Faculty of Engineering and Built Environment

Permanent URI for this communityhttp://ir-dev.dut.ac.za/handle/10321/9

Browse

Now showing 1 - 14 of 14

Assessment of UV index using artificial neural networks
(2002) Human, Sep; Bajic, Vladimir B.; Duffy, Kevin Jan
Contribution to the analog simulation of particular dynamic phenomena in rock mass
(2001) Glasspoole, Errol Edward; Bajic, Vladimir B.
Design of a non-linear analog PID controller
(1997) Govender, Poobalan; Bajic, Vladimir B.
In this study we propose an analogue nonlinear PID controller with antiwindup and dead-time compensation to optimise the control of loops experiencing degradation in the control performance as a result of dead-time and saturation nonlinearity. Loops containing a significant dead-time are notoriously difficult to control. The proposed controller optimises the control of loops experiencing the negative effects of saturation and dead-time.
Design, fabrication, production and erection of elements for a low cost housing unit
(1999) Nxumalo, Mutshutshu; Bajic, Vladimir B.
Housing is currently one of the most critical development areas in the Republic of South Africa (RSA). This developmental project forms part of a possible solution in the delivery of housing through mass production. The objective of this study is to establish a empathetic connection between modem geodesic construction and the traditional dome and to design an effective and efficient prefabricated dwelling unit suited to be easily built by semi-skilled persons. The success of this project will fulfill one of the major objectives of the RSA's Reconstruction and Development Programmes (RDP) - 'A better lifefor all'.
Development of an interface monitor to operate over a data transmission line
(1994) Crankshaw, Brett Robert; Bajic, Vladimir B.; Galbraith, Bruce G.
One of the telecommunication facilities provided by Telkom S.A. is connection to the National and International X.25 data network. This network follows the international data protocol standard based upon the CCITT Recommendation X.25. The network allows data calls to be made from the X.25 user in very much the same fashion as a caller on the Public Switched Telephone Network (PSTN), in that the caller must specify the number of the called user on the network in order to be connected. It is in this environment, specifically in the testing and maintaining of these data circuits, that this study is based.
Electrocardiogram, heart rate and temperature monitoring system
(2000) Malindi, Phumzile; Bajic, Vladimir B.
The purpose of this study is the development of an affordable computer-based electrocardiogram, heart rate and temperature monitoring system, that would complement those that are available on the market and contribute to the reduction of the shortage of these medical instruments in South African hospitals and clinics. Electrocardiogram (ECG) refers to the graph that results from time versus voltage in a patient's chest. It reflects the rhythmic activity of the heart. For this reason the electrocardiogram has a diagnostic value that can be used by medical personnel to examine the biological (hence, clinical) behavior of the heart. The electrocardiogram can also be used to get the heart rate. This thesis explained how to acquire ECG signals from the patient and also how to achieve a cheaper way of providing galvanic isolation, which is required for sensors that are attached to the human body. It also explains computer interfacing using the parallel port and computer-based processing of these ECG signals to determine the instantaneous value of the heart rate and also to reduce the interference that contaminates these signals. In reducing interference, the performance of traditional IIR notch and adaptive filters, as noise cancelers, has been analyzed and compared. Least Mean Squares (LMS) and Normalized Least Mean Squares (NLMS) algorithms are the two algorithms that were considered in this study for adaptive noise canceling and their performance is evaluated and is compared based on their convergence rate, complexity and noise reduction.
Hierarchical identification of large-scale system models
(1997) Jankovic, Boris R.; Bajic, Vladimir B.
In this study we propose a new concept and methodology of hierarchical identification. The need for such a methodology comes from the fact that identification of large-scale systems (LSSs) by one-shot approach may be numerically very complex. The analysis of LSSs is, in general, not approached by the one-shot methodologies normally associated with non-LSSs. The proposed method of hierarchical identification can be therefore viewed as an extension of LSS methodologies to system identification. LSS methodology aims at breaking up the initial, complex problem into a set of smaller size subproblems.
The integration of a high voltage cable fault location instrument with modern information technology
(2002) Kelly, Roger James; Bajic, Vladimir B.
Modern society as a whole seems destined to have an ever-increasing demand for power for both industrial and domestic use, as continued population growth means that cities, suburbs and industrial areas become larger and denser. At the same time the trend toward increased productivity in all segments of industry is influencing the development and techniques employed at locating faults in power cables and networks to ensure only limited downtime and reduced direct and indirect costs associated with the location of faults
An investigation into the feasibility of the use of an electrostatic technique for the measurement of mass flow rate of pneumatically conveyed solids
(1997) Hole, Evan D.; Bajic, Vladimir B.
This report details the findings of work carried out over the course of 1994 and 1995 to determine if it is feasible to use an electrostatic technique for the measurement of mass flow rate of pneumatically conveyed solids. This includes results from tests done on a flow rig and at various power stations.
A model review and proposed mechanistic tiller model for the canegro sugarcane crop model
(2000) Bezuidenhout, Carel Nicolaas; Bajic, Vladimir B.
The CANEGRO model is a mechanistic model that describes environmental, physiological and managerial features of the agricultural sugarcane production system. The model originated from many sources of crop, climatic and soil research work. This caused a fragmentation in the model's components. The objectives of this study was to give a detailed model overview after an investigation into the model's code, published and unpublished documentation and to propose a mechanistic approach in the modelling of tiller populations.
Neural networks approach to process control : the case of processes with long dead times
(1999) McLeod, Charles Meredith; Bajic, Vladimir B.
