Hierarchical Control of a Hybrid Solar Air Conditioning Plant

This paper describes the development and practical application of a hierarchical scheme for the global control of a solar air conditioning plant. The plant is a variable configuration hybrid process characterised by nonlinearities and time delays, which uses two energy sources for its daily operation, namely solar energy and gas. A control algorithm able to handle these characteristics ensures that the plant gets constantly reconfigured for the most suitable operating mode. The envisaged global control encompasses the starting and stopping phases, as well as operating the plant with and without cooling demand. The hierarchical structure proposed is composed of two main levels, namely the configuration and the regulatory control levels. The configuration level selects the operating mode by means of minimizing a linear function with variable weights. The weights assigned depend on the current state of the plant and on the weather conditions, since the main energy source is solar radiation and it directly influences the selected operating mode. The regulatory control level instead adjusts the variables of the process related to each operating mode by using model predictive control in several structures of control loops. We show the results of the implementation of the hierarchical scheme over the real plant. The scheme exhibits a satisfactory behaviour even in presence of adverse weather conditions, thus proving able to satisfy the cooling demand throughout the day.