一种井下人员无线定位算法研究

针对传统井下指纹定位算法存在需要采集大量指纹数据和定位精度不高的问题,提出了一种差分鱼群优化最小二乘支持向量机(DEAFSA-LSSVM)的井下人员无线定位算法.首先将井下实验区域划分为多个小区域,并利用克里金插值算法建立指纹数据库;然后利用差分进化与人工鱼群混合智能算法优化正则化参数和核函数宽度,建立最小二乘支持向量机算法模型,利用无线采集接收终端采集待定位点的无线信息数据,通过最小二乘支持向量机算法模型计算出其所属小区域;最后利用小区域内无线信息数据,通过加权K近邻算法进行实时定位.实验结果表明:该定位算法的收敛速度快,分类准确,准确率达到98.87%;定位精度高,平均定位误差为1.51 m,比未经优化的最小二乘支持向量机算法的定位精度提高18.82%.

Research on a wireless positioning algorithm for underground personnel

For problems that traditional underground fingerprint positioning algorithm needs to collect a large number of fingerprint data and positioning accuracy is not high, a wireless positioning algorithm for underground personnel based on differential evolution and artificial fish swarm algorithm optimization least square support vector machine( DEAFSA-LSSVM) was proposed. Firstly, the underground experimental area is divided into several small areas, and the fingerprint database is established by Kriging interpolation algorithm. Secondly, the hybrid intelligent algorithm of differential evolution and artificial fish swarm is used to optimize regularization parameters and width of kernel function, and the least squares support vector machine algorithm model is established. The wireless acquisition and reception terminal is used to collect wireless information data of undetermined site, and its small area is calculated by the least squares support vector machine algorithm model. Finally, the wireless information data in the small area is used for real-time positioning by weighted K-nearest neighbor algorithm. The experimental results show that the algorithm has high convergence speed and high classification accuracy, the classification accuracy is 98.87%;and has high positioning accuracy, the average positioning error is 1.51 m, which is 18.82%higher than that of the least squares support vector machine algorithm without optimization.