Root Mean Square Error Research Articles

Reconstruction of PM2.5 Concentrations in East Asia on the Basis of a Wide–Deep Ensemble Machine Learning Framework and Estimation of the Potential Exposure Level from 1981 to 2020

Satellite observations are widely used to estimate the concentrations of surface air pollutants, but the temporal coverage of these datasets is relatively short. To overcome this limitation, we propose a wide–deep ensemble machine learning framework to reconstruct the fine particulate matter (PM2.5) dataset of east Asia (EA) over the past four decades (1981–2020). The results indicate that the framework effectively leveraged the advantages of satellite observations (higher accuracy) and model-based estimations (longer temporal coverage) of surface air pollutants. The reconstructed PM2.5 concentrations agreed well with the ground measurements, with coefficient of determination (R2) and root-mean-square error (RMSE) values of 0.99 and 1.38 μg·m−3, respectively, which outperformed the satellite-based PM2.5 estimates. As more ground measurements were incorporated into the model for training, the average RMSE in Japan and the Korean Peninsula decreased to 0.83 and 1.50 μg·m−3, respectively. Simultaneously, on the basis of the reconstructed datasets, we investigated the exposure level to PM2.5 in EA from 1981 to 2020. Since 2000, the increase in anthropogenic emissions has substantially worsened the air quality in EA, and nearly 50% of the population resided in areas where the annual average PM2.5 concentrations exceeded 50 μg·m−3 from 2009–2010. Despite the implementation of various mitigation strategies by local authorities to lower the ambient PM2.5 concentrations, the entire exposure level in EA is still implausible to meet the WHO air quality guidelines. In addition, population aging and climate change have the potential to increase PM2.5 exposure risk in the future. For policy-makers in EA, it is essential to consider the effects of these factors and develop more effective mitigation strategies that aim to lessen the health impact associated with PM2.5 exposure.

A novel ensemble machine learning exposure model system for ground-level ozone at the national scale: A case of mainland China from 2013 to 2020

In epidemiological research, accurate estimation of historical ground-level ozone (O3) concentrations with enhanced spatiotemporal resolution is crucial for effective exposure assessment. The current state-of-the-art for estimating air pollutant concentrations is a two-stage ensemble method that integrates outputs from multiple machine learning algorithms. Despite its effectiveness, opportunities exist to refine this approach for more precise O3 estimation. In this study, we propose an enhanced ensemble method that incorporates four key strategies. First, we employ high-resolution spatiotemporal predictors derived from prior machine learning studies for refined secondary learning. Second, we use sophisticated algorithms, including categorical gradient boosting, deep neural network, random forest, stochastic variable Gaussian process, transformer, and a combination of convolutional neural network and long short-term memory neural network, as sublearners to enhance learning capabilities. Third, we spatiotemporally split the sample set and then train submodels separately on each subset to eliminate the unobserved spatiotemporal heterogeneity. Finally, we apply a complex machine learning algorithm, rather than the generalized additive model, for integrating sublearner predictions, enabling the capture of intricate nonlinear relationships beyond basic spatiotemporal linear weights. To validate these improvements, we estimated daily maximum 8-h moving average O3 concentrations ([O3]MDA8) across Chinese mainland from 2013 to 2020 at a 1 km spatial resolution. The proposed method demonstrated notable accuracy, achieving an out-of-station determination coefficient (R2) of 0.943 and a root-mean-square error (RMSE) of 10.197 μg/m3. This performance marks a nearly 15% improvement over the best existing Chinese O3 exposure model based on a single algorithm and also surpasses previous studies utilizing traditional ensemble methods for other air pollutants. Our enhanced ensemble approach significantly bolsters the reliability and robustness of future environmental epidemiological studies by further mitigating “misclassification” errors.

A StarGAN and transformer-based hybrid classification-regression model for multi-institution VMAT patient-specific quality assurance.

