Electrifying cars will help us reduce emissions, but they need to drive further on a single charge of the batteries to truly take the mantel from combustion engine vehicles. Connecting them to the internet of things could help them achieve this.
The automotive industry is going through huge upheaval as it develops the technologies that will keep us mobile for decades to come but without the huge impact on the environment that we currently have.
At the forefront of this change are the electrification of the car and greater levels of connectivity.
Electrified cars are slowly growing their market share, and showing the first signs of being a true replacement for the combustion engine, but in these initial embryotic times how far we can drive on a fully charged battery has become a sticking point.
Batteries are getting better and you can now reach 400km in a battery electric vehicle, and well into double figures for a plug-in electric vehicle (PHEV) before needing to plug in.
Work is moving at a relatively quick pace, with new battery technology being developed, including Ultra Fast Carbon Batteries as I wrote about in a previous article, but humans always want more and tend to be impatient.
So how can we get even greater efficiency – and range – from EVs without having to wait for the next battery technology. The answer is in another growing area of research and development in the automotive industry: connectivity.
We spend our lives connected, for the majority of us that’s done through our smartphones, but also includes personal computers, smart devices, even our fridges as we constantly upload and receive information and data from the internet of things. And the humble car is next on the list to get connected.
Researchers at Michigan Technological University – with backing from the US Department of Energy, ARPA-E agency and in partnership with General Motors – are looking at how using data from different sources could be used to alter powertrain dynamic control to improve overall efficiency.
Put simply, they want a connected electrified vehicle to use data to improve powertrain efficiency so we can drive further on a single charge. The target is a 20% reduction in energy consumption.
The base vehicle that researchers used in the project was a Chevrolet Volt PHEV with no modifications, and they developed a test route to track the vehicle through and see how it coped in real world driving conditions rather than using testbeds.
The route was 38.6km long with numerous obstacles: five sets of traffic lights, 13 stop signs, two potential stops at lift bridges and various elevation changes – all of which could impact significantly on vehicle efficiency – not to mention the differing traffic levels.
To overcome this long list of variables, researchers used GPS data to get a clearer picture of route and traffic data, vehicle-to-vehicle information to understand traffic congestion and vehicle-to-infrastructure data for traffic light information.
Researchers then developed four control technologies to apply to the test vehicle to try and improve efficiency over the test route. The control systems worked either over the entire length of the route or predicted obstacles shortly ahead to improve efficiency.
The first system is a long-term prediction tool to forecast best use of battery and combustion engine in the of the PHEV for the entire driving route. The process begins by establishing the route and getting average velocity and grade to forecast energy usage of the vehicle powertrain; determining the best use of the battery and combustion engine over the test route. Over the researcher’s test route this system has been shown to reduce energy consumption by an average of 6% and as high as 12%.
The second system uses a nonlinear model predictive control and knowledge of the route, road grade and speed limits, to develop a 10 second prediction of optimal vehicle velocity. Not only can this approach boost efficiency by over 4%, on average, the algorithm can also reduce transit time by 1.1%.
Next, as many plug-in hybrids have different driving modes – combustion engine only, battery power only, and a blend of the two – connected data can be used to better forecast which mode should be used, so the vehicle is in the optimal operating mode ahead of the driving situation occurring. The average energy reduction noted with the implementation of this method is approximately 6%.
The final technology involves a 10 second prediction of the torque split between the combustion engine and electric motors for a given operating mode based on information such as vehicle speed and road gradient, allowing the car to smooth out the power requested to minimise transients and excessive energy loads. On the researcher’s test route, the system shows a 4 to 6% benefit in energy consumption.
There’s still work to be done, as although testing shows that individually the developed technologies can save between 4% and 12% energy on the test route, the next step for the team is to couple all four technologies and determine the impact on efficiency, still with the objective of reducing energy consumption by the targeted 20%.
If the researchers can hit that point then it won’t be long before car manufacturers are taking the teams work onboard for the next generation of electrified vehicles, taking us a little bit further on each charge of the battery.
Are passenger drones about to hit the skies?
To enable comments sign up for a Disqus account and enter your Disqus shortname in the Articulate node settings.