Connect with us

Solar Energy

The future of AI with solar-powered synaptic devices

Published

on

The future of AI with solar-powered synaptic devices


The future of AI with solar-powered synaptic devices

by Riko Seibo

Tokyo, Japan (SPX) Nov 26, 2024






Artificial intelligence (AI) is increasingly relied upon for predicting critical events such as heart attacks, natural disasters, and infrastructure failures. These applications demand technologies capable of rapidly processing data. One such promising approach is reservoir computing, particularly physical reservoir computing (PRC), known for its efficiency in handling time-series data with minimal power consumption. Optoelectronic artificial synapses in PRC, mimicking human neural synaptic structures, are poised to enable advanced real-time data processing and recognition akin to the human visual system.

Existing self-powered optoelectronic synaptic devices, however, struggle to process time-series data across diverse timescales, which is essential for applications in environmental monitoring, infrastructure maintenance, and healthcare.



Addressing this challenge, researchers at Tokyo University of Science (TUS), led by Associate Professor Takashi Ikuno and including Hiroaki Komatsu and Norika Hosoda, have developed an innovative self-powered dye-sensitized solar cell-based optoelectronic photopolymeric human synapse. This groundbreaking device, featuring a controllable time constant based on input light intensity, represents a major advancement in the field. The study, published on October 28, 2024, in ‘ACS Applied Materials and Interfaces’, highlights the potential of this technology.



Dr. Ikuno explained, “To process time-series input optical data with various time scales, it is essential to fabricate devices according to the desired time scale. Inspired by the afterimage phenomenon of the eye, we came up with a novel optoelectronic human synaptic device that can serve as a computational framework for power-saving edge AI optical sensors.”



The new device integrates squarylium derivative-based dyes, incorporating optical input, AI computation, analog output, and power supply at the material level. It demonstrates synaptic plasticity, exhibiting features such as paired-pulse facilitation and depression in response to light intensity. The device achieves high computational performance in time-series data processing tasks while maintaining low power consumption, regardless of the input light pulse width.



Remarkably, the device achieved over 90% accuracy in classifying human movements, including bending, jumping, running, and walking, when used as the reservoir layer of PRC. Its power consumption is only 1% of that required by traditional systems, significantly reducing carbon emissions. Dr. Ikuno emphasized, “We have demonstrated for the first time in the world that the developed device can operate with very low power consumption and yet identify human motion with a high accuracy rate.”



This innovation holds significant promise for edge AI applications, including surveillance cameras, automotive sensors, and health monitoring systems. “This invention can be used as a massively popular edge AI optical sensor that can be attached to any object or person,” noted Dr. Ikuno. He further highlighted its potential to improve vehicle energy efficiency and reduce costs in standalone smartwatches and medical devices.



The novel solar cell-based device could redefine energy-efficient edge AI sensors across various applications, marking a significant leap forward in both technology and sustainability.



Research Report:Self-Powered Dye-Sensitized Solar-Cell-Based Synaptic Devices for Multi-Scale Time-Series Data Processing in Physical Reservoir Computing


Related Links

Tokyo University of Science

All About Solar Energy at SolarDaily.com





Source link

Continue Reading
Click to comment

Leave a Reply

Solar Energy

Existing EV batteries may last significantly longer under real-world conditions

Published

on

By

Existing EV batteries may last significantly longer under real-world conditions





Existing EV batteries may last significantly longer under real-world conditions

by Clarence Oxford

Los Angeles CA (SPX) Dec 10, 2024






Electric vehicle (EV) batteries subjected to typical real-world driving scenarios-such as heavy traffic, urban commutes, and long highway trips-could last up to 40% longer than previously projected, according to new research from the SLAC-Stanford Battery Center, a collaboration between Stanford University’s Precourt Institute for Energy and SLAC National Accelerator Laboratory. This finding suggests EV owners may delay the costly replacement of battery packs or the purchase of new vehicles for several more years than expected.

Traditionally, battery scientists have tested EV batteries in labs using a constant charge-discharge cycle. While effective for quick evaluations of new designs, this method does not accurately reflect the varied usage patterns of everyday drivers, the study published in *Nature Energy* on Dec. 9 reveals.



Although battery costs have fallen by approximately 90% over the past 15 years, they still represent about one-third of an EV’s price. This research could provide reassurance to current and prospective EV owners about the longevity of their vehicle’s batteries.



“We’ve not been testing EV batteries the right way,” said Simona Onori, the study’s senior author and an associate professor at Stanford’s Doerr School of Sustainability. “To our surprise, real driving with frequent acceleration, braking, stopping for errands, and extended rest periods helps batteries last longer than previously thought based on industry-standard tests.”

