Dejan P. Jovanovic

Approach to frequency estimation in self-mixing interferometry: multiple signal classification

Based on the nature of self-mixing signals, we propose the use of the multiple signal classification (MUSIC) algorithm in place of the fast Fourier transform (FFT) for processing signals obtained from self-mixing interferometry (SMI). We apply this algorithm to two representative SMI measurement techniques: range finding and velocimetry. Applying MUSIC to SMI range finding, we find its signal-to-noise ratio performance to be significantly better than that of the FFT, allowing for more robust, longer-range measurement systems. We further demonstrate that MUSIC enables a fundamental change in how SMI Doppler velocity measurement is approached, letting one discard the complex fitting procedure and allowing for a real-time frequency estimation process.

Torque Ripple Verification in PM Machines

In this paper, the problem of the ripple torque experienced with permanent magnet synchronous motors (PMSM) is discussed and solved by adopting advanced simulation tools. The two pulsating torque components are identified: the cogging (detent) torque and the slot harmonics caused ripple. Simulated results are verified by using finite element method (FEM), taking into account the nonlinearity of magnetic material and secondary effects. The results are than proved on the experimental setup. Simplified model for the torque disturbance is proposed, suitable for designing active and passive compensating measures

Fault diagnosis using consensus of Markov chains

A fault diagnosis procedure is proposed based on consensus in a group of local agents/experts. Local models are represented by Markov chains and modelling consensus as a mixture of these allows estimation of optimal ratings using an EM framework. To deal with the unobservable case the procedure is extended to accommodate Hidden Markov models (HMMs).

Distributed Fault Detection Using Consensus of Markov Chains

We propose a fault detection procedure appropriate for use in a variety of industrial engineering contexts, which employs consensus among a group of agents about the state of a system. Markov chains are used to model subsystem behaviour, and consensus is reached by way of an iterative method based on estimates of a mixture of the transition matrices of these chains. To deal with the case where system states cannot be observed directly, we extended the procedure to accommodate Hidden Markov Models.

Remote Controlled Low-Cost Single-Phase Motor System Using Infrared Communication; IEEE student contest - Future Energy Challenge 2005

The goal of this paper is to present advantages of remote controlled 500W integrated electrical drive using infrared communication. A cost-effective, low pin count device for implementing IrDA standard wireless connectivity is used. Placing this device between Pocket PC (graphical user interface) and microcontroller (the heart of the speed motor system) is explained in details. Paper also shows performance and reliability of proposed communication showing advantages comparing to standard RS-232 interface