Building operations account for approximately 40% ofUS energy use and carbon emissions, and vapor compressioncycles are the primary method by which refrigeration and air conditioningsystems operate. Representing a significantportion of commercial and residential building energyconsumption, vapor compression cycles are a target forimprovement in efficiency and savings. This paper presents adata-driven approach to find the optimal operating conditionsof single- and multi-evaporator systems to minimize energyconsumption while meeting operational requirements such asconstant cooling or constant evaporator outlet temperatures.The problem lies in the development of a control architecture thatwill minimize the energy consumed without requiring any modelsof the system or expensive mass flow sensors. The application ofthe presented approach improves efficiency and is demonstratedin simulation and on an experimental system.