The coming battery boom will need better software, sensors & data


Lithium-ion batteries have become the power backbone for our always-on gadgets, laptops and cell phones. Now, thanks to Tesla Motors, they’re moving into our cars and potentially onto the power grid and into buildings, too. With the potential for so many batteries becoming embedded in everything — providing both short bursts of power and long drawn-out energy trickles — the need for smart software, predictive algorithms and temperature management systems for batteries will only continue to rise.

Tesla itself has already spent a lot of time and money building its innovative battery pack configuration for electric cars, which strings together thousands of small-format lithium-ion batteries. The pack uses software to help manage individual battery cooling, management and charging, as well as monitoring for safety issues. An important part of Tesla’s intellectual property lies in this battery pack design and software.


But beyond Tesla, there are many other big companies, startups and even researchers in labs working on new types of battery software, sensing and predictive algorithms. It will be these computing and communication technologies that will help enable lithium-ion batteries (and other batteries) — which can be flammable and volatile under the wrong conditions — become safe enough to allay fears of having them in vehicles and in huge battery banks near power lines.

In the labs

At the U.S. Department of Energy’s ARPA-E Summit last month, there were a dozen projects that focused on next-generation battery management systems that can continuously sense and monitor the internal workings of individual batteries and packs. The DOE’s ARPA-E Program gives small grants to high-risk early-stage research efforts and many of these projects were represented at the recent show.

BYD batteries in Santa Rita jail

Improved battery sensors are a particularly hot area for the ARPA-E-backed university and government lab research. The idea behind these projects is to get as much granular data as possible — in as close to real-time as possible — out of the batteries.

Researchers at Battelle Memorial Institute are working on an optical fault sensor for lithium-ion batteries that can detect internal faults in batteries long before they can lead to a battery failure. Similarly GE’s Global Research group is developing tiny, low-cost sensors that can detect battery pressure and temperature in real time and help predict battery problems before they happen.

Engineers at Lawrence Livermore National Labs have been working on a wireless sensor system for the battery pack itself, which would reduce the use of more expensive pack cables that connect the pack. PARC, in conjunction with battery maker LG-Chem, has been creating a fiber optic sensing system for electric car battery packs that detects shifts in wavelength and combines that tech with smart algorithms. Washington University in St. Louis has a team focused on building modelling software to create a predictive battery management system, which could optimize battery cell use and charging.

Startup Imprint Energy makes a flexible battery that could be used in wearable devices.

Startup Imprint Energy makes a flexible battery that could be used in wearable devices.

From the minds of entrepreneurs

I was actually surprised that there were so many projects at ARPA-E focused on battery software and sensing, given the program’s mission is to fund “moonshot” technology. Information technology is readily available and doesn’t have to be invented. Many startups and entrepreneurs are trying to build similar technologies with what’s already available off the shelf.

Startup GELI — which I once described as a sort of Android for grid batteries — is already selling its software to a handful of battery companies, which are in turn using it to manage grid batteries. The software can help a company monitor and control a group of batteries in real time for applications like providing energy storage for solar systems, storing and discharging energy when the demand for energy becomes out of balance with supply.

Berkeley Lab batteries

Another startup, Stem (founded back in 2009 as Powergetics) installs batteries at a building and uses software to convert the building over to battery power (instead of using power from the grid) when energy from the power grid is expensive (like a hot summer day). The company launched its first battery and analytics system with a hotel in San Francisco over a year ago, and now they’re working on installing about a megawatt of systems across Hawaii.

Remember the batteries already here

Beyond the newer markets of lithium ion batteries for electric cars and the power grid, better battery-management software will also be increasingly important for the markets that are already here and huge: cell phones, gadgets and laptops. Cell phones are becoming slimmer and more powerful, but are also connecting to increasingly powerful networks and using more and more energy-sucking apps. Better software to manage the device, the battery, and the wireless network itself has been an important area for investment in recent years.

All the major consumer electronics companies — from Apple to Samsung to Google — are searching for ways to extend battery life by whatever means possible for devices, whether that’s software, alternative energy sources or new battery chemistries.

But let’s face it: software is clearly a cheaper and quicker option than a new type of battery or a new untapped energy source.


Dick Beckman

In the early eighties I worked in the aerospace industry building low earth orbit (LEO) satellites, launching them, and building/operating their lifetime mission management and support systems. Managing batteries in these 90 minute cycles was critical and very challenging.

I was surprised to learn at that time that fully understanding batteries was one of the least advanced of the many edgy technological challenges we faced. As a result, little was left to chance or unchallenged human judgement. We needed these systems to effectively operate on-orbit for a decade and more.

So even if three decades has passed, I am not surprised that significant effort is still focused on this critical need. I’m confident current market demand will ultimately yield enough understanding to build adequately sophisticated metrics and software tools to achieve what we all require for electric-powered homes and transportation options.

Gaining understanding and the appropriate IP will require our legislature overcoming the grips of the powers that be.


You will want to quit your job tomorrow when you try this system and start getting paid online $7362 in 1 month. Go now ᵂᵂᵂ.fiscalpen.ᶜₒᴹ

Comments are closed.