Automobile manufacturers are constantly searching for more efficient and robust batteries for electric vehicles (EVs) — but it’s not easy.
The performance of the battery and the vehicle depends on the output the electrolytes generate, and there is scope to improve the output.
To get there, scientists or researchers must try various permutations and combinations of molecules and test the performance of the electrolyte the combination produces.
It’s a long-drawn process, and there are two potential results — you either find a great combination, or it’s a dud.
The main problem with this approach is the amount of time it takes. Given that there are 10 billion commercially procurable molecules today, it’s a daunting task to create and test so many combinations of five molecules.
Here, Generative AI (GAI) plays an important role. GAI can quickly generate accurate molecule combinations that produce better output and provide either efficient charging or better power management (or both).
Let us explore the impact of generative AI in the world of EVs.
The Top Two Problems Electronic Vehicles Need to Solve
– Range issues
Range means the distance an EV can travel on a single charge. There is a lot of confusion and anxiety for EV owners when they take their cars out, especially on long road trips, in case they run out of charge in the middle of nowhere.
Range depends on batteries that provide better output, a problem the industry is grappling with.
– Charging speed
Slow charging is a problem, but fast chargers are expensive. In the US, three chargers are available – Level 1, Level 2, and Level 3, with Level 3 being the fastest and the most expensive.
How Generative AI can help
– Molecular Management
Take the case of Aionics, a startup working on providing clean energy with the help of AI.
It is using Generative AI to produce better batteries for EVs by using GAI to recognize the molecules that have been tested and move on to others.
Two, Aionics trains Generative AI models on existing batteries to learn about existing combinations and produce new ones.
Third, AI models are trained on resources around chemistry and physics to winnow down many possible combinations of molecules that might not be useful.
The shortlisting continues until a few samples or combinations are found. The combinations are sent for validation. If the samples don’t work, they’re subject to further iterations until they can go to the market.
“If we don’t get it on the first round, we iterate and we can run some clinical trials to prove it until we get to the winner. And once we find the winner, we work with our manufacturing partners to scale that manufacturing and bring it to market.”
– Battery Performance Optimization
Generative AI can quickly and efficiently try various combinations of chemicals to determine which combination can enable EV batteries to produce optimum performance, longer life, and faster charging.
– Efficient Charging Algorithms
Generative AI can learn about the batteries and identify charging algorithms that enable these batteries to optimize their output. This will likely positively affect the cost of production of EV batteries, making life cheaper for customers.
– Predictive Maintenance
When Generative AI learns about EVs and their batteries, it picks up much information on the maintenance cycles. These circumstances can lead to premature issues, tools that can impact EV performance, speeds that affect EV performance, and more.
Accordingly, GAI can provide EV makers with analytics, insights, and dashboards that enable predictive and proactive maintenance information.
For example, EV makers can have dynamic analytics on the battery life and the nearest charging station of different battery types. They can fit the EVs with systems that alert the drivers when they should proactively charge their batteries to avoid problems.
Generative AI appears to be a tool that can not only solve many problems the EV industry faces but also ignite an innovation spree.
The EV industry is still taking baby steps toward adoption, and various legal, economic, infrastructural, logistical, and geopolitical issues provide a lot of hindrances.
Companies and startups are doing some trend-setting work, and in turn, this may inspire a concerted and widespread effort toward making EVs mainstream.