基于BCSSA-SVM的煤矿瓦斯安全预警模型
Model of Coal Mine Gas Safety Early Warning System Based on SSA-SVM
张朝 1冯锋1
作者信息
- 1. 宁夏大学信息工程学院,宁夏 银川 750021
- 折叠
摘要
目前煤矿瓦斯预警主要是在瓦斯浓度监测上,考虑到其它环境因素对煤矿瓦斯安全的影响,提出了基于改进的麻雀搜索算法(BCSSA)优化SVM参数,融合多种环境参数对瓦斯安全等级进行预警分类.首先,BCSSA针对麻雀搜索算法种群初始化不均匀、易陷入局部最优问题,引入了伯努利映射和柯西变异;其次,建立BCSSA-SVM安全预警模型,根据《煤矿安全规程》将煤矿瓦斯安全等级分为四个等级;最后,分别对算法性能和模型进行了仿真.仿真结果表明,BCSSA算法性能有着明显的提升,BCSSA-SVM模型相比其它几种模型,精度有着明显提高.
Abstract
At present,coal mine gas early warning mainly focuses on gas concentration monitoring.Considering the impact of other environmental factors on coal mine gas safety,an improved Sparrow Search Algorithm(BCSSA)is proposed to optimize SVM parameters and integrate multiple environmental parameters to classify gas safety levels for warning.Firstly,BCSSA introduced Bernoulli mapping and Cauchy mutation were used to solve the problem of uneven initialization and susceptibility to local optima in the sparrow search algorithm.Secondly,the BCSSA-SVM safety warning model was established,and the coal mine gas safety level was divided into four levels according to the ″coal mine safety Regulations″.Finally,the algorithm performance and model were simulated respectively.The simulation results show that the performance of BCSSA algorithm is significantly improved,and the accuracy of BCSSA-SVM model is significantly improved compared with other models.
关键词
瓦斯/麻雀搜索算法/支持向量机/预警模型Key words
Gas/Sparrow search algorithm/Support vector machines/Early warning model引用本文复制引用
基金项目
宁夏重点研发计划(2022BEG02016)
宁夏回族自治区自然科学基金重点项目(2021AAC02004)
出版年
2024
NETL
NSTL
万方数据