基于气候变量的苎麻产量SSA-BP预测模型

doi:10.13304/j.nykjdb.2023.0557

摘要/Abstract

摘要：

苎麻产量与生长期间的气候因子具有极高相关性，基于气候变量构建的苎麻产量预测模型能够有效精准预测最终产量。BP（back propagation）神经网络具有强大的数据分析能力，在作物产量预测建模中得到广泛应用，然而传统BP神经网络存在精度低、鲁棒性差等问题，可采用麻雀搜索算法（sparrow search algorithm，SSA）对BP神经网络模型进行优化。基于2010—2019年苎麻长期定位试验采集的纤维产量、鲜皮产量和气候数据，分析气候因子在10年内的变化趋势及其对多年生苎麻产量的影响，对比构建的BP神经网络模型及优化后的SSA-BP神经网络模型预测苎麻产量的性能，确定最佳的苎麻产量预测模型。结果表明，苎麻产量与季平均气温、季极端最高气温均值、季极端最低气温均值、季日照时数均值4项气候因子具有极显著相关关系。SSA算法能有效优化BP神经网络，基于SSA-BP的苎麻纤维产量预测模型和鲜皮产量预测模型的R²分别为0.591 3和0.679 1，高于BP神经网络的苎麻纤维产量预测模型（R²=0.405 7）和鲜皮产量预测模型（R²=0.551 8）。因此，SSA-BP模型能够更加科学、合理地预测苎麻产量，对于苎麻生产的田间管理及统筹规划具有重要指导意义。

关键词: 产量预测, 气候因子, 麻雀搜索算法, BP神经网络

Abstract:

The yield of ramie has a high correlation with the climate factors during the growth period， and the final yield can be effectively and accurately predicted by constructing a ramie yield prediction model based on climate variables. The BP neural network has strong data analysis capabilities and is widely used in crop yield prediction modeling. However， traditional BP neural networks have problems such as low accuracy and poor robustness. The sparrow search algorithm （SSA） can be used to optimize the BP neural network model. Based on the fiber yield， fresh skin yield and climate data collected in ramie long-term positioning experiment from 2010 to 2019， this study analyzed the changing trend of climate factors in 10 years and their impacts on the perennial ramie yield by comparing the performance of BP neural network model and the optimized SSA-BP neural network model in predicting ramie yield to determine the best prediction model. It showed that there were extremely significant correlations between the yield of ramie and 4 meteorological factors including the seasonal average temperature， the average seasonal extreme maximum temperature， the seasonal extreme minimum temperature， and the seasonal average sunshine hours， among which the ramie yield had the highest correlation with the seasonal average temperature. SSA algorithm could effectively optimize the BP neural network. R² of the ramie fiber yield prediction model and fresh skin yield prediction model based on SSA-BP were 0.591 3 and 0.679 1， respectively， which were higher than that of the ramie fiber yield prediction model （R²=0.405 7） and fresh skin yield prediction model （R²=0.551 8）. Therefore， the SSA-BP model could predict ramie yield more scientifically and reasonably， which was of great guiding significance for field management and overall plan of ramie production.

Key words: yield forecast, climatic factor, sparrow search algorithm, BP neural network

中图分类号:

S126

王辉, 付虹雨, 岳云开, 崔国贤, 佘玮. 基于气候变量的苎麻产量SSA-BP预测模型[J]. 中国农业科技导报, 2024, 26(1): 110-118.

Hui WANG, Hongyu FU, Yunkai YUE, Guoxian CUI, Wei SHE. Ramie Yield SSA-BP Prediction Model Based on Climate Variables[J]. Journal of Agricultural Science and Technology, 2024, 26(1): 110-118.

