A large building with an open space is exposed to substantial influence of the outdoor conditions on the exterior side of the walls and roof, which in return a effects the indoor temperature conditions. This thermal process within the building is characterised by highly non linear behaviour and slow dynamics, presenting additional challenges of indoor climate control mechanism that the small to mid size constructions do not encounter. Many existing commercial and industrial buildings require these indoor climate control solutions to ensure that the indoor temperature remains within the specified boundaries. Currently it is often achieved through the use of heating, ventilation and air conditioning(HVAC) systems, which tend to be energy intensive and contribute to the peak demand for gas and electricity. Many of the existing solutions, which are based on Proportional-Integral(PI) controller method, despite achieving some energy-saving, it is still inefficient particularly when operating to satisfy several indoor climate requirements concurrently. The research presented in this thesis has been conducted to develop an innovative approach for indoor climate control. The proposed reliable and energy efficient control solution adopts Model Predictive Control (MPC) architecture, which optimises the energy consumption by altering the air entering the Air Handling Unit (AHU) between the recirculated indoor air and the fresh outdoor air. The proposed control strategy is able to utilise only the mechanical ventilation (i.e. damper blades position) to pre-cool and pre-heat the indoor space, at the same time contributing to the indoor climate requirements satisfaction. The model derived for this purpose is a State-Dependent Parameter(SDP) model that is capable of responding to the changes in the model parameters caused by the variation in the supply air mass flow rate and alteration of the source air, which are nonlinear. The prediction of the indoor conditions is made with prior knowledge of the weather forecast. This prediction is used by the Genetic Algorithm(GA) to find the optimal control action for the position of damper blades on the AHU entrance. The results of using optimisation with the MPC via simulation approach indicate the ability of the proposed method to lower the HVAC system energy consumption achieved through redesign of the control strategy. In the summer time it is possible to decrease energy consumption and save around8%, whereas in other season it varies between 0% and 3%. This approach benefits from the fact that it doesn’t require additional mechanical equipment to the existing solutions other than a controller that can handle the algorithm locally or remotely, offering a reliable and robust energy efficient indoor air temperature control system.
|Date of Award
|Jet Environmental Systems Ltd
|Mathias Foo (Supervisor), Hafiz Ahmed (Supervisor) & Ivan Zajic (Supervisor)