Urban Waterlogging Detection and Severity Prediction Using Artificial Neural Networks. Gupta, A., Bansal, A., Gupta, R., Naryani, D., & Sood, A. In 2017 IEEE 19th International Conference on High Performance Computing and Communications; IEEE 15th International Conference on Smart City; IEEE 3rd International Conference on Data Science and Systems (HPCC/SmartCity/DSS), pages 42–49, December, 2017.
doi  abstract   bibtex   
In many developing countries, rapid urbanisation of cities coupled with the lack of proper urban planning has made various streets susceptible to waterlogging during heavy rains. This severely affects the traffic movement across an entire city and leads to disruption in work. Since an important component of smart cities is developing an efficient urban mobility system, the authors have developed a method which helps in detection of areas prone to waterlogging and prediction of severity of waterlogging in these areas in the future. The areas susceptible to waterlogging are detected with the help of elevation of the area and the past travel time data. Elevation of an area is an indicator of the level or height of an area, so the low-lying areas are more prone to accumulation of water when it rains. Similarly, the past travel time data of an area also serves as a measure to find out the intensity of water logging as the larger the accumulation of water in an area, the more is the travel time. The past data pertaining to waterlogging severity in an area with respect to parameters such as the amount of rainfall and day of the week is used to train a neural network, which is then used to predict the possibility of waterlogging and its intensity in that area in the future.
@inproceedings{gupta_urban_2017,
	title = {Urban {Waterlogging} {Detection} and {Severity} {Prediction} {Using} {Artificial} {Neural} {Networks}},
	doi = {10.1109/HPCC-SmartCity-DSS.2017.6},
	abstract = {In many developing countries, rapid urbanisation of cities coupled with the lack of proper urban planning has made various streets susceptible to waterlogging during heavy rains. This severely affects the traffic movement across an entire city and leads to disruption in work. Since an important component of smart cities is developing an efficient urban mobility system, the authors have developed a method which helps in detection of areas prone to waterlogging and prediction of severity of waterlogging in these areas in the future. The areas susceptible to waterlogging are detected with the help of elevation of the area and the past travel time data. Elevation of an area is an indicator of the level or height of an area, so the low-lying areas are more prone to accumulation of water when it rains. Similarly, the past travel time data of an area also serves as a measure to find out the intensity of water logging as the larger the accumulation of water in an area, the more is the travel time. The past data pertaining to waterlogging severity in an area with respect to parameters such as the amount of rainfall and day of the week is used to train a neural network, which is then used to predict the possibility of waterlogging and its intensity in that area in the future.},
	booktitle = {2017 {IEEE} 19th {International} {Conference} on {High} {Performance} {Computing} and {Communications}; {IEEE} 15th {International} {Conference} on {Smart} {City}; {IEEE} 3rd {International} {Conference} on {Data} {Science} and {Systems} ({HPCC}/{SmartCity}/{DSS})},
	author = {Gupta, A. and Bansal, A. and Gupta, R. and Naryani, D. and Sood, A.},
	month = dec,
	year = {2017},
	keywords = {积水风险评估},
	pages = {42--49},
}

Downloads: 0