Accurate prediction of PM2.5 concentration is essential for public health and environmental protection, and specifically crucial for the management of the availability of sufficient health personnel during adverse health episodes. However, its nonlinearity, variability, and complexity make this task challenging. This study proposes a long short-term memory (LSTM) weighted by K-nearest neighbor (KNN) algorithm (namely Weighted KNN-LSTM Model) that can effectively predict the PM2.5 concentration time series. Firstly, the K-nearest neighbors of each time point are sought based on the Euclidean distance within the data time range. Given that neighboring observations typically exert a more pronounced influence than distant ones in spatial processes, weights are accordingly assigned to these neighbors to quantitatively reflect their relative importance in the analysis. Subsequently, after the initial data is processed by the weighted KNN algorithm, it is reorganized and transformed into a reconstructed dataset with a size K times that of the original data. The data used for model training and the data used for evaluating the model’s prediction performance are completely independent, and the test dataset is never involved in the model training process to ensure the authenticity and reliability of the prediction performance evaluation. Then, the LSTM neural network model is trained on this new dataset to enhance its generalization ability. The experimental results show that the weighted KNN-LSTM model exhibits excellent predictive performance in predicting PM2.5 concentration. It is important to note that the dataset used to evaluate the model’s performance was strictly independent from the data used to train the model. This separation ensures that the reported accuracy reflects true predictive capability rather than mere fitting quality. The model provides a technical reference for hourly PM2.5 concentration prediction in Nanchang City, and the prediction results can be used as an auxiliary reference for regional air quality monitoring; the application of the model in heavy pollution warnings needs to be further optimized and verified by combining multi-source data such as meteorology, which provide reliable data support for the formulation of dynamic emission reduction policies.

Accurate forecasting of urban haze, especially fine particulate matter with an aerodynamic diameter of 2.5 μm or less (PM2.5), is important for air-pollution control and public-health protection. However, the strong nonstationarity of PM2.5 series, together with mixed trend–fluctuation patterns and nonlinear interactions among pollutant and meteorological variables, makes multi-step prediction challenging for conventional models. To address this issue, this study develops an artificial-intelligence-based forecasting framework that integrates double Holt linear-trend exponential smoothing decomposition with an inverted Transformer for short-term multi-step forecasting of PM2.5. The proposed framework first decomposes the input sequence into trend and seasonal-fluctuation components through Holt-based double exponential smoothing, and then uses a dual-stream inverted Transformer to learn their complementary temporal characteristics and inter-variable dependencies in parallel. Using hourly pollutant and meteorological data from Nanchang, Jiangxi Province, the proposed framework was compared with Long Short-Term Memory, Gated Recurrent Unit, Convolutional Neural Network-Long Short-Term Memory, and Time Series Transformer with Exogenous Variables. The results show that it generally achieves lower Root Mean Square Error, Mean Absolute Error, Mean Absolute Percentage Error, and Theil Inequality Coefficient, together with higher Index of Agreement, than the benchmark models. Additional analyses further confirm its accuracy and robustness. Supplementary validation across the 11 prefecture-level cities in Jiangxi Province also supports its regional applicability. Overall, the proposed framework provides an effective solution for short-term PM2.5 forecasting and province-scale air-quality early warning.

