Morgan Kiani

Effects of Voltage Unbalance and System Harmonics on the Performance of Doubly Fed Induction Wind Generators

Renewable energy harvesting is of prime interest in lieu of increasing demand for energy and high price of fossil fuels. Albeit the diminishing resources of petroleum and natural gas has given a renewed attention to the topic. Wind energy is by far the most accessible and efficient form of renewable energy forms given the state of technology in other renewable resources i.e., solar, fuel cells, etc. Over the past decade there have been significant advances on capturing wind energy using doubly fed induction generators (DFIG). The versatility of the DFIG in providing a controllable flow of active and reactive power to the system as well as its inherent low cost and minimum size power electronic converters has been the driving force in superior adequacy of this family of generators. However, as it has been noted in the literature, voltage unbalance and system harmonics can deteriorate the performance of DFIG by introducing unwanted torque harmonics and inaccuracy in generation of commanded active/reactive power. The present article will address this phenomenon from an electromagnetic and system analysis points of view. Simulation results are given to validate the claims.

Detection of Rotor Faults in Synchronous Generators

Synchronous generators are subject to a variety of failures which may occur in various parts of their structure. Furthermore, these faults may be categorized as partial failure or catastrophic faults. One may note that most partial faults can eventually result in a permanent lack of service. The present digest deals with a class of failures which may happen in the rotor of a turbo-generator. These faults include electrical failures in the Held n inding and mechanical faults in the form of dynamic eccentricity. Since the electromechanical conversion of energy is closely related to magnetic response of the machine any alteration of the excitation and geometry will result in detectable changes in stator quantities such as no-load voltage. The present digest is based on a detailed analysis of the electromagnetic response of a two pole turbo generator under various types of rotor faults. In the first step existence of detectable time-domain signatures have been investigated. Next, uniqueness of these signatures is studied. In the last part, a systematic method for just-in-time detection and classification of various rotor faults has been developed. Finite element analysis has been used for this study.

Optimal Design of Doubly Fed Induction Generators Using Field Reconstruction Method

Doubly fed induction generator is attracting attention among options in distributed wind energy harvest. Traditional design and analysis of doubly fed induction generator (DFIG) dominantly rely on lump-parameters and finite-element models. Although finite-element analysis (FEA) method is available, the computational time limits its application in iterative optimal design practices. This paper is based on the use of field reconstruction method (FRM) which can greatly reduce the computation cost, whereas maintaining acceptable accuracy. In order to conduct efficiency optimization, the procedure to calculate flux density and core losses are described. Finally, an optimal design method of DFIG towards maximum annual energy production in a given area with available wind speed information is presented.

Rotor Shape Investigation and Optimization of Double Stator Switched Reluctance Machine

This paper presents an optimal design for the rotor of a double stator switched reluctance machine. The effect of the rotor shape on the torque profile is analyzed first. Due to the nonlinearity of magnetic core and complex geometry, an iterative approach is developed to automatically shape the rotor geometry using static finite element method. The rotor of a 12/8/12 prototype is optimized. Simulation results validate the effectiveness of the optimal design.

Online detection of faulty battery cells in energy storage systems via impulse response method

Just-in-time detection of anomalies in automotive batteries can proibit a viral propagation of the problem within the energy storage system. Given the cost associated with replacement of energy storage system in electric and hybrid electric vehicles, an effective condition monitoring can translate to substantial savings and reduced downtime. In this paper a condition monitoring technique based on the impulse response of the electrochemical batteries has been proposed. This method is capable of detection of abnormal operating characteristics in battery cells within an energry storage system while the unit is performing its intended function. Simulation and experimental results are provided to illustrate the practicality of the proposed method.

