{"title":"适合不同地区居民的智能天气警报系统","authors":"A. Durrani, M. Khurram, H. R. Khan","doi":"10.1109/IBCAST.2019.8667190","DOIUrl":null,"url":null,"abstract":"With changing climate, heatstroke has proved to be disastrous for few countries especially. The dwellers of different areas are not warned of the consequences to come specifically in their areas as they are told the average of the whole city, while temperature varies at different altitudes and over short distances. The solution provided in the paper, is a smart weather station that not only monitor weather data but also predict it and generate instant alerts for dwellers of different areas, to help them be warned of the future hazard, using the combination of Internet of things and Machine Learning. It is deployed with different sensors that collect weather data from the environment, which are sent to cloud, where predictions are made, for which certain neural network models have been compared to find out which gives the most accurate results. Those values, as well as the real-time values can be displayed on the mobile Application 24/7. Also, alerts are generated in the form of Tweets, which are accessible to everyone, as shown in Figure-1. It is also discovered that Nonlinear Autoregressive Exogenous Neural Network (NARXNET) Algorithm is the best to be implemented for prediction of Weather, with the mean squared error of 0.084% in 1.55 seconds for training a model and producing predictions for next 24 hours.","PeriodicalId":335329,"journal":{"name":"2019 16th International Bhurban Conference on Applied Sciences and Technology (IBCAST)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Smart Weather Alert System for dwellers of different Areas\",\"authors\":\"A. Durrani, M. Khurram, H. R. Khan\",\"doi\":\"10.1109/IBCAST.2019.8667190\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"With changing climate, heatstroke has proved to be disastrous for few countries especially. The dwellers of different areas are not warned of the consequences to come specifically in their areas as they are told the average of the whole city, while temperature varies at different altitudes and over short distances. The solution provided in the paper, is a smart weather station that not only monitor weather data but also predict it and generate instant alerts for dwellers of different areas, to help them be warned of the future hazard, using the combination of Internet of things and Machine Learning. It is deployed with different sensors that collect weather data from the environment, which are sent to cloud, where predictions are made, for which certain neural network models have been compared to find out which gives the most accurate results. Those values, as well as the real-time values can be displayed on the mobile Application 24/7. Also, alerts are generated in the form of Tweets, which are accessible to everyone, as shown in Figure-1. It is also discovered that Nonlinear Autoregressive Exogenous Neural Network (NARXNET) Algorithm is the best to be implemented for prediction of Weather, with the mean squared error of 0.084% in 1.55 seconds for training a model and producing predictions for next 24 hours.\",\"PeriodicalId\":335329,\"journal\":{\"name\":\"2019 16th International Bhurban Conference on Applied Sciences and Technology (IBCAST)\",\"volume\":\"8 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 16th International Bhurban Conference on Applied Sciences and Technology (IBCAST)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/IBCAST.2019.8667190\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 16th International Bhurban Conference on Applied Sciences and Technology (IBCAST)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IBCAST.2019.8667190","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Smart Weather Alert System for dwellers of different Areas
With changing climate, heatstroke has proved to be disastrous for few countries especially. The dwellers of different areas are not warned of the consequences to come specifically in their areas as they are told the average of the whole city, while temperature varies at different altitudes and over short distances. The solution provided in the paper, is a smart weather station that not only monitor weather data but also predict it and generate instant alerts for dwellers of different areas, to help them be warned of the future hazard, using the combination of Internet of things and Machine Learning. It is deployed with different sensors that collect weather data from the environment, which are sent to cloud, where predictions are made, for which certain neural network models have been compared to find out which gives the most accurate results. Those values, as well as the real-time values can be displayed on the mobile Application 24/7. Also, alerts are generated in the form of Tweets, which are accessible to everyone, as shown in Figure-1. It is also discovered that Nonlinear Autoregressive Exogenous Neural Network (NARXNET) Algorithm is the best to be implemented for prediction of Weather, with the mean squared error of 0.084% in 1.55 seconds for training a model and producing predictions for next 24 hours.