Erik Hoedemaekers MSc
- Electric vehicles
Part of TNO’s Power Trains department pursues the support of OEMs and TIERs in their developments regarding Energy Storage Systems (eg. batteries). For this purpose TNO investigates and applies current and future technologies related to batteries in vehicles. This research is performed together with OEMs, TIERs and other research institutes in different kinds of partnerships.
Battery models form the basis of most battery related research topics, ranging from basic simulation models up to complex electro-thermal pack models combined with prediction, state and/or parameter estimation.
The most rudimental purpose of battery models is to model the terminal voltage of the battery during a typical load-cycle applicable for that specific use-case. Additionally, the models can be expanded in different directions, eg. including thermal behaviour of model parameters is an often used addition.
Typically, the batteries behaviour is represented by (Electrical) Equivalent Circuit Models ((E)ECM), that are capable of capturing both dynamic and steady-state behaviour of the battery.
With the current focus of the Automotive Industry on accurate electrical range estimation of (PH)EVs as well as a strong focus on battery ageing topics, investigating online state and parameter estimation techniques is a topic that is a logical part of TNO's battery research.
The most common estimators that TNO develops are State-of-Charge (SoC) estimators, designed to accurately estimate the current remaining amount of ‘charge’/energy in the battery. These types of estimators are applied under demanding conditions, where straight-forward current integration does not suffice. The SoC estimators are making use of a parametrized battery model, which is cleverly combined with both voltage and current measurement by an Extended Kalman-Filter (EKF) based estimator. The SoC forms then the basis for electric range estimation, safety limit algorithms and other diagnostic purposes.
Additionally, TNO also developed several online parameter estimation techniques, both for electrical and thermal battery model parameters. These parameter estimators quantify the batteries performance (both in terms of power and energy) during the complete lifetime of the battery. Thereby providing the SoC estimators with model parameter updates during the lifetime, thus ensuring accurate SoC estimation, even at the end-of-life of the battery. Besides, they also provide diagnostic metrics, which compared to the initial values are commonly known as State-of-Health (SoH).
Closely linked to the online state and parameter estimations, TNO also develops different types of BMS algorithms. Most of these algorithms require inputs from one or more of the estimators to function correctly. As an example, the list of the BMS algorithms developed by TNO in the past includes 'Maximum power prediction' and 'Thermal prediction'.
For all previously mentioned developments TNO creates data with a variety of Battery Test Setups (BTS), ranging from small setups for cell level testing up to large setups for module and pack level testing. All of these battery cyclers can be combined with temperature (and humidity) controlled environments.
The testing mostly focuses on electro-thermal characterisation and validation, while in some special cases also abuse testing belongs to the possibilities. If necessary, the abuse testing can be combined with gas analysis (FTIR, GC-MS, GC-TCD, etc.).
In addition to the standard available measurements, like voltage, current and temperature, thermal images can also be obtained to observe/identify temperature homogeneity.
All mentioned expertise and knowledge is either developed internally, together with B2B customers or within partnerships like ABattReLife, AMBER, 3Ccar, etc.