Dynamic prediction model of soil moisture in rocky desertification region based on BP neural network optimized by genetic algorithm
YANG Jiaqi1,2, GUO Jianbin1,2, TANG Minghua1,2, ZHOU Jinxing1,2, WAN Long1,2
1. Jianshui Research Station, School of Soil and Water Conservation, Beijing Forestry University, 654300, Jianshui, Yunnan, China; 2. Key Laboratory of National Forestry and Grassland Administration on Soil and Water Conservation, Beijing Forestry University, 100083, Beijing, China
Abstract：[Background] Jianshui county of Yunnan province is a typical karst faulted basin landform, which is an important type area for the comprehensive control of karst rocky desertification in China. Because of the severe seasonal drought and rocky desertification, the barren water holding capacity of soil is poor and heterogeneity is high, and the prediction of soil water dynamic is difficult. It is of great significance for soil and water conservation and ecological restoration to reveal the process of soil water dynamic change and its influencing factors in this area.[Methods] Based on the daily meteorological data from April 16, 2006 to December 1, 2020 and soil moisture data from two different degrees of rocky desertification areas in Jianshui Karst rift basin of Yunnan province, a dynamic prediction model of soil moisture volume based on BP neural network was established for 0-10, 10-20 and 20-30 cm soil layers. Genetic algorithm was used to optimize the weights and thresholds of the model. The default factor method was used for sensitivity analysis to identify the main meteorological factors affecting the prediction of soil water dynamics in this area.[Results] The BP neural network model optimized by genetic algorithm was used to predict the soil volume moisture content of mild rocky desertification area and moderate rocky desertification area from September 13, 2019 to December1, 2020. The results showed that the predicted value of the model was close to the measured value. YMARE increased by 45% and 63%, YRMSE by 3% and 12%, R2 by 27% and 17%, respectively. The simulation accuracy of soil water in 20-30 cm depth was improved most obviously, The sensitivity analysis showed that the sensitivity index of soil moisture to rainfall was the highest (1.317-1.735), followed by the sensitivity index to average temperature (0.880 9-1.071 2), followed by atmospheric pressure and solar radiation.[Conclusions] The simulation result of genetic algorithm optimization was improved obviously. The results show that the BP neural network model optimized by genetic algorithm can be well applied to soil moisture simulation in rocky desertification area, and the simulation accuracy is significantly improved compared with the non-optimized model. It is proved that the prediction accuracy of soil water in moderate rocky desertification plot is higher than that in mild rocky desertification plot, and the prediction accuracy decreases with the increase of soil depth. Sensitivity analysis was used to determine that precipitation was the main meteorological factor, followed by air temperature. The sensitivity analysis showed that the soil at 0-10 cm surface layer was the most sensitive to meteorological factors, while the soil at >10-20 cm middle layer was the most sensitive to meteorological factors.
杨佳琦, 郭建斌, 汤明华, 周金星, 万龙. 基于遗传算法优化BP神经网络的石漠化区土壤水分动态预测模型[J]. 中国水土保持科学, 2022, 20(3): 109-118.
YANG Jiaqi, GUO Jianbin, TANG Minghua, ZHOU Jinxing, WAN Long. Dynamic prediction model of soil moisture in rocky desertification region based on BP neural network optimized by genetic algorithm. SSWC, 2022, 20(3): 109-118.
蒋忠诚, 罗为群, 童立强, 等.21世纪西南岩溶石漠化演变特点及影响因素[J].中国岩溶,2016, 35(5):461. JIANG Zhongcheng, LUO Weiqun, TONG Liqiang, et al. Evolution features of rocky desertification and influence factors in karst areas of southwest China in the 21stcentury[J]. Carsologica Sinica,2016,35(5):461.
[2]
付奔, 胡关东, 杨帆, 等. 云南干旱"常态化"的分析[J]. 水文, 2014, 34(4):82. FU Ben, HU Guandong, YANG Fan, et al. Analysis of drought normalization in Yunnan province[J]. Journal of China Hydrology, 2014, 34(4):82.
[3]
JIANG Zhongcheng, LIAN Yanqing, QIN Xiaoqun. Rocky desertification in southwest China:Impacts, causes,and restoration[J]. Earth-Science Reviews, 2014, 132(3):1.
