Exploiting PSO-SVM and sample entropy in BEMD for the prediction of interval-valued time series and its application to daily PM2.5 concentration forecasting

Jiang, Liyuan; Tao, Zhifu; Zhu, Jiaming; Zhang, Junting; Chen, Huayou

doi:10.1007/s10489-022-03835-3

Cited by 15 publications

(4 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, satisfactory forecasting results are obtained. However, for interval time series, such a forecast framework has only been used over the past 2 years (Jiang et al, 2023; Wang, Gao, et al, 2023; Zhu, Wan, & Wang, 2022).…”

Section: Resultsmentioning

confidence: 99%

“…Algorithm‐based studies are focused on algorithmic improvement for enhancing forecasting performance, which can be achieved through the following techniques: reducing the volatility of the time series through decomposition techniques (Hu et al, 2015; Liu et al, 2022; Maia et al, 2008, 2011; Wang, Li, et al, 2022; Xiong et al, 2014a; Zhang et al, 2020; Zhu, Wan, & Wang, 2022), improving the robustness of the prediction results by combining prediction methods (Arroyo et al, 2011; Maia & Carvalho, 2019; Xiong et al, 2017; Wang, Chudhery, et al, 2023; Zhang et al, 2020), using the optimization algorithm to avoid the blindness of parameter selection (Jiang et al, 2023; Liu et al, 2019; Wang, Gao, et al, 2023; Xiong et al, 2014b; Zhou et al, 2019), or using a combination of these techniques.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Interval time series forecasting: A systematic literature review

Wang,

Gurmani,

Tao

et al. 2023

Journal of Forecasting

Self Cite

View full text Add to dashboard Cite

Interval time series forecasting can be used for forecasting special symbolic data comprising lower and upper bounds and plays an important role in handling the complexity, instability, and uncertainty of observed objects. The purpose of this research is to identify the most widely used definition of interval time series; classify existing research into mature research, current research focus, and research gaps within the defined framework; and recommend future directions for interval forecasting research. To achieve this goal, we have conducted a systematic literature review, comprising search strategy planning, screening mechanism determination, document analysis, and report generation. During the search strategy planning stage, eight literature search libraries are selected to obtain the most extensive studies (total of 525 targets). In the screening‐mechanism determination stage, through the inclusion and exclusion mechanism, the literature that is repetitive, of low‐relevance, and from other fields are discarded, and 125 studies are finally selected. In the document analysis stage, tag‐based methods and classification grids are selected to analyze the shortlisted studies. The results show that there are still numerous research gaps in interval time series forecasting, such as the establishment of hybrid models, application of multisource information, development and application of evaluation techniques, and expansion of application scenarios. In the report‐generation stage, the problems that have been solved and encountered in interval forecasting are summarized, and future research directions are proposed. Finally, the most significant contribution of this research is to provide an overview of interval time series forecasting for easy reference by researchers and to facilitate further research.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Interval time series forecasting: A systematic literature review

Wang,

Gurmani,

Tao

et al. 2023

Journal of Forecasting

Self Cite

View full text Add to dashboard Cite

show abstract

“…Inspired by the way that cells communicate, In order to estimate the performance and production parameters of carbon fibers, Xiao et al [18] proposed IPSO to develop a bidirectional prediction model based on SVM. For a hybrid interval time series analysis model, Jiang [19] suggested a hybrid zone optimization technique based on enhanced BEMD and PSO-SVM. The interval time series-related BEMD's edge effect problem was resolved, and lower data volatility was also attained.A unique SVM ensemble creation technique based on clustering analysis was put forth by Zhou [20].…”

Section: Prediction and Development Of Support Vector Machinesmentioning

confidence: 99%

Research on wheel wear prediction and algorithm improvement based on Data-Driven technology analysis

Zhang

Niu

Chen

et al. 2023

Preprint

View full text Add to dashboard Cite

Because they support the entire weight of the train and carry substantial safety consequences, rail train wheels are vital bogie components. As a result, it is crucial to predict tire wear so that early wheel turning or replacement may effectively improve the stability of train operation and the economy of railroad operation. A substantial percentage of wheel wear will occur during travel, impacting safety. Data intelligence may be able to anticipate tire wear more precisely and timely than traditional prediction based on historical experience. As data-driven analysis advances, numerous intelligent algorithms are being employed to the prediction of tire wear. Despite the rapid advancement of data-driven technology, predicting wheel and rail. Regression is used in this paper to forecast wheel wear using complex algorithms. In order to better accurately predict future wheel wear data, this study is based on data-driven technologies for wheel data collection, analysis, filtering, and acquisition. The Particle Swarm Optimization Support Vector Machine (PSO-SVM) model is improved by employing the grid search strategy to minimize the two parameters of the support vector machine's kernel parameter g and the punishment function C. while employing the cross-validation method to prevent overfitting and achieve the regression prediction of wheel wear data. The results of the experiments demonstrate that the improved model generates more reliable results for each prediction. In order to demonstrate how the model affects optimization, the Genetic algorithm for optimization of support vector machines (GA-SVM) mathematical model is optimized for comparative analysis in terms of iteration rate, fitness, and mathematical metrics.

show abstract

“…While the use of BPNNs to predict ESP efficiency is rare, Wang et al [14] proposed a pork supply prediction method based on an improved mayfly optimization algorithm and back-propagated artificial neural network. Jiang et al [15] established a mixed interval time-series prediction model to achieve the high-precision interval PM2.5 concentration prediction by considering the daily variation in pollutant concentrations. Considering the problem of the unstable prediction results of BPNNs, the optimization of the PSO-SVM prediction model emerged.…”

Section: Introductionmentioning

confidence: 99%

Prediction and Comparative Analysis of the Influence of Magnetic Field Effect on PM2.5 Trapping Efficiency in Electrostatic Precipitator (ESP) under Different Temperatures

Zhang,

Zhang

2023

Applied Sciences

View full text Add to dashboard Cite

Charged particles have high momentum under high-temperature conditions, which helps to promote their movement towards a dust collector in a magnetic field environment, making it possible to improve the efficiency of the high-temperature wire-plate electrostatic precipitator (ESP) in this environment. A multi-field coupling model was established to numerically simulate PM2.5 dust-removal efficiency in an ESP under different working conditions. Combining the particle swarm optimization (PSO) algorithm with the support vector machine (SVM) model, the PSO-SVM prediction model is presented. Simulated data were used as training data, and PSO-SVM and back-propagation (BP) neural network models were utilized to predict collection efficiency under different working conditions, respectively. The results show that introducing a magnetic field can effectively improve the PM2.5 collection efficiency of wire-plate ESP, and the effect of a magnetic field on the dust-removal efficiency is more obvious at higher temperatures and higher flue gas velocities. When changing the working conditions, the predicted results of the magnetic field effect conform to simulated ones, and the PSO-SVM predicted values have a smaller relative error than those of the BP model, which can better adapt to different working conditions. All of the above conclusions can be utilized as a simple and adequately efficient example of the ESP model for follow-up research.

show abstract

Exploiting PSO-SVM and sample entropy in BEMD for the prediction of interval-valued time series and its application to daily PM2.5 concentration forecasting

Cited by 15 publications

References 39 publications

Interval time series forecasting: A systematic literature review

Interval time series forecasting: A systematic literature review

Research on wheel wear prediction and algorithm improvement based on Data-Driven technology analysis

Prediction and Comparative Analysis of the Influence of Magnetic Field Effect on PM2.5 Trapping Efficiency in Electrostatic Precipitator (ESP) under Different Temperatures

Contact Info

Product

Resources

About