基于IACO-BP算法的洪涝灾害应急物资需求预测
|
摘要
为了提高洪涝灾害应急物资需求预测的准确性,提出了一种改进蚁群优化BP神经网络智能算法.以受灾转移人数为预测对象,选取受灾人口、最大降雨量、洪水等级、降雨等级、受灾范围、房屋倒塌数、降雨时长和预报水平等洪涝灾害指标为研究因素,获得基于IACO-BP算法的受灾转移人数预测模型.结合库存管理知识间接预测洪涝灾害应急物资需求量.结果表明:IACO-BP算法获得预测值的均方误差比BP和PSO-BP算法获得的均方误差分别小93. 62%和90. 91%; IACO-BP、PSO-BP和BP网络运行时间分别为3、10和33 s; IACO-BP算法具有更高的精度和网络迭代效率.
In order to improve the accuracy of demand prediction of emergency supplies for the flood disaster,an improved BP neural network intelligent algorithm(IACO-BP) by ant colony optimization was proposed. Taking the number of disaster-related migrants as the prediction object,such indicators of flood disaster as disaster-related population,maximum rainfall,flood grade,rainfall grade,disaster area,number of collapsed houses,rainfall duration and forecast level were selected as the research factors to obtain the prediction model for the disaster-related migrants based on the IACO-BP algorithm. Combined with the inventory management knowledge, the demand amount of emergency supplies for flood disaster was indirectly predicted. The results showthat the mean square error of prediction value obtained with the IACO-BP algorithm is lower by 93. 62% and 90. 91% than that with BP and PSO-BP algorithms,respectively. The network running time of IACO-BP,PSO-BP and BP is 3 s,10 s and 33 s,respectively.The IACO-BP algorithm has higher accuracy and network iteration efficiency.
In order to improve the accuracy of demand prediction of emergency supplies for the flood disaster,an improved BP neural network intelligent algorithm(IACO-BP) by ant colony optimization was proposed. Taking the number of disaster-related migrants as the prediction object,such indicators of flood disaster as disaster-related population,maximum rainfall,flood grade,rainfall grade,disaster area,number of collapsed houses,rainfall duration and forecast level were selected as the research factors to obtain the prediction model for the disaster-related migrants based on the IACO-BP algorithm. Combined with the inventory management knowledge, the demand amount of emergency supplies for flood disaster was indirectly predicted. The results showthat the mean square error of prediction value obtained with the IACO-BP algorithm is lower by 93. 62% and 90. 91% than that with BP and PSO-BP algorithms,respectively. The network running time of IACO-BP,PSO-BP and BP is 3 s,10 s and 33 s,respectively.The IACO-BP algorithm has higher accuracy and network iteration efficiency.
引文
[1]Spencer S,Evans R,Allen T,et al.A multivariate time series approach to modeling and forecasting demand in the emergency department[J].Journal of Biomedical Informatics,2009,42(1):123-139.
[2]蔡玫,曹杰.应急物资需求量的二型模糊集合预测方法[J].中国安全科学学报,2015,25(9):165-170.(CAI Mei,CAO Jie.A type-2 fuzzy set based approach to predicting emergency material demand[J].China Safety Science Journal,2015,25(9):165-170.)
[3]曾波,孟伟,刘思峰,等.面向灾害应急物资需求的灰色异构数据预测建模方法[J].中国管理科学,2015,23(8):84-91.(ZENG Bo,MENG Wei,LIU Si-feng,et al.Prediction modeling method of grey isomerism data for calamity emergency material demand[J].Chinese Journal of M anagement Science,2015,23(8):84-91.)
[4]詹沙磊,傅培华,李修琳,等.基于马尔科夫决策的应急物资动态分配模型[J].控制与决策,2018,33(7):1312-1318.(ZHAN Sha-lei,FU Pei-hua,LI Xiu-lin,et al.Dynamic programming approach for relief goods allocation based on M arkov decision[J].Control and Decision,2018,33(7):1312-1318.)
[5]李沁鲜.基于需求分析的灾害应急物资配送问题研究[D].兰州:兰州交通大学,2013.(LI Qin-xian.Study on disaster emergency material distribution based on demand analysis[D].Lanzhou:Lanzhou Jiaotong University,2013.)
