摘要
为更准确地预测瓦斯浓度,提高煤矿传感器瓦斯浓度监测数据的精准度,提出基于信息融合技术与遗传支持向量机(GA-SVM)相结合的算法。首先,利用信息融合技术对原始瓦斯浓度数据进行关联性重构;然后,通过对基于遗传算法(GA)优化的支持向量机(SVM)惩罚因子C和回归参数w寻优,建立煤矿瓦斯浓度多传感器预测模型。结果表明:基于信息融合和GA-SVM的煤矿瓦斯浓度多传感器性能得到较大提升,使煤矿瓦斯浓度传感器在复杂的井下环境中,能够较为准确地预测出浓度范围,并在此基础上拟合出理想曲线,有效追踪瓦斯浓度趋势。
A algorithm based on information fusion technology and GA-SVM is worked out in order to predict the gas concentration value more accurately. First, using fusion technology, the original correlation data are reconstructed and viewed as the data base. According to the optimization of penalty factor C and regression parameter w of the support vector machine based on genetic algorithm optimization, a coal multi- sensor prediction model is built. The results show that the performance of multi-sensor based on GA-SVM and information fusion technology is improved greatly, and the range of gas concentration can be predict accurately. The ideal curve can be fitted based on the predicted values, thus the trend of gas concentration can be tracked effectively.
出处
《中国安全科学学报》
CAS
CSCD
北大核心
2013年第9期33-38,共6页
China Safety Science Journal
基金
国家自然科学基金资助(51274118)
