- 學位論文
龍門電廠起動測試暫態研究
Start-up Test Transient Analyses for Lungmen ABWR Plant
若您是本文的作者，可授權文章由華藝線上圖書館中協助推廣。
摘要
本論文研究目的，在於建立合適且精確之龍門電廠RETRAN輸入模式，並進行起動測試暫態之預測分析。龍門電廠RETRAN模式起動預測分析結果，除了可以提供電廠工作人員起動測試前，預先瞭解系統參數行為與暫態反應，作為測試過程決策之依據；更可以提供管制單位與台電公司審視、平行驗證燃料廠家分析數據之憑據。本項研究並植入龍門電廠主要控制系統模組，包括了：再循環水流量控制系統、飼水控制系統、以及蒸汽旁通與壓力控制系統。為了確認控制系統模組之正確性與最佳化，進行控制參數（設定點）階變校驗，以及根據相關控制系統運作之暫態進行評估。分析結果顯示：控制系統模組之反應都能符合規範之要求，並且已經調整控制參數設定值至最佳化設定。 本研究以龍門電廠最終安全分析報告（FSAR）之中度頻率暫態為對象，共計進行了八項暫態校驗，包括了：(1) 喪失飼水加熱、(2)飼水控制器失效達最大需求、(3)單一汽機控制閥關閉、(4)發電機棄載無旁通、(5)汽機跳脫無旁通、(6)所有主蒸汽隔離閥關閉、(7) 喪失飼水流、以及(8)所有爐內泵跳脫暫態。由相關案例之系統暫態趨勢與熱限值（ΔCPR）結果比較，可以確認所有案例的主要爐心與系統參數反應，均有一致之趨勢與接近的數值，而最限制暫態之案例也有相同結論（單一控制閥關閉暫態）。在喪失飼水加熱暫態分析，證實RETRAN 的確具備此類反應度變化暫態之分析能力，可以完整驗證反應爐實際的動態變化。龍門電廠RETRAN分析模式之控制閥操作採取半開模式（Partial ARC），較FSAR之單一控制閥關閉暫態（全開模式，Full ARC），有較為限制的熱限值結果。整體而言，FSAR 暫態驗證結果可以確認RETRAN 分析模式，確實具備龍門電廠暫態分析能力。 至於，龍門電廠起動測試之預測分析研究，本論文選擇了九項起動測試，這些起動測試包括了：冷卻水溫度降低之測試（喪失飼水加熱）、飼水流量降低（飼水泵跳脫）、增壓類型測試（汽機跳脫及負載棄載與急速降載）以及爐內泵跳脫等。本項研究之起動預測分析結果，皆可以確實掌握系統參數之變化；相關預測分析結果與靈敏度計算可以提供龍門電廠進行起動測試之參考依據。一台飼水泵跳脫測試，當MDRFP啟動失效情形發生，反應器產生L3急停訊號；即使考慮最小棒位線效應（Rod Line Effect），反應器低水位急停訊號雖未產生，然而測試規範之水位餘裕要求仍未符合。三台爐內泵跳脫測試，爐內泵之馬達慣量使用硬體最小值（19.5 kg-m2）恰巧違反三台爐內泵跳脫之泵轉速遞降要求，至於分析最小慣量值（17.5 kg-m2）之結果則符合規範。所有起動預測分析結果都可以符合測試規範之要求。 本論文建立了龍門電廠RETRAN 最佳化估算以及安全分析模式，相關植入模式之主要控制系統，經過階變測試與暫態分析之驗證，確實具備暫態分析能力。惟諸如：汽機控制閥之動態模擬、反應器爐槽水位計算之精確性、爐內泵之運轉特性、SCRRI 設計功能之驗證…..等等工作內容，必須等待龍門電廠進行起動測試或商業運轉後，才可以取得相關數據進行模式之精進。未來龍門電廠發生與FSAR暫態不盡相同之運轉暫態時，本研究建立之RETRAN 模式仍可提供作為評估工具。在修改相關數入數據之後，可以迅速提供此類暫態之系統參數變化趨勢，增加台電公司或核管單位對事件之瞭解，解除相關疑慮。甚或龍門電廠必須進行設備或設計變更評估（例如設定點變更…等）工作時，也可以利用RETRAN 程式分析變更案對安全限值之影響，瞭解安全餘裕的變化程度。
並列摘要
This paper is to develop an applicable and precise RETRAN input for Lungmen Plant in order to predict the behavior of transient in Start-up Test Program (STP). The predictions of RETRAN can be referred to acquaint the staff of STP with the system responses in advance before the actual test, which is the basis to evaluate the appropriate decision during the processes. And it also provides the Regulatory and TPC a parallel tool to benchmark the results of Vendor. In this study the LRM is transferred into an advanced version of RETRAN-3D, and the major control systems are remodeled in RFCS, FWCS and SBPCS. Several verification and validation calculations are conducted for individual control system according to how the manipulation of the control system is close to the related transient. The benchmark results show that the responses of control modules are accepted in acceptance criteria and the Gain setting are already tuned to the optimal value. In this dissertation eight moderate transients in Lungmen FSAR are used to benchmark the ability to evaluate the system responses of Lungmen, including: (1) Loss of Feedwater Heating, (2) Feedwater Controller Failure with Maximum Demand, (3) Single Control Valve Closure in Fast Mode, (4) Generator Load Rejection with No Bypass, (5) Turbine Trip with No Bypass, (6) Reactor Full Isolation, (7) Loss of Feedwater, and (8) All RIPs Trip. From the FSAR benchmark, the RETRAN results are highly in