基于磁感应的油藏裂缝内地下无线传感网络定位

针对在油藏裂缝中地下无线传感网络节点定位问题，提出一种基于可变方向增强拉格朗日方法和粒子群优化相结合的定位算法。锚节点布置在井筒固定位置，传感器节点随压裂过程进入裂缝具有位置随机分布特性，节点间采用三线圈磁感应方式通信。推导了基于接收信号磁感应强度的节点间距离估计公式，据此获得全部节点与锚节点及与其邻居节点的距离集合。然后将定位问题转化为半定规划问题，并采用可变方向增强拉格朗日方法求解上述凸优化问题，获得初步定位，再将其作为粒子群优化算法的初始值，在上述初始值小邻域内局部搜索获得最优解作为最终定位。仿真结果表明该算法相对定位误差低于0.6，且定位精度受测量噪声变化影响较小。

Magnetic induction-based localization wireless underground sensor networks in oil reservoir

A magnetic induction (MI) based positioning scheme for accurately locating randomly deployed micro sensor nodes inside reservoir fracture is proposed. Firstly, the MI-based localization problem is presented based on the distance sets, which are estimated from the received magnetic flux density between transmitter and receiver nodes. Then the localization problem is further formulated into the semi-definite programming problem with the designated error function. Next, this paper proposes a two step positioning strategy for getting accurate localization results, which combines the alternating direction augmented Lagrangian (ADM) method and particle swarm optimization algorithm (PSO). The ADM method is first used to obtain the rough sensor locations, and then the fine-grained positioning is performed via the PSO algorithm. Simulation results show that the proposed method can achieve accurate sensor locations with relative error <0.6 and less sensitive to measurement noise.