New Intelligent Model of Cuttings Logging Based on Grey Clustering

With the deepening of exploration and development, the lithology of the drilling strata becomes more complex. Using the digital technology for processing the data obtained by the cuttings logging helps to provide accurate lithological data and evaluate clamping of the formation interface. However, the existing logging digitization technology relies on element logging and is restricted by the large error of the cuttings logging instrument, the disunity of multi-source data, and the poor pertinence of data. In this paper, we propose an intelligent identification model of the cuttings logging based on grey clustering analysis. First, the grey prediction method is used for processing the in-depth instrument data, and then the extended Kalman filter is used to standardize and unify the multi-instrument data. Finally, the identification model based on the grey clustering method is applied to identify the cuttings. The results of the simulation analysis and field application show that the identification model proposed in this paper can accurately identify the rock strata. Compared with the traditional methods, the accuracy of the proposed has been greatly improved. The field applications show that the model provides important theoretical support for the development of rock-cutting digital technology.