图/表 11

参考文献 23

1	杨洁, 唐昀, 申香英, 等. 我国苎麻产业现状与振兴发展[J]. 中国麻业科学, 2022, 44(4): 253-256.
	YANG J, TANG Y, SHEN X Y, et al.. Current situation and development of ramie industry in China [J]. Plant Fiber Sci. China, 2022, 44(4): 253-256.
2	付虹雨, 崔国贤, 李绪孟, 等. 基于无人机遥感图像的苎麻产量估测研究[J]. 作物学报, 2020, 46(9): 1448-1455.
	FU H Y, CUI G X, LI X M, et al.. Estimation of ramie yield based on UAV remote sensing image [J]. Acta Agron. Sin., 2020, 46(9): 1448-1455.
3	王辉, 付虹雨, 王继龙, 等. 基于多元回归和BP神经网络的苎麻产量估测模型比较研究[J]. 中国麻业科学, 2020, 42(5): 227-238.
	WANG H, FU H Y, WANG J L, et al.. Comparative study of ramie yield estimation models based on multiple regression and BP neural network [J]. Plant Fiber Sci. China, 2020, 42(5): 227-238.
4	王辉, 白玉超, 付虹雨, 等. 3个苎麻品种三年间产量及其构成因子的变化研究[J]. 中国麻业科学, 2020, 42(2): 76-82.
	WANG H, BAI Y C, FU H Y, et al.. Study on changes of yield and component factors of three ramie varieties in three years [J]. Plant Fiber Sci. China, 2020, 42(2): 76-82.
5	于珍珍, 邹华芬, 于德水, 等. 融合田间水热因子的甘蔗产量GA-BP预测模型[J]. 农业机械学报, 2022, 53(10): 277-283.
	YU Z Z, ZOU H F, YU D S, et al.. GA-BP prediction model of sugarcane yield with field hydrothermal factors [J]. Trans. Chin. Soc. Agric. Mach., 2022, 53(10): 277-283.
6	张文婧, 赵锦, 崔文倩, 等. 气候平均态和极端态变化对东北玉米气象产量的影响[J]. 中国农业科学, 2023, 56(10): 1859-1870.
	ZHANG W J, ZHAO J, CUI W Q, et al.. Effects of climate mean and extreme changes on meteorological yield of maize in Northeast China [J]. Sci. Agric. Sin., 2019, 56(10): 1859-1870.
7	刘焕才, 史书琦, 李曼, 等. 疏勒河流域中游地区玉米性状及单产的影响因子研究[J]. 干旱区地理 2023, 46(9): 1453-1466.
	LIU H C, SHI S Q, LI M, et al.. Study on influencing factors of maize character and yield in Middle reaches of Shule River Basin [J]. Arid Land Geography 2023, 46(9): 1453-1466.
8	刘头明, 汤清明, 朱四元, 等. 长江流域5个地区间苎麻产量差异的气候影响因素分析[J]. 安徽农业科学, 2011, 39(25): 15559-15562+15734.
	LIU T M, TANG Q M, ZHU S Y, et al.. Analysis of climatic factors affecting the difference of ramie yield in five regions in the Yangtze River Basin [J]. J. Anhui Agric. Sci., 2011, 39(25): 15559-15562, 15734.
9	MELILLO J M, RICHMOND T C, YOHE G W. Climate change impacts in the United States [R].The Third National Climate Assessment, 2014.
10	唐俊, 赵成萍, 周新志, 等. 基于EVI-RBF的玉米长势监测及产量预测[J]. 江苏农业学报, 2020, 36(3): 577-583.
	TANG J, ZHAO C P, ZHOU X Z, et al.. Maize growth monitoring and yield prediction based on EVI-RBF [J]. Jiangsu J. Agric. Sci., 2019, 36(3): 577-583.
11	徐敏, 吴洪颜, 张佩, 等. 基于气候适宜度的江苏水稻气候年景预测方法[J]. 气象, 2018, 44(9): 1200-1207.
	XU M, WU H Y, ZHANG P, et al.. Annual climate prediction method for Jiangsu rice based on climate suitability [J]. Meteorol. Monthly, 2018, 44(9): 1200-1207.
12	刘振洋, 赵家松, 胡仁傑, 等. 基于关联规则与多元线性回归的云南省甘蔗产量预测模型[J]. 广东农业科学, 2022, 49(12): 160-166.
