Application Of Multi - Attribute Decision - Making Theory In Optimal Oil Production

Artificial lift is a supplementary energy source for down-hole pressure reduction, which can keep down-hole fluid flowing to the surface. Therefore, the proper selection of artificial lift method is very important for an oil field; it can not only affect the oil production rate and enhance the field recovery but also reduce the cost of the production process. The most common use of artificial lift methods are reciprocating rod lift systems, progressing cavity pumping systems, gas lift systems, hydraulic lift systems, and electric submersible pumping systems. Each method has different application considerations which including the affects in reservoir condition, fluid property, field application conditions, and economic limitations. So the decision makers should based on the production expeditions and real conditions to make a choice for each well.There are several technique to select a best artificial lift method, with the rapid development of computer science and fuzzy math theory, multiply attribute decision making (MADM) has been successfully used in optimize the artificial lift methods (ALM), SAW, TOPSIS, ELECTRE(Elimination and Choice Expressing REality), WPM(Weighted Product Model), VIKOR(VlseKriterijumslka Optimizacija I Kopromisno Resenje in Serbian) models are used to select a ALM worldwide, and reported on public journals commonly.This paper will study the relevant theory of MADM and fuzzy math; use the SAW principle to realize the optimization of ALM, technique for order performance by improved AHP_TOPSIS method. The process can be described briefly as follow:input the basic data into a program, form a decision making matrix by membership function of each criteria, then calculate the subjective weight and objective weight of each criteria respectively by using the AHP method and TOPSIS model, and calculate the combination weight using the SSD (Sum of Squares of Deviations) technique to improve the decision weight. Finally the decision maker can use the combination weight victor and the decision making matrix to calculate the performance of each ALM. then obtain the order performance and get the final selection.The research consists of such items as following:(1) Studied the AHP_TOPSIS method, using the analytic hierarchy process built a artificial lift module;(2) Studied the fuzzy MADM theory, using the Chen-Hwang5point linguistic approach replaced the Rank and prioritize to improve the reasonable selection.(3) Studied the attributes of each artificial lift method, based on practical and proven technology on the performance and operating capabilities of the methods.(4) Studied the fuzzy math theory, built each criteria membership functions;(5) Studied the combination of decision weight vectors by using the SSD method to re-join the weight vectors in traditional AHP and TOPSIS models, the combination weight vector will be used in the decision making which can not only reduce the subjectivity in AHP model, but also revise the objectivity in TOPSIS model;(6) Programming the’Artificial Lift Optimization Software Platform’...