摘要
建立花期预报模型,发布观赏性植物的精准花期预报,为旅游活动提供重要参考依据,已经成为气象服务领域一个新的发展方向。为了解高淳旅游区油菜花期的变化规律,探索其预报方法,指导高淳油菜花节旅游活动,根据1985-2010年高淳站日最高气温、日最低气温、日平均气温、日降水量、日日照时数、日平均5 cm地温、日平均相对湿度和日小型蒸发量等气象观测数据,利用主成分分析法,得到其与油菜花期相关系数较大的3个主成分,即温度因子、天气因子和辐射因子,以此为输入因子,建立基于BP神经网络的油菜花期预报模型,探讨BP神经网络在花期预报领域的应用。结果表明,传统的有效积温方法预报结果与实际开花期平均相差4.25天,BP神经网络方法预报结果与实际开花期平均相差1.5天,与有效积温预报油菜花期的方法相比,BP神经网络技术具有预测结果准确率高和操作简单等特点,在花期预报领域具有广阔的应用前景。
Developing flowering phenological models is conducive to the accurate simulation of flowering periods of ornamental plants and could provide basis for seasonal flowering tourism events. This has become a new direction for the business of meteorological services. This paper analyzes the variation of oilseed rape florescence in Gaochun tourist area and explores its florescence forecasting methods to guide the tourism activities during the Gaochun rape flower festival. Based on meteorological observations, such as daily maximum temperature, daily minimum temperature, daily mean temperature, daily mean rainfall, daily sunshine hours, daily mean 5 cm soil temperature, daily mean air humidity and daily mean evaporation from 1985 to 2010, three principal components of meteorological elements including temperature factor, weather factor and radiation factor are obtained by using the principal component analysis(PCA) method. Then, taking these as input factors, we establish a prediction model of rape flower florescence based on BP neural network and discuss the application of BP neural network in the prediction of flower florescence. The results demonstrate that, compared with 4.25 d flowering forecast error by the traditional effective accumulated temperature method, the error by the BP neural network method is 1.5 d on average, so the BP neural network method has higher accuracy in predicting florescence of flower, and is easy to utilize. It can be widely used in the operation of florescence forecasting.
作者
孙家清
张志薇
艾文文
Sun Jiaqing;Zhang Zhiwei;Ai Wenwen(Meteorological Service Center of Jiangsu Province,Nanjing 210009,China)
出处
《气象与环境科学》
2019年第4期22-26,共5页
Meteorological and Environmental Sciences
基金
青年基金项目(Q201704)
关键词
BP神经网络
主成分分析
油菜花期
预报模型
BP neural network
principal component analysis
oilseed rape florescence
prediction model