Technique In Well Testing Program Based On Intelligent Data Optimization

Oil well and water well testing is the key to understand the situation of oil production,to prevent and resolve production problems.However,standards and programs on the test wells' election haven't been formed.Thus,the use of data mining technology for oil production testing which provide a suitable classification method has very important significance.Through the analysis of the oilfield existing data structure,the further study can be classified into the field of time series clustering based on features.Content includes:(1)Pointing out the clustering method of feature set is more applicable than original time series,and constructing the feature set after extracting the features.During the procedure of data processing,a wavelet transform method which can peal off various factors that effect time sequence and is benefit for the analysis of time series as well as the feature extraction.(2)After the construction of feature set,a method combined means clustering and hierarchical clustering is applied to make sure the clustering result satisfy the requirement of program management,adjust the number of test well based on production situation,and achieve the optimization of overall test program.(3)Building optimization software of well program based on data intelligence.With the help of the valuable data resources and the software,we can extract the feature set from the history data.Then the feature set would be employed in wellhead clustering and state division.Finally,realize the value of theoretical research.This article builds a basic framework for the developmenton the Windows platform with the idea of mixture programming of Java and Matlab,designs an oil wells testing program optimization model.Experiments are performed on oil wells and water injection wells' data set.And the result shows that the method of first feature extracting and then clustering can get a good classification for most well.In a conclusion,it can play the role of auxiliary decision-making for the test well selection in oil field production management.