Application of T-S Fuzzy Model in Candidate-well Selection for Hydraulic Fracturing
Hydraulic fracturing (HF) is the key technology of increasing production and injection for low permeable reservoirs. The candidate-well selection for HF is essential to oil and gas wells stimulation potential evaluation, which is crucial to improve fractu
- PDF / 136,968 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 46 Downloads / 136 Views
Abstract Hydraulic fracturing (HF) is the key technology of increasing production and injection for low permeable reservoirs. The candidate-well selection for HF is essential to oil and gas wells stimulation potential evaluation, which is crucial to improve fracturing operation efficiency and reduce HF investment risk. The candidate-well selection model is a high dimension, nonlinear, strong coupling, multi-input single-output system. However, the conventional methods, such as production performance comparisons can not be easy to use for this nonlinear model. As a solution, the advanced methods such as T-S models in this paper can be effectively used in the candidate-well selection for HF. First, the subtractive clustering (SC) algorithm is employed to partition the fuzzy space of the given input–output data, which is adopted as the initial premise structure and parameters. Second, the clusters obtained on the first stage are used to initialize the fuzzy c-means (FCM) algorithm, which can obtain optimal cluster number and cluster centers. Third, the consequent parameters are identified by using the orthogonal least-squares (OLS) algorithm. Finally, the proposed approach is successfully applied to candidate-well selection for HF in Hechuan gas field in Sichuan basin, and validation results have demonstrated the effectiveness of the proposed method. Keywords Hydraulic fracturing · T-S fuzzy model · Stimulation potential · Candidate-well selection.
1 Introduction The formation of Hechuan in Sichuan basin is a typical gas reservoir with low permeability and low porosity. Since the geological condition is so complex, natural flow yields low production. However, such reservoirs are capable of producing at X. Xiang (B) · Y. Ting School of Science, Southwest Petroleum University, Chengdu 610500, China e-mail: [email protected] B.-Y. Cao and H. Nasseri (eds.), Fuzzy Information & Engineering and Operations Research & Management, Advances in Intelligent Systems and Computing 211, DOI: 10.1007/978-3-642-38667-1_55, © Springer-Verlag Berlin Heidelberg 2014
563
564
X. Xiang-jun and Y. Ting
commercial rates with the help of HF technology. The candidate-well selection for HF is crucial. Therefore, it is evident that to adopt this technology, considerable efforts have to be made in candidate-well selection. The model of candidate-well selection is a high dimension, nonlinear, strong coupling, multi-input single-output system. However, the conventional methods, including production performance comparisons, pattern recognition technology, production type curve matching [1–3], can not be easy to use for this nonlinear model. On the other hand, advanced methods such as T-S fuzzy systems have been proved to be useful in complex nonlinear system [4], especially, showing excellent ability in describing complicated dynamic of nonlinear behaviors of a process [5]. Therefore, T-S fuzzy model may be a good choice to describe such systems. In identification of fuzzy models, in order to obtain model structure and parameter estimation,
Data Loading...
