Multivariate KPI for energy management of cooling system in food industry

Within EU, the food industry is currently ranked among the energy-intensive sectors, mainly as a consequence of the cooling system shareover the total energy demand. As such, the definition of appropriate key performance indicators (KPI) for ammonia chillers can play a strategic role for the efficient monitoring of the energy performance of the cooling systems. The goal of this paper is to develop an appropriate management approach, to account for energy inefficiency of the single compressors, and to identify the specific variables driving the performance outliers. To this end, a new KPI is proposed which correlates the energy consumption and the different process variables. The construction of the new indicator was carried out by means of multivariate statistical analysis, in particular using Kernel Partial Least Square (KPLS).This method is able to evaluate the maximum correlation between dataset and energy consumption employing nonlinear regression techniques. The validity of the new KPI is discussed on a case study relevant to the cooling system of a frozen ready meals industry. The assessment of the proposed metric is one against Specific Energy Consumption (SEC) like indicator, typically used in the context of the Energy Management Systems.