Paul Büschel

Calculation of the distribution of relaxation times for characterization of the dynamic battery behavior

Impedance spectroscopy is widely used for characterization of batteries. Beside model based approaches for data analysis, the impedance spectra can be used to calculate a distribution of relaxation times (DRT). These DRT spectra can be seen as the time domain representation of the linear battery behavior as it describes the relaxation for different time constants. In contrast to the impedance spectra, it can be used to directly simulate the linear response of the battery. In this paper a calculation method for the distribution of relaxation times is shown that employs an iterative regularization scheme combined with stochastic methods for determination of the regularization parameters. For preserving peaks in the relaxation spectra during calculation edge preserving techniques are implemented.

Distribution of relaxation times for effect identification and modeling of impedance spectra

Impedance spectroscopy provides an increased amount of information with one measurement. The drawback is that a suitable model as well as a reliable fitting is needed for data evaluation. Therfore a suitable data representation is needed that serves two purposes. First it should help to identify processes within the device under test as a support for building a model and it should give some initial values for parameterization. The distribution of relaxation times (DRT) provides such a data representation. It transforms impedance data from the frequency domain to a time constant domain were different relaxation mechanisms can be separated more easily and sometimes their type can be identified using the shape of the DRT. In an example the DRT is used to identify mechanisms and to build a suitable model for the impedance data analysis of battery aging data.

Online cellphone battery entropy measurement for SoH estimation

The widespread use of batteries requires knowledge of their state of health (SoH). In this paper we present a method to use entropy evolution as an aging indicator for batteries. Voltage, current and temperature are monitored in battery cells used in cellphones and in real-time operation. Irreversible entropy is estimated from irreversible energy dissipation in the battery for long periods of use. Simple open circuit battery voltage Uocv models were evaluated and compared to lab measured Uocv. It was found that the entropy generation rate is always increasing during battery operation and that irreversible entropy is a fingerprint of aging for batteries.

Study of excitation signals parameters for portable bioimedical devices

The aim of this paper is to select the suitable excitation signal for a portable biomedical system based on the simulations of different signals while respecting the constraints on the system including excitation time, energy consumption, cost and ease of generation and at the same time guarantying the safety of the patient by limiting the currents magnitude. In order to get the optimum working point that minimize the trade-off between the quality of the reconstructed signal, the time consumption and the hardware complexity when applying the perturbation signal to the system under test, the interaction between pertinent parameters is determined. These parameters are the number of the samples, the sampling frequency and the noises level. The number of samples (N) is directly related to the memorys size of the hardware used in the measurement, the sampling frequency (Fs) is associated to the clock and the noises level is affiliated to the algorithm or the measurement hardware. To achieve a good signals reconstruction, the ratio (Fs/f) should be lower than (~2*N) and the windows size which is proportional to (N, Fs/f) as large as possible.

Investigation of the electrode surface of a liquid quality sensor by local impedance spectroscopy

Local electrochemical impedance spectroscopy (LEIS) is a suitable technique for the investigation of electrode surfaces. In this paper, we investigate the correlation between changes on the electrode surface with the results of LEIS. For LEIS measurements, a novel measurement set-up is developed with high mechanical stability and a sufficient space resolution of 200nm. Three identical liquid quality sensors were exposed to aging with different immersion time. Using scanning electronic microscopy (SEM), the results depict that the deposits increase with immersion time and have typical dimensions of 0.7 mm2. The LEIS investigation shows a good sensitivity to deposits on electrode surfaces and results are confirmed by the SEM investigation.

Effiziente Parameterschätzung impedanzbasierter Sensoren durch lokale, lineare Transformation

Zusammenfassung In diesem Beitrag wird der Ansatz zur Auswertung von spektralen Datensätzen durch eine lineare Transformation vorgestellt. Ziel dieser Transformation ist die effiziente Bestimmung eines Eingangsparameters des Sensors, um einen zuvor definierten Arbeitspunkt, unabhängig von weiteren im Datensatz repräsentierten Eingangsparametern. Die Ersparnis an Rechenzeit und Rechenleistung bei vergleichbaren Ergebnissen – gegenüber der sonst üblichen nichtlinearen Regression – ist besonders bei rechnenintensiven Modellen und leistungsschwacher Hardware, wie beispielsweise in Embedded-Anwendungen, vorteilhaft. Die Parametrierung der linearen Abbildung durch das Ausgangskennlinienfeld selbst erlaubt die Berechnung der Eingangsparameter ohne explizite Kenntnis der Modellgleichung. Alle Prozessschritte wie Datenauswahl, Bestimmung der linearen Koeffizienten und Anwendung der Transformation werden erläutert und an einem Beispiel für zwei Parameter demonstriert. Die Ergebnisse der Parameterschätzungen werden anschließend mit verwendeten Eingangsparametersätzen verglichen und diskutiert.

Eingebettetes Impedanzmesssystem für das Batteriemanagement in Elektrofahrzeugen

Zusammenfassung Für Batteriemanagementsysteme (BMS) sind Umfang und Verlässlichkeit der Datenbasis von besonderer Bedeutung. Bisher kommt ein Großteil dieser Daten aus Prüfständen und wird dem BMS als Tabelle bereitgestellt. Um den Einfluss dieser Offline-Daten im BMS zu reduzieren und so die Leistungsfähigkeit eines Batteriesystems zu steigern, wurde eine Hardware entwickelt, welche eine Online-Messung des Impedanzspektrums von Batterien ermöglicht. Der Fokus liegt hierbei auf der Adaption der im Labor häufig eingesetzten Impedanzspektroskopie, um diese in mobilen Anwendungen mit Low-cost-Hardware verfügbar zu machen.

Impedance spectroscopy in solid state electrolyte characterization

In this work, a method for the determination of the ion concentration and the diffusion coefficient of lithium ions in a phosphorus oxynitride will be presented. The reproducibility of the parameter extraction and the entire measurement process will be evaluated. The obtained results are in the value range of publications of similar sample properties.