Mary E. Kaye

An FPGA-Based Singular Value Decomposition Processor

A two-sided rotation Jacobi SVD algorithm is used to compute the SVD and is implemented on a two million gate FPGA. A mesh-connected array structure is proposed based on Brent, Luk, and Van Loans idea of an expandable square systolic array of simple 2x2 processors to compute the SVD of a large matrix, so as to shorten the iteration time and thus increase the implementation speed. The array consists of an n/2xn/2 array of 2x2 processor elements to compute the SVD of an nxn matrix. The trigonometric functions and the vector multiplication in the algorithm are tailored to the use of CORDIC (coordinate rotation digital computer) algorithms for hardware-efficient solutions. Two SVD processors, the basic SVD processor and the extended SVD processor, were developed. The algorithms to decompose the matrix were first evaluated in Matlab and then the processors were implemented using the Virtex-II FPGA from Xilinx as the target device. The basic SVD processor utilizes the proposed mesh-connected array structure and CORDIC algorithm. The implementation concentrates on utilizing the features of the FPGA to speed up operations and reduce the area required. In order to compute a large SVD without increasing the size of the FPGA, the extended SVD processor was developed to reuse the SVD array of the basic SVD processor. These two processors were successfully implemented on the FPGA device. Speed data and comparisons are presented

Advanced building blocks of power converters for renewable energy based distributed generators

The power converter is a critical component of a distributed generation (DG) system, particularly for a renewable energy based distributed generator. With the rapid development and growing applications of DG systems, power converters have evolved from a traditional “power conversion device” to a “system integrator”. This paper outlines the basic and advanced building blocks of power converters for distributed generation systems, with an emphasis on the new algorithms developed especially for grid-connected single-phase power converters in small wind turbines and photovoltaic systems. A grid-connected power converter should feed power of a high quality into a grid with a low total harmonic distortion (THD), and fast dynamic response to the variations of a renewable energy resource. The newly developed space vector based pulse-width-modulation (PWM) with predictive current control provides a direct cycle-by-cycle regulation of the current waveform feeding into the grid. Effective anti-islanding is an essential function of a power converter, in compliance with the grid interconnection requirements. As active islanding detection methods exert disturbances to the grid and passive islanding detection methods generally have large non-detection zones, new and hybrid islanding detection methods have been developed. A power spectral density method has been developed to continuously monitor and identify the possible features of an islanding formation. Maximum power extraction from renewable energy resources is vitally important for distributed generators. Many algorithms have been developed for maximum power point tracking (MPPT) of photovoltaic and wind energy systems. A new MPPT method has been developed by integrating the fast extremum-finding method of Golden Section Search (GSS) into a multi-stage MPPT algorithm. The proposed MPPT algorithm provides fast response (in the order of 40 msec) and high tracking accuracy (99.9%) as compared to the existing algorithms.

Aggregated domestic electric water heater control - building on smart grid infrastructure

A pilot project is underway in New Brunswick, Canada, to demonstrate direct control of thermostatically controlled loads as a means of providing ancillary services. This paper will focus on one of the load classes, aggregated control of domestic electric water heaters (DEWHs) as a means of shifting load profiles. The approach not only strives to satisfy utility level objectives by curtailing individual DEWHs in a strategic manner but also takes account of the thermal dynamics of individual heaters and estimates of hot water usage in an attempt to maintain customer comfort and satisfaction. The aggregated load controller is unique in its topology and operation, consisting of an individual load extractor, a water demand model, and a temperature estimator for each DEWH, together with a reserve capacity forecast unit and aggregated load control logic. Details of the aggregated load controller are developed, the relationship to the system operator is described and a summary of preliminary results is provided.

Parameter identification of thermal models for domestic electric water heaters in a direct load control program

This paper investigates an approach for identification of physical models of domestic electric water heaters (DEWHs) that are used to provide aggregated regulation services. The model is used within a direct load control (DLC) algorithm to estimate and forecast the water usage and temperature of individual water heaters. Individual physical models are used instead of aggregated models to allow the DLC algorithm to minimize the impact on customers by customizing the control of each water heater to individual water usage patterns. A single-zone lumped-parameter thermal model was considered for a single element DEWH. Two scenarios are investigated: i) when measurements of water temperature and heater power consumption are available, and ii) when measurements of only heater power consumption are available. The problem is challenging because water usage patterns cannot be measured directly and are governed by the behaviour of individual customers.