Given the limited research on high-quality development (HQD) at the county level, especially in major agricultural provinces such as Henan, this study constructs a multidimensional evaluation system tailored to local heterogeneity. Guided by the Five Development Concepts and incorporating the space of flows perspective to reveal inter-county spatial linkages, we introduce a coupling detrended fluctuation analysis (CDFA)-based assessment method to develop a temporally informed county-specific weighting mechanism. This enables the calculation of county-specific indicator weights and overcomes the limitation of traditional methods that assign fixed and uniform weights. Using panel data on 19 indicators for 102 counties in Henan Province from 2010 to 2021, we examine the spatiotemporal trends and structural heterogeneity of county-level HQD. Results indicate broad-based improvement in county-level HQD, while high- and medium-high-quality counties became more concentrated around core urban areas and the remaining weaker counties were mainly located in peripheral areas, demonstrating a clearer core-periphery spatial pattern under overall upgrading. The CDFA-based weighting results reveal significant county-level differences in the relative structural roles of economic vitality, green development, social welfare, and inter-county association, highlighting substantial heterogeneity in county-level indicator-weight profiles rather than a uniform indicator structure across the province. Comparative analysis shows that the proposed method is broadly consistent with a conventional objective weighting approach in overall classification patterns, while offering additional insight into county-level heterogeneity. The optimal parameters-based geographical detector further indicates that locational conditions, industrial foundations, and topographic constraints are important explanatory factors for the spatial differentiation of county-level HQD, and that factor interactions generally produce stronger explanatory power than single factors. Based on these findings, this study proposes differentiated development strategies for core counties, transitional counties, agricultural counties, and peripheral ecologically constrained areas, thereby providing a methodological reference for HQD assessment in spatially heterogeneous regions and supporting regional spatial governance.

With the acceleration of urbanization in China, haze pollution has become a problem that cannot be ignored. PM2.5 is one of the main components of haze, and this paper aims to find a stable and accurate prediction method for PM2.5 prediction. Combined with existing studies, BP neural network is commonly used for prediction and optimization, but its accuracy is not satisfactory due to the randomness of the initial parameters of BP neural network. In order to solve this problem, this study proposes a new type of fusion model-improved particle swarm optimized backpropagation neural network (IPSO-BP) model. In this paper, we use the BP neural network to predict the value of PM2.5, and at the same time, we use the improved particle swarm algorithm to optimize the initial parameters of the BP neural network, which makes the prediction performance improved. Taking a simulation experiment in Nanchang City as an example, the prediction accuracy is 86.76%, the correlation coefficient R2 is 0.95734, and the root-mean-square error (RMSE) is 5.2407. Compared with a single BP neural network model, the advantages of the IPSO-BP model are: (1) Asynchronous learning factor is used, particle swarm algorithm (PSO) exists two learning factors, individual learning factor c1 and population learning factor c2, the former affects the local search ability while the latter affects the global search ability. Through the iterative formula proposed in this paper, the algorithm can be made to satisfy the strong global search ability in the early stage and the strong local search ability in the later stage. (2) Adaptive inertia weights are introduced, where larger values of inertia weights mean that it is more difficult to change the direction of the particles. In the initial stage of the model, a larger inertia weight helps to improve the global search ability of the algorithm, while a smaller inertia weight helps to improve the local search ability of the algorithm as it enters the end of the search. Adaptive inertia weights are the iterative formulas proposed in the paper that make the inertia weights of the model large at the beginning and small at the end. (3) Incorporating the Levy flight search strategy, which aims to solve the shortcomings of traditional particle swarm algorithms that often fall into the suboptimal solution, it can be judged according to the evolutionary effect of the particle position, and if the particles are still unable to enter the more optimal position in many iterations, the Levy flight will be used to update the position of the particles, which is a strategy that increases the vitality of the particles. In summary, the IPSO-BP model proposed in this study has excellent predictive ability and, makes some positive contributions to the cause of air pollution prevention.

To mitigate the adverse effects of air pollution, accurate PM2.5 prediction is particularly important. It is difficult for existing models to escape the limitations attached to a single model itself. This study proposes a hybrid PM2.5 prediction model utilizing deep learning techniques, which aims to complement each other’s strengths through model fusion. The model integrates the transformer and LSTM architectures and employs parameter optimization through the particle swarm optimization (PSO) algorithm. The proposed model achieves superior performance by utilizing the gating mechanism of the LSTM model, the positional encoding and self-attention mechanism of the Transformer model, and PSO’s robust optimization capabilities. Experimental results show that the new model outperforms both the traditional LSTM model and the PSO-LSTM model in the PM2.5 prediction task, and its evaluation metrics, R2, MAE, MBE, RMSE, and MAPE, are all improved. Furthermore, the model demonstrates stable performance across different cities and various periods. This study offers a robust approach to improving the accuracy and reliability of PM2.5 forecasting.

