论文部分内容阅读
井斜是评价成井质量的重要指标之一,钻井工艺参数是影响竖井井斜的重要因素。基于自适应神经网络模糊推理(ANFIS,adaptive neural-network-based fuzzy inference system)建立了竖井井斜预测模型,将钻压和转速两个主要钻井工艺参数作为输入变量,并选用某竖井部分录井数据作为基础数据对该模型进行了训练和评价。结果表明,该预测方法的相对误差为0～22.3%,平均相对误差为7.6%,预测效果较好,在工程实践应用中具有现实意义。
Well inclination is one of the important indexes to evaluate well quality. Drilling process parameters are important factors that affect shaft deviation. Based on the adaptive neural network-based fuzzy inference system (ANFIS), a prediction model of shaft deviation is established. Taking the two main drilling parameters of WOB and RPM as input variables, some wells are selected for logging The data were used as the basic data to train and evaluate the model. The results show that the relative error of the forecasting method is 0-22.3% and the average relative error is 7.6%. The forecasting effect is good and has practical significance in engineering practice.