Development and Optimization of Electric Drivetrains

Rotor position sensor systems based on resolvers are state-of-the-art in electric drive train applications, e.g. for permanent magnet synchronous motors, due to their robustness and accuracy. However, the effort for signal processing in terms of angle determination has a great complexity. Electric powertrains in automotive applications make high demands on the quality of control, which requires a precise detection of the current motor shaft position and speed. Different types of influences, like sensor signal cables, interfaces, sensor signal quantization, conditioning (e.g. filtering) and processing, cause error accumulation that interacts with the general mechanical-based disturbing factors, e.g. displacements, shock, vibrations and temperature influences. This work introduces a holistic view of the sensor signal measurement chain in terms of analysis of signal processing and error characteristics, exemplary shown on a state-of-the-art resolver system of an electric powertrain architecture. The second emphasis of this work investigates the mechanisms of signal processing and the signal processing - based effects that provoke additional errors to the measured rotor position information. The methods developed in the scope of this work are based on both, simulations of the entire electric drive train sensor system and verifications under realistic conditions by use of a highly accurate reference system on a specifically developed position sensor system test bench. The developed methods support measures of optimization to reduce the angular rotor position error of the entire sensor system based on an integration of computational and experimental investigations. As a result, causes and consequences of signal processing characteristics are pointed, which provides fundamental knowledge about the different contributing factors in terms of rotor position distortion. In this way, an approach is introduced that enables improvement and optimization of the resolver-based rotor position sensor system and thus the overall quality of control of electric traction motors even in the early design phase of electric drivetrain systems.

This paper presents a proposed high speed data acquisition and processing system that can measure electrical power quantity as well as other physical parameters from electric drivetrain of two wheeler and three wheeler electric vehicle and digitally process these captured data to calculate power in real time. Hardware requirements for this proposed system is discussed in this paper with major focus on developed signal processing algorithm which uses DFT (Discrete Fourier Transform) based power calculation algorithm that can calculate inverter's output and input power in real time. Data measured from this acquisition system can be used for electric drivetrain performance analysis, mileage estimation, drive cycle analysis and diagnostic for two wheeler and three wheeler electric drivetrain.

The electric drivetrain is a key technology for future mobility. Safety, efficiency and cost reduction are main factors to influence the market acceptance of electrical vehicles. Within the MotorBrain project, the partners are developing a new drivetrain for an electric car with approx. 1000 kg weight. The drivetrain will avoid the use of rare earth magnets. The paper presents the electrical machine design from electromagnetic point of view in combination to the inverter design, with a combined view on motor and inverter. Different technology shifts are addressed: SMC rotor with ferrite magnets to reduce cost and avoid rare earth magnets; 9 phase system to reduce phase currents and utilize smaller power modules; new capacitor technology (ceramics) for better temperature performance and lower size; and enhanced safety technology with special safety measures in hardware (peripherals) and software (monitoring functions). The paper highlights the benefits of the different technologies in context of the electric, mechanical and electromagnetic system design.

In this paper we study the effects of increased unsprung mass that result from the use of in-wheel electric motors used in hybrid and/or electric drive trains. For this purpose we reduce the problem to the analysis of the quarter car suspension model to assess how increasing unsprung mass affects the ride comfort. In order to see the change in the ride comfort level according to ISO 2631 the analysis was done in frequency domain using random road profile inputs generated according to ISO 8608. Also the results were assessed using a commercial multi body vehicle simulation software. Finally we suggest empirical solutions regarding modified suspension and tire parameters to compensate the detrimental effects of increased unsprung mass.

3 - The market for battery electric vehicles

University fleets represent an enticing opportunity to explore the near-term feasibility of achieving net-zero-carbon emissions in transportation. In many instances, universities operate much like a small, self-contained ecosystem with all the same transportation needs as a larger municipality, but with a smaller geographic footprint. Their fleets often include a wide variety of vehicle types serving the campus, including low-speed vehicles (e.g., golf carts), light-duty sedans, SUVs, and pickups, as well as medium-duty trucks and delivery vehicles. The mix of vehicle and operational needs combined with broader activities related to net-zero campuses makes universities and colleges unique microcosms to determine the feasibility of and path to achieving net-zero fleets. As the availability of electric drivetrains expands beyond light-duty sedans, fleets need to understand when it will be operationally and financially appropriate to start adding electric drivetrains to their fleets. To better understand these opportunities, NREL contracted Sawatch Labs to analyze the role electric vehicles (EVs) can have in helping universities meet net-zero emissions and fleet sustainability goals they have instituted.

<div class="section abstract"><div class="htmlview paragraph">The prime factor which influences noise and vibrations of electro-mechanical drives is wear at the components. This paper discusses the numerical methods developed for abrasion, vibration calculations and the coupling between wear and Noise Vibration and Harshness (NVH) models of the drive unit. The vibration domain model, initially, focuses on the calculations of mechanical excitations at the gear shafts which are generated via a nonlinear dynamic model. Furthermore, the bearings are studied for the influences on their stiffness and eventually their impact on the harmonics of the drivetrain. Later, free and forced vibrations of the complete drivetrain are simulated via a steady-state dynamic model. Consequently, the paper concentrates on the abrasion calculations at the gears. Wear is a complex process and understanding it is essential for determining the vibro-acoustics characteristics. As part of wear modeling, a finite element static model used for abrasion calculations is described. The special subroutines developed aid in the coupling of the wear equations, various contacts and friction formulations to the numerical model. The abraded solution model is then inputted into the vibration module for further analyzing changes in the velocities of the housing surfaces. As part of the method validation, a drivetrain with a three-stage (gear) drivetrain model is developed. The simulation model allows studying the influences of wear at the gear components and its impact on the vibrations of the whole structure. Eventually, the experimental calculations for wear on the gear surfaces and surface vibrations of housing are performed with an optical microscope and laser vibrometer, respectively. Finally, it is shown that the developed simulation method allows for a better comprehension of the influences of abrasion on the vibro-acoustics of drive units.</div></div>

