Abstract: Based on minimization of Gibbs free energy，a genetic algorithm (GA) was developed for the calculation of phase equilibria and stability in multiphase reaction systems. The total optimization problem was decomposed into two sub-problems，i.e. the minimization of Gibbs free energy for any postulated phases was first solved and then the phase stability analysis for each postulated phases was performed based on the tangent plane distance criterion. By adding sequentially new phases into the system and solving the two sub-problems with the genetic algorithm，the final phase distribution and equilibrium compositions would be obtained. Two complex examples were solved. One was the synthesis of methanol in a high pressure system and the other was the esterification reaction. The results obtained showed that the suggested approach was effective for the addressed problem.

摘要： 基于吉布斯自由能最小法原理，研究用遗传算法求解复杂化学体系的平衡问题。将相数和相态未知体系的平衡计算问题分解为两步最优化问题：假定相态数目下的吉布斯自由能最小化问题和相稳定性判据问题。通过序贯加入新相和用遗传算法交替求解两个最优化子问题，可得到最终体系的相数、相态和平衡组成。对合成甲醇和酯化反应两个多相复杂体系进行了的计算，结果验证了本研究方法的有效性。