基于改进麻雀搜索算法优化BP神经网络的PM2.5浓度预测

doi:10.13474/j.cnki.11-2246.2022.0292

测绘通报 ›› 2022, Vol. 0 ›› Issue (10): 44-48,104.doi: 10.13474/j.cnki.11-2246.2022.0292

基于改进麻雀搜索算法优化BP神经网络的PM_2.5浓度预测

赵侃¹, 师芸^2,3, 牛敏杰^2,3, 王虎勤⁴

1. 中煤地航测遥感局, 陕西西安 710199;
2. 西安科技大学测绘科学与技术学院, 陕西西安 710054;
3. 自然资源部煤炭资源勘察与综合利用重点实验室, 陕西西安 710021;
4. 西安捷达测控有限公司, 陕西西安 710054

收稿日期:2021-09-21 发布日期:2022-11-02
通讯作者: 师芸。E-mail:shiyun0908@hotmail.com
作者简介:赵侃(1997-),男,硕士,助理主程师,主要研究方向为环境遥感。E-mail:19210061015@stu.xust.edu.cn
基金资助:
国家自然科学基金(41674013;41874012)

Prediction of PM_2.5 concentration based on optimized BP neural network with improved sparrow search algorithm

ZHAO Kan¹, SHI Yun^2,3, NIU Minjie^2,3, WANG Huqin⁴

1. Aerial Photogrammetry and Remote Sensing Bureau, Xi'an 710199, China;
2. College of Geomatics, Xi'an University of Science and Technology, Xi'an 710054, China;
3. Key Laboratory of Coal Resources Exploration and Comprehensive Utilization, Ministry of Land and Resources, Xi'an 710021, China;
4. Xi'an Jieda Measurement and Control Co., Ltd., Xi'an 710054, China

Received:2021-09-21 Published:2022-11-02

摘要/Abstract

摘要： 针对传统BP神经网络模型收敛速度慢、易陷入局部极值等问题,本文采用分段线性混沌映射(PWLCM)和萤火虫算法(FA)改进麻雀搜索算法(SSA),并优化BP神经网络模型初始权值和阈值,对西安市PM_2.5浓度进行预测。通过比较不同模型预测结果的评价指标,并与性能较优的SSA-BP模型对比,ISSA-BP模型预测结果的RMSE、MAPE、MAE分别下降了3.70、3.73、3.34。试验结果表明,改进后的麻雀搜索算法具有高效的全局最优搜索能力,优化后的ISSA-BP神经网络预测稳定性高,精度优于BP、SSA-BP神经网络模型,可用于预测PM_2.5浓度。

关键词: 麻雀搜索算法, 分段线性混沌映射, 萤火虫算法, BP神经网络, PM_2.5浓度预测

Abstract: Aiming at the problems of slow convergence speed and easy to fall into local extremum of traditional BP neural network model, this paper uses PWLCM(piece wise linear chaotic map)and FA(firefly algorithm)to improve SSA(sparrow search algorithm)and optimize the initial weights and thresholds of the BP neural network model to predict the PM_2.5 concentration in Xi'an. By comparing the evaluation indicators of the prediction results of different models, compared with the better-performing SSA-BP model, the RMSE, MAPE and MAE of the ISSA-BP model prediction results decrease by 3.70, 3.73 and 3.34, respectively. The experimental results show that the improved sparrow search algorithm has efficient global optimal search ability. The optimized ISSA-BP neural network has high prediction stability and accuracy, which is better than BP and SSA-BP neural network models and can be used to predict PM_2.5 concentration.

Key words: sparrow search algorithm, piece wise linear chaotic map, firefly algorithm, BP neural network, PM_2.5 concentration prediction

中图分类号:

P237

赵侃, 师芸, 牛敏杰, 王虎勤. 基于改进麻雀搜索算法优化BP神经网络的PM_2.5浓度预测[J]. 测绘通报, 2022, 0(10): 44-48,104.

ZHAO Kan, SHI Yun, NIU Minjie, WANG Huqin. Prediction of PM_2.5 concentration based on optimized BP neural network with improved sparrow search algorithm[J]. Bulletin of Surveying and Mapping, 2022, 0(10): 44-48,104.

