Estimation of the Lost Circulation Rate Using Fuzzy Clustering of Geological Objects by Petrophysical Properties

Abstract

Рекомендовано членом редакційної колегії д-ром геол. наук, проф. С.А. Вижвою) Currently, cluster-analysis or automatic classification problems are widely used in various fields, in particular economics, sociology, medicine, geology, and other sectors, where there are sets of arbitrary kinds of objects to be automatically divided into groups of similar objects based on their "similarities-differences" features. In recent years, these methods have been widely used in data analysis problems. Conventional methods of cluster-analysis suggest a clear partition of the original set into subsets, in which each point is included only in one cluster after the partition. However, it is well known that such a restriction is not always true. It is often necessary to make such kind of partition, which allows determining the degree of membership of each object for each set. In this case it is advisable to use fuzzy clusteranalysis methods. Problems in this formulation arouse interest of specialists dealing with geology, geophysics, oil-and gaswell drilling and oil and gas production. One of the most important results of the study of lost circulation zones is determination of the coefficient of lost circulation intensity. Purpose. Estimation of drilling mud lost circulation during drilling and emerging risks. Methodology. The solution of the problems posed in the work was carried out using methods known from mathematical statistics and the theory of fuzzy sets. The technique of processing the results, as well as fuzzy cluster analysis, was used for that purpose. Findings. As a result of the research, 5 classes were obtained, each of which characterizes the rate of mud lost circulation, expressed by linguistic variables. On the basis of this, fuzzy models are constructed, expressing the relationship between the indices of petrophysical properties and the volume of the absorbed solution. Originality. A method based on fuzzy cluster analysis has been developed, which makes it possible to predict drilling mud lost circulation of different rate at an early stage during drilling. Practical value. The obtained results allow making decisions on prevention of lost circulation and timely liquidation of their consequences.

Complex fuzzy-probabilistic analysis of information on drilling mud losses

“…Using a fuzzy clustering algorithm, drilling conditions were classifi ed according to the severity of losses based on data on the petrophysical properties of the rocks being drilled. This is very important for early diagnosis of complications and assessment of their risk [9][10][11].…”

Evaluating the Degree of Complexity of Tight Oil Recovery Based on the Classification of Oils

“…We also examined the results of HTR reserves classification and revealed a need for dividing the whole set into homogeneous groups according to a combination of classification criteria. A cluster analysis is best suited for this task (Efendiyev et al, 2018). In order to substantiate the technique of choice this paper discusses the core concept of the cluster analysis and the results of its application, and evaluates the degree of oil recovery complexity and its effect on performance.…”

“…The method elaborated in (Efendiyev et al, 2018) is based on a fuzzy cluster analysis and allows predicting lost circulation of any scale at an early stage of drilling operations. A total of five classes were differentiated in the course of this study, with each class defining the severity of lost circulation expressed as linguistic variables.…”

“…The results obtained facilitate decision-making regarding lost circulation prevention and a timely response to eliminate its consequences. In most cases, specifically when it comes to oilfield practices (Efendiyev et al, 2018;Akhmetov et al, 2018;Koilybayev et al, 2018;Зайченко и Гончар, 2007;Bezdek et al, 1984), a partition must be done in such a way as to determine the degree to which each object belongs to, i.e. the degree of recognition.…”

Prediction of the Well Perfomance Indicators With the Use of Fuzzy Cluster Analysisys

“…In this case, it is advisable to use fuzzy methods of cluster analysis. The tasks in this formulation attract the interest of specialists involved in oil field practice [20]. One of the most important results in studying the operation of pumps during field exploitation is the determination of the delivery rate and the turnaround time.…”