Summary

This project addressed the assessments related to design and operational safety investigations performed for the PAKS-VVER 213 project. In order to improve the safety situation, a Hungarian Technical Safety Organization proposed a near term safety improvement program based, among others, on the IAEA recommendations given for VVER 440/213 reactors. Such a program and its implementation was reviewed by the Hungarian Nuclear Safety Authority in order to verify that the plant will achieve an acceptable safety level to be allowed to operate further.

The intent of the project was to cooperate with and to assist the Hungarian Nuclear Safety Authority and its Technical Support Organization in licensing assessments of the improvement program of the PAKS NPP Units 1 to 4, and in particular in primary system depressurization methods following steam generator primary collector break.

On the basis of already elaborated results and relevant information which has to be provided from the AGNES project a. o. the material was be reviewed and assessed against further relevant analytical results or experimental data.

The general objective consisted of 3 tasks:

• Task 1:Based on already performed analytical results possible primary pressure reduction (PPR) procedures following a steam generator primary collector rupture should be assessed.
• Task 2: Detailed analysis of the most effective procedures by PMK experiments and code calculations with the aim to demonstrate their applicability.
• Task 3: Guidelines for the development of an emergency operating procedure for steam generator rupture based on results of Tasks 1 and 2.

In addition 3 tests were been performed at the PMK test facility:

• Experiment on the PMK facility for the full opening of the steam generator collector cover with the conditions defined in Task 1 and code validation on the test
• />Two experiments on the PMK facility for the rupture of one (1) heat transfer tube and ten (10) heat transfer tubes to verify that these events will not lead to more serious consequences than the full opening of the steam generator collector cover and code validation on the tests.

The main technical results according to the 3 tasks are given below:

Task 1:

The work consist of two parts: Assessing the PPR procedures and additional recommendations of automatic and manual actions for PPR.

The following system function failures have been regarded:

• Failure of the steam safety valve at the damaged SG in open and closed position; Failure of the steam dump valves in open and closed position;
• Failure of the feedwater system to start or to stop injection;
• Failure of safety system to start or to stop injection to primary side;
• Failure of the main steam isolation valves to close;
• Failure of the let down line to close;
• Failure of the pressurizer emergency spray.

Additional recommendations of automatic and manual actions for PPR:

• The isolation of the steam line should be automatically reset and safety injection systems should be shut off, if the pressure at secondary side of the damaged SG exceeds the design value.
• The recovery strategy in case of a stuck open safety valve should be a fast cool down and a primary pressure reduction by shut down of safety injection system trains. There should not occur neither core uncovery, after the loss of the subcooling margin nor a vessel failure due to brittle fracture. Manual actions are necessary in order to control the cool down, the shut down of safety injection systems and the pressure reduction by spraying.
• To prevent deboration and pressurization of the primary side during cool down a procedure based upon manual measures has to be defined in order to shut down the feed water supply by closing the lines or by disconnecting the main-, auxiliary- feed water pumps and by taking into operation the emergency feedwater supply to single SGs.
• A PRISE leak in combination with a total loss of feed water or a steam release from the steam safety valves only can be coped with bleed and feed of e.g. the secondary side. Not only the procedures for bleed and feed have to be developed, verified, validated and trained, but also actions against the pressure rise and level increase at the secondary side after the initiation of the safety injection system have to be defined.

Licensing of nuclear power plants is based upon deterministic and probabilistic safety analysis. The deterministic design criteria for primary to secondary (PRISE) leakage resulting from steam generator failure are in agreement with international standards. But finally there have not been determined any probabilistic criteria in the AGNES project. In the course of the Phare 2-02/94 project a nearly complete set for principles and strategies has been defined for small and large PRISE in combination with necessary plant modifications. The assessment of system function failures needed for the recovery at PRISE leaks has shown, that the primary pressure reduction procedures are quite dependant from the event sequence. A stuck open steam safety valve at the damaged steam generator as the single failure in case of the postulated accidents is most important for the conservation of the design limits. An alternative to reduce the radioactive release e.g. at low pressure should be taken into consideration. More attention has to be given to the consequences of the single failure, that the feedwater supply to the isolated SG cannot be or has not been shut off. Special attention should be paid after the SG-isolation also to the failure of the two safety valves at the damaged SG in order to avoid overpressurisation of secondary side components. An automatic rest of isolation and shut off of the high pressures safety system should be performed under these conditions. The recommendations for a remote operation of the steam safety valve for an alternative steam release and the use of the TK-system for compensating the leakage are supported. If the make-up pumps of the TK-systems will be also connected to the emergency power grid and to borated water storages, then it can be used as an auxiliary spray system for the primary pressure reduction

Task 2:

Guidelines for the development of an Optimal Response Procedure for PRISE events (Small PRISE procedure, Large PRISE to include collector cover lifting) have been defined taking into account the results of the previous tasks performed in the PHARE/Hu/TS101 and previous activities performed by AEKI and IVO.

The guidelines have been based on the assumption that the following plant modifications are implemented:

• A N-16 detection system has been implemented (one detector for each steam line);
• A Reactor Coolant System subcooling margin measurement is continuously available in the control room;
• Provision for automatic isolation of the steam generator on high water level (this signal also trip the main coolant pump of the affected loop and pressurizer heaters);
• Provision of means of emergency pressurizer spray using HPSI pumps;
• Provision of a means of terminating HPSI from the main control room when criteria for termination are met;
• Increase in the water reserves for the emergency injection (borated water from containment bubble trays);
• Protection system modification to avoid tripping of the TK pumps when offsite power is available to permit accident mitigation with the Main Coolant Pump operable.

Those improvements have already been discussed in the frame of PHARE PH 2.02/94 and are strongly recommended for the mitigation of PRISE events.

Other improvements (e.g. Variable set-point Power Operated Relief Valve for each steam generator, qualification of the MSL upstream of the MSIVs for water solid operation) are still strongly recommended since their implementation significantly reduces the frequency of loss of secondary system integrity coincident with a PRISE event.

In particular, the implementation of power operated relief valves improves the cooldown control in the isolated loop providing a major flexibility during the PRISE recovery.

In addition to the above, it is suggested to implement collector cover design improvements, as suggested by Russian Gidropress. The design solution permits to decrease the equivalent diameter of the leak size from 107 mm to 37 mm, Figure 5-6. This will provide the operator with a much larger time to detect and mitigate the initial PRISE event and so decrease the possibility of concomitant events.

The guidelines have been developed following a symptom based approach. Optimal recovery guidelines have been outlined. Furthermore, it is recommended to link them to more generic Critical Safety Function Restoration guidelines to be sure that the operator is properly guided to face the possible contingency assuring the respect of all the critical safety functions while is recovering the plant from a PRISE event.

Task 3:

The work of task 3 consisted of several subtasks. The three most important ones are listed below:

• Evaluation of report PH2.02194
• Evaluation of specification and Pre-test analysis report for PKM-2 tests
• Evaluation of the PKM-2 experimen
  s