This work deals with the energy management of belt conveyor systems (BCS) under various demandside
management (DSM) programmes. The primary objective of this work is to model the energy
consumption and energy related cost of operating troughed belt conveyor systems under different
electricity pricing tariffs. This research is motivated by the increasing need for energy efficiency and
energy cost reduction in the operation of BCS. This is as a result of technological improvements in
BCS technology leading to increasingly longer belts being commissioned and as a result of rapidly
rising electricity costs.
An energy model derived from established industry standards is proposed for long conveyors. The
newly proposed model uses a first-order partial differential equation (PDE) in order to capture the
state of material on the belt. This new model describes the conveyor's power requirement using
an equation with two parameters. A system identification set-up involving a recursive parameter
estimating algorithm is simulated for measurements with varying degrees of noise. The results show
that the proposed model estimates conveyor power and material delivered by long conveyors more
accurately than the existing steady-state models.
Downhill conveyors (DHCs) are important potential energy sources that can be tapped to improve the
overall energy efficiency of BCSs. A generic optimisation model that is able to optimally schedule three
configurations of BCS with DHC is proposed. The economic assessment of implementing dynamic
braking and regenerative drives technology on downhill conveyors is undertaken with the help of the
model. The assessment shows that combining regenerative drives and optimal operation of BCS with
DHC generates energy savings that give attractive payback period of less than 5 years.
A chance-constrained model predictive control (cc-MPC) algorithm is proposed for scheduling belt
conveyor systems with uncertain material demand on the output storage. The chance-constraints are
based on the modelling of material demand by a sum of known mean demand and, zero-mean and
normally distributed random component. The cc-MPC algorithm is shown to produce schedules that
give a smaller number and smaller magnitude of storage limit violations compared to normal MPC and
chance-constrained optimal control algorithms. An equation that gives the amount of effective storage
required to meet storage constraints for a given value of standard deviation is established.
The optimal scheduling of BCS under the real-time pricing (RTP) tariff is considered. This study
develops a methodology for establishing the economic value of price forecasting schemes for loads
capable of load-shifting. This methodology is used to show that the economic benefit obtained from
a forecast is highly dependent on the volatility of the electricity prices being predicted and not their
mean value. The methodology is also used to illustrate why the commonly used indices mean absolute
percentage error (MAPE) and root mean square error (RMSE) are poor indicators of economic benefit.
The proposed index using Kendall's rank correlation between the actual and predicted prices is shown
to be a good indicator of economic benefit, performing far better than RSME and MAPE. / Thesis (PhD)--University of Pretoria, 2016. / Electrical, Electronic and Computer Engineering / PhD / Unrestricted
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:up/oai:repository.up.ac.za:2263/61311 |
| Date | January 2016 |
| Creators | Mathaba, Tebello Ntsiki Don |
| Contributors | Xia, Xiaohua, tmathaba@gmail.com |
| Publisher | University of Pretoria |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Rights | © 2017 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. |
Page generated in 0.0022 seconds