Bibliography: leaves 117-125. / Falling commodity prices have reduced the profit margins of Southern African sugar producers. Although these price falls have been severe, they reflect a long-term trend of reducing margins for basic commodity producers during the 20th Century. This trend has forced. producers to closely examine their processes and to look for areas in which improvements in productivity, yield and efficiency can be achieved. Evaporation is the most energy intensive unit operation in the sugar factory, and it is responsible for the removal of most of the water from sugar solution, or juice, which is extracted from the sugar cane. There is also a large potential to lose sucrose at the evaporators due to the high temperatures and long residence times employed there. The smooth control of the evaporators is thus vital to consistent factory operation, and the evaporators are commonly a sugar factory bottleneck. This study developed a control strategy for the particular evaporator configuration found at Triangle Sugar Mill in south eastern Zimbabwe. There are currently several evaporator control strategies being used in the sugar industry. Most of these are an assembly of single loop Proportional Integral Derivative (PID) controllers, which cannot optimally account for the interactions encountered in most evaporator stations. Ideally, any evaporator control system should be able to handle the multiple input multiple output problem while anticipating and handling constraints on inputs and outputs. Several multivariable approaches have been tried, but these usually require a great deal of expensive instrumentation.After a review of the multivariable control literature and testing of several alternative control systems, Dynamic Matrix Control (DMC) was chosen as the bestwsuited control algorithm for the Triangle control problem. A dynamic model of the Triangle evaporator station was, developed to formulate and test the DMC and other controllers. The model was based on a set of differential equations involving mass and energy balances through the evaporators. Real plant data were collected from the SCADA system and the model was tested against this data. After validation the model was. used to record step responses of the process to key input variables. The control system had nine (9) measurable inputs, and three (3) controlled outputs. The objective of the control system was to deliver the maximum amount of consistently high quality symp, within plant constraints. This was formulated in an objective function which seeks to minimize a weighted sum of the errors of syrup concentration from a setpoint, and the fluctuations in juice flowrates. Two alternative formulations were developed, and tested on the plant model.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uct/oai:localhost:11427/5397 |
| Date | January 2000 |
| Creators | Smith, Patrick D |
| Contributors | Swartz, Chris, Harrison, STL |
| Publisher | University of Cape Town, Faculty of Engineering and the Built Environment, Centre for Bioprocess Engineering Research |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Master Thesis, Masters, MSc |
| Format | application/pdf |
Page generated in 0.0018 seconds