影响水力压裂效果的因素及人工神经网络评价

FACTORS INFLUENCING THE EFFECTS OF HYDROFRACTURING IN A LOW-PERMEABILITY OILFIELD AND POTENTIAL EVALUATION USING ARTIFICIAL NEURAL NETWORK

水力压裂是低渗透油气田提高开采效益的主要技术手段之一,但是影响水力压裂效果的因素较多,如地质特征、储层物性和地层能量等。为了达到理想的压裂效果,就要综合考虑各个影响因素之间的相互关系,找出影响压裂效果的主要因素。本文利用人工神经网络方法建立了数学评估模型并对已有的大量生产数据进行了网络训练和方法验证。结果证明所建立的压裂井潜能评估模型稳定性好,预测精度较高,对油田水力压裂的选井评层及产能预测具有一定的指导意义。

Hydrofracturing is one of the main technical means for improving the recovery efficiency in low-permeability oil/gas fields. However, there are many factors that influence the hydrofractufing effects, including geological characteristics, physical properties of reservoirs and energy of strata. In order to obtain ideal hydrofracturing results, it is necessary to give a comprehensive consideration of the relationships between various influence factors and find out the main factors that influence the hydrofractufing effects. The authors constructed a mathematic evaluation model by using the artificial neural network method and performed net training and method check and verification of a wealth of available production data. The results prove that the constructed potential evaluation model using hydrofracturing wells has good stability and a high precision of prediction. It has certain guiding significance for choosing wells and evaluating layers for hydrofracturing and forecasting of the production capacity.