Filip Kulic

Support vector machine classifier for diagnosis in electrical machines

Highlights? System for broken bar detection for wide slip range. ? Detection based on measuring only a motor current. ? No need for mathematical models, classifier is trained on acquired data. ? Detection of broken bar at low slip, where classical broken bar detection classifiers are not applicable. ? Reliable, mobile, and cost effective system that can be successfully applied in real working environment. This paper presents a support vector machine classifier for broken bar detection in electrical induction machine. It is a reliable online method, which has high robustness to load variations and changing operating conditions. The phase current is only physical value to be measured. The steady state current is analyzed for broken bar fault via motor current signature analysis technique based on Hilbert transform. A two dimensional feature space is proposed. The features are: magnitude and frequency of characteristic peak extracted from spectrum of Hilbert transform series of the phase current. For classification task support vector machine is used due to its good robustness and generalization performances. A comparative analysis of linear, Gaussian and quadratic kernel function versus error rate and number of support vectors is done. The proposed classifier successfully detects a broken bar in various operational situations. The proposed method is sufficiently accurate, fast, and robust to load changes, which makes it suitable for use in real-time online applications in industrial drives.

Design of the speed controller for sensorless electric drives based on AI techniques: a comparative study

Abstract The paper investigates applicability of different artificial intelligence (AI) techniques in the design of a speed controller for electric drives. A speed-sensorless drive system is considered. A controller structure consisting of a load torque observer, a speed estimator and a speed predictor is developed. Next, different AI based approaches to speed controller design are investigated. The speed controllers based on (1) feed-forward neural network, (2) neuro-fuzzy network, and (3) self-organising Takagi–Sugeno (TS) rule based model are designed. A comparative analysis of the drive behaviour with these three types of AI based speed controllers is performed. In addition, a comparison is made with respect to the drive performance obtained with a conventional optimised PI controller. A detailed simulation study of a number of transients indicates that the best performance, in terms of accuracy and computational complexity, is offered by the self-organising Takagi–Sugeno controller. The controllers are developed and tested for a plant comprising a variable-speed separately excited DC motor.

Software architecture for Smart Metering systems with Virtual Power Plant

This paper presents a novel architecture for Smart Metering systems which enables their seamless, secure and efficient integration in wider SmartGrid software structures. Smart metering solutions represent one of the fastest evolving areas in the field of power distribution systems. There is an extensive interest of leading software vendors in the field, for development of architectures that can efficiently manage transmission, processing and storing of tremendous amount of data produced by such metering devices deployed at the end-end side. The integration of these systems into existing power system software architectures (outage management, workforce management, etc.) represents a major challenge for research community. In such an environment it is of fundamental importance to adopt standardized data exchange mechanisms. The proposed architecture is conceived as modular and scalable structure so that it can support implementation of novel power distribution concepts as Virtual Power Plants (VPPs). The proposed architecture has been successfully tested and verified in real life operation as one of modules of Smart Metering system named Meter Data Management (MDM).

Design of a near-optimal, wide-range fuzzy logic controller

The paper describes a novel method for the design of a fuzzy logic controller with near-optimal performance in the wide range of operating conditions. The approach is based on the analysis of the system behaviour in the error state-space. The final control structure, in a form of a dual fuzzy logic controller, is arrived at in two stages. The first stage encompasses design and tuning of the PI-like fuzzy controller. The second stage consists of placing an additional fuzzy controller, of a structure similar to that of the first one, in parallel with the PI-like fuzzy controller designed in the first stage. The final structure of the dual fuzzy controller is obtained by synthesis and tuning of those segments of the second controller that are different. The resulting dual controller is characterised with high performance in the wide range of operating conditions, and with small number of parameters that can be adjusted using simple optimisation methods. The controller is developed and tested for a plant comprising a variable-speed DC motor drive.

Artificial neural networks broken rotor bars induction motor fault detection

Paper deals with application of online rotor broken bar fault detection via artificial neural networks. Fault can be detected by monitoring abnormalities of the spectrum amplitudes at certain frequencies in the motor current spectrum. These discriminative features are used for training of feed-forward backpropagation artificial neural network. Trained network is capable to successfully classify induction motor rotor condition. Results are presented in tables and figures.

A smart metering architecture as a step towards Smart Grid realization

Emerging concept of Smart Grids aims at increasing visibility and controllability of electricity grids boosting their operational efficiency, enabling novel enhanced services to customers and utilities at a same time. Successful realization of this concept will in great part depend on efficient management of tremendous amounts of data to be gathered and processed in very short time periods. In this work we propose a novel smart metering architecture to manage data collected from deployed smart meters logically encapsulated in form of virtual meters. The metering infrastructure is structured in the form of Advanced Metering Infrastructure (AMI). The architecture of Meter Data Management (MDM) system and its integration in Control Center structure of power system is described in details. The testing and verification of proposed solution is performed on real life operation data obtained in collaboration with power distribution company Elektrovojvodina, Serbia.

Artificial neural network as a gain scheduler for PI speed controller in DC motor drives

The paper proposes an application of artificial neural network (ANN) as a gain scheduler for a conventional PI speed controller. A comparative analysis of the DC motor drive behaviour, controlled by a conventional PI speed controller with and without ANN based gain scheduling, is performed. It is shown that the gain scheduling by a suitably trained ANN enables very good quality of the drive performance over a wide range of operating conditions. The achievable quality of performance is superior to the one obtainable without gain scheduling. Verification of the proposed DC motor speed control system is provided by extensive simulation.

Artificial Neural Networks Eccentricity Fault Detection of Induction Motor

This paper deals with eccentricity fault detection in a induction motor via artificial neural networks. Discriminative features are extracted from magnitudefrequency plot of line current spectra at characteristic frequencies. Based on this data, training and test sets for used artificial neural networks are made. Feedforward and radial basis function neural networks are used for tasks of rotor condition classification. Well trained artificial neural networks are capable to successfully classify rotor condition at medium and full shaft load for choosen features. Simple structure and implementation made them suitably for practical usage.

DETERMINING TOPOLOGICAL CHANGES AND CRITICAL LOAD LEVELS OF A POWER SYSTEM BY MEANS OF ARTIFICIAL NEURAL NETWORKS

ABSTRACT In this paper, an investigation is done of the possibilities of implementing multi-layered artificial neural networks in analyzing the dynamic stability of an power system during load and topology changes. To solve this problem, a multi-layered neural network is used whose inputs are the state vector components, and whose outputs are encoded line outages and the real component of dominant eigenvalues of the system state matrix of the power system. The neural network is trained through the error back-propagation method. The proposed methodology is tested on an power system with ten nodes and four generators. The obtained results indicate the attractiveness of “on-line” application possibilities of multi-layered neural networks in order to efficiently evaluate the stability of an power system during conditions of load and topology change.

A solution for CIM based integration of Meter Data Management in Control Center of a power system

Modern power systems, in particular Control Center structures, involve more and more software applications in their normal operation. Such scenario urges for standardization of inter and intra processes communication and data exchange. In this work we propose a solution for seamless Meter Data Management (MDM) integration with Control Center structures through Common Information Model (CIM). The solution is implemented in form of a wrapper that adopts messages (i.e. payloads) to the standard requested form. The proposed solution has been verified using a simulation framework which emulates regular control and data flow through predefined set of request patterns.