Tag Archives: Stanford


Stanford designed software to spot every solar panel in the US (there’s a lot of them)

A new open-access tool developed at Stanford University reveals that, in certain U.S. states, solar panels now account over 10% of total energy generation.


The interactive map of the United States on the DeepSolar website.
Image credits DeepSolar / Stanford University

Policy-makers, utility companies, researchers, and engineers currently have a hard time estimating just how many solar panels installed throughout the country. Stanford University researchers have come to their aid, however, with a new algorithm that makes it easier than ever before to quantify them and analyze development. The tool (accompanied by an open-access website) draws on high-resolution satellite data and automated image analysis.

Sunnyside up

“With these methods, we can not only maintain and update a high-fidelity database of solar installations, but also correlate them at the census-tract level with the amount of incoming solar radiation as well as non-physical factors such as household income and education level,” says co-senior author Arun Majumdar, a mechanical engineering professor at Stanford and co-Director of the Precourt Institute for Energy.

The tool, dubbed DeepSolar, offers unprecedented insight into the trends that drive solar power adoption by society at large, the team says. The algorithm works by analyzing high-resolution images across the U.S., looking for solar panels. When it finds a match, the program records the location and calculates its size.

In stark contrast to its predecessors, DeepSolar isn’t painfully slow. “Previous algorithms were so slow that they would have needed at least a year of computational time” to identify most of the solar panels in the U.S., says co-senior author Ram Rajagopal, a civil engineering professor at Stanford. Meanwhile, DeepSolar only needs a “fraction” of that time.

The team reports using DeepSolar to locate roughly 1.47 million individual solar installations across the country. These included rooftop panels, solar farms, and utility-scale installations. The software should help optimize solar development at the aggregate level, the team explains, especially since decentralization of solar power made it hard to keep track of all the panels being installed.

DeepSolar city.

DeepSolar interactive map showing solar panel distribution by county in the region surrounding Chicago.
Image credits DeepSolar / Stanford University.

One area the team hopes to make an immediate impact with DeepSolar is in the U.S. power grid. The tool, they say, could be used to better integrate solar into the grid by accounting for daily and seasonal fluctuations in incoming sunlight.

“Now that we know where the solar panels are, or are likely to be in the future, we can feed that information into questions of modeling the electricity system and predicting where storage units and substations should go,” says Majumdar.

DeepSolar could also help in pinpointing new areas for solar deployment. The team used the program to establish correlations between solar installation density and variables such as population density or household income — which, when pooled together, allowed them to create a model predicting which areas are most likely to adopt solar in the future.

“Utilities, companies that install solar panels, even community planners that are thinking about sustainability, they all can benefit from this high-resolution spatial data and a website where they can explore and analyze the different trends involved,” Rajagopal says.

The team plans to expand the DeepSolar database to include solar installations in other countries with suitably high-resolution satellite images and to improve its ability to estimate energy output based on characteristics such as the angle of incoming light.

The paper “DeepSolar: A Machine Learning Framework to Efficiently Construct Solar Deployment Database in the United States” has been published in the journal Joule.

The world could go 100% renewable by 2050, Stanford study finds

A new study from Stanford University found that the world could realistically go 100% renewable in a few decades, using only wind, water, and sunlight (WWS).

Image from the Stanford Study.

Right now, we’re still a long way to go. Less than 5% of the planet’s energy demands are satisfied with renewables, but things are improving fast.

“As of the end of 2014, 3.6% of the WWS energy generation capacity needed for a 100% world has already been installed in these countries, with Norway (58%), Paraguay (54%), and Iceland (46%) the furthest along,” the study reads. “The roadmaps envision 80% conversion by 2030 and 100% conversion of all countries by 2050.”

A climate deal was signed at Paris, but now the real work begins. Is a renewable energy future really feasible, and if yes – how could we reach it? This is where this study steps in.

“These are basically plans showing it’s technically and economically feasible to change the energy infrastructure of all of these different countries,” says Mark Z. Jacobson, director of the Atmosphere/Energy Program at Stanford University, who worked with University of California colleagues to analyze energy roadmaps for 139 countries.

Basically, the study made estimates for the 2050 energy consumption of these 139 countries for agriculture, transportation, electricity, heating, cooking and forestry. They then moved on and calculated how these energy requirements could be fulfilled, showing that such a future is indeed not only possible – but within grasp.

