Deep Spatio-Temporal Graph Network with Self-Optimization for Air Quality Prediction

Abstract

The environment and development are major issues of general concern. After much suffering from the harm of environmental pollution, human beings began to pay attention to environmental protection and started to carry out pollutant prediction research. A large number of air pollutant predictions have tried to predict pollutants by revealing their evolution patterns, emphasizing the fitting analysis of time series but ignoring the spatial transmission effect of adjacent areas, leading to low prediction accuracy. To solve this problem, we propose a time series prediction network with the self-optimization ability of a spatio-temporal graph neural network (BGGRU) to mine the changing pattern of the time series and the spatial propagation effect. The proposed network includes spatial and temporal modules. The spatial module uses a graph sampling and aggregation network (GraphSAGE) in order to extract the spatial information of the data. The temporal module uses a Bayesian graph gated recurrent unit (BGraphGRU), which applies a graph network to the gated recurrent unit (GRU) so as to fit the data’s temporal information. In addition, this study used Bayesian optimization to solve the problem of the model’s inaccuracy caused by inappropriate hyperparameters of the model. The high accuracy of the proposed method was verified by the actual PM2.5 data of Beijing, China, which provided an effective method for predicting the PM2.5 concentration.

Optimization Algorithm of Moving Object Detection Using Multiscale Pyramid Convolutional Neural Networks

“…CNN is a brand-new network that combines convolutional operation with multilayer artifcial neural networks (NNs) [7,8]. It automatically instructs the computer to perform a convolution operation to extract the desired features from the image, resulting in more universal and natural-looking features.…”

Filtered auxiliary model recursive generalized extended parameter estimation methods for Box–Jenkins systems by means of the filtering identification idea

“…By means of the filtering identification idea, this paper investigates and presents a filtered auxiliary model (multi-innovation) generalized extended SG identification method, a filtered auxiliary model (multi-innovation) recursive generalized extended gradient identification method, and a filtered auxiliary model (multi-innovation) recursive generalized extended least squares identification method for equation-error autoregressive moving average systems (i.e., Box-Jenkins systems). The proposed filtered auxiliary model recursive identification methods can be generalized to other linear systems, bilinear systems and nonlinear multivariable stochastic systems with colored noises [102][103][104][105][106][107][108][109] and can be applied to some engineering dynamical systems with colored noises [110][111][112][113][114][115] such as signal modeling and communication networked systems.…”

Multi‐innovation gradient‐based iterative identification methods for feedback nonlinear systems by using the decomposition technique

“…The MIGI algorithm and the D-MIGI algorithm in this article can be extended to polynomial nonlinear models, rational models, switching models, and exponential autoregressive models. The proposed multi-innovation gradient-based iterative identification methods for feedback nonlinear systems by using the decomposition technique in this paper can combine other identification idea and methods [108][109][110][111][112][113][114][115][116][117] for develop new identification algorithms of dynamical stochastic linear and nonlinear systems [118][119][120][121][122][123][124][125][126] such as chemical process control systems, robot control systems, information processing systems [127][128][129][130][131][132][133][134][135] and so on.…”