An Optimization Method Of Heat Exchanger Networks Based On Pressure Drop

As an important sub-process system and energy recycling system of chemical oilrefining process industries, the level of design has an important effect on energy consumptionand economy of process systems. Therefore, integration and optimization of heat exchangernetwork(HEN) has became a hot research issue.In many integration researches about HEN, the pressure drop of the network isconsidered little. Pressure drop, as a major influencing parameter of HEN, by which theinvestment cost of power consumption and power equipment are in a large proportion in thetotal network cost, should be considered as optimal. Genetic algorithm(GA) as an adaptiveprobabilistic search algorithm, with a strong global search ability, is widely used in theintegrated HEN which are characterized by non-convexity, nonlinear and many local minima.But the GA for solving HEN problems exist defects of falling into a local optimum value.Considering the pressure drop, this paper integrates non-diversion HEN through usingimproved GA. Around this theme, the main research work are carried out as follows:1. An aspect about taking pressure drop as a HEN optimization parameter has beenintroduced in details. Power energy consumption costs and power-equipment investmentcosts caused by pressure drop is a part of the network cost optimization goal. The pressuredrop calculation method is discussed. Based on the Esso law to solve pressure drop ofshell-side, the pressure drop formula is deprived. And discussed the impact of the shells onthe performances of the HEN, as well as the methods for calculating the shells.2. Integrated heat exchanger networks using genetic algorithms and pinch methodtogether. Firstly, the optimal ΔTminis determined through pinch method at the targeting stage.Secondly, at the optimal design stage, the HEN problems are solved by GA and the globaloptimum network structure is obtained.3. Two cases are analyzed and studied using GA. The first one is a HEN system of anaromatic plant. The results show that the total annual cost of HEN system has reduced compare with that without considering pressure drop. The utility consumption area andpressure drop are optimized simultaneously. Relative to the block decomposition, the paperapproach can obtain the global optimal more easier. The second one comes from anatmosphere-vacuum distillation for crude oil. The results indicated that the model andalgorithm of this paper proposed can quickly get the global optimal solution or approximateglobal optimal solution for more complex heat exchanger networks.