Detection of faults in PMSM using Field Reconstruction Method and Mechanical Impulse Response

A novel approach for quick evaluation of the electromagnetically induced vibration in a Permanent Magnet Synchronous Machine (PMSM) is introduced. Field Reconstruction Method (FRM) and Mechanical Impulse Response (MIR) are used for this purpose. The electromagnetic forces and vibration emissions are analyzed in time and frequency domain for healthy and faulty cases. These analyses allow for detecting specific properties of forces and vibration which are later on used for identifying various fault conditions. The proposed concept has been experimentally verified. Comparison of the vibration emissions for different operating conditions under various types of faults are presented as well.

Frequency Domain Methods for Detection of Rotor Faults in Synchronous Machines under No-Load Condition

Reliable and efficient generation of electricity is of paramount importance in an advanced electric power network. Whether in large synchronous generators that are installed in power plants or for distributed generators in wind farms, survivable performance of these electric machines plays a vital role in accomplishing this goal. The worldwide economical impact of achieving a reliable generation is immeasurable. As a result real time monitoring and just-in-time maintenance of synchronous generators deserves due attention. The present paper deals with this outstanding issue. Synchronous generators are subject to a variety of failures which may occur in various parts of their structure. The present paper deals with a class of failures which may happen in the rotor of a turbo-generator. Since the electromechanical conversion of energy is closely related to magnetic response of the machine any alteration of the excitation and geometry will result in detectable changes in stator quantities such as voltage and current. The present paper is based on a detailed finite element analysis (FEA) of the electromagnetic response of a two pole synchronous generator under various types of rotor faults. In the first step existence of detectable time-domain signatures have been investigated. Next, uniqueness of these signatures is studied. Accordingly, a systematic method for just-in-time detection and classification of various rotor faults has been developed.

A Novel High Energy Density Double Salient Exterior Rotor Permanent Magnet Machine

In this paper, a double salient exterior rotor permanent magnet machine is proposed and analyzed. This machine introduces the use of permanent magnet (PM) into a switched reluctance machine and has the same envelope as that of Toyota Prius third-generation (P3G) 60 kW interior PM synchronous machine (IPMSM). Simulation results show that the continuous torque and peak torque of the proposed machine are 97% and 30.4% more than that of P3G IPMSM under the same current density.

Elimination of System-Induced Torque Pulsations in Doubly-Fed Induction Generators Via Field Reconstruction Method

Effects of system unbalance and system harmonics on the operation of doubly-fed induction generator (DFIG) used in wind energy harvesting are of great concern. This is primarily due to the fact that system unbalance and harmonics can generate unwanted torque undulations that can potentially undermine the mechanical integrity of the tower, and reduce the lifetime of the moving components that are attached to the generator shaft. This paper focuses on the development of a solution for the above problem by judicious selection of the rotor currents to actively eliminate/mitigate these undesirable vibrations. The enabling technology for optimal calculation of the rotor currents is based on the field reconstruction method (FRM). FRM is an analytical tool for the approximation of the magnetic field distribution in the middle of the air gap. Once the FRM formulation is setup, it is capable to predict the tangential/normal components of the magnetic forces. In this paper, the FRM is applied to compute the rotor phase currents in lieu of the availability of the real-time stator currents such that the resultant field will generate a smooth torque.

Model predictive control of stator currents in Switched Reluctance Generators

Control of phase current is an integral part of operation in Switched Reluctance Generators (SRG). Since machine windings experience a substantial back-emf during generation, stator phase currents tend to increase even after the phase is turned off. This in turn necessitates an oversized converter, thereby increasing the cost and overall size of the system. Moreover, due to variations in speed of the prime mover, the power electronic converter should be designed for the worst possible case. This will magnify the additional cost and size issues. Furthermore, implementation of a current profiling scheme for specific reasons such as active vibration cancellation, etc. requires an online estimation of the motional back-emf. In the present paper, a novel method for controlling the current in SRG is proposed. This method enhances the cost and size of the power electronics converter. In addition, using an online estimation of motional back-emf, it makes current profiling in the stator windings possible. This will pave the way for advanced applications of SRG in automotive, aerospace and other challenging places. It must be noted that this method does not require any extra circuitry or memory for its implementation.