testing, using a more accurate and reliable environment in terms of performance and safety.Copyright

Digital Nonlinear Control for a Pressurizer in a Pressurized Water Reactor

Studying criteria ensuring better implementations of control laws and logics on digital programmable devices is today a crucial point in order to improve performance and safety in nuclear plants of novel design. To this aim, a mathematical model of a PWR is considered, describing the dynamics of the primary circuit. This model is simple enough to allow the determination of a control policy, but it is also sufficiently accurate to capture the nonlinear, time–varying, and switching nature of the system. A dynamic controller for the pressurizer pressure and water level is hence designed. Its main advantage is the reduction of the control response. This latter is a classical drawback, which may cause some troubles to pressure control, due to the long response of temperature sensors.The digital implementation of the controller has been taken into account to study the important aspects of the influence of the implementation on the performance of the control law. A particularly effective and flexible environment where it is possible to analyze these aspects is Matlab/Simulink, a general–purpose tool for analysis and simulation of multi–domain dynamic systems, practically a standard in control implementations, also in industrial applications. The designed controller ensures a good performance when applied to the model used to derive them, also in the presence of unmodeled uncertainties and disturbances. Its nonlinear switching nature, reflecting the real pressurizer dynamics, ensures better transient behaviors. Since it contains a PI action, it represents an evolution and an improvement with respect to classical PID controllers, usually implemented in standard control actions. To better analyze the control performance, three steps have been considered: first, the designed nonlinear controller has been compared with a linear one. Second, it has been tested on a more detailed model, with more realistic dynamics. Finally, the digital implementation of the controller has been simulated, in order to optimize the electronic implementation on the FPGA–CPU hybrid platform.A simulation study has been conducted with sampling times in ascending order, to further test its robustness against delays. LabVIEW development tool and technology has been chosen to create the electronic layout and to obtain preliminary results in terms of timing, resources availability and power consumption. Due to the same graphical mode of programming, an immediate link between Matlab/Scilab and LabVIEW is possible in order to optimize the use of each language where it guarantees the best performance.The methodology here presented could be very useful to better integrate the different features of control engineering and IC design, thus giving the possibility of realizing more accurate simulations and permitting to perform a hardware in the loop (HIL) testing of a wide variety of control algorithms.Copyright

The Role of FPGA-Based Architectures in the Control Room Modernization Process: Preliminary Results of a Case-Study

In the context of a plant modernization, developing digital I&C technology is a crucial challenge to improve nuclear plants safety and reliability. Digital technology is usually oriented to achieve functions such as plant control, monitoring, simulation and protection in a user-friendly way. On the other hand, the analogue instrumentation implemented in the so-called “old generation consoles” is often essential and not immediately or completely replaceable. As a consequence, the interaction between the analogue and digital data seems to be a necessary step before starting the digital I&C licensing process. The fundamental difference between analogue and digital technologies relies on the fact that digital logic is based on processors, hence it can be customized by programming its software. However, introducing new code can result in a new set of potential failure modes to be accounted for. As a consequence, original analogue systems mostly assure a higher level of protection with respect to digital systems. In this scenario, a benefit could arise from the use of Field-Programmable Gate Arrays (FPGAs), based on a hardware architecture whose routing is made via software, thus resulting in a variety of possible tasks. FPGAs’ employment ranges from automotive and industrial applications, ASIC prototyping, software defined radios, radar, image and DSP. In this work a critical analysis of FPGA fundamental features and potentialities in nuclear plant I&C design is achieved in conjunction with some practical applications. Troubles arising from coping with processor-based system are presented and compared to benefits and potentialities offered by FPGA real-time architectures: indeed, FPGAs comprise a higher number of logic blocks and functions able to manage parallel processes with self triggering, and provided into a “non-frozen” structure but easily reconfigurable. This characteristics of being in-system programmable (ISP), i.e. a device capable of being programmed while remaining resident in a high-level system, can be considered as the main advantage of using FPGA. The employment on a large scale is also justified by its high determinism and testability, leading to high performance in terms of reliability. As a case-study, we propose a supervisory full-digital system that has been designed, realized, tested and validated implementing a FPGA architecture to be used in parallel to the TRIGA nuclear reactor RC-1 analogue console at the ENEA Casaccia Research Centre in Rome. We report on the design choices, and on pros and cons of using FPGA instead of the classical processor-based architectures. This preliminary apparatus has been developed using the LABVIEW environment and FPGA-based technology, an appropriate tool to get across simulation to Hardware-in-the-Loop (HIL) technique allowing to move on production from prototyping.Copyright

Influence of cognitive human factor on nuclear reactor safety: a simple decision support system for operators in emergency conditions

In this paper a prototype for the TRIGA Research Reactor Decision Support System is proposed. Its framework is realised on the basis of three models: the Universal Management Paradigm, the Information-Preference-Knowledge Model from the TOGA meta-theory, and the Finite State Machine Model. Such an application uses data from the instrumentation and control system and according to the human factor state recognises the type of operative situation and suggests to the operator a decision action as an output. The tool can be used for helping the operator in accident conditions. Here, for illustrative purposes only, an accidental variation of the control rod position has been simulated and results from the proposed tool are analysed.

Using the Reflectometric technique for monitoring the boiling flow dynamics in a once through steam generator

In control systems design for nuclear facilities, the correct choice for sensors, transducers and actuators is not an easy task when different options must be evaluated. In particular, for a Once Through Steam Generator (OTSG), the control of its boiling flow dynamics is usually performed with sensors that may be slow in response, must be placed all through the zone of interest (or need heavy mechanical modifications) and may not distinguish correctly when the liquid/steam interface is not so well defined at high temperatures and pressures. In this paper, a theoretical study about a reflectometric technique applied to an OTSG is proposed. This technique can be used also for the liquid/steam levels monitoring of the boiling flow. The overall behaviour of the variables of interest and the first theoretical results show the benefits of such an innovative approach.Copyright