fueled by intelligent technology, the traditional industries are evolving rapidly in recent years. great time and efforts are devoted by nuclear workersworldwide to combine advanced smart technology with relatively conservative nuclear industry, which is believed to beenormously revolutionized.
in this context, a panel was held on the 25thinternational conference on nuclear engineering (icone25) on july 4th, situated on the yangtze river hall of shanghai international convention center, theming on intelligent technology application in npp, chaired by jun ding from terra power and co-chaired by yinsheng li from japan atomic energy agency.
intelligent cognitive systems: a new partnership between humans and technology
mr. paul fechtelkotter from ibm watson mentioned that intelligent cognitive systems have the ability of understanding, reasoning, learning and interacting through which can produce personalized data based on the diverse collections of user’s habit and information such as call logs and emails.
the intelligent cognitive systems of watson enable five classes of cognitive services (ask, explore, discover, decide and visualize) working across structured and unstructured data with natural language processing and machine learning, which augment human operators and can look for potential issues in npp.
do more with less: digital nuclear power plant (dnpp) simulation
panelist dr. kin wah wong from a-d technology gave us thoughts on the implementation of digital nuclear power plant (dnpp)based on the fast growth in computational speed, less cost on data processing and rapid development in network performance.
with the help of open source software such as cmake, more efficient methodology and solver libraries are developed to solve very large matrix equations, and object-oriented programming method and languages are used to enhance software modularity and reusability as well. the dnpp will be usedto assist and verify designs for new nuclear power plant, develop training modules of the new nuclear power plant and generate large amount of virtual plant data in many different scenarios to gain better.
combining reliability, performance, and maintenance data for asset management decision making
dr. wes hines from the university of tennessee pointed out that 3 sources of information play vital role on npp:
1.reliability data: past failure times;
2.performance data: past, recent, and current performance;
3.maintenance data: maintenance data and trends and maintenance records.
through deep look and analysis of these basic data, goals as follows can be achieved:
2.reduce unplanned outages;
3.optimize the timing of capital replacement and maintenance activities;
5.achieve component and system health monitoring;
6.get accurate prediction of remaining useful life;
7.reduce prediction bias.
by the means of big data collection on npp maintenance and performance, systems can be customized for individuals to assist management decision making.
virtual plant helps operating &maintenance (o&m) in npp
mr.fumihikoishibashi from toshiba corporation demonstrated the major role of 3d virtual plant in enhancing operating &maintenance (o&m) in npp. with 3d cad and laser scan technology, a virtual plant of similar systems, functions and structures of a real npp can be simulated right in the computer, capable of working, operating and debugging.
examples are made in electrical isolation with deep learning. typically,planning electrical isolation will take 2 weeks for a skilled engineer with knowledge and experience, going through hundreds of the circuit diagrams. however, with the speed-up of computing time with deep learning, planning time can be narrowed down to less than 1 day.
mr. ishibashi also exampled the adoption of wireless control system in npp.
perfect match: mix reality (mr) and npp
mr. jie li from datamesh technology corp., ltd showed great expansibility of an intelligent npp. the operability and visuality can be dramatically improved by external hardware and software systems.
mr. li exhibited a demo of how the microsoft hololens be applied in an intelligent npp by overlaying virtual displays on real components such as pipes and equipment, intuitively showing the live status and condition of the things you are watching.
investment and outlook of intelligent energy
mr.haojie li from censtar science and technology corp., ltd indicated that world economic growth requires more energy and china will be one of the largest energy growth markets. the new development and application of networking technology (iot) and artificial intelligence (ai) can serve the energy market in two aspects:
1.operation side: from the energy product reserves, wholesale, transportation, inventory management toretail, through the energy sales process, the whole business chain of intelligenthardware and software systems, as well as fine management model.
2.consumer side: smart filling devices for mobilephones and energy (such assmart tanker, gas dispenserand smart charging pile) areintelligent interactive terminalsfor energy consumption.
realization of energy consumption ecological systemcan promote the comprehensive application of green electricity from hydrogen and nuclear power in even broader field. mr. li also outlooked a brighter future of intelligent and clean energy consumption system in the next ten years.
in q&a section of this panel, the panelists answered questions concerning data analysis and processing and uncertainty from the audience, showing the immense potentials and extensive foreground of intelligent technology application in npp.