To address the performance degradation in existing PM2.5 prediction models caused by excessive complexity, poor spatiotemporal efficiency, and suboptimal parameter optimization, we employ stacking ensemble learning for feature weighting analysis and integrate the ant colony optimization (ACO) algorithm for model parameter optimization. Combining meteorological and collaborative pollutant data, a model (namely the stacking-ACO-LSTM model) with a much shorter consuming time than that of only long short-term memory (LSTM) networks suitable for PM2.5 concentration prediction is established. It can effectively filter out feature variables with higher weights, thereby reducing the predictive power of the model. The prediction of hourly PM2.5 concentration of the model is trained and tested using real-time monitoring data in Nanchang City from 2017 to 2019. The results show that the established stacking-ACO-LSTM model has high accuracy in predicting PM2.5 concentration, and compared to the same model without considering time and space efficiency and defective parameter optimization, the mean square error (MSE) decreases by about 99.88%, and the coefficient of determination (R2) increases by about 2.39%. This study provides a new idea for predicting PM2.5 concentration in cities.

Despite recent progress in controlling PM2.5 pollution, identifying its spatiotemporal distribution characteristics, key pollution sources, and regional transmission patterns remains a critical task in addressing PM2.5 pollution in Jiangxi Province. However, accurately identifying the seasonal community divisions, core contribution sources, and transmission hub cities for PM2.5 pollution in this region still poses significant challenges. This study employs the optimized community detection method of Community Detection by Motif-aware Label Propagation (MWLP) to explore the seasonal community segmentation of PM2.5 pollution in Jiangxi Province and the key pollution sources within these communities. Additionally, the Hungarian algorithm is used to identify the hub cities involved in the PM2.5 pollution transmission process. The results show that the PM2.5 pollution network in Jiangxi Province is segmented into 4, 6, 4, and 3 communities during spring, summer, autumn, and winter, respectively; the key pollution sources for each season are Nanchang, Xinyu, Ganzhou, and Shangrao; Pingxiang, Yingtan, Ganzhou, Shangrao, Jiujiang, and Jingdezhen; Jingdezhen, Pingxiang, Ganzhou, and Nanchang; and Pingxiang, Ganzhou, and Nanchang, respectively. Additionally, the core cities driving long-distance PM2.5 transmission are primarily concentrated in central Jiangxi Province, while cities with lesser transmission roles are distributed along the periphery. Further analysis indicates that PM2.5 pollution in Jiangxi Province exhibits significant spatial variability and seasonal changes, with clear spatiotemporal correlations between polluted cities.

The problem of haze pollution, mainly caused by delicate particulate matter (PM2.5), is becoming increasingly severe. The coverage of haze weather is constantly expanding, leading to increasing pressure on the urban atmospheric environment. At the same time, it seriously restricts the sustainable development of China's economy and damages the people's health. In response to the problem of poor prediction accuracy of delicate particulate matter (PM2.5) concentration, this work proposes a PM2.5 concentration prediction model based on Whale Optimization Algorithm (WOA) and Attention Mechanism (AM) optimized Bidirectional Long Short Term Memory Network (BiLSTM), namely the WOA-BiLSTM-ATT model. This model can effectively alleviate the problem of gradient vanishing, better adapt to multiple learning tasks and further enable AM to allocate weights to features to achieve the accurate prediction of PM2.5. The empirical results indicate that the stability and prediction accuracy of the WOA-BiLSTM-ATT model are superior to other models.

In recent years, with the increase of pollution control and ecological environmental protection, the quality of the ecological environment in China has continued to improve, but the problem of haze pollution with PM2.5 characteristics should not be underestimated. Daily PM2.5 data of 11 prefecture- level cities in Jiangxi Province from January 1 to December 31, 2022 were selected, and the spatial spillover effects of haze pollution from a network perspective. Main findings are as follows: (1) The spatial correlation network of haze pollution for the 11 prefecture-level cities in Jiangxi Province is constructed using the Granger causality test. (2) the network characteristics are get by utilizing social network analysis. Partial result Nanchang emerges as the most significant media city in this network. (3) The network can divided into four distinct sectors, namely a sector representing net benefits, a sector indicating net overflow, a bilateral overflow sector and broker sector. These four sectors have distinct geographical distributional characteristics. (4) The spatial correlation network of haze pollution in the 11 prefecture-level cities was found to be correlated with the level of urbanization difference by QAP regression analysis, but the single urban population proportions did not pass the significance level test of 10%. Based on the analysis above, this paper considers that it is an important strategy for the Jiangxi Province haze control to establish a regional collaborative prevention and control mechanism, implement the differentiated regional governance strategy, and achieve the high-quality development of the whole province.

