Jovitha Jerome

Adaptive PSO for optimal LQR tracking control of 2 DoF laboratory helicopter

LQR weight optimization problem is solved using adaptive particle swarm optimization (APSO) algorithm.The convergence speed and precision of conventional PSO is improved by introducing an adaptive inertia weight strategy based on the success rate of the particles.The performance of APSO tuned LQR is validated on a benchmark 2 DoF laboratory helicopter for trajectory tracking application. This paper deals with the attitude tracking control problem for a 2 DoF laboratory helicopter using optimal linear quadratic regulator (LQR). As the performance of the LQR controller greatly depends on the weighting matrices (Q and R), it is important to select them optimally. However, normally the weighting matrices are selected based on trial and error approach, which not only makes the controller design tedious but also time consuming. Hence, to address the weighting matrices selection problem of LQR, in this paper we propose an adaptive particle swarm optimization (APSO) method to obtain the elements of Q and R matrices. Moreover, to enhance the convergence speed and precision of the conventional PSO, an adaptive inertia weight factor (AIWF) is introduced in the velocity update equation of PSO. One of the key features of the AIWF is that unlike the standard PSO in which the inertia weight is kept constant throughout the optimization process, the weights are varied adaptively according to the success rate of the particles towards the optimum value. The proposed APSO based LQR control strategy is applied for pitch and yaw axes control of 2 Degrees of Freedom (DoF) laboratory helicopter workstation, which is a highly nonlinear and unstable system. Experimental results substantiate that the weights optimized using APSO, compared to PSO, result in not only reduced tracking error but also improved tracking response with reduced oscillations.

A novel control strategy for the boost DC - AC inverter

Boost DC-AC inverter naturally generates in a single stage an AC voltage whose peak value can be lower or greater than the DC input voltage. The main drawback of this structure deals with its control. Boost inverter consists of Boost DC-DC converters that have to be controlled in a variable-operation point condition. The sliding mode control has been proposed as an option. However, it does not directly control the inductance averaged-current. This paper proposes a control strategy for the Boost inverter in which each Boost is controlled by means of a double-loop regulation scheme that consists of a new inductor current control inner loop and an also new output voltage control outer loop. These loops include compensations in order to cope with the Boost variable operation point condition and to achieve a high robustness to both input voltage and output current disturbances. As shown by simulation results, the proposed control strategy achieves a very high reliable performance, even in difficult transient situations such as nonlinear loads, abrupt load changes, short circuits, etc., which sliding mode control cannot cope with.

Intuitionistic Fuzzy Expert System based Fault Diagnosis using Dissolved Gas Analysis for Power Transformer

In transformer fault diagnosis, dissolved gas analysis (DGA) is been widely employed for a long period and numerous methods have been innovated to interpret its results. Still in some cases it fails to identify the corresponding faults. Due to the limitation of training data and non-linearity, the estimation of key-gas ratio in the transformer oil becomes more complicated. This paper presents Intuitionistic Fuzzy expert System (IFS) to diagnose several faults in a transformer. This revised approach is well suitable to diagnosis the transformer faults and the corresponding action to be taken. The proposed method is applied to an independent data of different power transformers and various case studies of historic trends of transformer units. It has been proved to be a very advantageous tool for transformer diagnosis and upkeep planning. This method has been successfully used to identify the type of fault developing within a transformer even if there is conflict in the results of AI technique applied to DGA data.

LabVIEW based Intelligent Controllers for Speed Regulation of Electric Motor

LabVIEW (laboratory virtual instrument engineering workbench) is gaining its popularity as a graphical programming language not only for data acquisition and measurement but also for designing intelligent controllers for real time control purpose. This paper describes about the intelligent controller designed and developed in LabVIEW for speed regulation of DC motor. Artificial neural network, fuzzy logic and neuro-fuzzy controller are developed to achieve accurate trajectory control of speed especially when DC drive and load dynamics are unknown. The nonlinear unknown dynamics are capture by training the Artificial neural network and tuning the fuzzy control algorithm. The developed controller utilizes the same chopper driven DC motor. Experimental results are presented to demonstrate the effectiveness of these controllers. Comparison reveals the advantage of neuro-fuzzy controller over artificial neural network controller and fuzzy controller

Algebraic approach for selecting the weighting matrices of linear quadratic regulator

This paper proposes an algebraic approach for selecting the weighting matrices of linear quadratic regulator (LQR) for trajectory tracking application. One of the important problems in designing the state feedback controller via LQR is the choice of Q and R matrices. Normally, the weights of LQR controller are chosen based on trial and error approach to obtain the optimum state feedback controller gains, but it is often cumbersome and tedious to tune the controller gains via trial and error method. Hence to address the weight selection problem of LQR, a novel algebraic approach, which relates the time domain specifications of the system to be controlled to the weighting matrices of LQR, is proposed in this paper. The key idea of the proposed approach is the synthesis of time domain design specifications for the formulation of cost function of LQR, which directly translates the system requirement into cost function, so that an optimal performance can be obtained via a systematic approach. A nonlinear magnetic levitation system is used to validate the efficacy and robustness of the proposed methodology, and a detailed simulation results are presented.

