網路論壇高飛鷂和蘇志銓溝通『數位儀控系統是不是核四的罩門？』

Please send this email to the People of Taiwan (not just for Senior Gao (NE1971)).
This email is long, but please read it completely before you jump to conclusions.
I sincerely hope someone in Taiwan can translate this email to and type it in Traditional Chinese for the People of Taiwan because I cannot type Chinese using PCs.

常識與學位之-4 ~~ 勿把台灣老百姓當傻瓜 (September 03, 2013)
*這篇文章 (this email) 歡迎登載傳閱，作者本人負起所有責任 ~~ 電機博士蘇志銓
Tsing-Hua University, NEB1976, NEG1980

Some people do not think it is very dangerous of using Digital Network Real Time Control (with 400,000 input signals) on the TaiPower Fourth Nuclear Power Plant because they do not really understand the technical issues on this.
Please read this long email to understand why I have been so worrying about the very dangerous designs over the past 10 years on this.
Especially, Dr. Way Kuo (NE1972), Dr. Chris Wang (NE1976, see the middle of this email) and those self-claimed Nuclear Engineering (NE) experts thought they could guarantee the Safety of the TaiPower Fourth Nuclear Power Plant without detailed, unbiased, non-conflict-of-interest, and non-cover-up (non-making-data) testings.

South Korean nuclear power plants (over 30 including those will be exported to Indonesia and Middle East ) still use Analog Control Systems (not Digital Network Real Time Control Systems) today.

Digital Network Real Time Control Systems have been used for Chemical Plants, traditional power plants and others, and I knew this.
When Digital Network Real Time Control Systems for those chemical plants, traditional power plants and others do not work correctly, the whole systems might just be shut-downed and restarted hours or days later (such as Taiwan High Speed Train Incidents several months ago in Taiwan due to unknown causes of both Primary and Secondary Control Systems).

However, when Digital Network Real Time Control Systems (Primary and Backup Systems) with 40,000 signal inputs for the TaiPower Fourth Nuclear Power Plant crash (I use the word "crash"), it will be disasters because the Cooling Water Circulations (or even the Control Rods of the Nuclear Reactors) might not work, too.
Why do I use the word Crash?
The Software for the Digital Network Real Time Control Systems with 40,000 signal inputs will be big and very complicated because there are Combinations of the 40,000 input signal Real Time Control Conditions (not just 40,000 Real Time Control Conditions) to cover all Real Time Control Conditions.

Japan has four nuclear reactors using Digital Network Real Time Control Systems (with 28,000 input signals) successfully, but a lot of countries including France and South Korea FAILed on this and gave up.
About 10 years ago, when I knew the TaiPower Fourth Nuclear Power Plant would use Digital Network Real Time Control Systems, I asked why they would not ask Japan to Design the new Digital Network Control Systems for the TaiPower Fourth Nuclear Power Plant.
Remember, I knew Senior Dr. OuYang (NE1971), and I recommended him for the head of Taiwan AEC position, via the Honorable Dr. Yuan-Tse Lee, in 2000.
I still have the copy of the letter, and some NE-ESS Alumni saw the letter in my house, too.

The answer was "because Japan asked too much money (know-how costs)".
Then Taiwan Power Company, AEC, and GE (with GE permissions) hired a lowest bid (US$ 50 millions, instead of US$ 300 millions) of a small, unqualified, and then bankrupted US consulting company to do the design jobs.

Japan has real experience on this with just 28,000 input signals.
But, Taiwan Power Company, AEC and GE thought they would be smarter than Japan , and then they could create this World-Number New Digital Network Real Time Control Systems with 400,000 input signals by this small US consulting company.
They did not have any real experience on this (not like Japan ), and they did not understand why France and South Korea failed on this and gave up.

