Search

Estimating the impedance of lithium-ion cells is of increasing interest, especially on-board electric vehicles to enlarge their diagnostic capabilities. Therefore, an innovative approach for cell impedance measurements is presented in the context of intelligent battery management systems. Hereby, the load current excitation injected by the electric powertrain is switched by power MOSFETs. This quasi-active impedance spectroscopy can be performed with a reconfigurable battery system to shape the excitation signal without further hardware. A broadband data processing algorithm that handles both pulsing and non-pulsing data in the frequency domain is introduced. It generates impedance data of high quality while reducing the computational efforts compared to previous implementations. To evaluate the proposed methods under real world driving conditions, an experimental setup that emulates the application within a traction battery system is constructed. The obtained measurement results are in great accordance with those of reference laboratory equipment.

Besides the performance and range requirements, the breakthrough of electromobility depends crucially on the safety of battery systems. Cell faults that can lead to thermal runaway of the energy storage reduce customer acceptance. Thermal runaways are often preceded by an Internal Short Circuit (ISC). Thus, there is a necessity for a method that is able to detect the formation of ISCs, before a significant temperature rise is observed and the risk of thermal propagation arises. To achieve this, a data-driven method that enables early detection of soft ISCs is presented. The unsupervised machine-learning method is based on linear Principal Component Analysis (PCA) and nonlinear Kernel PCA (KPCA). Since the method only requires fault-free voltage measurements for training, it is directly applicable in conventional battery systems. The nonlinear KPCA is thoroughly compared with the linear PCA using experimental data. The data originates from a module consisting of twelve automotive cells. While the linear method has advantages in computational complexity, the nonlinear method detects ISCs earlier due to its high sensitivity and specificity.

Embedding impedance spectroscopy into onboard applications such as electric vehicles increases the diagnostic capabilities of battery management systems largely. A novel approach is presented and experimentally investigated throughout this work, whereby the load current excitation injected by the electric powertrain is pulsed by transistors. With a reconfigurable battery system, this form of active impedance spectroscopy can be used without further hardware. To obtain high-quality impedance data, a broadband data processing algorithm in the frequency domain is employed that deals inherently with both pulsing and non-pulsing data. An experimental setup is constructed to evaluate the proposed methods under real world conditions and the measurement results are in great accordance with laboratory reference measurements. Finally, a linear regression model acts as filter to suppress outliers, smooth the impedance spectra and thus describe the time-varying impedance of each cell.

To increase the monitoring capabilities of battery management systems in electric vehicles, innovative sensor technologies and system identification methods are implemented. Impedance spectroscopy is one of the most promising techniques, especially for the estimation of battery cell aging. Interpreting the impedance spectra is not trivial due to severe disturbances on board electric vehicles that distort the measurement results. In this work, the impact of temperature on aging estimation is investigated by sensitivity analysis. Optimal test frequencies that represent relevant aging effects and minimize the temperature influence are identified and their uncertainty is quantified. Various experiments with multiple lithium-ion cells show the general applicability of the proposed methods.

Nonlinear impedance spectroscopy is a novel method for the diagnosis of lithium-ion battery cells. To thoroughly evaluate the application of nonlinear impedance measurements within this contribution, the topic is theoretically investigated at first. The ordinary frequency response function is extended to the Volterra model to incorporate nonlinear behavior. The applied data-driven approach simplifies the assessment of nonlinear distortions based on Fourier transforms. To experimentally analyze the frequency response of battery cells in a defined manner, a measurement setup is constructed that features a mechanical clamping device. Next to controlling the temperature and the state of charge (SoC), the force compressing the cell can be specifically adjusted with this test rig. Thereby, the influence of compressive force on the impedance spectrum and its harmonic content is examined. In addition, it is analyzed if and how the mechanical influence can be distinguished with nonlinear impedance spectroscopy from other impedance-changing conditions such as temperature and SoC. The investigation based on pouch battery cells results in diverse diagnostic scenarios, where nonlinear impedance measurements are a valuable extension to traditional battery characterization tools. The comparison is complemented by evaluating drawbacks of the method and the impact of the measurement procedure on the device under test.

An important issue towards cell individual battery monitoring is the employment of new sensor technologies and efficient algorithms to enlarge the diagnostic capabilities of battery systems. Especially for the onboard application in electric vehicles, impedance spectroscopy is one of the most promising methods to gather useful information about the battery. In this paper, a new framework for onboard impedance spectroscopy on cell level is developed, which incorporates a passive method without the need of an additional active signal generation. The derived estimator is based on the weighted overlapped segment averaging algorithm and is optimized for the application in a battery management system. The theoretical considerations are experimentally examined with a battery module consisting of parallel connected cells. Therefore, a battery test rig with high bandwidth is designed to enable a realistic replication of road driving scenarios. Statistical quality criterions for the online validation of impedance spectra using linear Kramers-Kronig tests are established and verified with corresponding experiments. A significant comparison with reference measurements based on the accuracy of estimated model parameters shows very good agreement. Next to model parameter estimation, diverse other application scenarios for the monitoring and diagnosis of battery systems based on impedance data are outlined.

In recent years, intelligent battery systems came into researchers’ focus, which comprise sensors and actuators on cell level. These architectures allow the battery management system to observe and to control the current flow within the battery system, which is particularly promising for battery electric vehicles. Besides, insight into battery cells’ states and model parameters is essential for valuable battery management and is often achieved by online state and parameter estimation. Though, during real-world operation the system excitation is often insufficient for an accurate estimate. Within this contribution, we present strategies that utilize the actuators to improve the system’s excitation and thereby enhance the observability. Controlling the current flow with the objective of enhanced state and parameter estimation in time domain is a novel approach. The benefit of switching for state and parameter estimation is investigated simulatively and experimentally with NMC/graphite lithium-ion cells. Furthermore, the switching operation’s influence on degradation is discussed. Results show that the investigated switching strategies enhance the accuracy of state and parameter estimation by 4% to 30% depending on the inspected parameter, while a negligible impact on cell aging is expected considering the cell heating caused by switching operation is below 1.5%.

In the field of signal filtering sensor noise variance estimation is of high interest. Various approaches exist for different applications. Within this work, we propose a novel online noise variance estimation scheme based on an online algorithm with exponential forgetting. The approach serves as an extension of a sensor data fusion algorithm that was presented earlier for the application within multi-cell battery systems equipped with cell-individual sensors. Utilizing measurements of electrical-linked sensors the signals and their noises are separated, and the noise variance is adaptively determined. Experiments show that sensor data fusion is equivalent to common methods like low-pass filtering to gain the target signal. Consequently, the variance is estimated with high accuracy especially with regard to signals featuring high dynamic range. Moreover, the results are on a par with difference-based noise estimation. Furthermore, the influence of relevant parameters on the method is investigated namely the adaptivity of the algorithm and the necessary number of involved sensors. As a result, with just eight sensors decent results are achieved within an exemplary application.