Control of dead-time systems using derivative free particle swarm optimisation

Particle swarm optimisation (PSO), a population-based nature inspired algorithm has mostly been used for solving continuous optimisation problems, discrete variants also exist. It finds application in most of the engineering design problems. This paper introduces two improved forms of PSO algorithm applied to PID controller and Smith predictor design for a class of time delay systems. In this paper, derivative free optimisation methods, namely simplex derivative pattern search and implicit filtering are used to hybridise PSO algorithm with improved convergence than original PSO. The effectiveness of the proposed algorithms namely SDPS-PSO, IMF-PSO are demonstrated using unit step set point response for a class of dead-time systems using PID controller and Smith predictor designed using the proposed hybrid PSO algorithms. The results are compared with earlier controller tunings proposed by Kookos, Syrcos, Chidambaram, Kanthaswamy and Luyben.

An FPGA Based Control Algorithm for Cascaded Multilevel Inverters

In recent years, thanks to the various developments in VLSI, Field-Programmable Gate Arrays (FPGAs) have become key components in implementing high performance digital signal processing (DSP) systems, especially in the areas of digital communications, networking, video and imaging but its potential is not fully utilized in the area of power control and conversions. The logic fabric of todays FPGAs consists not only of look-up tables, registers, multiplexers, distributed and block memory, but also dedicated circuitry for fast adders, multipliers, and I/O processing (e.g., giga-bit I/O). The memory bandwidth of a modern FPGA far exceeds that of a microprocessor or DSP processor running at clock rates two to ten times that of the FPGA. Coupled with a capability for implementing highly parallel arithmetic architectures, this makes the FPGA ideally suited for creating high-performance custom data path processors for tasks such as digital filtering, fast Fourier transforms, and forward error correction. In this paper a XILINX FPGA based multilevel PWM three phase inverter test rig was constructed by adding bi-directional switches to the conventional bridge topology and its performance is suitably analysed. The inverter can produce three and five different output voltage levels across the load. XILINX FPGA is a programmable logic device developed by XILINX which is considered as an efficient hardware for rapid prototyping. It is used as a SVPWM generator to apply the appropriate signals to inverter switches. In addition to XILINX FPGA, Matlab/Simulink, system Generator software was used for simulation and verification of the proposed circuit before implementation, Simulation and experimental results show that both are in close agreement. The present PWM signal generation scheme can be used for any multilevel inverter configuration with minimum changes in driver circuit.

Weighted fuzzy fault tolerant model predictive control

This paper proposes a new active fault-tolerant control (FTC) using fuzzy predictive logic. The FTC approach is based on two steps, fault detection and isolation (FDI) and fault accommodation. The fault detection is performed by a model-based approach using fuzzy modeling and fault isolation uses a fuzzy decision making approach. The information obtained on the FDI step is used to select the model to be used in fault accommodation, in a model predictive control (MPC) scheme. The fault accommodation is performed with one fuzzy model for each identified fault. The FTC scheme is used to accommodate the faults of real-time CSTR level process. The fuzzy FTC scheme proposed in this paper was able to detect, isolate and accommodate correctly the considered faults of the system.

Real-Time Vision Based Driver Drowsiness Detection Using Partial Least Squares Analysis

Robust eye state classification in real-time is very crucial for automatic driver drowsiness detection to avoid road accidents. In this paper, we propose partial least squares (PLS) analysis based eye state classification method and its real-time implementation on resource constraint digital video processor platform, to monitor the eye state during all time driving conditions. The drowsiness is detected using percentage of eye closure (PERCLOS) metric. In this approach, face in the infrared (IR) image is detected using Haar features based cascaded classifier and within the face, eye is detected. For binary eye state classification, PLS analysis is applied to obtain the low dimensional discriminative subspace, within which simple PLS regression score based classifier is used to classify test vector into open and closed. We compared our algorithm to recent methods on challenging test sequences and the result shows superior performance. The results obtained during on-vehicle testing show that the proposed system achieves significant improvement in classification accuracy at nearly 3 frames per second.

Application of matrix converter in wind energy conversion system employing synchronous generator

Renewable energy is catching peoples attention with increasing global energy consumption and noticeable environmental pollution across the globe. The increasing size of wind farms requires power system stability analysis including dynamic wind generator models. For turbines above 1 MW, doubly-fed induction machines are the most widely used concept. However, direct-drive wind generators based on converter-driven synchronous generator concepts have reached considerable market penetration. This paper presents converter driven synchronous generator models of various order that can be used for simulating transients and dynamics in a very wide time range. The matrix converter is used for conversion of variable voltage to a constant value that can be connected to a grid.