	LIU Z Y, ZHAO J S, HU R J, et al.. Sugarcane yield prediction model based on association rules and multiple linear regression in Yunnan Province [J]. Guangdong Agric. Sci., 2022, 49(12): 160-166.
13	XUE J K, SHEN B. A novel swarm intelligence optimization approach: sparrow search algorithm [J]. Syst. Sci. Control Eng., 2020, 8(1): 22-34.
14	E A P, D. B. D, Narendra K, et al.. A comprehensive investigation of Genotype-Environment interaction effects on seed cotton yield contributing traits in Gossypium hirsutum L. using multivariate analysis and artificial neural network [J/OL]. Computers Electron. Agric., 2023, 211:107966 [2023-09-06]. .
15	SONG S, XIONG X, WU X, et al.. Modeling the SOFC by BP neural network algorithm [J]. Int. J. Hydrogen Energy, 2021, 46(38): 65-77.
16	娄高中, 谭毅, 白二虎. 主成分分析与改进PSO-BP神经网络的下沉系数预测[J]. 测绘科学, 2023, 48(2): 124-130, 147.
	LOU G Z, TAN Y, BAI E H. Principal component analysis and subsidence coefficient prediction of improved PSO-BP neural network [J]. Sci. Survey. Map., 2023, 48(2): 124-130, 147.
17	郭倩, 魏嘉豪, 张健, 等. 基于无人机多光谱影像和随机森林的蔬菜识别[J]. 中国农业科技导报, 2023, 25(2): 99-110.
	GUO Q, WEI J H, ZHANG J, et al.. Vegetable recognition based on UAV multi-spectral image and random forest [J]. J. Agric. Sci. Technol., 2023, 25(2): 99-110.
18	薛建凯. 一种新型的群智能优化技术的研究与应用[D]. 上海: 东华大学, 2020.
	XUE J K. Research and application of a new swarm intelligence optimization technology [D]. Shanghai: Donghua University, 2020.
19	顾清华, 姜秉佼, 常朝朝, 等. 求解大规模优化问题的改进麻雀搜索算法[J]. 控制与决策, 2023, 38(7): 1960-1968.
	GU Q H, JIANG B J, CHANG C C, et al.. An improved Sparrow search algorithm for large-scale optimization problems [J]. Control Decision, 2023, 38(7): 1960-1968.
20	胡建华, 黄宇龙, 张坚, 等. 基于麻雀搜索算法优化双隐含层BP神经网络的张力减径钢管壁厚预测[J]. 塑性工程学报, 2022, 29(8): 145-151.
	HU J H, HUANG Y L, ZHANG J, et al.. Prediction of wall thickness of tension reduced steel pipe by double-hidden layer BP neural network based on sparrow search algorithm [J]. J. Plasticity Eng., 2022, 29(8): 145-151.
21	刘头明, 汤清明, 朱四元, 等. 6个不同生态区域苎麻产量及环境影响因素分析[J]. 耕作与栽培, 2011, No.178(1): 30-32.
	LIU T M, TANG Q M, ZHU S Y, et al.. Analysis of ramie yield and environmental impact factors in six different ecological regions[J]. Tillage Cultivation, 2011, 178(1): 30-32.
22	袁晋琰, 吴春燕, 阮启亮, 等. 分宜气象因子对苎麻生长的影响及种植区域的划分[J]. 中国麻业科学, 2013, 35(4): 210-213, 218.
	YUAN J Y, WU C Y, RUAN Q L, et al. Influence of meteorological factors on ramie growth and division of planting area [J]. Plant Fiber Sci. China, 2013, 35(4): 210-213+218.
23	崔兴华, 靳晟, 姚芷馨, 等. 基于麻雀搜索算法和广义回归神经网络的玉米产量预测[J]. 数学的实践与认识, 2022, 52(7): 88-96.
	CUI X H, JIN S, YAO Z X, et al.. Maize yield prediction based on sparrow search algorithm and generalized regression neural network [J]. Math. Practice Theory, 2022, 52(7): 88-96.