With the rapid expansion of industrialization and urbanization, fine Particulate Matter (PM2.5) pollution has escalated into a major global environmental crisis. This pollution severely affects human health and ecosystem stability. Accurately predicting PM2.5 levels is essential. However, air quality forecasting currently faces challenges in processing vast data and enhancing model accuracy. Deep learning models are widely applied for their superior learning and fitting abilities in haze prediction. Yet, they are limited by optimization challenges, long training periods, high data quality needs, and a tendency towards overfitting. Furthermore, the complex internal structures and mechanisms of these models complicate the understanding of haze formation. In contrast, traditional Support Vector Regression (SVR) methods perform well with complex non-linear data but struggle with increased data volumes. To address this, we developed CUDA-based code to optimize SVR algorithm efficiency. We also combined SVR with Genetic Algorithms (GA), Sparrow Search Algorithm (SSA), and Particle Swarm Optimization (PSO) to identify the optimal haze prediction model. Our results demonstrate that the model combining intelligent algorithms with Central Processing Unit-raphics Processing Unit (CPU-GPU) heterogeneous parallel computing significantly outpaces the PSO-SVR model in training speed. It achieves a computation time that is 6.21-35.34 times faster. Compared to other models, the Particle Swarm Optimization-Central Processing Unit-Graphics Processing Unit-Support Vector Regression (PSO-CPU-GPU-SVR) model stands out in haze prediction, offering substantial speed improvements and enhanced stability and reliability while maintaining high accuracy. This breakthrough not only advances the efficiency and accuracy of haze prediction but also provides valuable insights for real-time air quality monitoring and decision-making.

The concentration prediction of PM2.5 plays a vital role in controlling the air and improving the environment. This paper proposes a prediction model (namely EEMD-ALSTM) based on Ensemble Empirical Mode Decomposition (EEMD), Attention Mechanism and Long Short-Term Memory network (LSTM). Through the combination of decomposition and LSTM, attention mechanism is introduced to realize the prediction of PM2.5 concentration. The advantage of EEMD-ALSTM model is that it decomposes and combines the original data using the method of ensemble empirical mode decomposition, reduces the high nonlinearity of the original data, and Specially reintroduction the attention mechanism, which enhances the extraction and retention of data features by the model. Through experimental comparison, it was found that the EEMD-ALSTM model reduced its MAE and RMSE by about 15% while maintaining the same R2 correlation coefficient, and the stability of the model in the prediction process was also improved significantly.

With the rapid development of industry, haze pollution has become an urgent environmental problem. This study innovatively utilizes network-based methods to investigate the spatial correlation effects of haze pollution transmission between urban clusters in the Yangtze River Economic Belt. A spatial correlation network of haze pollution in the Yangtze River Economic Belt was constructed using 328 urban meteorological data collection points as research samples, and its structural characteristics were examined. Main findings are as follows: (1) The spatial correlation network of PM2.5 in the Yangtze River Economic Belt urban agglomeration exhibits typical network structural characteristics: obvious spatial correlation within the network. (2) Chengdu, Chongqing, Wuhan, Nanchang, Yichang, Changsha and Yueyang are located at the center of the spatial network. They have more receiving and sending relationships. (3) 36 cities can be divided into four types: bilateral overflow, net beneficiary, net overflow and broker. Each type has different functional characteristics and linkage effects in the network. (4) Haze pollution positively correlates with the city’s synergistic development capacity and urbanization rate, the higher the city’s development level and the higher the Urbanization rate, the stronger its haze pollution capacity. This study provides new insights into the study of the spatial correlation and impact of haze pollution.