agreement with the FSAR and come to the conclusion of the identical limiting transient (Single Control Valve Closure in Fast Mode). In LFWH, the capability of RETRAN to analyze the slow transient has been verified and can predict the dynamic behavior in transient. In single control valve closure transient, the part arc mode of TCV in LRM is different from the full arc mode in FSAR analysis. The different operation mode leads to the more conservative thermal limit in RETRAN than that of FSAR. On the whole, it is concluded that the LRM is certainly able to catch the trend of the system parameters of core in transient. In the predictions of STP, there are nine tests selected to be analyzed, including those transient type of: (1) decrease in reactor coolant temperature, (2) decrease in reactor feedwater flow, (3) increase in reactor pressure, and (4) RIP trip. The prediction analyses of STP show that the results of RETRAN can exactly catch the system responses to the transient, and the sensitiveity studies can be employed to a reference to the staff of STP. In one feedwater pump trip test the low level scram will be actuated in the case of no auxilliary MDRFP function. In the prediction results of three RIPs trip show that the coastdown speed of tripped RIPs with the minimu hardware inertia of motor/generator is just against the requirement for the coastdown speed, while the case with the minimum analytical inertia comes out to the acceptable coastdown rate between the bounding curves. This dissertation develop the best estimate and safety analysis RETRAN input for Lungmen Plant, and implant the major control system modules into the model. The functions of the control modules have been identified to fulfill the design requirement through the individual step change benchmark and related transient evaluations. But there are still several points like water level prediction, the characteristic of RIP, SCRRI feature etc., which needed to upgrade or verify against the STP database or plant operational data. In general, the LRM show the good agreement with the FSAR cases in transient responses. Based on this conclusion the LRM can be employed to support the operational transient analysis and even the evaluation for equipment or design change cases if the model benchmark further against the appropriate plant data after Lungmen plant has been commercialized.
並列關鍵字
無資料參考文獻
1. Takaya SEKO et al, “Modern Technology Applied in Advanced BWR (ABWR)”, Journal of NUCLEAR SCIENCE and TECHNOLOGY, Vol. 25 No. 11, pp. 837~847, 1988
17. Taiwan Power Company, “Final Safety Analysis Report for Lungmen Nuclear Power Station Units 1&2”, August 2007
22. 臧逸群，「進步型沸水式反應器運轉暫態熱水力特性研究」，國立清華大學工科系博士論文，中華民國九十九年
37. Taiwan Power Company, “LUNGMEN NUCLEAR POWER STATION STARTUP TEST PROCEDURE- Turbine Trip and Load Rejection”, STP-31-HP, Revision 1, 08/07/2008
40. S.S. Ma, C.K. Shih, Y.R. Yuann, “RETRAN analysis results of feedwater pump trip transient for Lungmen ABWR Plant”, Annals of Nuclear Energy, 38 (2011), 1598–1608