Performance Optimization of Industrial Refrigeration Systems

Abstract

Industrial refrigeration systems can be found in applications ranging from ice making to food processing and preservation to industrial chemical processes. These systems typically consist of many different components, each component may be produced by a different manufacturer. The operational data provided by the different manufacturers for each component is used by system designers to specify installation and operational procedures of the system. Often times, the optimum control of an individual piece of equipment results in sub-optimal system performance due to unforeseen interactions between the different system components. It is important to identify and monitor key parameters of the system, such as power consumption and refrigeration effect, in order to optimize the performance.

The efforts of this research focused on modeling an operating, ammonia vapor compression, refrigeration system serving a two-temperature food storage and distribution facility located near Milwaukee, WI. This system utilized a combination of both single-screw and reciprocating compressors operating under single-stage compression, an evaporative condenser, and both liquid overfeed and direct expansion evaporators. The model was verified with experimental data recorded from the system and then used to identify alternative designs and operating techniques that lead to optimum system performance.

Changes in system operation such as variable frequency (VFD) or multi-speed motor control on condenser and evaporator fans, head pressure control, refrigerant temperature control, and aspects of load sharing between compressors were investigated. Also, the performance of several alternative system designs was investigated. The aspects of alternative system design that were examined are condenser sizing, two-stage compression, load separation by addition of another suction level, and thermosiphon.

A 31 percent reduction in annual energy usage and a 21 percent reduction in annual peak electrical demand over the current system operation is predicted to be possible with the most feasible of the optimization techniques and designs examined implemented.