“People who are trying to prevent this change would argue that it’s too expensive, or there’s just not enough power, or they try to say that it’s unreliable, that it will take too much land area or resources,” Jacobson said in an interview to the website Co.Exist

Image via Flickr.

Image via Flickr.

With 100% WWS, heating and ground transportation has to be completely electric, relying on electric and hydrogen vehicles, but that’s just the beginning. We’ll need to redesign much of our infrastructure, but according to the analysis, it would be profitable to do so. Renewable energy is getting cheaper and cheaper, and when you factor in the negative externalities of fossil fuels, the difference becomes even more striking. The transition could lead to $5 trillion savings from an environment, climate damage and fuel cost.

Interestingly, Stanford released a similar report on how the US could go fully renewable by 2050 –  a state-by-state analysis. The future can be green, and should be green.

New Technology for Monitoring Glaucoma: Microfluidic Implant And Smart Phone App Monitoring

Image credits: Araci et al.

Stanford Professor of Bioengineering and Applied Physics, Stephen Quake, and Head of the Ophthalmic Science and Engineering Lab at Bar Ilan University Dr. Yossi Mandell teamed up and created a new device which allows glaucoma patients to continuously monitor pressure levels in their eyes – this provides not only a better monitoring, but it also means that patients don’t have to go to the doctor every week.

Glaucoma is a term describing a group of ocular disorders characterized by abnormal pressure in the eye. The nerve damage involves loss of retinal ganglion cells in a characteristic pattern. Glaucoma affects one in 200 people aged 50 and younger, and one in 10 over the age of 80. If it is picked up in its early stages, it can be treated, slowing its development or, in some cases, even stopping it. But monitoring glaucoma is quite a hassle, and in most countries, glaucoma treatments are (let’s say) less than ideal. This is where this device steps in.

The design is very elegant and effective – it features a tiny tube, capped at one end and opened on the other, filled with gas. As the fluid pressure pushes against the gas, a marked scale permits reading of the intraocular pressure. It has absolutely no effect whatsoever on the patient’s vision and it was made to fit inside a commonly used intraocular lens prosthetic, and implanted through simple surgery such as for cataracts which many glaucoma patients already receive. A smart phone (or laptop, or even Google Glass) enables the wearer to take snapshots, reporting the pressure.

Close-up showing fluid-gas interface and tick-marks indicating intraocular pressure. Araci et al.


Currently, patients have to go to the doctor to have their intraocular pressure tested every week, and cannot monitor spikes or sudden changes in pressure. Even so, measurements are sometimes not accurate, because pressure is affected by several external factors, such as posture, medication, and even tightly worn clothing; these wrong measurements can lead to misdiagnosis, and consequently, mistreatment.

“For me, the charm of this is the simplicity of the device,” Professor Quake said. “Glaucoma is a substantial issue in human health. It’s critical to catch things before they go off the rails, because once you go off, you can go blind. If patients could monitor themselves frequently, you might see an improvement in treatments.”

Before the implant can be tested in humans however, they still have to work on the durability of the materials, ensuring that it won’t degrade in the human eye. However, due to the simple design, this is not really expected to be a problem – there’s a myriad of materials which can be successfully applied.

Journal Reference: Ismail E Araci,Baolong Su,Stephen R Quake& Yossi Mandel. An implantable microfluidic device for self-monitoring of intraocular pressure. Nature Medicine 20, 1074–1078 (2014) doi:10.1038/nm.3621

Stanford scientists split water with device that runs on an ordinary AAA battery

Researchers from Stanford have found a way to split water into oxygen and hydrogen using very little energy; the hydrogen they obtain could be used to power fuel cells in zero-emissions vehicles.

I’m quite excited for cars that run on hydrogen, which are set to hit the market in 2015; but while they are always presented as “zero emission cars”, many of the hydrogen cars will actually use hydrogen obtained with natural gas – which is still a fossil fuel and still has considerable emissions. Hopefully, that will only be a temporary stage, and pretty soon, manufacturers will move on to greener, more sustainable solutions – like this project from Stanford University.

