On account of intense technological advances regarding Electric Vehicles, the state evaluation and prediction issues of Li-ion cells have become increasingly important for ensuring the competitiveness in terms of feasible performance and range. Albeit the wide investigation of various standard modelling and estimation techniques, only limited researches focus on their precision and applicability under heavy transient working conditions. This paper is concerned with Li-ion battery terminal voltage and State-of-Charge (SoC) prediction for two types of dynamic loads. Attention is focused on the investigation of the applicability of direct multi-step forecasting strategy in combination with Machine Learning. Beside that, a feature bank is composed of discharge profiles obtained at different C-rates. The set of discharge curves is proposed to complement the feature extraction, i.e. the additional historical data is considered for model building. Special care is devoted for the design of appropriate training data. Hence, a battery cell model is built for simulating intensive dynamic load scenarios in addition to the experimental setup. The cell model is validated by using measurement data. Results have demonstrated, that in case of WLTP-type discharge load of 0.3C-rate the forecasting performance is highly efficient on measurement data. Under dynamic loads of 1C-rate, or when small historical data is available, the application of feature bank improves the performance. We have obtained comprehensive results proving that the application of direct multi-step forecasting strategy using XGBoost represents a viable alternative to capture real-time the cell dynamics and predict the terminal voltage and SoC of Li-ion batteries exposed to dynamic loads.

Extensive investigation and prediction of the effects of dynamic battery loading is key to on-board Battery Management Systems (BMS) of Electric Vehicles (EVs) in order to ensure reliable operation and efficient energy management. In this paper, measurements of WLTP discharge tests at different temperatures are conducted on a Lithium Nickel Manganese Cobalt Oxide (LiNiMnCoO2) cell. Terminal voltage, discharge rate and temperatures at four points are taken into consideration. After, historical measurement data is used to build ensemble of boosted tree models and then predict cell voltage outcome sequence into the future. The efficiency of the performance is compared in case of various measurement sets. The results support the efficiency and applicability of direct multi-step-ahead forecasting strategy with standard Machine Learning techniques in battery SoC prediction.

The purpose of a Multiphysics Analysis is the modeling of complex physical systems where different kinds of physical effects act simultaneously. We would like to predict the behavior of a machine with the least number of neglected effects. Nowadays more Finite Element Method (FEM) based simulation software are readily available to study the physics of an electrical machine. Although, there are no standardized solutions that would fit every kind of electric motor analysis. In this paper we will present a method how to go through step-by-step on the different physical phenomena of an electrical machine and therefore analyze the different design and working aspects.

Modern electric drive systems require more sophisticated monitoring and diagnosis methods for determining the expected life-Time and reliability. Recently, continuous assessment of the electrical condition of electrical machines has become increasingly important. Investigations to predict battery life are of the utmost importance for Electric Vehicles (EVs). The evaluation of the instantaneous State-of-Health (SoH) of the battery cell packs requires a precise measurement system that encompasses the especially important specifications related to the operating conditions. In addition modern Battery Management Systems (BMS) rely on empirical battery models whose parametrization is crucial from the reliability's point of view. The model parameters are necessary to be indentified via the appropriate measurement data. To address these considerations, this paper presents a new test system for Li-ion batteries and control software based on LabView, which is capable of conducting high quality and completely automated battery cell measurements defined by user input test vectors and test suites. The test system allows conducting also real time condition monitoring in Li-ion batteries by discharge impulse responses. Finally several measurements on the experimental setup are performed. Analysis of the measured data validate that the designed device ensures high performance testing and post-processing.