Researchers to present breakthrough method at the end of the month

A team of researchers from North Carolina State University have developed a breakthrough method for one of the biggest headaches associated with using electronics — accurately determining the amount of charge left on a device’s battery.

It’s a frustration that everyone has had to deal with, and while numerous methods have been tried and tested, the NCSU methodology is far and above the most accurate one to date.

Problems with modern-day methods

Believe it or not, determining the amount of charge left on a battery is pretty difficult. This is due in large part to there being so many things contributing to the expenditure of the charge itself: the capacity of a battery to hold a charge decreases with time, so that’s one factor. Also, temperature, the rate at which the battery is charged, and a battery’s history of use are all contributing factors as well.

Modern-day methodologies allow the data from these variables to be plugged into a charge measurement model / algorithm just once. Obviously, this is a problem because these variables change in real-time situations.

The team’s research was presented in a paper entitled “Adaptive Parameter Identification and State-of-Charge Estimation of Lithium-Ion Batteries.” In it, the group created a list of all the pros and cons of modern-day methodologies:

Electromechanical approaches: The pro is that it is accurate. The con is that it is very difficult to implement.

Coulomb counting: The pro is that is easy to implement. The cons are twofold: 1) Implementation is dependent upon the initial state of charge, and 2) it’s not suitable for plug-in electrical vehicles, what with the technology’s frequent charging / discharging profiles, and its need for accurate initial conditions.

Open Circuit Voltage measurement: The pro is that there’s no need for an algorithm to implement. The con is that it needs the battery to be in rest mode for a long time in order to provide a measurement.

Extended Kalman filter: There are two pros to this one: 1) It’s very accurate, and 2) it can handle white noise. The cons are 1) there’s a large amount of computational memory and time needed in order for everything to be processed, and 2) the algorithm that needs to be implemented is incredibly complicated.

Sliding-mode observer: Pros are that 1) It’s accurate and 2) It can handle system-modeling errors. The cons are 1) the method is nonlinear, and 2) it is not easy to implement.

The NCSU way

So, with all of that being said, the NCSU group created a software program capable of identifying and processing data that can update in real time. This, in turn, allows it to more accurately estimate the remaining charge in a battery.

In fact, the team’s new technique allows models to estimate remaining charge within just 5%. How does this translate into the real world? Well, let’s say a battery’s state of charge is 45%: Using the new NCSU methodology, the real state of charge would be between 40% and 50% (that’s 5% below or above the estimate charge).

Good news for everyone

While we’ve only really mentioned the benefits for the the actual battery users so far, the new method is great for battery developers, too. “This improved accuracy will also give us additional insight into the dynamics of the battery, which we can use to develop techniques that will lead to more efficient battery management,” says Dr. Mo-Yuen Chow, a professor of electrical and computer engineering at NC State and co-author of the paper. “This will not only extend the life of the charge in the battery, but extend the functional life of the battery itself.”

The team’s method was primarily developed for batteries in plug-in electric vehicles, but it can be applied to battery use in other applications, too. It’s all described in the group’s paper, which can be accessed via the link below, and will be formally presented at the 38th Annual Conference of the IEEE Industrial Electronics Society in Montreal, October 25th-28th.

Story via: ncsu.edu