The seminar
entitled “Design of an Accident Prevention System for Light Water Reactors Using
the Artificial Neural Network and a Highspeed Simulator” was organized by Department
of Mechanical Engineering, CityU on 27th Feb 2019. Seven years after Fukushima, the last missing
pieces of the jigsaw puzzle of light water reactor (LWR) safety have been put
together. In the United States, the nuclear power industry has implemented a
diverse and flexible strategy, FLEX (NEI 2012), for severe accidents. The system makes rapid predictions using a
high-speed simulation code and artificial intelligence (AI) diagnostics was
proposed by Dr. Li-Chi Cliff Po. I would
like to summarize today speech for sharing below.
In the beginning, Dr. Li-Chi Cliff Po (Founder president of Micro-Simulation Technology) introduced his company “Micro-Simulation Technology” which established in 1985 and had developed PC-Windows based nuclear power plant simulator PCTRAN.
Then Dr. Po
briefed “Risks” for Nuclear Power Plant Safety and discussed the deficiencies on
Nuclear plant of Probabilistic Risk Assessment (PRA). He proposed to use Artificial Intelligence
(AI) machine learning of analyzed cases to build diagnosis algorithms and using
Fast-speed simulation software to predict consequence way ahead actual
damage. Dr. Po also told us
some severe accident in LWR under different situations.
After that he
demonstrated the PC TRAN CRP1000 DBA and Severe Accident Simulator as
follows.
Changed the failure fraction to 2000%!
The core of
reactor was very hot!
And then Dr.
Po quoted Ayo-Imoru R.M. and Cilliers A.C. (2018) paper named “Continuous
machine learning for abnormality identification to aid condition-based
maintenance in nuclear power plant” to the introduction of a machine-learning
tool like artificial neural networks (ANN) to train both the simulator and
plant parameters.
Finally, he
introduced that the US nuclear industry, working through the Nuclear Energy
Institute (NEI), developed a framework for addressing this order, as documented
in NEI 12-06 "Diverse and Flexible Coping Strategies (FLEX) Implementation
Guide," published in August 2012. SAFER
program and equipment for accident was mentioned.
Lastly, newly developed ultimate response guidelines for BWR was also discussed.
At the end,
Dr. Li-Chi Cliff Po concluded that not just for training and education, but
also the entire technical staff entering into the industry. US FLEX and SAFER programs were merely
documents and methodologies that no practical hardware and/or software. Online AI Accident Prevention System is the
first of a kind. That could be
considered to resolve Inland Nuclear Construction Concerns.
Q&A
Session
Dr. BL Luk and
I had asked about the training of ANN for LWR because of limited cases data. *From Ayo-Imoru’s paper (2018), the nuclear
plant outputs are compared with the simulator outputs in steady state. Based on
the results from the first stage, in the second stage the plant and simulator
outputs are passed through an ANN, which are then compared for better fault identification
in transient.
Reference:
Department of
Mechanical Engineering, CityU - http://www.cityu.edu.hk/mne/
Seminar
leaflet - http://www.cityu.edu.hk/mne/pdf/mneseminar1819_25.pdf
Ayo-Imoru R.M.
and Cilliers A.C. (2018) “Continuous machine learning for abnormality
identification to aid condition-based maintenance in nuclear power plant” Annals of Nuclear Energy, 118, pp.61-70.
CityU
AI related activities:
20181114: CityU the
1st Workshop on Financial Data Analysis - https://qualityalchemist.blogspot.com/2018/11/cityu-1st-workshop-on-financial-data.html
20181227: CityU Distinguished Lecture - How to Make an Artificial Vision
System Smart? - https://qualityalchemist.blogspot.com/2018/12/cityu-distinguished-lecture-how-to-make.html
20190122: CityU Seminar on The
First Step for AI-based Human-Like Language Understanding – Sentiment Analysis
of Text - https://qualityalchemist.blogspot.com/2019/01/cityu-seminar-on-first-step-for-ai.html
20190128: CityU Distinguished Lecture on AI Enabled Personalized
Theranotics - https://qualityalchemist.blogspot.com/2019/01/cityu-distinguished-lecture-on-ai.html
沒有留言:
發佈留言