补偿模型误差的采油过程多目标优化

Multi-objective optimization for oil production process with compensating model error

中文关键词  采油过程  多目标优化  单位产油量综合能耗  误差补偿  高斯混合模型  非支配排序遗传算法

英文关键词  oil production process  multi-objective optimization  comprehensive energy consumption for per-ton oil  error compensation  Gaussian mixture model  NSGA--II

中文摘要

      通过对采油过程的分析, 本文建立了以最大化区块产油量和最小化单位产油量综合能耗为目标的优化模 型. 针对单位产油量综合能耗模型的输出与实际值存在较大误差, 利用高斯混合模型(GMM)对单位产油量综合能 耗混合模型误差特性进行描述, 实现对模型的误差补偿, 并将误差补偿后的单位产油量综合能耗引入到已建的优化 模型中, 使得优化结果更接近实际最优值. 在此基础上, 采用带精英策略的快速非支配排序遗传算法(NSGA--II)用 于所建的多目标优化模型求解. 最后, 以某采油作业区一区块生产过程为例进行仿真验证, 结果表明了所建模型和 优化算法的有效性.

英文摘要

      Through analyzing the oil production process, we build an optimization model for maximizing the oil production of block and minimizing comprehensive energy consumption for per-ton oil as the performance index. Considering the large difference between the model output of comprehensive energy consumption for per-ton oil and the actual value, we adopt a Gaussian mixture model (GMM) to describe the error characteristics of comprehensive energy consumption for per-ton oil hybrid model. The model error compensation is first implemented, and then the comprehensive energy consumption for per-ton oil is introduced into the built optimization model to make the optimization results further closer to the actual optimum. On this basis, a fast elitist non-dominated sorting genetic algorithm (NSGA-II) is adopted for solving the established multi-objective optimization model. A production process in a block of an oil production operation area is taken as an example for simulation; the results show the effectiveness of the built model and the proposed optimization algorithm.