The Quantitative Characterization Methods And Application To Diagenesis Of Ultra-low Permeability Reservoir

The research of ultra-low permeability is one of the hottest directions of futureenergy development. Traditional research of diagenesis usually is that geologistanalyze known information and study lithogenous phase qualitatively through richgeological experience. This method takes a lot of manpower and time, and its resultsometimes can not express lithogenous phase articulated, because of the limit ofgeologist’s knowledge and experience. Traditional methods of lithogenous phaseresearch increasingly inadequate with the enhancing of well logging and miningtechnology. There is no completely research method of quantitative analysis of thelithogenous phase domestically. Research of predecessors analyzed types ofdiagenesis quantitatively, put forward to the concept of visual compaction rate, visualdissolution rate, etc. Then the diagenetic comprehensive coefficient which can dividediagenesis through geological verification and mathematical analysis is raised.This paper automatic identify the diagenetic comprehensive coefficient whichcan reflect diagenesis through the study of bayes discriminant analysis. Therecognition accuracy is above95%, indicating that this method can reduce work loadand time of research diagenesis remarkably. The influencing factors of diagenesis ismore than10variable indicators reflected in the test sample data such as rock physics,content of cement, interstitial volume, dissolution pore volume, etc. Main factoranalysis is used in this research. The main factor analysis can determine thecomprehensive principal components of diagenesis and give the componentsgeological meaning. Replacing original variables with comprehensive factor canreduct dimension and also can analyze relation of variables and contribution rate offactors. Considering the main factor analysis do not have direct classification tosample, it is not sufficient to do research by simply divide these factors. But theconclusion of this method will support the diagenesis research in this paper.Because of the uncertainty of test sample, we applied fuzzy clustering to dividelithogenous phase of Hongliugou Youfangzhuang reservoir. In this paper, we getcomprehensive factors as variables from the main factor analysis and divide the lithogenous phase of Hongliugou Youfangzhuang zone through fuzzy clustering. Andat the same time we compared the results which is got from lithogenous phaseclustering about more than10indicators that affect diagenesis. Comparingclassification results analyzed from two different variables and the coinciding analysisof sands, we found that the clustering results which used comprehensive factor had apositive effect on coinciding flat graph with sands, and it is conform to the geologicalexplanation. And in this paper, after the core proper-horizon restoration of58wells instudy zone, we obtained the electrical interpretation chart about lithogenous phasecombining the analysis of single log lithogenous phase’s electrical character.