Pablo M. de la Barrera

High-Resistance Connection Detection in Induction Motor Drives Using Signal Injection

Two new methods for high-resistance connection (HRC) diagnosis in induction motor (IM) drives are presented in this effort. These automatic offline methods are based on a signal injection strategy applied to the IM at standstill. By measuring the voltage in the machine neutral point and phase currents, it is possible to detect and isolate connection problems. Experimental results showing a good sensitivity to HRCs and immunity to symmetrical stator resistance variations were obtained in a laboratory setup.

Multi-Domain Model for Electric Traction Drives Using Bond Graphs

In this work the Multi-Domain model of an electric vehicle is developed. The electric domain model consists on the traction drive and allows including faults associated with stator winding. The thermal model is based on a spatial discretization. It receives the power dissipated in the electric domain, it interacts with the environment and provides the temperature distribution in the induction motor. The mechanical model is a half vehicle model. Given that all models are obtained using the same approach (Bond Graph) their integration becomes straightforward. This complete model allows simulating the whole system dynamics and the analysis of electrical/mechanical/thermal interaction. First, experimental results are aimed to validate the proposed model. Then, simulation results illustrate the interaction between the different domains and highlight the capability of including faults.

Online Voltage Sensorless High-Resistance Connection Diagnosis in Induction Motor Drives

A new method for online high-resistance connection (HRC) diagnosis in induction motor (IM) drives is presented in this effort. This voltage sensorless method is based on a signal injection strategy applied to the IM during its normal operation. The signal is injected in the d-axis in such a way that it provides the phase currents with a dc component while avoiding torque perturbation. By measuring phase currents, it is possible to detect and isolate HRC problems. Experimental results showing high sensitivity to HRCs and immunity to symmetrical stator resistance variations have been obtained in a laboratory setup. The low perturbation produced by the injection method over the IM variables has also been experimentally demonstrated.

Multi-domain modeling of electric traction drives using Bond Graphs: Application to fault diagnosis

In this work a dynamic model of the induction machine that considers the thermal behavior is developed using bond graphs (BG). The thermal model is based on considering the slots and teeth as independent elements with their own temperature. It corresponds to a real motor used in the tractive wheel of an electric vehicle. The model of the vehicle is also obtained in BG and is coupled to the motor. The complete model allows simulating the whole dynamics in a single model and the analysis of electrical/mechanical/thermal interaction is performed. Simulation results are intended to illustrate this interaction under faulty and healthy conditions in the stator winding. The impact on the mechanical and thermal domain produced by a fault in the electrical domain is also analyzed.

Modeling of electric vehicles dynamics with Multi-Bond Graphs

The construction of a model that represents the behavior of an Electric Vehicle is studied in detail. The contribution of this work is twofold. On one hand an efficient and compact way to model dynamical systems is introduced. On the other hand, the power interchange between the electrical and mechanical sub-models allows a deep understanding of the dynamics involved in electrically driven vehicles. The approach used to model the mechanical parts (chassis, suspension, wheels) and the induction motors is the Multi-Bond Graph based in models discussed in recent literature. Then these models are integrated in order to form the complete model that simulates the whole system dynamics. Simulation results are aimed to illustrate the electromechanical interaction (ABS, regenerative braking, etc) as well as the evolution of certain variables during a risky situation. Conclusions are obtained based on these results.

Induction machine models for efficiency studies in EV design applications

Different induction machine (IM) models are evaluated, with the aim to determine its usefulness in efficiency studies for electric vehicles (EV) traction drives. In particular, two simple IM models proposed in the literature, which take into account the additional losses, are compared. With this aim, model parameters are adjusted from experimental tests performed on a 4 kW IM. Then, obtained results from both models at different operating points are compared. Even when the estimation errors are similar using both models, one of them is chosen because of its simplicity. Finally, a simple application of this model is proposed, which consists in use the model for obtaining a complete efficiency map of the IM, without need further essays. Such efficiency map can be used to evaluate the overall efficiency of the traction drive of an EV or in EV design.

Behavior of electric vehicles and traction drives during sensor faults

The behavior of an electric vehicle and its electric traction drive under sensor faults is studied in this paper. With this aim, simulations over a vehicle model with rear traction, using an induction machine, were carried out. The drive model can switch between three different conventional control strategies. With this setup, the longitudinal dynamics behavior of the vehicle and the electrical variables of the drive against sensor faults and changes of the control strategy are analyzed. The obtained results are useful in the design of a sensor fault tolerant architecture.

Multi-Domain Modeling of Electric Vehicles Including Lead-Acid Battery Dynamics

Abstract In this paper a multi-domain model of an Electric Vehicle (EV) including lead-acid battery dynamics is presented. The model is composed by the batteries, the electric traction system and the mechanical model. The modeling formalism chosen is Bond Graph (BG) that allows representing each EV sub-system in a compact and modular way. A key benefit of BG is that the sub-systems developed can be easily interconnected independently of their physical nature. Once the complete model is constructed the whole system performance can be evaluated. Simulation results are presented to demonstrate the multi-domain feature of the proposal and the energetic performance with/without regenerative braking.

Stator Winding Fault Detection Using High Frequency Signal Injection for Induction Motors with Closed Rotor Slots

Abstract A stator winding turn fault detection strategy for an induction motor (IM) with close rotor slots is proposed in the present effort. This strategy is based on the effect of the faults over the zero sequence voltage of the IM when a preset exploratory signal is injected by a three phase inverter in the machine. The strategy allows decoupling it from the motor fundamental excitation. For the evaluation of the feasibility of this proposal, a simplify multiple-coupled circuit model of the IM is used. In addition, experimental results are included to validate the strategy for detecting incipient faults.

Multi-Domain Model of Faulty Stator Core for Thermal Effects and Losses Evaluation

Abstract In this work, a multi-domain model of the stator core of an electric machine is presented. Its electromagnetic behavior and the thermal dynamics are represented using Bond Graph. This approach makes apparent the topology of the machine and the energy exchange. This feature enables the model for the easy representation of faults and analysis of their consequences in the thermal/magnetic/electric domain. The proposed model is used for reproducing the behavior of the stator core during an inter-laminar insulation test. The model is capable to reproduce the thermal distribution when a fault in the stator core occurs. This contribution helps to estimate the additional iron losses due to the stator core fault based on the thermal distribution. Additionally, this methodology could help to define the proper maintenance intervention to be performed on the machine.