A Prediction Model of IoT Data Using Long Short-Term Memory Neural Network
Download as PDF
Meiyu Wen, Dandan Che, Jean-Pierre Niyigena, Ruohan Li, and Qingshan Jiang
In order to ensure the freshness stability of different ingredients in the cold storage room of commercial hotels' kitchen, the real-time monitoring of temperature and humidity is required. Thus, the establishment of a model to predict the temperature and humidity for future and for conducting early warning analysis on the temperature that may exceed the threshold is needed, so that relevant personnel can take defense measures before the temperature changes drastically. This paper detects the processing of abnormal value and the missing value of temperature and humidity according to the sensors’ receiving time. Long Short-Term Memory (LSTM) model is used for temperature and humidity time series prediction. Then, the result is compared with the prediction result using traditional statistical model of Autoregressive Integrated Moving Average (ARIMA). The final findings show that the predictive accuracy of the LSTM model is significantly better than the traditional model of ARIMA and the final temperature prediction result error is quite small.
Component, commercial hotel kitchens, data analysis, prediction models, anomaly detection