气候因子 Climatic factor	头麻产量 Frist crop yield		二麻产量 Second corp yield		三麻产量 Thrid corp yield
气候因子 Climatic factor	纤维 Fibre	鲜皮 Fresh skin	纤维 Fibre	鲜皮 Fresh skin	纤维 Fibre	鲜皮 Fresh skin
季平均气温 Average seasonal temperature	-0.29	0.45^**	-0.35^**	-0.27	0.08	0.26
季极端最低气温均值 Average seasonal extreme minimum temperature	-0.40^**	-0.48^**	-0.33^*	-0.37^**	-0.18	0.15
季降水量均值 Mean seasonal precipitation	-0.06	0.41^**	-0.19	-0.12	0.08	0.31^*
季日照时数均值 Average seasonal sunshine hours	0.08	0.53^**	-0.17	-0.13	0.32^*	0.54^**
季极端最高气温均值 Average seasonal extreme maximum temperature	0.09	-0.26	-0.12	-0.06	0.37^**	0.35^**

气候因子 Climatic factor	头麻产量 Frist crop yield		二麻产量 Second corp yield		三麻产量 Thrid corp yield
气候因子 Climatic factor	纤维 Fibre	鲜皮 Fresh skin	纤维 Fibre	鲜皮 Fresh skin	纤维 Fibre	鲜皮 Fresh skin
季平均气温 Average seasonal temperature	-0.29	0.45^**	-0.35^**	-0.27	0.08	0.26
季极端最低气温均值 Average seasonal extreme minimum temperature	-0.40^**	-0.48^**	-0.33^*	-0.37^**	-0.18	0.15
季降水量均值 Mean seasonal precipitation	-0.06	0.41^**	-0.19	-0.12	0.08	0.31^*
季日照时数均值 Average seasonal sunshine hours	0.08	0.53^**	-0.17	-0.13	0.32^*	0.54^**
季极端最高气温均值 Average seasonal extreme maximum temperature	0.09	-0.26	-0.12	-0.06	0.37^**	0.35^**

气候因子 Climatic factor	鲜皮产量 Fresh skin yield	纤维产量 Fibre yield
季平均气温 Average seasonal temperature	-0.59^**	-0.54^**
季极端最高气温均值 Average seasonal extreme maximum temperature	-0.52^**	-0.51^**
季极端最低气温均值 Average seasonal extreme minimum temperature	-0.61^**	-0.55^**
季降水量均值 Mean seasonal precipitation	0.10	0.08
季日照时数均值 Average seasonal sunshine hours	-0.42^**	-0.46^**
季相对湿度均值 Average seasonal relative humidity	0.03	0.10

气候因子 Climatic factor	鲜皮产量 Fresh skin yield	纤维产量 Fibre yield
季平均气温 Average seasonal temperature	-0.59^**	-0.54^**
季极端最高气温均值 Average seasonal extreme maximum temperature	-0.52^**	-0.51^**
季极端最低气温均值 Average seasonal extreme minimum temperature	-0.61^**	-0.55^**
季降水量均值 Mean seasonal precipitation	0.10	0.08
季日照时数均值 Average seasonal sunshine hours	-0.42^**	-0.46^**
季相对湿度均值 Average seasonal relative humidity	0.03	0.10

模型Model	MSE	RMSE	MBE	MAE	MAPE/%
BP	1.93	1.39	-0.23	1.06	12.61
SSA-BP	1.33	1.15	-0.21	0.89	10.73