Plant-wide dynamic and static optimisation of supermarket refrigeration systems

• Performance function for food quality, energy efficiency, and system reliability.
• Performance function is divided into static and dynamic parts.
• We propose a strategy for choosing set-point for optimal performance in steady state.
• Dynamic performance optimisation uses dedicated excitation signals.
• We design the excitation signals (online) based on available system information.

Optimising the operation of a supermarket refrigeration system under dynamic as well as steady state conditions is addressed in this paper. For this purpose an appropriate performance function that encompasses food quality, system efficiency, and also component reliability is established. The choice of setup parameters, which are necessary for system performance optimisation, depends on whether the system operates under steady state or dynamic conditions. While operating under steady state conditions, the total system performance is shown to predominantly be influenced by the suction pressure. The dynamic optimisation requires use of dedicated excitation signals. A method for designing such signals under realistic operational conditions is proposed. A derivative free optimisation technique based on Invasive Weed Optimisation (IWO) is utilised to optimise the parameters of the controllers in the system. Simulation results is used to substantiate the suggested methodology.