Machinarium is an exploration of industrial bio-mechanical hybridity as part of the 21st Century paradigm. The dissertation investigates the potential of industry as urban catalyst - a mechanism with which to regenerate urban environments and re-integrate socio-ecological systems. In an attempt to redefine modern concepts of waste and mitigate the flood of pollution emanating from 20th century industrialisation, the investigation is contextually based in an ‘urban wasteland’ - which is re-programmed as part of a new industrial ecology. The dissertation therefore blurs present-day distinctions between ‘social’, ‘productive’ and ‘natural’ space, while at the same time placing focus on the global cultural dependence on waste. If humankind is to survive the predicted crises of the our time, a 21st-century approach to design must shift the modern understanding of architecture as ‘machines for living in’ towards that of architecture as living machines. Machinarium alludes to new ways of architectural place-making in a rapidly changing world. / Dissertation MArch(Prof)--University of Pretoria, 2014. / Architecture / MArch(Prof) / Unrestricted

A Left Ventricular Assist Device (LVAD) is a mechanical pump that helps patients with heart failure conditions. This rotary pump works in parallel to the ailing heart and provides an alternative path for blood flow from the weak left ventricle to the aorta. The LVAD is controlled by the power supplied to the pump motor. An increase in the pump motor power increases the pump speed and the pump flow. The LVAD is typically controlled at a fixed setting of pump power. This basically means that the controller does not react to any change in the activity level of the patient. An important engineering challenge is to develop an LVAD feedback controller that can automatically adjusts its pump motor power so that the resulting pump flow matches the physiological demand of the patient. To this end, the development of a mathematical model that can be used to accurately simulate the interaction between the cardiovascular system of the patient and the LVAD is essential for the controller design. The use of such a dynamic model helps engineers and physicians in testing their theories, assessing the effectiveness of prescribed treatments, and understanding in depth the characteristics of this coupled bio-mechanical system. The first contribution of this dissertation is the development of a pump power-based model for the cardiovascular-LVAD system. Previously, the mathematical models in the literature assume availability of the pump speed as an independent control variable. In reality, however, the device is controlled by pump motor power which, in turn, produces the rotational pump speed. The nonlinear relationship between the supplied power and the speed is derived, and interesting observations about the pump speed signal are documented. The second contribution is the development of a feedback controller for patients using an LVAD as either a destination therapy or a bridge to transplant device. The main objective of designing this controller is to provide a physiological demand of the patient equivalent of that of a healthy individual. Since the device is implanted for a long period of time, this objective is chosen to allow the patient to live a life as close to normal as possible. The third contribution is an analysis of the aortic valve dynamics under the support of an LVAD. The aortic valve may experiences a permanent closure when the LVAD pump power is increased too much. The permanent closure of the aortic valve can be very harmful to the patients using the device as a bridge to recovery treatments. The analysis illustrates the various changes in the hemodynamic variables of the patient as a result of aortic valve closing. The results establish the relationship between the activity level and the heart failure severity with respect to the duration of the aortic valve opening.

Cancer study and drug selection research attract more and more researchers, which need a significant laboratory technique, named cell injection. Hundreds of cells are loaded on devices and injected to investigate the behavior of the cells. Traditionally, cell injection is performed manually, which leads to human fatigue, is time-consuming and has a low success rate. Therefore, a system which can replicate the actions of what technicians do, such as to aspirate cells, transfer cells, immobilize cells, and release cells automatically, is needed. This system must be accurate, reliable, and efficient and operate without human intervention. A cell-transfer-cover and a cell-holder have been fabricated and a cell injection system has been set up to investigate the performance of the newly created device. Simulations and experiments have proven that this system would carry out the entire process of cell injection with the result of enhancing the speed of this important activity.

Cancer study and drug selection research attract more and more researchers, which need a significant laboratory technique, named cell injection. Hundreds of cells are loaded on devices and injected to investigate the behavior of the cells. Traditionally, cell injection is performed manually, which leads to human fatigue, is time-consuming and has a low success rate. Therefore, a system which can replicate the actions of what technicians do, such as to aspirate cells, transfer cells, immobilize cells, and release cells automatically, is needed. This system must be accurate, reliable, and efficient and operate without human intervention. A cell-transfer-cover and a cell-holder have been fabricated and a cell injection system has been set up to investigate the performance of the newly created device. Simulations and experiments have proven that this system would carry out the entire process of cell injection with the result of enhancing the speed of this important activity.

No description available.