[4]
彭淑惠, 袁忠玉, 张华. 滇东南南丘河流域石漠化特征及其防治研究[J]. 昆明理工大学学报(自然科学版),2014,39(1):26. PENG Shuhui, YUAN Zhongyu, ZHANG Hua. Research on characteristics and prevention of rocky desertification in Nanqiu river basin in southeastern Yunnan[J]. Journal of Kunming University of Science and Technology(Natural Sciences), 2014,39(1):26.
[5]
王赛男. 基于SWAT模型的气候变化与土地利用/覆被变化对岩溶断陷盆地水资源量的影响研究[D]. 重庆:西南大学, 2020:1. WANG Sainan. Impacts of climate change and land use/cover change on water resources in a karst Graben basin based on SWAT model[D]. Chongqing:Southwest University, 2020:1.
[6]
路璐. 淮北平原区农田土壤水分动态模拟实验研究[D]. 河北邯郸:河北工程大学,2020:5. LU Lu. Experimental study on dynamic simulation of farmland soil moisture in Huaibei plain[D]. Handan,Hebei:Hebei University of Engineering, 2020:5.
[7]
SHI Yajie, REN Chao, YAN Zhiheng,et al. High spatial-temporal resolution estimation of ground-based global navigation satellite system interferometric reflectometry (GNSS-IR) soil moisture using the genetic algorithm back propagation (GA-BP) neural network[J]. ISPRS International Journal of Geo-Information, 2021, 10(9):623.
[8]
林洁, 陈效民, 张勇等. 基于BP神经网络的太湖典型农田土壤水分动态模拟[J]. 南京农业大学学报, 2012, 35(4):140. LIN Jie,CHEN Xiaomin,ZHANG Yong,et al. Simulation of soil moisture dynamics based on the BP neural network in the typical farmland of Tai lake region[J]. Journal of Nanjing Agricultural University, 2012, 35(4):140.
[9]
丁从慧. 土壤水分对夏玉米生理生态特征的影响及动态模拟研究[D]. 南京:南京信息工程大学,2015:17. DING Conghui. Effects of soil moisture on ecophysiological characteristics of summer maize and dynamic simulation research[D]. Nanjing:Nanjing University of Information Science & Technology, 2015:17.
[10]
郝姗姗, 李梦华, 马永强, 等. 黄土丘陵区土壤侵蚀因子敏感性分析[J]. 中国水土保持科学, 2019, 17(2):77. HAO Shanshan,LI Menghua,MA Yongqiang,et al. Significance analysis of soil erosion factors in loess hilly gully region[J]. Science of Soil and Water Conservation,2019,17(2):77.
[11]
李柳阳, 朱青, 刘亚, 等. 基于气象因子的长三角地区农田站点土壤水分时间序列预测[J]. 水土保持学报, 2021, 35(2):131. LI Liuyang, ZHU Qing, LIU Ya, et al. Time series prediction of insitu soil moisture based on meteorological datain the Yangtze River Delta[J]. Journal of Soil and Water Conservation, 2021, 35(2):131.
[12]
刘威, 刘尚, 白润才, 等. 互学习神经网络训练方法研究[J]. 计算机学报, 2017, 40(6):1291. LIU Wei, LIU Shang, BAl Runcai, et al. Research of mutual learning neural network training method[J]. Chinese Journal of Computers, 2017, 40(6):1291.
[13]
LI Dongjie, LI Yangyang, LI Junxiang, et al. Gesture recognition based on BP neural network improved by chaotic genetic algorithm[J]. International Journal of Automation and Computing, 2018, 15(3):267.
[14]
HUANG Mingzhi, MA Yongwen,WAN Jinquan, et al. Simulation of a paper mill wastewater treatment using a fuzzy neural network[J]. Expert Systems with Applications, 2009,36(3):5064.
[15]
冯智莉, 易国洪, 李普山, 等. 并行化遗传算法研究综述[J]. 计算机应用与软件,2018,35(11):1. FENG Zhili, YI Guohong, LI Pushan, et al. Review of parallel genetic algorithm[J]. Computer Applications and Software, 2018, 35(11):1.
