Research Of The Quality Classification In Surrounding Rock Of Storage Cavern Based On The Improved AHP Fuzzy Comprehensive Evaluation

Oil is known as "industrial blood." In recent decades, global energy issues become increasingly prominent. It is very necessary to establish a strategic oil reserves for every country in the world. With the fast development of China's national economy, China's oil demand is also growing year by year. Establishing the strategic oil reserves has become a pressing task in our country.Internationally, the oil reserves from storage facilities and forms can be divided into onshore storage tanks, underground storage tanks, underground caverns. Comparing to other storage methods, water sealed caverns storage have a series of advantages, such as less land occupied, safety, low maintenance and low investment. It is also a promising new storage technology.The engineering characteristics of water sealed caverns determine its high quality requirements on the rock. Accordingly, accurate and reliable evaluation of the rock mass quality is the key to the success of such projects.This article relies on the national oil reserve bases—the water sealed caverns project in Huizhou. Through expert analysis, five influencing factors were identified. Then the improved AHP method was taken to determine the index weight, next the fuzzy mathematics method was taken to calculate and determine the classification of surrounding rock mass quality. In the calculation, the evaluation criteria values of the subordinate degree based on the two standards—"Standard for underground water enclosed cavern investigation of geotechnical engineering"(SY/T06610-2008),"Standard for engineering classification of rock masses"(GB50218-94). The improved AHP used 0 to 2 scale method, but not 0 to 9 scale method. This method did not appear logical error when made up the judge matrix of evaluation factor. And there is no need to do a consistency test. In fuzzy evaluation model selection, the weighted average model considered all factor of U i (i=1,2,…,m) when it decided the degree of membership b j of level V jevaluated by various factors. Comparatively, other model only consider the factor which affected b j most or the most prominent factor. Therefore, this paper used the weighted average model M (? ,⊕), and judged according to the principle of maximum degree of membership.Finally, compared with the values obtained from the common BQ rock classification method and the AHP fuzzy comprehensive evaluation method, both values were very approximate. Especially when there was a loss of real-time measured data, improved AHP fuzzy comprehensive evaluation can use other factors to revise the results, which made the evaluation results more objective than the BQ classification. Practice showed that the proposed method was feasible, and through the constant improvement it could be a practical engineering method in the future.