Skip to main content
Nuclear Safety Cooperation

R2.07/95 Functional Approach Assessment

Status
  • Closed
All Countries
Benefitting Zone
Worldwide
€ 585,430.45
EU Contribution
Contracted in 1998
TACIS
Programme
Technical Assistance to the Commonwealth of Independent States

Details

Type of activity

Design Safety

Nature

Services

Contracting authority

European Commission

Method of Procurement

Direct Agreement & AV DA

Duration

02/03/1998 - 02/02/2000

Partner

Rosenergoatom

Contractor

ANSALDO ENERGIA SPA

Project / Budget year

WW9508 Nuclear Safety 1995 / 1995

Background

In addition to existing operator information systems (parameter display, fixed alarm systems), an Early Fault Diagnostic System (EFDS) may be considered for a nuclear power plant. The EFDS would report to the operators any deviation from normal plant parameters along with suggestions on the possible causes. This would give the operator an opportunity to take appropriate and timely actions to control and limit any adverse consequence of the identified fault, malfunction etc., before any threshold is exceeded leading to a potentially dangerous plant condition.

Such a system, however, is achievable only by using a plant simulator, which would be able to adequately reproduce immediate plant response to a deviation and would have the capability to quickly generate and evaluate a large number of possible evolution scenarios. The incidents, which should be considered, may include leakages, flow/pressure disturbances, control systems malfunctions including sensors, component failures etc.

To help developing the EFDS for VVER-1000 type nuclear power plant, the R2.07/95 project was established. It was implemented in several phases:

  1. Phase 1 – Functional approach assessment of a frame product development
  2. Phase 2 – Conceptual definition of the frame product
  3. Phase 3 – Early Fault Diagnostic System, frame product development

As the last phase, the EDFS implementation at a pilot plant was foreseen out of the scope of the project R2.07/95.

The Phase 1 has been covered by a separate contract No. 24895 (formerly 97-0410). More information on the contract scope and results can be found in the related summary

Objectives

The contract No. 25306 covered specifically the Phase 3 of the R2.07/95 project – “EFDS frame product” development.

As the “frame product”, a representative version of the EFDS, a “demonstrator” was meant, which had to be incorporated in a plant simulator, including a man-machine interface (MMI). The demonstrator had to represent a reduced-scope EFDS suitable both for the personnel training in operation of an EFDS and for use as a self-contained EFDS.

The “reduced scope” of the EFDS has meant that only the following systems were to be covered:

  • Primary circuit;
  • Make up / let down system with some auxiliaries;
  • Pressurizer system;
  • Simplified secondary circuit (including main turbine);
  • Main associated control systems (analogue and binary).

Moreover, only certain power operation modes (30-100% power) were considered for the demonstrator, only safety-significant initiating events were selected and the system could be partly generic (i.e. not specifically representing VVER-1000).

The contract No. 25306 was aimed at development and testing of an EFDS software package relevant to VVER-1000 and its integration in a plant simulator as well as on supply of an appropriate simulator work station.

Results

The contract was implemented by the ANSALDO Nuclear Division with the support of local subcontractors: Atomenergoprojekt Moscow (AEP) and Institute of Control Sciences Moscow (ICS). The project started on 2 March 1998 and was originally planned for 12 months. However, due to difficulties in delivery of the EFDS hardware to Russia, it had to be extended by 10 months and the project was finally completed in 22 months.

The main project result was development, validation and documentation of the EFDS demonstrator software (VVER-1000 plant simulator models) including MMI, which was then installed at target workstation in Moscow. The project results were in full accordance with the original plan, including objectives, performed activities and outputs.

The project work was divided in 5 tasks:

  • Task 1 – Preliminary activities (general design specifications, data package definition and collection, identification of initiating events)
  • Task 2 – Process simulator development and testing (adaptation of existing simulation codes, development of plant system and process models, MMI development, simulator integration, testing and documentation)
  • Task 3 – Diagnostic system development and testing (development of communication interface with the simulator, development and testing of diagnostic rules, development of diagnostic system MMI, diagnostic system integration and documentation
  • Task 4 – EFDS integration and validation
  • Task 5 – Preparation of plans for EFDS project completion (activities to obtain Rosenergoatom and GAN approval for EFDS implementation at Balakovo NPP, preparation of detailed activity plan and cost estimate for the implementation).

An EFDS demonstrator has been developed and documented; it was installed at the target workstation at ICS in Moscow, applying the “simulator + expert system” approach and sharing the MMI with the plant simulator. Two SW packages have been developed: a reduced-scope simulator suitable for the EFDS training and the EFDS itself including the MMI. The target workstation (WS) has been supplied as a part of the project. While installing the EFDS software in the WS and its integration with the reduced-scope VVER-1000 simulator models in October 1999, the EFDS demonstrator was tested, satisfactorily validated and properly documented. The EFDS rules have been developed based on results of a training performed at the VVER-1000 reduced-scope simulator. Finally, the plan for EFDS prototype installation and testing at a pilot VVER-1000 plant has been produced, including inquiry at the Russian Regulatory Authority (GAN) concerning the associated regulatory requirements.
Within the project work, AEP provided the technical data on VVER-1000 plant necessary for the simulator model development and identification of initiating events (knowledge base of the diagnostic system). The ICS developed the VVER-1000 specific diagnostic system, including the MMI and operating rules and incorporated it in the simulator models provided by ANSALDO. The cooperation between the Contactor and its Russian counterparts and subcontractors was very good.

The project had to be extended as a result of enormous delay in the target workstation hardware delivery due to customs procedural difficulties (beyond Contractor’s control). Instead of in June 1998, the HW was delivered only in mid July 1999. The Contractor has therefore recommended for the future projects to avoid import of any standard hardware to Russia and procure it rather directly in Russia through a local partner, which could yield big time and cost savings. Another reason for the delay was unforeseen problem in the simulator integration requiring development of more complex communication interfaces between ANSALDO and ICS software.

The project results have been presented to Rosenergoatom and GAN personnel at an annual meeting in Podolsk on 2 December 1999, involving participants from all Russian NPPs and other Russian nuclear organisations and research institutes. The position and the role of the EFDS in the plant monitoring and control systems were widely discussed. In addition to the early diagnostic function, the dual-monitor design and multi-window MMI of the EFDS, developed within the project, was highly appreciated and inquired for possible wider application at NPP main control room (MCR) within the envisaged overall reconstruction and modernisation of MCR at Russian NPPs.

The system in general proved its adequacy in alerting operator on parameter deviations and providing information of possible causes for typical plant process malfunctions, as indicated in the testing report. However, the EFDS applications have been in some cases limited due to lack of appropriate sensors to give enough information for detecting the malfunction. It has been moreover discovered that the automated diagnostics may be sometimes too straightforward, not always providing useful information to the operator. Due to generally positive results, however, the Contractor recommended to the EC to support the finalization of the current project work in the frame of next TACIS projects.