A team working there found a way to separate hydrogen from water cheaply and efficiently, producing water electrolysis only powered by a battery. The battery sends an electric current through two electrodes that split liquid water into hydrogen and oxygen gas. Unlike other water splitters that use precious-metal catalysts, the electrodes in the Stanford device are made of inexpensive and abundant nickel and iron.

“Using nickel and iron, which are cheap materials, we were able to make the electrocatalysts active enough to split water at room temperature with a single 1.5-volt battery,” said Hongjie Dai, a professor of chemistry at Stanford. “This is the first time anyone has used non-precious metal catalysts to split water at a voltage that low. It’s quite remarkable, because normally you need expensive metals, like platinum or iridium, to achieve that voltage.”

In addition to producing hydrogen, the same technique could be used to obtain chlorine gas and sodium hydroxide, an important industrial chemical.

Hydrogen cars and carbon emissions

Stanford scientists have developed a low-cost device that uses an ordinary AAA battery to split water into oxygen and hydrogen gas. Gas bubbles are produced by electrodes made of inexpensive nickel and iron.

The auto industry has considered developing hydrogen fuel cell as a promising alternative to the gasoline engine for decades, using fuel cell technology. Fuel cell technology is basically water splitting in reverse – it’s like creating water, and getting energy in the process. Basically, the fuel cell stores hydrogen which reacts with the oxygen from the air to create electricity which powers the car. The only by-product is water – no emissions whatsoever.

Earlier this year, Hyundai began leasing fuel cell vehicles in Southern California, but it’s still a local thing. In 2015, Toyota and Honda will hit the market, selling fuel cell cars. The only problem with this technology is a cheap way of obtaining hydrogen – something for which the Stanford team proposes a simple yet surprising solution.

“It’s been a constant pursuit for decades to make low-cost electrocatalysts with high activity and long durability,” Dai said. “When we found out that a nickel-based catalyst is as effective as platinum, it came as a complete surprise.”

This could save time and a lot of money, potentially taking gas guzzling cars out of the streets in the long run. The discovery wouldn’t have been possible without Stanford graduate student Ming Gong, co-lead author of the study.

“Ming discovered a nickel-metal/nickel-oxide structure that turns out to be more active than pure nickel metal or pure nickel oxide alone,” Dai said.  “This novel structure favors hydrogen electrocatalysis, but we still don’t fully understand the science behind it.”

Water electrolysis was, of course is not a new thing. The novely comes with the nickel/nickel-oxide catalyst, which significantly reduces the voltage necessary for electrolysis.

“The electrodes are fairly stable, but they do slowly decay over time,” he said. “The current device would probably run for days, but weeks or months would be preferable. That goal is achievable based on my most recent results”

The next step in their research is to make the entire process fully sustainable – that is, obtain the energy for the batteries through solar panels – and there’s no reason why they shouldn’t be successful in their attempts.

“Hydrogen is an ideal fuel for powering vehicles, buildings and storing renewable energy on the grid,” said Dai. “We’re very glad that we were able to make a catalyst that’s very active and low cost. This shows that through nanoscale engineering of materials we can really make a difference in how we make fuels and consume energy.”

Journal Reference: Ming Gong,Wu Zhou,Mon-Che Tsai,Jigang Zhou,Mingyun Guan,Meng-Chang Lin,Bo Zhang,Yongfeng Hu,Di-Yan Wang,Jiang Yang,Stephen J. Pennycook,Bing-Joe Hwang& Hongjie Dai. Nanoscale ​nickel oxide/​nickel heterostructures for active ​hydrogen evolution electrocatalysis. Nature Communications 5, Article number: 4695 doi:10.1038/ncomms5695

Stanford scientists build a ‘brain stethoscope’ to turn seizures into music

“My initial interest was an artistic one at heart, but, surprisingly, we could instantly differentiate seizure activity from non-seizure states with just our ears,” Chafe said. “It was like turning a radio dial from a static-filled station to a clear one.”


When Chris Chafe and Josef Parvizi from Stanford University began transforming recordings of brain activity into music, they had artistic pursuits in mind – but they quickly understood they could use the data in scientific purposes – developing a powerful tool for identifying seizures – even for people without experience.

Josef Parvizi was enjoying a performance by the Kronos Quartet, a concert in which the melodies were based on radio signals from outer space when the idea hit him – he began wondering what the brain’s electrical activity might sound like set to music. He turned to Chris Chafe for help – one of the world’s leading minds in terms of “musification” – the science of transforming natural signals into music.