参考文献

[1] DI Qian, KOUTRAKIS P, SCHWARTZ J. A hybrid prediction model for PM_2.5 mass and components using a chemical transport model and land use regression[J]. Atmospheric Environment, 2016, 131:390-399.
[2] ZHAO Rui, GU Xinxin, XUE Bing, et al. Short period PM_2.5 prediction based on multivariate linear regression model[J]. PLoS One, 2018, 13(7):e0201011.
[3] 贾澎涛,何华灿,刘丽,等.时间序列数据挖掘综述[J].计算机应用研究, 2007, 24(11):15-18.
[4] 高愈霄,汪巍,黄永海,等.基于神经网络和数值模型的重点区域PM_2.5预报比较分析[J].环境科学, 2022, 43(2):663-674.
[5] 孙宝磊,孙暠,张朝能,等.基于BP神经网络的大气污染物浓度预测[J].环境科学学报, 2017, 37(5):1864-1871.
[6] 杨云,付彦丽.关于空气中PM_2.5质量浓度预测研究[J].计算机仿真, 2016, 33(3):413-418.
[7] 张旭,杜景林.改进PSO-GA-BP的PM2.5浓度预测[J].计算机工程与设计, 2019, 40(6):1718-1723.
[8] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088):533-536.
[9] RUMELHART D E, MCCLELLAND J L,GROUP P R. Parallel distributed processing:explorations in the microstructure of cognition, Volume 1:foundations[Z]. Cambridge, Ma:MIT Press, 1986.
[10] 黄丽. BP神经网络算法改进及应用研究[D].重庆:重庆师范大学, 2008.
[11] CHEN Yegang. Prediction algorithm of PM_2.5 mass concentration based on adaptive BP neural network[J]. Computing, 2018, 100(8):825-838.
[12] 胡娟,郑军,许文龙,等.利用多源数据建立GA-BP算法模型估算PM_2.5的研究[J].气象科学, 2021, 41(3):314-322.
[13] 薛建凯.一种新型的群智能优化技术的研究与应用:麻雀搜索算法[D].上海:东华大学, 2020.
[14] XUE Jiankai, SHEN Bo. A novel swarm intelligence optimization approach:sparrow search algorithm[J]. Systems Science&Control Engineering, 2020, 8(1):22-34.
[15] ZHANG Jiangnan, XIA Kewen, HE Ziping, et al. Semi-supervised ensemble classifier with improved sparrow search algorithm and its application in pulmonary nodule detection[J]. Mathematical Problems in Engineering, 2021:1-18.
[16] 毛清华,张强.融合柯西变异和反向学习的改进麻雀算法[J].计算机科学与探索, 2021, 15(6):1155-1164.
[17] PAREEK N K, PATIDAR V, SUD K K. Image encryption using chaotic logistic map[J]. Image and Vision Computing, 2006, 24(9):926-934.
[18] LI Chunhu, LUO Guangchun, QIN Ke, et al. An image encryption scheme based on chaotic tent map[J].Nonlinear Dynamics, 2017, 87(1):127-133.
[19] XIANG Tao, LIAO Xiaofeng, WONG K W. An improved particle swarm optimization algorithm combined with piecewise linear chaotic map[J].Applied Mathematics and Computation,2007,190(2):1637-1645.
[20] 段玉先,刘昌云.基于Sobol序列和纵横交叉策略的麻雀搜索算法[J].计算机应用,2022,42(1):36-43.
[21] YANG Xinshe, HE Xingshi. Firefly algorithm:recent advances and applications[J]. International Journal of Swarm Intelligence, 2013, 1(1):36.
[22] YANG Xin she. Firefly algorithm, stochastic test functions and design optimization[J]. International Journal of Bio-Inspired Computation, 2010, 2(2):78-84.

基于改进麻雀搜索算法优化BP神经网络的PM_2.5浓度预测

Prediction of PM_2.5 concentration based on optimized BP neural network with improved sparrow search algorithm

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	李坤, 赵俊三, 林伊琳, 周豹. 基于不同斜坡单元划分方法和BP神经网络的泥石流易发性评价[J]. 测绘通报, 2022, 0(8): 68-74.
[2]	刘璐, 罗年学, 赵前胜. 基于历史数据的黄海浒苔爆发规模预测[J]. 测绘通报, 2022, 0(7): 7-11.
[3]	冯俊俊, 周立, 欧阳犬平, 周珍. 优化短期余水位组合预测模型[J]. 测绘通报, 2022, 0(3): 96-100.
[4]	丁仁军, 王友昆, 张君华, 刘晨. 基于BP神经网络的昆明天顶湿延迟模型[J]. 测绘通报, 2022, 0(3): 107-110.
[5]	毛斌, 韩文泉, 谢宏全, 吕海扬. 基于北京二号影像辅助nDSM的建筑物自动提取[J]. 测绘通报, 2022, 0(3): 132-137.
[6]	苏艺凡, 党建武, 王阳萍, 杨景玉. 改进的区间二型模糊聚类遥感影像变化检测[J]. 测绘通报, 2021, 0(7): 44-51,58.
[7]	李紫微, 马庆勋, 吕杰. BP神经网络的近地面臭氧估算及时空特征分析[J]. 测绘通报, 2021, 0(6): 28-32,126.
[8]	宋宝, 柯福阳, 赵兴旺. 面向LiDAR/Radar松组合的迭代加权IEKF-BP组合算法精度分析[J]. 测绘通报, 2021, 0(2): 44-48.
[9]	鲁小红. 最优权组合预测法在采煤沉陷变形预测中的应用[J]. 测绘通报, 2020, 0(4): 111-115.
[10]	郭云开, 许敏, 张晓炯, 刘雨玲. 结合PRO-4SAIL和BP神经网络的叶绿素含量高光谱反演[J]. 测绘通报, 2020, 0(3): 21-24.
[11]	冯绍权, 花向红, 陶武勇, 宣伟, 吴伟, 续东. 一种基于GA-BP-MC神经网络的高铁桥墩沉降预测模型[J]. 测绘通报, 2019, 0(7): 50-53,82.
[12]	周访滨, 邹联华, 张晓炯, 孟凡一. 栅格DEM微地形分类的BP神经网络法[J]. 测绘通报, 2019, 0(10): 101-104,132.
[13]	韩红超. 量子粒子群BP神经网络在GNSS高程转换中的应用分析[J]. 测绘通报, 2019, 0(1): 85-88.
[14]	杨丁亮, 刘志平, 仲崇武. 百度地图坐标解密方法精度分析[J]. 测绘通报, 2018, 0(7): 43-47.
[15]	丰秋林, 郑南山. 机器学习算法辅助的GPS信噪比观测值土壤湿度反演[J]. 测绘通报, 2018, 0(7): 106-111.

基于改进麻雀搜索算法优化BP神经网络的PM2.5浓度预测

Prediction of PM2.5 concentration based on optimized BP neural network with improved sparrow search algorithm

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

基于改进麻雀搜索算法优化BP神经网络的PM_2.5浓度预测

Prediction of PM_2.5 concentration based on optimized BP neural network with improved sparrow search algorithm