基于量子衍生涡流算法和T⁃S模糊推理模型的储层岩性识别

Reservoir lithology identification based on quantum vortex search algorithm and T⁃S fuzzy reasoning model

鉴于梯度下降法易陷入局部极值、普通群智能优化算法易早熟收敛,提出一种基于量子衍生涡流算法(Quantum Vortex Search Algorithm,QVSA)和T⁃S模糊推理模型的岩性识别方法,QVSA具有操作简单、收敛速度快、寻优能力强等优点,有助于T⁃S模糊推理模型获得最优参数配置,从而实现储层岩性的准确识别。首先利用具有全局搜索能力的QVSA优化T⁃S模糊推理模型的各种参数;然后利用主成分分析方法降低获取的地震属性维度;再利用优化的T⁃S模糊推理模型识别储层岩性。实验结果表明,利用反映储层特征的8个地震属性识别储层岩性时,所提方法的识别正确率达到92%,比普通BP网络方法高5.1%,同时查准率、查全率、F1分数等指标也较BP网络方法提升明显。

Since the gradient descent method is prone to local extremes,and ordinary swarm intelligence optimi⁃zation algorithms are prone to premature convergence,a lithology identification method based on the quantum vortex search algorithm(QVSA)and T⁃S fuzzy reasoning model is proposed.QVSA has the advantages of simple operation,fast convergence speed,and strong optimization ability,which helps the T⁃S fuzzy reasoning model obtain the optimal parameter configuration and achieve accurate identification of reservoir lithology.Firstly,QVSA with global search capability is used to optimize various parameters of the T⁃S fuzzy reasoning model.Then,the principal component analysis method is used to reduce the dimensionality of the acquired seis⁃mic attributes,and the optimized T⁃S fuzzy reasoning model is utilized to identify the reservoir lithology.The experimental results show that when the eight seismic attributes reflecting the reservoir characteristics are used to identify the reservoir lithology,the identification accuracy of the proposed method reaches 92%,which is 5.1%higher than that of the ordinary BP network method.At the same time,the precision,recall,F1 score,and other indicators are improved significantly compared with those of the BP network method.