Air Quality Forecasting Based on Gated Recurrent Long Short Term Memory Model in Internet of Things

With the continuous development of the Chinese economy and the gradual acceleration of urbanization, it has caused tremendous damage to the environment. The bad air environment seriously damages the physical and mental health of the people. The change in smog concentration will be affected by many realistic factors and exhibit nonlinear characteristics. The method proposed in this paper is to use the Internet of Things (IoT) technology to monitor the acquired data, process the data, and predict the next data using a neural network. The existing prediction models have limitations. They don't accurately capture the law between the concentration of haze and the factors affecting reality. It is difficult to accurately predict the nonlinear smog data. One algorithm proposed in this paper is a two-layer model prediction algorithm based on Long Short Term Memory Neural Network and Gated Recurrent Unit (LSTM&GRU). We set a double-layer Recurrent Neural Network to predict the PM2.5 value. This model is an improvement and enhancement of the existing prediction method Long Short Term Memory (LSTM). The experiment integrates data monitored by the IoT node and information released by the national environmental protection department. First, the data of 96 consecutive hours in four cities were selected as the experimental samples. The experimental results are close to the true value. Then, we selected daily smog data from 2014/1/1 to 2018/1/1 as a train and test dataset. It contains smog data for 74 city sites. The first 70% of the data was used for training and the rest for testing. The results of this experiment show that our model can play a better prediction.