So after they got a patients’ consent, they started working; first of all, they located the source of a seizure, by placing electrodes in patients’ brains to create electroencephalogram (EEG) recordings of both normal brain activity and a seizure state. Parvizi shared the EEG with Chafe, who began setting the electrical spikes of the rapidly firing neurons to music; he chose a tone close to a human voice, in an attempt to give the listener an empathetic connection to the patient as well as an intuitive understanding of what is happening inside the patient’s brain during a seizure. However, as they listened to the recording, they understood they had done more than create an interesting piece of music.

Here’s the audio, with a description:

Around 0:20, the patient’s seizure starts in the right hemisphere, and the patient is talking and acting normally. Around 1:50, the left hemisphere starts seizing while the right is in a post-ictal state.

Because a seizure can happen even without any immediate symptom, their work has the potential to help thousands (if nto more people). If they could achieve the same thing with real time brain data, then they could develop a tool to allow caregivers for people with epilepsy to quickly hear when an undetected seizure is occuring. They dubbed the device a “brain stethoscope”.

Taking care of somebody with seizures can be very difficult, because not all seizures are accompanied by behavioral changes.

“Someone – perhaps a mother caring for a child – who hasn’t received training in interpreting visual EEGs can hear the seizure rhythms and easily appreciate that there is a pathological brain phenomenon taking place,” Parvizi said.

Still, this innovative and potentially very helpful idea is still very far from becoming a clinical reality.

“We’ve really just stuck our finger in there,” Chafe said. “We know that the music is fascinating and that we can hear important dynamics, but there are still wonderful revelations to be made.”

Still, the potential is there, and even without it becoming a reality, this is a statement of what two fields which are apparently incompatible can accomplish together:

“This is what I like about Stanford,” Parvizi said. “It nurtures collaboration between fields that are seemingly light-years apart – we’re neurology and music professors! – and our work together will hopefully make a positive impact on the world we live in.”

18 Universities compete to build tech campus in NYC

View of Roosevelt Island, where Stanford University submitted their campus development plan. (c) AP

In an attempt to gain ground against renowned tech centers from around the country like Sillicon Valley, Boston or Texas, the city of New York has reached out to universities all over the world to submit their development plans for an applied science and engineering campus.

Higher-Education institutions were first aware of this signal in December, and now out of the slew of interested institutions, 18 universities submitted their plans which officials hope to increase New York’s profile in the realm of technological innovation. Schools interested include  Stanford University, Carnegie Mellon University and Korea Advanced Institute of Science and Technology, Abo Akademi University in Finland, the Korea Advanced Institute of Science and Technology, and Technion-Israel Institute of Technology, as well as schools in England, India and Switzerland. Some institutions partnered and submitted joint venture plans, such as the one encompassing New York University, Carnegie Mellon University, the City University of New York, the University of Toronto and I.B.M.

“We received a response that was certainly equal to or exceeded our expectations,” said Robert K. Steel, the deputy mayor for economic development.

In December, Bloomberg solicited proposals for science and engineering centers for several sites, including the Navy Hospital Campus at the Brooklyn Navy Yard, the Goldwater Hospital Campus on Roosevelt Island in Manhattan, several areas of Governor’s Island and Farm Colony on Staten Island.

Stanford was among the first elite school that said it would consider applying, which chose Roosevelt Island as it’s location, a 2-mile-long sliver of land in the East River between Manhattan and Queens served by a popular tram, inhabited only by 9,000 people living in apartments. If accepted, the Stanford NY tech satellite would begin construction in 2013, cost around $250mil and open in 2015. The aim of the project is of training 440 graduates starting 2015-2016 year, and accommodate 2,200 students and few hundred faculty/staff members.

“It’s not about bringing money to the home campus,” Stanford’s president, John L. Hennessy, said in a recent interview. “It will require an investment of institutional resources to be successful.”

Cost will be a major challenge. A new city campus will likely cost hundreds of millions of dollars, said Seth Pinsky, president of the city’s Economic Development Corp. While the city could offer some city land and “seed investments,” the majority would have to be paid for by the selected university and its partners.

“There are very serious academic institutions that are willing to undertake an effort such as that,” Mr. Pinsky said.