[16]
NIKBAKHT S, ANITESCU C, RABCZUK T. Optimizing the neural network hyperparameters utilizing genetic algorithm[J]. Journal of Zhejiang University Science A(Applied Physics & Engineering), 2021, 22(6):407.
[17]
CHEN J, ISLAMSLAM S, BISWAS P. Nonlinear dynamics of hourly ozone concentrations[J] Atmospheric Environment. 1998. 32(11):1839.
[18]
邓建强. 典型红壤区土壤水分动态变化模拟及预测模型研究[D]. 南京:南京农业大学, 2009:42. DENG Jianqiang. Soil water simulation and perdication in typical red soil region[D]. Nanjing:Nanjing Agricultural University, 2009:42.
[19]
郭守平. 不同固沙模式土壤水分动态变化及预测研究[D]. 北京:中国农业科学院, 2006. 48. GUO Shouping.The study on dynamic change and prediction of soil water content under different sand-fixing modes on shifting dune[D]. Beijing:Chinese Academy of Agricultural Sciences, 2006. 48.
[20]
白冬妹. 黄土丘陵半干旱地区柠条林地土壤水分预测模型研究[D]. 西安:西北农林科技大学, 2014:32. BAI Dongmei. The forecast model study of soil water content in semiarid region of the Loess Plateau[D]. Xi'an:Northwest A&F University, 2014:32.
[21]
刘长成, 刘玉国, 郭柯. 四种不同生活型植物幼苗对喀斯特生境干旱的生理生态适应性[J]. 植物生态学报, 2011, 35(10):1070. LIU Changcheng, LIU Yuguo, GUO Ke.Ecophysiological adaptations to drought stress of seedlings of four plant species with different growth forms in karst habitats[J]. Chinese Journal of Plant Ecology, 2011, 35(10):1070.
[22]
王宽, 李萍, 汤航, 等. 基于改进粒子群算法优化BP网络的短期风电功率预测[J]. 工业控制计算机, 2021, 34(11):119. WANG Kuan, LI Ping, TANG Hang, et al. Short term wind power prediction based on improved particle swarm optimization BP network industrial[J]. Control Computer, 2021, 34(11):119.
[23]
杨帆, 周敏, 金继民, 等. 智能优化算法及人工神经网络在催化裂化模型分析中的应用进展[J]. 石油学报(石油加工), 2020, 36(4):878. YANG Fan, ZHOU Min, JIN Jimin, et al. Research progress on application of intelligent optimization algorithm and artificial neural network in FCC model analysis[J]. Acta Petrolei Sinica(Petroleum Processing Section), 2020, 36(4):878.
[24]
何兴潼, 袁淑杰, 谷晓平, 等. 贵州省喀斯特区域土壤水分持续上升时期气象要素对土壤水分的影响[J]. 水土保持通报, 2018, 38(2):87. HE Xingtong, YUAN Shujie, GU Xiaoping, et al. Response of soil moisture in arising term to meteorological factors over karst area of Guizhou province[J]. Bulletin of Soil and Water Conservation, 2018, 38(2):87.
[25]
韩湘云, 景元书, 李根. 低丘红壤坡面土壤水分变化的气象因素冗余分析[J]. 生态学杂志, 2013, 32(9):368. HAN Xiangyun, JING Yuanshu, LI Gen. Relationships between soil moisture variability and meteorological factors on low hill red soil slope:A redundancy analysis[J]. Chinese Journal of Ecology, 2013, 32(9):368.
[26]
林洁, 陈效民, 张勇等.基于BP神经网络的太湖典型农田土壤水分动态模拟[J].南京农业大学学报,2012,35(4):140. LIN Jie, CHEN Xiaomin, ZHANG Yong, et al. Simulation of soil moisture dynamics based on the BP neural network in the typical farmland of Tai Lake region[J]. Journal of Nanjing Agricultural University, 2012, 35(4):140.
[27]
梁香寒. 毛乌素沙地不同固定程度油蒿群落土壤水分变化过程及影响因素研究[D]. 北京:北京林业大学, 2020:23. LIANG Xianghan. Soil moisture dynamics and its mechanism of Artemisia ordosica community in Mu Us sandy land[D]. Beijing:Beijing Forestry University, 2020:23.
2022, Vol. 20 