With the continuous development of embodied cognition theory and virtual reality (VR) technology, its application in teaching has been paid more and more attention by researchers. However, there are still few practical studies on the combination of VR technology and embodied learning. Starting from literature research, the paper analyzes the factors that affect the embodied learning in the VR environment; then, put forward the design principles under the support of VR based on the previous studies; finally, constructs the embodied learning design framework, combined with the two parts of study. The research chooses a course (namely Discrete Mathematics) as an example of embodied learning design case, which organizes the application facing college students in Computer Science and Technology major of Nanchang Institute of Technology. The study found that the body is indeed involved in the cognitive process, and the embodied learning based on VR can effectively improve the learning achievement and has a positive impact on learners’ learning participation and interest in learning during the experiment.

In order to obtain the pattern of variation of PM2.5concentrations in the atmosphere in Nanchang City, we build a Support Vector Regression(SVR) with modified Whale Optimization Algorithm(WOA) hybrid model (namely mWOA-SVR model) that can predict the PM2.5concentration. Firstly, according to the Pearson correlation coefficient (PCC) method to examine the dynamic relationship between air pollutants and meteorological factors together with them, PM10, SO2and CO were selected as air pollutant concentration characteristics, while daily maximum and minimum temperatures, and wind power levels were selected as meteorological characteristics; then, using modified WOA algorithm for parameter selection of SVR model, four sets of better parameter combinations were found; finally, the mWOA-SVR model was built by the four sets parameters to predict PM2.5concentration. The results show that the prediction accuracy of mixed mWOA-SVR model with pollutant concentration plus weather factors as the feature was higher than single pollutant concentration.

Particulate matter (PM) is defined by the Texas Commission on Environmental Quality (TCEQ) as “a mixture of solid particles and liquid droplets found in the air”. These particles vary widely in size. Those particles that are less than 2.5 μm in aerodynamic diameter are known as Particulate Matter 2.5 or PM2.5. Urban haze pollution represented by PM2.5 is becoming serious, so air pollution monitoring is very important. However, due to high cost, the number of air monitoring stations is limited. Our work focuses on integrating multi-source heterogeneous data of Nanchang, China, which includes Taxi track, human mobility, Road networks, Points of Interest (POIs), Meteorology (e.g., temperature, dew point, humidity, wind speed, wind direction, atmospheric pressure, weather activity, weather conditions) and PM2.5 forecast data of air monitoring stations. This research presents an innovative approach to air quality prediction by integrating the above data sets from various sources and utilizing diverse architectures in Nanchang City, China. So for that, semi-supervised learning techniques will be used, namely collaborative training algorithm Co-Training (Co-T), who further adjusting algorithm Tri-Training (Tri-T). The objective is to accurately estimate haze pollution by integrating and using these multi-source heterogeneous data. We achieved this for the first time by employing a semi-supervised co-training strategy to accurately estimate pollution levels after applying the U-air system to environmental data. In particular, the algorithm of U-Air system is reproduced on these highly diverse heterogeneous data of Nanchang City, and the semi-supervised learning Co-T and Tri-T are used to conduct more detailed urban haze pollution prediction. Compared with Co-T, which train time classifier (TC) and subspace classifier (SC) respectively from the separated spatio-temporal perspective, the Tri-T is more accurate with a and faster because of its testing accuracy up to 85.62 %. The forecast results also present the potential of the city multi-source heterogeneous data and the effectiveness of the semi-supervised learning. We hope that this synthesis will motivate atmospheric environmental officials, scientists, and environmentalists in China to explore machine learning technology for controlling the discharge of pollutants and environmental management.

Along with the progress of social civilization, human demand for production and life is also increasing. As a factor that affects people’s health all the time, the air environment has caused extensive research by scholars on the prediction of air environment quality. However, due to the different correlations of various factors affecting the prediction, the prediction results are affected. Thus, this study uses statistical methods to analyze the correlation of different factors in the prediction and uses the long-short-term memory network on basis of the attention mechanism to make predictions. Finally, we tested with the air data in Beijing and calculated that the accuracy of the model was 87.7%. The results show that the long-short-term memory network with the attention mechanism can accurately predict PM2.5, helping us to monitor and control air pollution better in cities.