This study relates to applications of static artificial neural networks (ANNs) to two basic problems of process control: (a) process model identification, and (b) optimal controller tuning. The emphasis is on model identification, where several novel techniques are introduced. A review of the use of ANNs for determining optimal controller settings is included as a logical adjunct which would make the complete system suitable for realisation as a portable or networked system. Three methods for obtaining good approximations for the parameters of first-order processes with long dead time using artificial neural networks (ANNs) are proposed and described. These are termed in this study: time-domain, frequency-domain and model-based methods. In each case the aim was to develop a brief one-shot test that could be applied with minimal disturbance to a closed loop control system. These methods build on existing techniques, but introduce the following novel aspects: 2. The frequency-domain method makes use of the first 81 components of the FFT without further selection as input to a static ANN to yield process parameter estimates. 3. The model-based method uses a simple single-neuron implementation of an ARX model and uses a static ANN to relate process parameter values to the weights of this neuron. In making the analysis, the process input and output are applied repetitively to the neuron model with delays getting progressively larger. Useful effects arising from this are explored. A technique in which ANN training sets are slightly distorted in a random way during training of a radial basis function is developed as part of the time- and frequencydomain methods. The benefits arising from this technique are demonstrated. These experimental ANN-based control methods are evaluated by means of simulations in which accuracy in the presence of measurement noise and performance with higher order processes is measured and analysed. Although the main theme of this study is first-order-plus-dead-time (FOPDT) processes, the full autotuning scheme is tested with some representative higher order processes. Finally, the composition of a complete autotuning scheme is proposed which includes the automatic generation of controller parameters by means of ANN s.
Optimization approach to the frequency design of compensators for nonlinear systems with dead time
(1998) Stavrou, Marios; Bajic, Vladimir B.
Designing compensators in the frequency domain is a complicated problem even for linear systems that have dead time. The situation is far more difficult if the system is also nonlinear. This study introduces a new method for the design of compensators for time-invariant, nonlinear systems that have dead time. The method is based on an optimization approach and utilizes large signal linearization methodology.
Pattern recognition in the case of strong background noise
(2001) Wang, Xingmei; Bajic, Vladimir B.
This study presents a development of a method for recognition of a class of patterns in signals contaminated by strong noise. The class of signals considered is described by a finite alphabet. The target class of patterns is assumed to have specific statistical properties that can be conveniently captured by the position weight matrices (PWM) description. Itis also assumed thatthe 'signals: contain numerous patterns si~ilar to the patterns of the target class, but which belong to different classes. These other patterns represent the noise in the signals. The method for-improved recogrrition of the target class of patterns is based on clustering of the target motifs with regard to distance form the reference point (event) in the signal. This positional clustering enables more precise description of the target class of patterns by means of the PWMs. However, it requires the use of as many PWMs as there are clusters of the target class. The method developed is of general nature, applicable to the situations described. It is however, applied to the recognition of the specific short motifs in DNA sequences. The short motif considered is the TATA-box,one of the most important docking sites for proteins in Eukaryotic polymerase II promoter regions. The reference point in the singals obtained form DNA sequences the transcription .start site (TSS). Thus the positional dustering of the TATA-box motif resulted in 20 different PWMs, instead of only one that describes the whole TATA motif class. This however, resulted in more discriminative PWMs and the recognition accuracy has increased by about a factor of two when compared to the recognition of the TATA moti f based on the original PWM.
Statistical pattern recognition based on LVQ artificial neural networks : application to TATA box motif
(2000) Wang, Haiyan; Bajic, Vladimir B.
The computational analysis of eukaryotic promoters are among the most important and complex research domains that may contribute to complete gene identification. The current methods for promoter recognition are not sufficiently developed. Eukaryotic promoters contain a number of short motifs that may be used in promoter recognition. Having good computational models for these motifs can be crucial for increased efficiency of promoter recognition programs. This study proposes a combined statistical and LVQ neural network system as a computational model of the TAT A box motif of eukaryotic promoters. The methodology used is universal and applicable to any short functional motif in DNA. The statistical analysis of the core TAT A motif hexamer and its neighboring haxamers show strong regularities that can be used in motif recognition. Moreover, the positional distribution of the TAT A motif in terms of its distance from the transcription start site is very regular and is used in the statistical modeling. Furthermore, the matching score of the position weight matrix for the motif was used as a part of the model. Based on these statistical properties. a novel LV Q classifier for TAT A motif recognition is developed. The characteristics of the method are that the genetic algorithm was used for finding good initial weights of the LV Q system, while fine tuning of two LVQ networks was done by the lvq? algorithm. The final computational model is developed for a recognition level of 67.8o/c correct recognition on the test set with less than 1% false recognition. This model is evaluated in the task of promoter recognition on an independent test set. The results in promoter recognition outperform three other promoter recognition programs. It is shown that the recognition of promoters based on the recognition of the TAT A motifs using this new model is superior to the recognition based on the currently used position weight matrix description of this motif.