Collaborating Authors


Fukushima College robot wins top prize for nuclear decommissioning

The Japan Times

Fukushima – A robot created by a team from a technology college in northeastern Japan recently won the top prize in a robotics competition that had the theme of decommissioning the Fukushima No. 1 nuclear power plant. The Mehikari robot of Fukushima College earned praise for its speed as well as ability to employ different methods to retrieve mock debris similar in size to that at the plant, the site of a nuclear disaster triggered by a massive earthquake and tsunami on March 11, 2011. The robot completed the set task in about 2 minutes, the fastest time, in the annual competition aimed at fostering future engineers that was attended by students from 13 colleges belonging to the National Institute of Technology. Sunday's competition was the fifth of its kind. Students in 14 teams from the colleges across the country such as in Osaka and Kumamoto prefectures were tasked this year with developing robots to remove fuel debris from the plant, organizers said.

Fukushima nuclear debris removal to be delayed due to pandemic

The Japan Times

The operator of the Fukushima No. 1 nuclear power plant, which suffered core meltdowns in 2011, has decided to delay the removal of nuclear debris by about one year from 2021 due to the coronavirus pandemic, sources said Wednesday. The process of removing the melted fuel, the most difficult part of cleaning up the facility, was to begin at the No. 2 reactor in 2021, but the virus spread has stalled tests in the U.K. of a robot arm that is to be used for the removal, the sources said. Of the Nos. 1 to 3 reactors that experienced meltdowns following a massive earthquake and tsunami, the removal procedure was to start at the No. 2 unit because the operator, Tokyo Electric Power Company Holdings Inc., had the best grasp of its internal condition, they said. Tepco had planned to insert a robot arm into the unit's containment vessel, from which it would initially extract around 1 gram of the debris at a time, then gradually expand the amount as it works toward removing several kilograms a day. The company was originally scheduled to verify in August the viability of the robot arm in the U.K. and transfer the equipment to Japan in February 2021 so that workers could start training with it.

Deep Reinforcement Learning for Long Term Hydropower Production Scheduling Artificial Intelligence

We explore the use of deep reinforcement learning to provide strategies for long term scheduling of hydropower production. We consider a use-case where the aim is to optimise the yearly revenue given week-by-week inflows to the reservoir and electricity prices. The challenge is to decide between immediate water release at the spot price of electricity and storing the water for later power production at an unknown price, given constraints on the system. We successfully train a soft actor-critic algorithm on a simplified scenario with historical data from the Nordic power market. The presented model is not ready to substitute traditional optimisation tools but demonstrates the complementary potential of reinforcement learning in the data-rich field of hydropower scheduling.

Estimation of Gas Turbine Shaft Torque and Fuel Flow of a CODLAG Propulsion System Using Genetic Programming Algorithm Artificial Intelligence

In this paper, the publicly available dataset of condition based maintenance of combined diesel-electric and gas (CODLAG) propulsion system for ships has been utilized to obtain symbolic expressions which could estimate gas turbine shaft torque and fuel flow using genetic programming (GP) algorithm. The entire dataset consists of 11934 samples that was divided into training and testing portions of dataset in an 80:20 ratio. The training dataset used to train the GP algorithm to obtain symbolic expressions for gas turbine shaft torque and fuel flow estimation consisted of 9548 samples. The best symbolic expressions obtained for gas turbine shaft torque and fuel flow estimation were obtained based on their $R^2$ score generated as a result of the application of the testing portion of the dataset on the aforementioned symbolic expressions. The testing portion of the dataset consisted of 2386 samples. The three best symbolic expressions obtained for gas turbine shaft torque estimation generated $R^2$ scores of 0.999201, 0.999296, and 0.999374, respectively. The three best symbolic expressions obtained for fuel flow estimation generated $R^2$ scores of 0.995495, 0.996465, and 0.996487, respectively.

Mission possible? The long road ahead for Fukushima cleanup.

The Japan Times

Nearly a decade after the three meltdowns at Fukushima No. 1 nuclear power plant, plans are underway to finally remove nuclear fuel debris from the three reactors. But in order to remove it, Tokyo Electric Power Company Holdings Inc. (Tepco), the operator of the plant, needs to ensure there is a place to store the debris once it is retrieved. This is thought to be the reason why the government is rushing to give the green light to releasing tritium-laced water piling up at the plant into the Pacific -- to give room for the storage of fuel debris. But removing the fuel debris -- a crucial step in the decommissioning process -- is an enormous task on its own, with measures that need to be resolved emerging one after another. At a three-day online meeting of the Atomic Energy Society of Japan from Sept. 16, an official from the International Research Institute for Nuclear Decommissioning (IRID) who is in charge of technical development regarding the decommissioning of the Fukushima plant, explained the plan, or the lack thereof, to remove the debris at reactor No. 2. "We will consider what kind of measures to take, comparing tactics and developing techniques," the official said, with a hint of frustration at not being able to come up with a specific way yet.