[6]钱枫林,崔健.BP神经网络模型在应急需求预测中的应用---以地震伤亡人数预测为例[J].中国安全科学学报,2013,23(4):20-25.(QIAN Feng-lin,CUI Jian.Application of BP neural netw ork model in emergency demand forecasting:a case study of earthquake casualty prediction[J].China Safety Science Journal,2013,23(4):20-25.)
[7]刘建华,张正,吴洁明.基于BP神经网络的城市水灾灾情预测模型[J].计算机工程与设计,2005,26(3):699-701.(LIU Jian-hua,ZHANG Zheng,WU Jie-ming.City flood forecast model based on BP netw ork[J].Computer Engineering and Design,2005,26(3):699-701.)
[8]程琼琼.基于GA的BP神经网络的地震应急物资需求预测[D].成都:西南财经大学,2016.(CHENG Qiong-qiong.Earthquake emergency material demand forecast based on BP neural netw ork of genetic algorithm[D].Chengdu:Southw estern University of Finance and Economics,2016.)
[9]亢丽君.粒子群优化BP神经网络在应急物资需求预测中的应用研究[D].兰州:兰州交通大学,2013.(KANG Li-jun.Research on application of particle sw arm optimization BP neural netw ork in forecasting emergency supplies demand[D].Lanzhou:Lanzhou Jiaotong University,2013.)
[10]Mohammadi R,Ghomi S M T,Zeinali F.A new hybrid evolutionary based RBF netw orks method for forecasting time series:a case study of forecasting emergency supply demand time series[J].Engineering Applications of Artificial Intelligence,2014(36):204-214.
[2]蔡玫,曹杰.应急物资需求量的二型模糊集合预测方法[J].中国安全科学学报,2015,25(9):165-170.(CAI Mei,CAO Jie.A type-2 fuzzy set based approach to predicting emergency material demand[J].China Safety Science Journal,2015,25(9):165-170.)
[3]曾波,孟伟,刘思峰,等.面向灾害应急物资需求的灰色异构数据预测建模方法[J].中国管理科学,2015,23(8):84-91.(ZENG Bo,MENG Wei,LIU Si-feng,et al.Prediction modeling method of grey isomerism data for calamity emergency material demand[J].Chinese Journal of M anagement Science,2015,23(8):84-91.)
[4]詹沙磊,傅培华,李修琳,等.基于马尔科夫决策的应急物资动态分配模型[J].控制与决策,2018,33(7):1312-1318.(ZHAN Sha-lei,FU Pei-hua,LI Xiu-lin,et al.Dynamic programming approach for relief goods allocation based on M arkov decision[J].Control and Decision,2018,33(7):1312-1318.)
[5]李沁鲜.基于需求分析的灾害应急物资配送问题研究[D].兰州:兰州交通大学,2013.(LI Qin-xian.Study on disaster emergency material distribution based on demand analysis[D].Lanzhou:Lanzhou Jiaotong University,2013.)
[6]钱枫林,崔健.BP神经网络模型在应急需求预测中的应用---以地震伤亡人数预测为例[J].中国安全科学学报,2013,23(4):20-25.(QIAN Feng-lin,CUI Jian.Application of BP neural netw ork model in emergency demand forecasting:a case study of earthquake casualty prediction[J].China Safety Science Journal,2013,23(4):20-25.)
[7]刘建华,张正,吴洁明.基于BP神经网络的城市水灾灾情预测模型[J].计算机工程与设计,2005,26(3):699-701.(LIU Jian-hua,ZHANG Zheng,WU Jie-ming.City flood forecast model based on BP netw ork[J].Computer Engineering and Design,2005,26(3):699-701.)
[8]程琼琼.基于GA的BP神经网络的地震应急物资需求预测[D].成都:西南财经大学,2016.(CHENG Qiong-qiong.Earthquake emergency material demand forecast based on BP neural netw ork of genetic algorithm[D].Chengdu:Southw estern University of Finance and Economics,2016.)
[9]亢丽君.粒子群优化BP神经网络在应急物资需求预测中的应用研究[D].兰州:兰州交通大学,2013.(KANG Li-jun.Research on application of particle sw arm optimization BP neural netw ork in forecasting emergency supplies demand[D].Lanzhou:Lanzhou Jiaotong University,2013.)
[10]Mohammadi R,Ghomi S M T,Zeinali F.A new hybrid evolutionary based RBF netw orks method for forecasting time series:a case study of forecasting emergency supply demand time series[J].Engineering Applications of Artificial Intelligence,2014(36):204-214.