基于卫星及气象信息的输电线路灾害监测与风险评估平台研究
|
摘要
为了有效地进行灾害预警,研究了一种基于卫星及气象信息的输电线路灾害监测与风险评估平台,该平台包括输电线路地理信息子平台、卫星遥感数据接收子平台、数据建模与风险评估子平台、灾害预警与处置子平台、支撑和运营子平台。数据建模与风险评估子平台包括火灾发展趋势预测子模块,该子模块采用自回归集成移动平均作为时间预测方法预测火灾数据的时序,将动态回归神经网络作为空间预测方法来寻找火灾数据间潜藏的空间相关性,再用马尔可夫链模型来模拟火灾数据的扰动,最后用统计回归得到时空集成、去扰动的预测结果。
In order to realize disaster warning effectively,a disaster monitoring and risk assessment platform of transmission line based on satellite and meteorological information is reaesrched,including transmission line geographic information platform,satellite remote sensing data receiving sub platform,data modeling and risk assessment sub platform,disaster warning and disposal platform,as well as the support and operation platform.The data modeling and risk assessment sub platform includes the fire development trend prediction module.The module uses the autoregressive integrated moving average as the time series of fire data predicted by the time predition method and the dynamic regression neural network method as the space prediction method to find the latent spatial correlation between fire data,and then uses the Markov chain model to simulate the disturbance of fire data.Finally,the spatiotemporal integrated and disturbance eliminated prediction results are obtained with statistical regression.
In order to realize disaster warning effectively,a disaster monitoring and risk assessment platform of transmission line based on satellite and meteorological information is reaesrched,including transmission line geographic information platform,satellite remote sensing data receiving sub platform,data modeling and risk assessment sub platform,disaster warning and disposal platform,as well as the support and operation platform.The data modeling and risk assessment sub platform includes the fire development trend prediction module.The module uses the autoregressive integrated moving average as the time series of fire data predicted by the time predition method and the dynamic regression neural network method as the space prediction method to find the latent spatial correlation between fire data,and then uses the Markov chain model to simulate the disturbance of fire data.Finally,the spatiotemporal integrated and disturbance eliminated prediction results are obtained with statistical regression.
引文
[1]刘振亚,张启平.国家电网发展模式研究[J].中国电机工程学报,2013,33(7):1-10.
[2]陆佳政,刘毓,吴传平,等.输电线路山火卫星监测与告警算法研究[J].中国电机工程学报,2015,35(21):5511-5519.
[3]徐雪仁,宫鹏,黄学智,等.资源卫星(可见光)遥感数据获取任务调度优化算法研究[J].遥感学报,2007,11(1):109-114.
[4]张力,马忠松.基于软件无线电的QPSK解调器仿真及实现[J].中国有线电视,2005(6):557-561.
[5]冷春艳.基于红外图像处理的森林火灾识别关键技术研究[D].成都:电子科技大学,2012.
[6]孙轶轩,邵春福,计寻,等.基于ARIMA与信息粒化SVR组合模型的交通事故时序预测[J].清华大学学报(自然科学版),2014(3):348-353.
[7]梅志雄,徐颂军,王佳璆.基于DRNN和ARIMA模型的森林火灾面积时空综合预测方法[J].林业科学,2009,45(8):101-107.
[8]蒋乐,刘俊勇,魏震波,等.基于马尔可夫链模型的输电线路运行状态及其风险评估[J].电力系统自动化,2015,39(13):51-57,80.
[2]陆佳政,刘毓,吴传平,等.输电线路山火卫星监测与告警算法研究[J].中国电机工程学报,2015,35(21):5511-5519.
[3]徐雪仁,宫鹏,黄学智,等.资源卫星(可见光)遥感数据获取任务调度优化算法研究[J].遥感学报,2007,11(1):109-114.
[4]张力,马忠松.基于软件无线电的QPSK解调器仿真及实现[J].中国有线电视,2005(6):557-561.
[5]冷春艳.基于红外图像处理的森林火灾识别关键技术研究[D].成都:电子科技大学,2012.
[6]孙轶轩,邵春福,计寻,等.基于ARIMA与信息粒化SVR组合模型的交通事故时序预测[J].清华大学学报(自然科学版),2014(3):348-353.
[7]梅志雄,徐颂军,王佳璆.基于DRNN和ARIMA模型的森林火灾面积时空综合预测方法[J].林业科学,2009,45(8):101-107.
[8]蒋乐,刘俊勇,魏震波,等.基于马尔可夫链模型的输电线路运行状态及其风险评估[J].电力系统自动化,2015,39(13):51-57,80.