Real-World Driving Profiles Improve Battery Lifespan

The researchers developed four distinct EV discharge profiles, ranging from constant discharge to dynamic patterns based on actual driving data. Testing 92 commercial lithium-ion batteries over two years, they found that batteries subjected to realistic driving scenarios demonstrated significantly improved longevity.

Machine learning algorithms were crucial in analyzing the extensive data, revealing that certain driving behaviors, like sharp accelerations, slowed battery degradation. This contradicted prior assumptions that acceleration peaks harm EV batteries. “Pressing the pedal hard does not speed up aging. If anything, it slows it down,” explained Alexis Geslin, one of the study’s lead authors and a PhD candidate in materials science and computer science at Stanford.

Aging from Use vs. Time

The study differentiated between battery aging caused by charge-discharge cycles and aging from time alone. While frequent cycling dominates battery aging for commercial vehicles like buses or delivery vans, time-induced aging becomes a larger factor for personal EVs, which are often parked and idle.



“We battery engineers have assumed that cycle aging is much more important than time-induced aging,” said Geslin. “For consumers using their EVs for daily errands but leaving them unused most of the time, time becomes the predominant aging factor.”



The researchers identified an optimal discharge rate balancing both time and cycle aging for the batteries tested, which aligns with typical consumer driving habits. Manufacturers could update battery management software to incorporate these findings, potentially extending battery lifespan under normal conditions.

Implications for the Future

Evaluating new battery chemistries and designs under realistic conditions is critical for future advancements, said Le Xu, a postdoctoral scholar in energy science and engineering. “Researchers can now revisit presumed aging mechanisms at the chemistry, materials, and cell levels to deepen their understanding,” Xu added.



The study’s principles could apply beyond EV batteries to other energy storage systems, plastics, solar cells, and biomaterials where aging is a key concern. “This work highlights the power of integrating multiple areas of expertise-from materials science and modeling to machine learning-to drive innovation,” Onori concluded.



Research Report:Dynamic cycling enhances battery lifetime


Related Links

SLAC-Stanford Battery

Powering The World in the 21st Century at Energy-Daily.com





Source link

Continue Reading

Solar Energy

So you want to build a solar or wind farm? Here’s how to decide where

Published

on

By

So you want to build a solar or wind farm? Here’s how to decide where





So you want to build a solar or wind farm? Here’s how to decide where

by David L. Chandler | MIT News

Boston MA (SPX) Dec 08, 2024






Deciding where to build new solar or wind installations is often left up to individual developers or utilities, with limited overall coordination. But a new study shows that regional-level planning using fine-grained weather data, information about energy use, and energy system modeling can make a big difference in the design of such renewable power installations. This also leads to more efficient and economically viable operations.

The findings show the benefits of coordinating the siting of solar farms, wind farms, and storage systems, taking into account local and temporal variations in wind, sunlight, and energy demand to maximize the utilization of renewable resources. This approach can reduce the need for sizable investments in storage, and thus the total system cost, while maximizing availability of clean power when it’s needed, the researchers found.



The study, appearing in the journal Cell Reports Sustainability, was co-authored by Liying Qiu and Rahman Khorramfar, postdocs in MIT’s Department of Civil and Environmental Engineering, and professors Saurabh Amin and Michael Howland.



Qiu, the lead author, says that with the team’s new approach, “we can harness the resource complementarity, which means that renewable resources of different types, such as wind and solar, or different locations can compensate for each other in time and space. This potential for spatial complementarity to improve system design has not been emphasized and quantified in existing large-scale planning.”



Such complementarity will become ever more important as variable renewable energy sources account for a greater proportion of power entering the grid, she says. By coordinating the peaks and valleys of production and demand more smoothly, she says, “we are actually trying to use the natural variability itself to address the variability.”



Typically, in planning large-scale renewable energy installations, Qiu says, “some work on a country level, for example saying that 30 percent of energy should be wind and 20 percent solar. That’s very general.” For this study, the team looked at both weather data and energy system planning modeling on a scale of less than 10-kilometer (about 6-mile) resolution. “It’s a way of determining where should we, exactly, build each renewable energy plant, rather than just saying this city should have this many wind or solar farms,” she explains.



To compile their data and enable high-resolution planning, the researchers relied on a variety of sources that had not previously been integrated. They used high-resolution meteorological data from the National Renewable Energy Laboratory, which is publicly available at 2-kilometer resolution but rarely used in a planning model at such a fine scale. These data were combined with an energy system model they developed to optimize siting at a sub-10-kilometer resolution. To get a sense of how the fine-scale data and model made a difference in different regions, they focused on three U.S. regions – New England, Texas, and California – analyzing up to 138,271 possible siting locations simultaneously for a single region.



