Department of Computer Science & IT, University of Azad Jammu and Kashmir, King Abdullah Campus, Muzaffarabad 13100, AJK, Pakistan.
College of Computer Science & Engineering, University of Jeddah, Jeddah 23890, Saudi Arabia.
Math Biosci Eng. 2021 Mar 2;18(3):1992-2009. doi: 10.3934/mbe.2021104.
Accurate prediction of particulate matter (PM) using time series data is a challenging task. The recent advancements in sensor technology, computing devices, nonlinear computational tools, and machine learning (ML) approaches provide new opportunities for robust prediction of PM concentrations. In this study, we develop a hybrid model for forecasting PM and PM based on the multiscale characterization and ML techniques. At first, we use the empirical mode decomposition (EMD) algorithm for multiscale characterization of PM and PM by decomposing the original time series into numerous intrinsic mode functions (IMFs). Different individual ML algorithms such as random forest (RF), support vector regressor (SVR), k-nearest neighbors (kNN), feed forward neural network (FFNN), and AdaBoost are then used to develop EMD-ML models. The air quality time series data from Masfalah air station Makkah, Saudi Arabia are utilized for validating the EMD-ML models, and results are compared with non-hybrid ML models. The PMs (PM and PM) concentrations data of Dehli, India are also utilized for validating the EMD-ML models. The performance of each model is evaluated using root mean square error (RMSE) and mean absolute error (MAE). The average bias in the predictive model is estimated using mean bias error (MBE). Obtained results reveal that EMD-FFNN model provides the lowest error rate for both PM (RMSE = 12.25 and MAE = 7.43) and PM (RMSE = 4.81 and MAE = 3.02) using Misfalah, Makkah data whereas EMD-kNN model provides the lowest error rate for PM (RMSE = 20.56 and MAE = 12.87) and EMD-AdaBoost provides the lowest error rate for PM (RMSE = 15.29 and MAE = 9.45) using Dehli, India data. The findings also reveal that EMD-ML models can be effectively used in forecasting PM mass concentrations and to develop rapid air quality warning systems.
使用时间序列数据准确预测颗粒物（PM）是一项具有挑战性的任务。传感器技术、计算设备、非线性计算工具和机器学习（ML）方法的最新进展为 PM 浓度的稳健预测提供了新的机会。在这项研究中，我们开发了一种基于多尺度特征和 ML 技术的 PM 和 PM 预测的混合模型。首先，我们使用经验模态分解（EMD）算法对 PM 和 PM 的多尺度特征进行分解，将原始时间序列分解为多个固有模态函数（IMF）。然后，使用不同的单个 ML 算法，如随机森林（RF）、支持向量回归（SVR）、k-最近邻（kNN）、前馈神经网络（FFNN）和 AdaBoost 来开发 EMD-ML 模型。利用沙特阿拉伯麦加的 Masfalah 空气站的空气质量时间序列数据对 EMD-ML 模型进行验证，并将结果与非混合 ML 模型进行比较。还利用印度德里的 PMs（PM 和 PM）浓度数据对 EMD-ML 模型进行验证。使用均方根误差（RMSE）和平均绝对误差（MAE）评估每个模型的性能。使用平均偏差误差（MBE）估计预测模型的平均偏差。结果表明，对于麦加的 Misfalah 数据，EMD-FFNN 模型在 PM（RMSE=12.25，MAE=7.43）和 PM（RMSE=4.81，MAE=3.02）方面提供了最低的误差率，而对于 PM（RMSE=20.56，MAE=12.87）和 PM（RMSE=15.29，MAE=9.45）方面，EMD-kNN 模型提供了最低的误差率。研究结果还表明，EMD-ML 模型可有效地用于预测 PM 质量浓度，并开发快速空气质量预警系统。
