EGS地热储层天然裂隙网络的非稳态温度场反演

储层天然裂隙网络是增强型地热系统（EGS）构建高渗透性人工储层的基础，是资源评价和开发利用的关键因素，已成为当下地热能研究热点。针对EGS热储层天然裂隙网络无法精确描述的难题，创建三维非稳态导热反问题的研究方法，对岩体裂隙网络参数进行反演计算。采用离散裂隙网络法、地球物理反演理论、传热学正/反问题的理论建模、高性能数值求解、勘测实验相结合的研究路线，将储层温度场局部信息（有限个测温孔的观测温度和大地热流）作为求解条件，采用模糊推理算法，对储层天然裂隙的方位、间距、延续性、隙宽、粗糙度、张开度、充填性、组数和连通性作参数反演，进而建立裂隙结构参数与温度场之间的数学关系；探索待反演参数的目标函数、灵敏度矩阵、分辨率计算公式，揭示温度测孔布置和误差水平对反问题解的存在性和唯一性的影响。为地球物理温度场反演和超大尺度三维非稳态导热反问题求解提供科学基础，并为EGS资源评估与优化开采提供工程指导。

The natural fracture system is essential to construct highly permeable reservoirs of the enhanced geothermal system (EGS), as well as to the resource evaluation and utilization. Therefore it is becoming a hot research topic. This project is to propose a research method based on the inverse problem of three-dimensional steady-state heat conduction, for accurately describing the natural fracture system of EGS reservoirs and inversely calculating rock fracture properties. Combining microscopic characterization, geophysical inversion theory, theoretical modeling of heat transfer modeling, high-performance numerical solution and experimental verification, and besides, using fuzzy inference algorithm, the reservoir natural fracture systems' orientation, spacing, continuity, gap width, roughness, opening degree, filling, group number and connectivity are inversely calculated according to limited information of reservoir temperature field (exploratory data including temperature and heat flux). And the mathematical relationship between fracture property parameters and temperature field will be established; the objective function, sensitivity matrix and resolution ratio calculation formula will be investigated, and the impacts of temperature exploratory placement and error level on inverse problem solution's existence and uniqueness will be revealed. This proposal could provide a scientific basis for the geophysical temperature field inversion and inverse problem solution of super-large-scale three-dimensional steady-state heat conduction. It would supply an engineering guide for EGS resource evaluation and utilization optimization.