Development and validation of a Kalman filter based load torque estimation method for electric axle drives. Issue 1 (22nd December 2022)

• Record Type: Journal Article
• Authors: Kalcher, Robert; Ellermann, Katrin; Kelz, Gerald; Reisinger, Karl Heinz
• Is Part Of: International journal of vehicle performance. Volume 9:Issue 1(2023)
• Page Start: 41
• Page End: 72
• Publication Date: 2022-12-22
• Subjects: load torque estimation -- electric rear axle drives -- ERAD -- electric rear axle drive -- virtual sensing -- Kalman filter -- multi-body systems -- multi-body simulations -- MBS -- multi-body system -- powertrain test rig -- simulation-measurement adjustment; Vehicles -- Design and construction -- Periodicals; 629.049
• Languages: English
• ISSNs: 1745-3194

Abstract

Based on the automotive development strategies towards electrification and autonomous driving, the need of enhancements regarding state estimations, extended diagnosis as well as redundant architectures are evident. Accurate, robust and cost-efficient load torque estimation methods can bring significant benefits in this respect. Thus, this work presents a Kalman filter based load torque estimation method for electric axle drives by mean of virtual sensing. The elastically mounted electric rear axle drive of a Renault Twizy 80 was installed at a powertrain test rig. Using the measurement results, the parameters of a multi-body system (MBS) model were adjusted, in order to obtain a reference model for method validation. Subsequently, the basic functioning of the proposed load torque estimation method was shown by several test manoeuvres based on the measurement-adjusted reference model. The investigations revealed favourable results even in the case of reasonable effective torque dynamics (up to 3 Hz) and realistic sensor noise.