Torsional drivetrain vibrations in hybrid and electric drivetrains, which occur during traction and braking control, are commonly encountered. This problem occurs mainly due to the weak damping characteristic of the hybrid/electric drivetrain and the fast time response characteristics of the electric motor. An active motor damping (AMD) control algorithm for a hybrid electric vehicle, considering ABS braking maneuvers on different road conditions, is developed in this work. The control problem is structured such that the control objective is disturbance rejection against the estimated brake torque signal received from the ABS module. Communication delay between the ABS and motor control module is handled within the control strategy. State feedback control strategy is used with the linearized version of the non-linear state-space equations together with an Extended Kalman filter. Gear backlash is taken into account without the need to estimate the mode of backlash. In the development of the controller, state feedback gains are set such that the ABS functionality on tire slip regulation is not altered. Existing studies in the literature feedback only the angle of twist and axle wrap angular speed for damping the vibrations. The structure of the controller in this study is different in that it estimates and feeds back tire slip, in addition to these two. Simulation results show the effectiveness of the controller in reducing the angle of twist without deteriorating ABS tire slip control and braking performance of the vehicle.

Integrated Architectures for Third Generation Electric Vehicles – First Results of the ELVA Project

Electrical energy storage systems in electricity generation: Energy policies, innovative technologies, and regulatory regimes

Torsional drivetrains oscillations is a well known problem in electrical or hybrid electric drivetrains which may greatly influence the driving comfort. An experimental testing setup that can be used to design and evaluate active oscillation damping schemes is proposed in this paper. Instead of in-vehicle test, the drivetrain transmission is emulated from mathematic models, executed by a servo drive which is connected to the physical electrical drive system. The proposed setup functions as a hardware-in-the-loop simulator and can be easily adjusted to simulate any drivetrains without additional cost.

The paper reviews new drivetrain concepts including plug-in full hybrid, mild hybrid designed for 48 V and zero emission vehicles like battery electric vehicle. The presented drivetrain concepts should lead to reduction of CO2 emission what is one of the most important goals of future transportation. Comparison between parallel and series hybrid drivetrain is specially delivered. The series hybrid drivetrain could be efficient alternative to parallel one. Such drivetrain concept, where electric drivetrain can be separated from power unit, can be adapted for fuel cell vehicle or battery electric vehicle. Mild hybrid solutions for 48V due to optimised costs should have more importance in the near future.

In electric drivetrains, the traction machines are often coupled to a gear transmission. For the noise and vibration analysis of such systems, linearised system models in the frequency domain are commonly used. In this paper, a system approach in the time domain is introduced, which gives the advantage of analysing the transient behaviour of an electric drivetrain. The focus in this paper is on the dynamic gear model. Finally, the modelling approach is applied to an exemplary drivetrain, and the results are discussed.

The dilemma between high control performance and high fault tolerance for a power electronic penetrated shipboard propulsion system (SPS) impedes the development of more-electric ships (MESs). In this article, with a doubly fed induction machine (DFIM) electric drivetrain adopted, a partially power decoupled SPS structure is presented to enhance the system robustness to solid-state device failure. Two electric machines, which are the synchronous generator (SG) and DFIM, generate and consume power in a DFIM-SPS, respectively. However, the coupling effect between their controls results in potential threats to the system stable operation. The phase-locked loop (PLL) functions as the key factor of interacting their operation, and its mechanism is illustrated in detail for DFIM-SPS to explain the deteriorated performance. This article proposes an emulated stator voltage-orientated vector control (ESVO-VC) strategy with coupling effect elimination to realize separate control of SG and DFIM electric drivetrain. The proposed ESVO-VC strategy is verified by simulation in MATLAB/Simulink for a 10-MW DFIM-SPS, in which three-stage mechanical load torque changes are presented to indicate different working conditions. Moreover, the power circuits of SG and back-to-back power converter (BTBPC) are simulated in a real-time digital simulator (RTDS), and the proposed ESVO-VC is further validated by a control-hardware-in-the-loop (CHIL) setup.

Electrifying road transport is seen as one of the key components in decreasing the carbon footprint of the society as a whole. Recent developments in electric drivetrain and battery technology have helped to design vehicles with ranges that make them independent of public charging infrastructure during most sub-urban and commuting trips. Once long-haul trips are planned, however, these vehicles require a dense network of charging infrastructure. In this paper, the impact of a large-scale electrification of vehicles in long-distance trips is evaluated by combining an agent-based long distance transport model of Sweden with a detailed model of energy consumption and battery charging. Energy consumption and charging schemes are simulated for different types of vehicles and chargers. In a first application, all vehicle traffic is electrified. Results demonstrate that the daily estimate for energy consumption is in the region of 150 GWh. This equals roughly 40% of the current Swedish electricity consumption. Energy consumption is the highest along in the motorway network connecting the south of the country (Malmo, Goteborg and Stockholm). Along these motorways, also the highest demand for charging infrastructure arises. Nationwide, two peak times for vehicle charging seem to exist: One is around lunch time and another in the mid-afternoon. During the first peak, overall energy demand is presumably the highest.