A new strategy for wind speed forecasting using hybrid intelligent models

Predicting wind power is considered as one of the most important tasks for the large-scale integration of intermittent wind-powered generators into power systems. Given the cubic relationship between wind speed and wind power, accurate forecasting of wind speed is important for the estimation of future wind power generation output. This paper presents a short-term wind speed forecasting technique using a hybrid intelligent algorithm that utilizes a data filtering technique based on wavelet transform (WT) and a soft computing model (SCM) based on fuzzy ARTMAP (FA) network. The effectiveness of the proposed hybrid WT+FA model is evaluated by comparing it with various other SCMs as well as hybrid models. The test results show that a significant reduction in forecast error of an individual FA network by more than 40% through the application of a combined FA and WT. The forecasting performance of the proposed WT+FA is not only robust and more effective than that of individual FA network but also it shows superiority over other considered SCMs. The forecasting techniques were tested using the real data from the North Cape wind farm located in PEI, Canada.

Identification and Estimation for Electric Water Heaters in Direct Load Control Programs

PowerShift-Atlantic (PSA) is a pilot project lead by Canadian Maritime utilities that demonstrates direct load control strategies for up to 20 MW of commercial and residential loads for the purpose of balancing the intermittency of renewable generation and supporting demand-side management programs. On the residential side, domestic electric water heaters (DEWHs) form a significant end use class. The ability to accurately estimate and predict the state of individual end use devices allows aggregated control systems to better ensure end-use performance and comfort levels. This paper presents a methodology for estimating and predicting the state of individual DEWHs from models of their thermodynamics and water consumption that are derived under two scenarios: 1) when measurements of both power consumption and water temperature are available; and 2) when only measurements of power consumption are available. The proposed methodology was implemented as part of the PSA pilot project for the DEWH load class to simulate the behavior of the load in presence of the controller and evaluate the performance of the controller. Experimental results show that the model and water usage profile mimic the actual behavior of DEWHs, and can predict the future power consumption when the thermostatic control of a DEWH is interrupted as part of a load control strategy.

Hardware/Software Codesign for a Fuzzy Autonomous Road-Following System

In this research, a fuzzy logic controller is designed for vision-based autonomous road-following. Because of its high-speed response, portability, and flexibility, a field programmable gate array is applied to implement this control system. Furthermore, a novel hardware/software partitioning method using the genetic algorithm is developed. This method is capable of finding the tradeoff among several evaluation factors under conditions of hard constraints. A small-scaled intelligent vehicle, which is capable of autonomous road-following is built and the proposed controller is tested in the real-world environment. Experiments of hardware implementation and codesign implementation are presented and compared.

Frequency measurement using a frequency locked loop

A phase locked loop method is proposed for fast estimation of utility grid frequency, for control and protection purposes in grid-connected power converters. Using a second order generalized integrator (SOGI) and a ‘novel’ frequency locked loop (nFLL), which makes the SOGI frequency adaptive, the proposed SOGI-nFLL detects small and large step changes in frequency with fast response and without large overshoot. In this paper, the SOGI-nFLL method is analyzed, simulated and implemented in a Field Programmable Gate Array (FPGA). Its performance is compared with the original SOGI-FLL using simulations and experimental data. The experimental results verify that the proposed method results in smoother transients for frequency detection that may be used to reliably activate over-frequency protection within a half cycle.

Design and Lyapunov Stability Analysis of a Fuzzy Logic Controller for Autonomous Road Following

Autonomous road following is one of the major goals in intelligent vehicle applications. The development of an autonomous road following embedded system for intelligent vehicles is the focus of this paper. A fuzzy logic controller (FLC) is designed for vision-based autonomous road following. The stability analysis of this control system is addressed. Lyapunovs direct method is utilized to formulate a class of control laws that guarantee the convergence of the steering error. Certain requirements for the control laws are presented for designers to choose a suitable rule base for the fuzzy controller in order to make the system stable. Stability of the proposed fuzzy controller is guaranteed theoretically and also demonstrated by simulation studies and experiments. Simulations using the model of the four degree of freedom nonholonomic robotic vehicle are conducted to investigate the performance of the fuzzy controller. The proposed fuzzy controller can achieve the desired steering angle and make the robotic vehicle follow the road successfully. Experiments show that the developed intelligent vehicle is able to follow a mocked road autonomously.

Aggregate Load Forecast with Payback Model of the Electric Water Heaters for a Direct Load Control Program

Domestic electric water heaters (DEWH) hold a large share of residential load in North America. The aggregated load profile of electric water heaters follows a similar pattern to the total household load profile, which means that changing the profile of DEWH load can significantly change the shape of the aggregated load profile. To change the load profile, the controller requires an estimation of future load profile and the payback effect of the control action on the forecasted load. This paper presents a load forecast module that uses a Kalman filtered neural network to forecast the aggregated controllable load combined with a statistical payback model to identify the impact of the control action on the load forecast. The proposed method was used by the University of New Brunswick as part of a pilot project named Power Shift Atlantic that aims to provide more than 11MW of ancillary services by controlling more than 1200 controllable loads. The experimental results on the real pilot project shows that the forecast method can be adapted with the dynamic behaviour of the customers. The payback model was also verified by applying various control signals on the pilot project.