Just like one Stupid retired senior vice president (not NE Alumni, I think) of Taiwan Power Company said "Taiwan Power Company could complete the TaiPower Fourth Nuclear Power Company without GE", they have done many stupid things over the past 20 years:

Firstly, they chose to use Digital Network Real Time Control Systems (with 400,000 input signals) on the TaiPower Fourth Nuclear Power Plant without any real experience on this.
GE was really happy on this because GE finally found a Stupid White Mouse to test the World Number-One New System to gain real experience for GE future business.
If it failed, GE would not get blames because the stupid designs were done by the small unqualified and bankrupted US consulting company, Taiwan Power Company, Taiwan AEC, and some of NE-ESS professors.

Secondly, Taiwan Power Company, AEC and some NE-ESS professors changed over 400 GE designs (not Digital Network Real Time Control Systems) without GE permissions (because that Stupid senior VP claimed Taiwan (not just Taiwan Power Company) could complete the TaiPower Fourth Nuclear Power Plant without any GE helps).
They even changed the GE structure designs (such as material used in welding the pipes, the steel rods, the thickness of the walls, and others).

Thirdly, they did not go through really detailed testings (and then even making data to cover their mistakes), and then they claimed (and guaranteed) that Total Safety of the TaiPower Fourth Nuclear Power Plant.
Those people including the current head (NE1973) of the TaiPower Fourth Nuclear Power Plant and Dr. Way Kuo who is a reliability expert.

Dr. Way Kuo admitted (on Page 182 of his recent book in Traditional Chinese published in Taiwan, May 2013) that it is very difficult (or even impossible) for the Software Reliability of large and complicated software projects (such as the Digital Network Real Time Control Systems for the TaiPower Fourth Nuclear Power Plant).
How could Dr. Way Kuo and those self-claimed Nuclear Engineering (NE) Experts guarantee the Safety of the Fourth Nuclear Power Plant without detailed unbiased and non-conflict-of-interest, non-cover-up testings?

Finally I got some kinds of answer form Dr. Chris Wang (NE1976) who is a NE professor in the USA .
Please read the emails from Dr. Chris Wang and my response to his email below.
==============================
Jim,

I am not worried about testings and retestings as long as there are truly needed. My worry is that the retesting saga is nothing but a strategy (or an excuse) used by the anti-nuke and anti-government freaks to delay and eventually abolish the construction of the Lungmen project. The word "guarantee" cannot be interpreted in the absolute sense. So, here is my question to you: Will you support completing the Lungmen project if TPC passes all the retestings?

Chris
====================================
Chris, Chairman Yeh and All NE-ESS Alumni,

If they cannot correct all messes, then the TaiPower Fourth Nuclear Power Plant should be converted into a natural gas power plant to dynamically support the peak hours of the on-demand power needs during Summer in Taiwan .

The converting costs are almost the same as or less than completing the non-hope and unsafe TaiPower Fourth Nuclear Power Plant.
Such conversions occurred in the USA for some uncompleted nuclear power plants.
There are some nuclear power constructions in the USA that did not NRC safety rules.
Then they were either abandoned or converted into the nature gas power plants in the USA .
Please check your knowledge about such events and conversions.

Please do not think that you could create messes (such as changing over 400 GE designs without GE permissions) on a nuclear power plant and claim it is still safe.
Please see the real example below of San Onofre Nuclear Generating Station in the USA .

http://en.wikipedia.org/wiki/San_Onofre_Nuclear_Generating_Station

The first unit of the nuclear power station was totally shutdowned only after 25 years of operations
If they constructed it correctly, the life of the first unit would be at least 40 years (or even 60 years).

Then the second and third units were shutdowned this year after they tried to extend them for another 20 years.
However, they (the Con-Edison? Power Company) did not follow the NRC rules and created really technical messes without solutions to rescue them with reasonable costs (just like the messes of the TaiPower Fourth Nuclear Power Plant now).

If you (Dr. Chris Wang) did not know the serious events (and/or incidents) above, I think you are not qualified to be a NE professor in the USA .