By comparing the results of siting based on a typical method vs. their high-resolution approach, the team showed that “resource complementarity really helps us reduce the system cost by aligning renewable power generation with demand,” which should translate directly to real-world decision-making, Qiu says. “If an individual developer wants to build a wind or solar farm and just goes to where there is the most wind or solar resource on average, it may not necessarily guarantee the best fit into a decarbonized energy system.”



That’s because of the complex interactions between production and demand for electricity, as both vary hour by hour, and month by month as seasons change. “What we are trying to do is minimize the difference between the energy supply and demand rather than simply supplying as much renewable energy as possible,” Qiu says. “Sometimes your generation cannot be utilized by the system, while at other times, you don’t have enough to match the demand.”



In New England, for example, the new analysis shows there should be more wind farms in locations where there is a strong wind resource during the night, when solar energy is unavailable. Some locations tend to be windier at night, while others tend to have more wind during the day.



These insights were revealed through the integration of high-resolution weather data and energy system optimization used by the researchers. When planning with lower resolution weather data, which was generated at a 30-kilometer resolution globally and is more commonly used in energy system planning, there was much less complementarity among renewable power plants. Consequently, the total system cost was much higher. The complementarity between wind and solar farms was enhanced by the high-resolution modeling due to improved representation of renewable resource variability.



The researchers say their framework is very flexible and can be easily adapted to any region to account for the local geophysical and other conditions. In Texas, for example, peak winds in the west occur in the morning, while along the south coast they occur in the afternoon, so the two naturally complement each other.



Khorramfar says that this work “highlights the importance of data-driven decision making in energy planning.” The work shows that using such high-resolution data coupled with carefully formulated energy planning model “can drive the system cost down, and ultimately offer more cost-effective pathways for energy transition.”



One thing that was surprising about the findings, says Amin, who is a principal investigator in the MIT Laboratory of Information and Data Systems, is how significant the gains were from analyzing relatively short-term variations in inputs and outputs that take place in a 24-hour period. “The kind of cost-saving potential by trying to harness complementarity within a day was not something that one would have expected before this study,” he says.



In addition, Amin says, it was also surprising how much this kind of modeling could reduce the need for storage as part of these energy systems. “This study shows that there is actually a hidden cost-saving potential in exploiting local patterns in weather, that can result in a monetary reduction in storage cost.”



The system-level analysis and planning suggested by this study, Howland says, “changes how we think about where we site renewable power plants and how we design those renewable plants, so that they maximally serve the energy grid. It has to go beyond just driving down the cost of energy of individual wind or solar farms. And these new insights can only be realized if we continue collaborating across traditional research boundaries, by integrating expertise in fluid dynamics, atmospheric science, and energy engineering.”



Research Report:Decarbonized energy system planning with high-resolution spatial representation of renewables lowers cost


Related Links

Department of Civil and Environmental Engineering

All About Solar Energy at SolarDaily.com





Source link

Continue Reading

Solar Energy

China to send batteries to Europe via route bypassing Russia: Kazakhstan

Published

on

By

China to send batteries to Europe via route bypassing Russia: Kazakhstan





China to send batteries to Europe via route bypassing Russia: Kazakhstan

by AFP Staff Writers

Almaty, Kazakhstan (AFP) Dec 6, 2024






China will soon send lithium-ion batteries to Europe via Kazakhstan on a trade route that bypasses sanctions-hit Russia, the Central Asian country said Friday.

Trade via the Trans-Caspian International Transport Route (TITR) that crosses the Caspian Sea has jumped since Moscow invaded Ukraine in 2022, as European countries seek to avoid imports that transit Russia.

Kazakhstan has agreed to “jointly develop” the route with Beijing, launching a “trial run for the transportation of lithium-ion batteries from China” in December, Kazakhstan’s transport ministry said Friday.

China is the world’s largest producer of lithium-ion batteries and among the top miners of the metal, which is used to power phones and electric vehicles.

“The volume of transportation from China along the TITR (in the direction of China to Europe) has exceeded the equivalent of 27,000 20-foot containers, which is 25 times more than in the same period last year,” the ministry said.

The ministry also noted an increase in goods transported between China and Kazakhstan, with both sides discussing the idea of opening new transport routes across their shared border.

Europe has looked to Central Asia as a key trading partner since Moscow launched its Ukraine offensive, triggering a barrage of Western sanctions on Moscow.

Beijing has also invested billions of dollars in building rail and road routes that traverse Central Asia, as it seeks to turn the region into a trading hub for its “New Silk Road”.

Construction is underway to build a China-Kyrgyzstan-Uzbekistan railroad that will shorten transport times between China and Europe.

Related Links

Powering The World in the 21st Century at Energy-Daily.com





Source link

Continue Reading

Trending