Two birds with one stone

Simulation technologies have always had a clear division between engineering applications, using extensively validated codes run in batch mode, and training simulators where real-time computation is a mandatory requirement. Thanks to the flexibility of modern simulation technology and increased computer performance, it now becomes possible to develop nuclear power plant simulators that can be used for both engineering and training purposes. The organization of such a project evolves in time, moving from flexible data and simulation platforms management in the first stage of the project (characteristic of an engineering simulator), and progressively freezing the reference data set and reinforcing the volume and scope of tests (characteristic of a training, or full-scale simulator, FSS). The great advantage of this approach is that it offers significant time and cost savings in the context of new plant build where ‘as built’ data are not available.

Roughly speaking, the first half of the project schedule is an engineering project and the second half of the project is a training simulator project, with two dedicated products at the output (engineering simulator and full-scope simulator), both available at the required time. This solution is possible for three reasons. First, the models used for the V&V projects are of similar scope and quality as those of FSS projects. Their accuracy can be refined by means of an additional tuning phase. This allows a smooth transition from a V&V simulator model to a FSS model. Second, the schedule of execution of a V&V project is such that only a few months after each data package transmission the platform is updated and tested. Therefore, the final version of the platform is ready sufficiently early to be an entry product to the formal FSS validation phases (pre-factory acceptance testing, FAT and site acceptance testing). Third, the global schedule of the V&V step-by-step design is roughly two years, and the validation phase of the FSS is one year. A combined project can therefore fit roughly in a 3-year schedule.

The combination of both systems brings six major advantages to the customer. First, the engineering simulation platform is available throughout the design process, leading to a significant increase in efficiency of engineering work, in the possibility to validate design hypotheses and detect defects and conflicts early. Second, the budget of the combined project is significantly lower than two separated projects as an important part of the models is common to V&V and FSS. Third, the progress of the project can be monitored very accurately by the different stages of V&V in the early phases of the project. Difficulties in model accuracy or data voids can be identified early and handled properly. Fourth, the quality of the models is very well-controlled, because at each stage of the V&V phases an acceptance test is performed. Fifth, the data and model tuning consistency between the V&V simulator and FSS is guaranteed. Sixth, the availability of an engineering station with a specifically designed man-machine interface (MMI) and functional features for the V&V simulator will enhance the efficiency of the validation phase of the FSS.

Implementation strategy

The overall project implementation is split into four main phases.

The phase 1 simulator will be used by the architect-engineer for defining and validating the operating rules in case of incidents or accidents. It comprises nuclear island process modelling, plus the corresponding I&C models, malfunctions and local commands. The instructor station and operator station will be in prototype version with MMI-enabling monitoring and control of the simulator.

The phase 2 simulator takes the actual instrumentation and control into account in its simulation of the performance of this equipment. Its use will enable the engineers to specify and validate normal, incident and accident operating procedures. Its scope, still limited to the nuclear island, is extended to include the update and addition of detailed design data on the main systems, the I&C, and a high-fidelity simulation of the operator workstation.

The phase 3 simulator will validate both the entire set of plant operating procedures (in both normal operation and during incidental and accidental transient) and training of the future plant operators. It will include full-scope modelling covering the entire plant and of the instrumentation and control, and simulation of the control room back-up hard panels.

During a fourth phase, the I&C simulation can be substituted by a digital control system (DCS) emulation once this is available from the DCS supplier. The simulated main control room can be removed and replaced with a real-plant replica.

Pascal Gain is vice president – power simulation at Corys T.E.S.S., 74, rue des Martyrs, F-38027 Grenoble Cedex 1, FRANCE. Jody Ryan is CEO of Corys Thunder Inc, 107 Industrial Drive, Building E, St. Marys, Georgia 31558, USA.