Energy-saving control method for NH3-CO2 cascade refrigeration system by directly regulating slide valve position in twin-screw compressor

Abstract

Intermediate temperature (Tmid) is the pivotal parameter for energy-saving operation of existing cascade refrigeration systems (CRS), while few studies conducted on the on-site control method required to attain the optimal Tmid. This study proposes an energy-saving optimization method by the direct regulation of slide valve position in CRS high-temperature-stage (HTS) screw compressor rather than indirect regulation of Tmid. The backpropagation neural network relied on Levenberg-Marquardt algorithm is used to establish two prediction models for COP and slide valve position in high-temperature-stage twin-screw compressor (Vi). By selecting the maximum COP as the optimization objective, the optimal Tmid under each working condition is investigated. Meanwhile the corresponding optimal Vi is calculated through the data-driven model, so as to realize the direct control of the HTS volumetric flow. Compared to the traditional fixed Tmid operation, setting the Tmid as a real-time variable and control Vi through a data-driven model which correlated Vi with Tmid, can effectively avoid the significant fluctuation magnitude of Vi and Tmid. Meanwhile, the CRS can achieve an impressive energy-saving rate of up to 17.55%, and the average COP increases from 2.11 to 2.30. This study presents a novel solution that ensures the reliability operation of CRS, which is the crucial for sustainable reduction of energy consumption.