BP- and GA-based energy consumption optimization of gas dehydration plants

This paper aims to reduce the energy consumption and operation cost in gas dehydration process. First, the ProMax simulation software was adopted to build process modeling of a natural gas dehydration plant, from which the data obtained was then trained and forecasted by the BP neural network. Then, the BP neural network model was incorporated into Genetic Algorithms (GA) to establish a BP- and GA-based energy consumption optimization model of gas dehydration plants. In a case study, the operating parameters were optimized by this model for a gas dehydration unit with the capacity of 600 X 104 m 3/d in a certain natural gas processing plant. The result demonstrated that the energy consumption of this dehydration unit was reduced by 12. 23% under the premise of ensuring purification gas product quality. In conclusion, this model is practical and can be also used for the operation parameter optimization of other process systems.