基于PSO-SVM的矿用CO传感器非线性补偿方法研究

发布时间：2018-08-20 15:36

【摘要】：随着矿井环境信息感知、危险源辨识等技术的发展,对气体传感器检测精度和可靠性的要求显著提高。为改善矿用气体传感器的性能,针对气体传感器补偿方法存在的技术难题,提出一种微粒群优化支持向量机(PSO-SVM)的非线性补偿方法。以CO传感器为例,采用Matlab软件进行数值仿真,BP神经网络方法将误差从18.48%降到8.51%,而采用微粒群优化支持向量机方法将误差降到5.28%。实验结果表明:PSO-SVM补偿方法能有效消除非目标参量对传感器输出结果的影响从而完成非线性补偿,提高了矿用CO传感器的可靠性与检测精度。
[Abstract]:With the development of mine environmental information perception and hazard source identification, the detection accuracy and reliability of gas sensors are greatly improved. In order to improve the performance of mine gas sensor, a nonlinear compensation method based on particle swarm optimization support vector machine (PSO-SVM) is proposed to solve the technical problems of gas sensor compensation. Taking CO sensor as an example, the error is reduced from 18.48% to 8.51% by using Matlab software and the error is reduced to 5.28% by using particle swarm optimization support vector machine method. The experimental results show that the proportion PSO-SVM compensation method can effectively eliminate the influence of non-target parameters on the output results of the sensor and thus achieve nonlinear compensation. The reliability and detection accuracy of the mine CO sensor are improved.
【作者单位】： 西安科技大学安全科学与工程学院;陕西省煤火灾害防治重点实验室;西安科技大学电气与控制工程学院;
【基金】：国家自然科学基金项目(51504186) 陕西省科技攻关项目(2016GY-191) 省教育厅科研专项项目(14JK1477)
【分类号】：TD711;TP18;TP212
，

本文编号：2194143