The indicator that is most commonly used to quantify the performance of cold generation plants is the “Coefficient of Performance” (COP). The maximum possible COP value for a certain compressor is given by the manufacturer for specific operating conditions. However, this indicator is not the real maximum performance that can be achieved by the entire plant, as the COP value differs with varying working conditions of the plant.
Therefore, the industrial refrigeration plant was modelled to determine the COP at each point and its real margin for improvement. If a plant does not operate with the maximum possible COP, the reasons will be determined and the operating instructions or setpoints that would maximise COP provided.
To develop the smart monitoring system, analytical techniques and multivariate modelling will be applied, based on artificial intelligence and data mining to recognise patterns in the variables that are analysed, associated with stages of plant operation. The aim is to model not only the stationary stages of operation, but also the transitions between points of operation, thus addressing energy optimisation for various operational situations.
Two main results are expected to be obtained. First, a Basic monitoring module will be obtained, which will be used to determine the reduction in COP and its causes. Second, an Optimisation module will be developed to complement the previous module and determine the optimum parameters for plant operation.
Currently, there are no smart monitoring systems for industrial cold generation plants that have the same features and functionalities as the one that is being developed.
The project is funded by the State Program for Research, Development and Innovation Oriented to the Challenges of Society 2017 of the Ministry of Economy and Competitiveness (State Plan for Scientific and Technical Research and Innovation).