You often used the terms "the anti-nuke and anti-government freaks" including me.
Actually, you are one of these "the pro-nuke and pro-government freaks" who want to be the running dogs for the current Taiwan government leaders who have PhD degrees but without Common Sense and without Basic Conscience.
--- Jim Su , NE 1976, NEG1980, PhD in ECE, CISSP

=======================================
確實的核能安全
Tsing-Hua University, NEB1976, NEG1980
常識與學位之-3 ~~ 勿把台灣老百姓當傻瓜 (1996)
*這篇文章歡迎登載傳閱，作者本人負起所有責任 ~~ 電機博士蘇志銓

今年是我從清華核工系(1976)畢業20年，以這篇文章作為回顧與前瞻。1972年大學聯考捨台大電機系，以第一志願進入清華核工系，我抱著造核子彈報效國家的幻想，1976年大學畢業，在海軍造船廠服預備軍官役，當少尉工業安全工程師，碰到修船單位不遵守油氣安全測試，以致油艙油氣爆炸，看到三修船技工（其中一位是海軍陸戰隊退役中校）被燒死，孤兒家人呼天搶地哭喊的慘事，深深感到生產第一，工業安全（或環境保護）次的潛在禍害。預備軍官役退伍，回到清華念核工碩士，接著出國，轉行念電機碩士、博士。畢業後到AT & T Bell Laboratories工作。我有兩個男孩，已是美國高中二年級，初中二年級，不會讀寫中文，我只好留在美國，當失根的蘭花，但內心還認為我是台灣國民，懷著回國做事的願望。

當年（1976年前）在清華，「核能發電安全便宜」幾乎是核工系學生（至少是我）的信條。這在當時可能是合理的假設，因為1971年中東石油禁運，油價上漲，而且當時尚未發生1979年美國三浬島核能電廠事件及1986年蘇聯核能電廠事件。但是到今天，台灣台電及原委會仍高喊「核能電廠百分之百安全及便宜」的口號，甚至台灣前行政院長說出「核能電廠百分之二百安全」的夢幻謊言。這些口號是不實在的，真實的數據可以證明「核能發電安全便宜」的假設在今天是不成立的。

偽造數據，是理工學者的大忌。台灣台電及原委會至今仍以核能發電「安全便宜」來誤導大眾百姓，我真不了解？尤其最近看了一篇報導，台灣核能二廠及三廠預算都追加到原初的2.5倍以上。工程預算追加，並不一定不可以，但是最近引起爭論的核能四廠預算，遠低於美國投標核能廠商的最低投標價錢（相差達美金20億以上），真不知台電如何訂低價？是否有清華校友參與？是否偽造數據?

===============================
常識與學位之-1 ~~ 勿把台灣老百姓當傻瓜 (October 1994)
*這篇文章歡迎登載傳閱，作者本人負起所有責任 ~~ 電機博士蘇志銓
Tsing-Hua University, NEB1976, NEG1980

這篇文章是有感於台灣政府迷信專家，忽略常識判斷，而花費大錢鬧了幾件世界級的大笑話，例如台電核能四廠、台北捷運系統、經國號IDF戰鬥機的事件。台灣政府動不動就用專家的意見來誤導大眾，其實很多事情，只要老百姓肯用”常識”分析，可以發覺專家講的話並不一定都是對的。我並不是說專家(或學位)不重要，但是我覺得”常識判斷”更是重要。我有核工學士、核工碩士、電機碩士、電機博士。在世界著名的研究機構 (AT&T Bell Laboratories) 工作，而且我是數個學業成績優秀的榮譽會員，如斐陶斐 (Phi Tau Phi)， Tau Beta Pi, Eta Kappa Nu會員。我強調”常識判斷”的重要，絕不是酸葡萄心理。現在我以”常識判斷”來談談台電核能四廠、台北捷運系統、經國號IDF戰鬥機的事。