Artificial intelligence can enhance natural gas delivery, NARUC reports


Artificial intelligence can provide value to natural gas utilities and customers, a National Association of Regulatory Utility Commissioners (NARUC) primer states. The primer from the US regulatory non-profit is aimed to improve awareness of artificial intelligence tools and practices, with a focus on the potential to enhance natural gas utility performance. It zeroes in on the three most common challenges being faced. These are ageing distribution infrastructure, excavator damage to underground infrastructure and customer participation in energy efficiency programmes. Regarding ageing infrastructure, artificial intelligence can assist in identifying and prioritising repair and replacement programmes.

pymgrid: An Open-Source Python Microgrid Simulator for Applied Artificial Intelligence Research Artificial Intelligence

Microgrids, self contained electrical grids that are capable of disconnecting from the main grid, hold potential in both tackling climate change mitigation via reducing CO2 emissions and adaptation by increasing infrastructure resiliency. Due to their distributed nature, microgrids are often idiosyncratic; as a result, control of these systems is nontrivial. While microgrid simulators exist, many are limited in scope and in the variety of microgrids they can simulate. We propose pymgrid, an open-source Python package to generate and simulate a large number of microgrids, and the first open-source tool that can generate more than 600 different microgrids. pymgrid abstracts most of the domain expertise, allowing users to focus on control algorithms. In particular, pymgrid is built to be a reinforcement learning (RL) platform, and includes the ability to model microgrids as Markov decision processes. pymgrid also introduces two pre-computed list of microgrids, intended to allow for research reproducibility in the microgrid setting.

How AI Will Make Nuclear Energy More Affordable


Nuclear power is one of the cheapest forms of generating carbon-free energy but is instead known for being the opposite. While constructing a nuclear plant is expensive and recent projects in the US and EU have suffered from overruns, operating it is cheaper than many other energy sources. It also turns out that the reason for expensive construction is not entirely technical and often has political factors outside the control of the maker. So how can these costs be reduced? A nuclear plant's costs are made up from capital and operation costs.

A New Inference algorithm of Dynamic Uncertain Causality Graph based on Conditional Sampling Method for Complex Cases Artificial Intelligence

Dynamic Uncertain Causality Graph(DUCG) is a recently proposed model for diagnoses of complex systems. It performs well for industry system such as nuclear power plants, chemical system and spacecrafts. However, the variable state combination explosion in some cases is still a problem that may result in inefficiency or even disability in DUCG inference. In the situation of clinical diagnoses, when a lot of intermediate causes are unknown while the downstream results are known in a DUCG graph, the combination explosion may appear during the inference computation. Monte Carlo sampling is a typical algorithm to solve this problem. However, we are facing the case that the occurrence rate of the case is very small, e.g. $10^{-20}$, which means a huge number of samplings are needed. This paper proposes a new scheme based on conditional stochastic simulation which obtains the final result from the expectation of the conditional probability in sampling loops instead of counting the sampling frequency, and thus overcomes the problem. As a result, the proposed algorithm requires much less time than the DUCG recursive inference algorithm presented earlier. Moreover, a simple analysis of convergence rate based on a designed example is given to show the advantage of the proposed method. % In addition, supports for logic gate, logic cycles, and parallelization, which exist in DUCG, are also addressed in this paper. The new algorithm reduces the time consumption a lot and performs 3 times faster than old one with 2.7% error ratio in a practical graph for Viral Hepatitis B.

How AI can help boost alternative and renewable energy use


Ten years ago, I was engaged in the writing of an energy power grid report that was part of a national initiative to assess the health of our electrical energy grid and its resilience. Assets like wind farms and contemporary fossil and nuclear fuel systems were in place for energy distribution, but to my surprise there was also equipment in the grid that dated back to the 1890s and was still in production. I began to understand the challenges of using renewable energy such as wind and solar when it came to assessing energy supply and demand and ensuring there is enough on-hand energy to power the homes and businesses that are relying on it. When utilities were using gas, coal, or nuclear energy to power the grid, the in-flow of that fuel from its source was consistent, so it was easy to assess supply and demand on any given day and to deliver the energy needed to power homes and businesses. What if the wind gusted to 40 mph one day, and was perfectly still on the next day?