Abstract
A soft sensor is presented that approximates certain health parameters of automotive rechargeable batteries from on-vehicle measurements of current and voltage. The sensor is based on a model of the open circuit voltage curve. This last model is implemented through monotonic neural networks and estimate over-potentials arising from the evolution in time of the Lithium concentration in the electrodes of the battery. The proposed soft sensor is able to exploit the information contained in operational records of the vehicle better than the alternatives, this being particularly true when the charge or discharge currents are between moderate and high. The accuracy of the neural model has been compared to different alternatives, including data-driven statistical models, first principle-based models, fuzzy observers and other recurrent neural networks with different topologies. It is concluded that monotonic echo state networks can outperform well established first-principle models. The algorithms have been validated with automotive Li-FePO4 cells.
Highlights
Battery costs comprise an estimated 25–50% of the electric vehicle [1]
The experiments were conducted on three different batteries: Battery #1 is a 42Ah pouch battery from European Batteries, and batteries #2 and #3 are cylindrical commercial Lithium Iron Phosphate (LFP) cells manufactured by A123 Systems (Livonia, MI, USA), with 2.3 Ah name plate capacity
The first part of the experimentation consists of a numerical comparison of the accuracies of the present approach and other Open Circuit Voltage (OCV) models in terms of the residual of the approximation of the model to the “ground truth” OCV measured at the laboratory
Summary
Battery costs comprise an estimated 25–50% of the electric vehicle [1]. Maximizing the life expectancy of a battery has undeniable economic implications, preventing the conditions that shorten Lithium battery life (high temperatures, overcharging, deep discharges, high current, etc.) is a major concern. A soft sensor is proposed that combines transformation models [5] with reservoir computing [6] in a new class of monotonic Echo State Networks (ESN). The new soft sensor is able to exploit the health-related information contained in operational records of the vehicle better than the alternatives, when the charge or discharge currents are moderate to high. To validate this assertion, the accuracy of the new sensor has been compared, over automotive Li-FePO4 cells, to a selection of model-based observers of the state of health, including data-driven statistical models, first principle-based models, fuzzy observers and recurrent neural networks with different topologies.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