The ACLPSO algorithm, based on dimension specification intervals, adaptively tunes maximum velocity, inertia weight, acceleration coefficient, and learning probability for each dimension. It has demonstrated excellent performance in benchmark function tests involving both single-modal and multi-modal functions. However, to obtain the global optimal or approximate optimal solutions for all executed benchmark functions, it is necessary to manually set appropriate values for the maximum velocity coefficient s and the learning probability coefficient v during the operation of the ACLPSO algorithm. This study introduces an automatic approach to assign values to s and v, relying on function iteration count and test function fitness convergence. This enhancement enables the improved ACLPSO algorithm to directly derive the global optimal or approximate optimal solutions for all benchmark functions, eliminating the need for manual parameter tuning.

Abstract Human life and development are affected by air quality, which means it is significant to accurately estimate air quality utilizing science. Considering PM 2.5 is an essential component of air pollution, accurate PM 2.5 prediction is crucial to maintaining air quality, preserving human wellness, and advancing urbanization's sustainability and scientific evolution. In this investigation, we provide a Particle Swarm Optimization (PSO) and Central Processing Unit-Graphics Processing Unit (CPU-GPU) heterogeneous parallel Support vector regression (SVR) method to forecast the airborne PM 2.5 content. The approach enhances computational effectiveness, accelerates solution times, and satisfies debugging and application requirements in realistic circumstances where the size and dimensionality of the pertinent data are increasing. by contrasting it with conventional support vector regression methods and conducting analysis of variance on various data sets. The simulation results demonstrate a rise of over fifty per cent in prediction speed with a high degree of accuracy for the CPU-GPU heterogeneous parallel support vector regression (PSO-CPU-GPU-SVR) model.

The atmospheric environment is of great importance to human health. However, its influencing factors are complex and variable. An efficient technique is required to more precisely estimate PM2.5 concentration values. In this paper, an enhanced Sparrow Search Algorithm (LASSA)-optimized Light Gradient Boosting Machine (LightGBM) is proposed for PM2.5 concentration prediction. This approach can provide accurate predictions while also reducing potential losses resulting from unexpected events. LightGBM is regarded as an outstanding machine learning approach; however, it includes hyperparameters that must be optimally mixed in order to achieve the desired results. We update the Sparrow Search Algorithm (SSA) and utilize it to identify the optimal combination of the most crucial parameters, using cross-validation to increase the reliability. Using limited air quality data and meteorological data as inputs, PM2.5 concentration values were predicted. The LASSA-LGB’s output was compared to normal LGB, SSA-LGB and ISSA-LGB. The findings demonstrate that LASSA-LGB outperforms the other models in terms of prediction accuracy. The RMSE and MAPE error indices were lowered from 3% to 16%. The concordance correlation coefficient is not less than 0.91, and the R2 reached 0.96. This indicates that the proposed model has potential advantages in the field of PM2.5 concentration prediction.

Abstract Human life and development are affected by air quality, which means it is significant to accurately estimate air quality utilizing science. Considering PM2.5 is an essential component of air pollution, accurate PM2.5 prediction is crucial to maintaining air quality, preserving human wellness, and advancing urbanization's sustainability and scientific evolution. In this investigation, we provide a Particle Swarm Optimization (PSO) and Central Processing Unit-Graphics Processing Unit (CPU-GPU) heterogeneous parallel Support vector regression (SVR) method to forecast the airborne PM2.5 content. The approach enhances computational effectiveness, accelerates solution times, and satisfies debugging and application requirements in realistic circumstances where the size and dimensionality of the pertinent data are increasing. by contrasting it with conventional support vector regression methods and conducting analysis of variance on various data sets. The simulation results demonstrate a rise of over fifty per cent in prediction speed with a high degree of accuracy for the CPU-GPU heterogeneous parallel support vector regression (PSO-CPU-GPU-SVR) model.