PFVAE: A Planar Flow-Based Variational Auto-Encoder Prediction Model for Time Series Data

Mathematics, Journal Year: 2022, Volume and Issue: 10(4), P. 610 - 610

Published: Feb. 16, 2022

Prediction based on time series has a wide range of applications. Due to the complex nonlinear and random distribution data, performance learning prediction models can be reduced by modeling bias or overfitting. This paper proposes novel planar flow-based variational auto-encoder model (PFVAE), which uses long- short-term memory network (LSTM) as designs (VAE) data predictor overcome noise effects. In addition, internal structure VAE is transformed using flow, enables it learn fit nonlinearity improve dynamic adaptability network. The experiments verify that proposed superior other regarding accuracy proves effective for predicting data.

Language: Английский

---

PM2.5 volatility prediction by XGBoost-MLP based on GARCH models

Journal of Cleaner Production, Journal Year: 2022, Volume and Issue: 356, P. 131898 - 131898

Published: April 25, 2022

Language: Английский

---

Haze Risk Assessment Based on Improved PCA-MEE and ISPO-LightGBM Model

Systems, Journal Year: 2022, Volume and Issue: 10(6), P. 263 - 263

Published: Dec. 19, 2022

With the economic development in China, haze risks are frequent. It is important to study urban risk assessment manage disaster. The indexes of 11 cities Fenwei Plain were selected from three aspects: sensitivity disaster-inducing environments, component hazards and vulnerability disaster-bearing bodies, combined with regional disaster system theory. hazard levels evaluated using matter-element extension (MEE) model, indicator weights determined by improving principal analysis (PCA) method entropy weight method, finally, five models established particle swarm optimization (IPSO) light gradient boosting machine (LightGBM) algorithm. used assess affected populations, transportation damage risk, crop area direct loss comprehensive before a event occurs. experimental comparison shows that index Xi’an city highest, full can improve evaluation accuracy 4–16% compared only causative factor index, which indicates proposed PCA-MEE-ISPO-LightGBM model results more realistic reliable.

Language: Английский

---

The spatial-temporal evolution mechanism of PM2.5 concentration based on China's climate zoning

Journal of Environmental Management, Journal Year: 2022, Volume and Issue: 325, P. 116671 - 116671

Published: Nov. 4, 2022

Language: Английский

---

Forecasting of Beijing PM2.5 with a hybrid ARIMA model based on integrated AIC and improved GS fixed-order methods and seasonal decomposition

Heliyon, Journal Year: 2022, Volume and Issue: 8(12), P. e12239 - e12239

Published: Dec. 1, 2022

Accurate particulate matter 2.5 (PM2.5) prediction plays a crucial role in the accurate management of air pollution and prevention respiratory diseases. However, PM2.5, as nonlinear time series with great volatility, is difficult to achieve prediction. In this paper, hybrid autoregressive integrated moving average (ARIMA) model proposed based on Augmented Dickey-Fuller test (ADF root test) annual PM2.5 data, thus demonstrating necessity first-order difference. The new method using akaike information criterion (AIC) improved grid search (GS) methods avoid bias caused by AIC alone determine order because data are not exactly normally distributed. comprehensive evaluation coefficient (CEC) used select optimal parameter structure considering multiple perspectives. entropy value decomposed obtained range A (RangeEn_A), reconstructed according value, finally predicted. We Beijing for validation results showed that ARIMA values RMSE 99.23%, MAE 99.20%, R2 118.61%, TIC 99.28%, NMAE 98.71%, NMSE 99.97%, OPC 43.13%, MOPC 98.43% CEC 99.25% compared traditional model. show does greatly improve performance provides convincing tool policy formulation governance.

Language: Английский

---

A novel two-stage seasonal grey model for residential electricity consumption forecasting

Energy, Journal Year: 2022, Volume and Issue: 258, P. 124664 - 124664

Published: June 30, 2022

Language: Английский

---

Grey prediction of carbon emission and carbon peak in several developing countries

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108210 - 108210

Published: March 12, 2024

Language: Английский

---

Understanding the distribution and drivers of PM2.5 concentrations in the Yangtze River Delta from 2015 to 2020 using Random Forest Regression

Environmental Monitoring and Assessment, Journal Year: 2022, Volume and Issue: 194(4)

Published: March 16, 2022

Language: Английский

---

ResInformer: Residual Transformer-Based Artificial Time-Series Forecasting Model for PM2.5 Concentration in Three Major Chinese Cities

Mathematics, Journal Year: 2023, Volume and Issue: 11(2), P. 476 - 476

Published: Jan. 16, 2023

Many Chinese cities have severe air pollution due to the rapid development of economy, urbanization, and industrialization. Particulate matter (PM2.5) is a significant component pollutants. It related cardiopulmonary other systemic diseases because its ability penetrate human respiratory system. Forecasting PM2.5 critical task that helps governments local authorities make necessary plans actions. Thus, in current study, we develop new deep learning approach forecast concentration three major China, Beijing, Shijiazhuang, Wuhan. The developed model based on Informer architecture, where attention distillation block improved with residual block-inspired structure from efficient networks, named ResInformer. We use quality index datasets cover 98 months collected 1 January 2014 17 February 2022 train test model. also proposed for 20 months. evaluation outcomes show ResInformer ResInformerStack perform better than original yield forecasting results. This study’s methodology easily adapted similar efforts fast computational modeling.

Language: Английский

---

An ensemble interval prediction model with change point detection and interval perturbation-based adjustment strategy: A case study of air quality

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 222, P. 119823 - 119823

Published: March 9, 2023

Language: Английский

---