A similarity measure recognized by morphological characteristics analysis of well logging curves: application to the kno
- PDF / 683,923 Bytes
- 7 Pages / 595.276 x 790.866 pts Page_size
- 49 Downloads / 160 Views
GMGDA 2019
A similarity measure recognized by morphological characteristics analysis of well logging curves: application to the knowledge domain of sandstone reservoir Ruishan Du 1,2,3 & Huanyu Chen 1 & Fuhua Shang 1 & Nan Ma 1 Received: 18 June 2020 / Accepted: 2 September 2020 # Saudi Society for Geosciences 2020
Abstract The identification of sandstone contrasting layers is a key step in the depth calibration of sandstone horizons, whose errors can cause large deviations in positioning results, which in turn affect subsequent drilling and production of oil wells. The identification of a sandstone contrast layer is calculated based on distinct response characteristics of well logging curves in the depth range of the layer. The identification of features of a well logging curve is generally made using the axial geological signals centered on the depth, combined with a time series similarity analysis of the data. However, when the local morphological features indicated and described by the curves corresponding to the depth of the sandstone reservoir are distinctly different, there will be differences in the features to obtained. To improve the accuracy in extracting similar features from well logging curves, we propose a similarity measurement method based on morphological characteristics to comprehensively analyze three morphological features of the curve sequence: the segmentation trend, fluctuation amplitude, and depth span. The similarity of the subsequence is measured according to the weighted slope difference of the comparative sequences. The method treats the attributes of similar waveform characteristics as a whole, which can overcome the shortcomings of conventional time series methods in describing the degree of morphological changes of the curves. Results from processing experimental data show that compared with the conventional method, the proposed method can effectively improve the accuracy in extracting the features of well logging curves and identifying the sandstone contrast layer. Keywords Time series . Distance measures . Sequence matching . Similarity measures . Knowledge
Introduction A time series is a collection of data resulting from orderly sampling according to temporal or spatial changes. Time series similarity is used to measure the degree of similarity of This article is part of the Topical Collection on Geological Modeling and Geospatial Data Analysis * Ruishan Du [email protected] 1
School of Computer and Information Technology, Northeast Petroleum University, Daqing 163318, China
2
Key Laboratory of Oil & Gas Reservoir and Underground Gas Storage Integrity Evaluation of Heilongjiang Province, Daqing 163318, China
3
FAPS Energy-Tech Ltd. of Heilongjiang Province, Daqing 163318, China
subsequences or the overall similarity between different sequences (Esling and Agon 2012; Fu 2011). The determination of time series similarity involves the search for a series or subsequence among the compared series that is closest to a benchmark series (Zhu et al. 2019). During the searc
Data Loading...