台灣行政院動不動就抬出”核能專家的決定”，來否定反對核四廠者。我自己擁有核工學士、核工碩士、電機碩士、電機博士，我相信也夠資格列入行政院的所謂核能專家。這裡我是要用”常識判斷”來討論台灣前行政院長沒常識的名言”核能電廠百分之二百安全。”世界各地(尤其台灣)核能電廠大小事故不斷，台灣核能電廠的跳機率高於世界核能電廠的跳機平均值，所以說核能電廠連百分之一百的安全都達不到。(常識就可以看出)。說出”核能電廠百分之二百安全”的話，這也太把台灣老百姓當傻瓜了。台灣的核能專家敎出(或許是誤導) 這麼一位”核能電廠百分之二百安全”的前行政院長，那麼這些 (或許只是一位) 核能專家也要自己檢討: 是不是因為核能專家老王賣瓜，自賣自誇，因為涉及本身的工作利益，而只重核能，不能做出客觀正確的決定或建議。

以上幾件台灣政府迷信專家，常忽略常識判斷，而造成世界級的大笑話。真正的專家(並不一定要有博士學位) 是重要，但是根據常識分析判斷更是重要。台灣的工程，常常自誇用世界最新的設計，其實這也不一定是好的。因為真正的工程技術設計，必須經過長時間的實際使用，才能看出它的可靠性及耐用性。若只是使用那些剛在實驗室實驗完成的技術，很可能是花了大錢而問題很多。若台灣的主政者 (尤其那位不知慚愧 (無愧) 的前行政院長, How, Po-Chieu) 能根據常識分析判斷，則”核能電廠百分之二百安全”這麼沒常識的名言就不會講出來。

==============================================
我不反對「核能發電」，但需要確實的「核能安全」
Tsing-Hua University, NEB1976, NEG1980
常識與學位之-2 ~~ 勿把台灣老百姓當傻瓜 (1994)
*這篇文章歡迎登載傳閱，作者本人負起所有責任 ~~ 電機博士蘇志銓

蘇校友雖然人在美國，但對國內核能電廠的安全問題備極關注。他說：「在
美國看到台中大火六十多人死亡的新聞，覺得實在悲慘。為了台灣經濟發展
需要，而有多座核能電廠，但核能事故後果比火災更可怕，希望台灣能確實
加強核能安全。」蘇校友去年為此問題，曾數度寫信給台電席總經理，
語重心長，特批露其中一封，饗我清華人。編者

席總經理鈞鑒：
謝謝您 12/31/94 的來信。
我相信台電絕對有溝通的誠意，不只是對一千多位的大學教授，更重要的是對一般
的大眾老百姓。
---- Jim Su, NE1976, NEG1980, PhD in ECE, CISSP

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

JIM認為核四的Digital Network Real Time Control 不會成功的原因有三:

1. 點數太多: 400,000個I/O

2. 設計公司不夠格

3. 沒有成功的先例

我的看法是這樣:

1是根本這個系統就是需要這麼多點I/O，跟控制設備的選擇無關(用類比系統也要這麼多)。

2 是設計者不夠qualified，更是與設備無關。

3點必須往下分析，沒有成功先例這件事與數位儀控或類比儀控的選擇是怎麼樣的關聯。是不是我之前的第一篇意見裡第2點，digital system因為零件用得多，所以reliability有問題? (當然任何系統前無古人，第一個採用是比較risky，但是我還是要強調核電系統的複雜程度與規模大小如同1中所說的，是系統本身決定的，不是控制系統決定的)。也許Jim願意再花一點時間再分析下去?

Kent Koh 高飛鷂

P.s. 意外發現Jim也曾經服役海軍，我也是，我是21期預官，考進海軍電子官科，在漢陽艦當末代電工官，我退伍不久，它就除役了(不是我搞壞的喔，實在是太老了)

好奇寶寶

2024PeoPo公民新聞獎