The field of artificial intelligence (AI)-based patient-specific quality assurance (PSQA) for volumetric modulated arc therapy (VMAT) faces challenges in terms of developing general models across institutions due to the prevalence of multi-institution data collection and multivariate heterogeneity. Building a general model that is capable of handling diverse multi-institution data is critical for enabling large-scale integration and analysis. This study aims to develop a star generative adversarial network (StarGAN) and transformer-based hybrid classification-regression PSQA framework to address unification of heterogeneous data from different institutions. A StarGAN and transformer-based hybrid classification-regression model was developed as a general PSQA framework to predict gamma passing rates (GPRs) and classify quality assurance (QA) results as "Pass" or "Fail" at multiple institutions. A total of 1815 VMAT plans were collected from eight institutions to develop the general PSQA framework and perform clinical commissioning and implementation. Among them, 20 independent clinical plans from each of eight institutions, for a total of 160 plans, were used for the clinical commissioning, and 205 new clinical plans from eight institutions were used for clinical implementation. For the 3%/3, 3%/2, and 2%/2mm gamma criteria, the sensitivity of the proposed PSQA framework with pretraining was 90.13%, 92.03%, and 95.84%, respectively, while the specificity was 76.01%, 76.12%, and 85.34%, respectively. The mean absolute errors (MAEs) of the proposed PSQA framework with pretraining were 1.36%, 2.37%, and 3.96%, respectively, while the root-mean-square errors (RMSEs) were 2.31%, 3.89%, and 5.17%, respectively. The results demonstrated visible improvement at multiple institutions. For clinical commissioning, the deviations between the predicted and measured results were all within 3% for 3%/3 and 3%/2mm at eight institutions. For clinical implementation, all failure plans were correctly identified by the proposed PSQA framework. The general PSQA framework enables diverse clinical data sources to be handled to achieve enhanced model performance and generalizability, and provides a solution to the unification of heterogeneous data from different institutions to construct robust QA models. This approach can be clinically deployed for VMAT QA.

Estimating monthly NO2, O3, and SO2 concentrations via an ensemble three-stage procedure with downscaled satellite remote sensing data and ground measurements

The gaseous air pollutants nitrogen dioxide (NO2), ozone (O3), and sulfur dioxide (SO2), in addition to particulate matter (PM2.5), are known to be associated with many adverse health effects. However, exposure estimates may not be available in rural or mountainous areas without monitoring stations. In this study, we retrieved satellite remote sensing data for NO2, O3, and SO2 from the Ozone Monitoring Instrument (OMI) L3 products. Together with ground measurements (air monitoring stations and meteorological and land use data), we estimated the monthly NO2, O3, and SO2 concentrations with a spatial resolution of 3×3 km across Taiwan from 2005 to 2019. A three-stage estimation procedure was utilized: in Stage 1, an ensemble generalized additive model (GAM) and machine learning method were used to determine the spatiotemporal variations; in Stage 2, the remote sensing data were downscaled; and in Stage 3, the downscaled concentrations were reused in the Stage 1 procedure for fine-tuning estimations. We obtained overall leave-one-out cross-validation (LOOCV) R2 values ranging from 0.927-0.950, 0.704-0.721, and 0.601-0.716, and root-mean-square-errors (RMSEs) ranging from 1.59-2.28, 3.81-4.18, and 0.67-1.32 ppb for NO2, O3, and SO2, respectively, via the random forest procedure. The annual NO2 and SO2 concentrations greatly improved from 2005-2019, especially in the western residential area of Taiwan. However, despite these improvements in air quality, the annual O3 concentrations tended to increase from 2015–2019. This might be due to the complex mixtures of precursors (e.g., NO2), atmospheric circulation, barriers of the Central Mountain Ridge, and increasing ground temperatures over the past decade. The proposed multistage estimation procedure performed well over the whole island with complex terrain and topography. The study outcomes may provide epidemiological information for long-term ambient exposure estimates and guidance for future administrative policies.

Fast non-destructive measurement of intramuscular fat in Australian beef and lamb using nuclear magnetic resonance (NMR) technologies

Nuclear magnetic resonance (NMR) is an excellent technique for non-destructive analysis of meat because it has high accuracy, a linear response, and insignificant drift over time, which removes the need for recalibration. Furthermore, single-side NMR devices have open geometries that enable measurements of subsections of larger samples without taking sub-samples. Here we demonstrated long-term reproducibility in a benchtop device and the utility of a single-sided NMR device. We validated long-term reproducibility of NMR measurements of lamb intramuscular fat (IMF) in a commercial processor boning room years after the original model was created. It was hypothesised that the NMR IMF% model would retain precision and accuracy on independent validation. The root mean squared (RMS) error of prediction of lamb IMF was 0.79 %. The R2 between reference measurements, predicted IMF% was 0.74, the slope of the chemical IMF% vs NMR predictions was 0.989, and the bias was 0.53 % IMF. In the second example, we showed that IMF% measurements of high value beef striploins could be measured off a commercial processing belt and returned without damaging the product. It was hypothesised that a commercial prototype single-sided NMR system would predict IMF% in beef M. longissimus thoracis et lumborum. Here the RMS error of the correlation was 1.58 % IMF and R2 was 0.97. We believe the long-term stability, high accuracy, and nondestructive nature make unilateral NMR devices ideal for applications in the red meat industry where intramuscular fat contributes to product value.

A framework for generating recommendations based on trust in an informal e-learning environment

Rapid advancement in information technology promotes the growth of new online learning communities in an e-learning environment that overloads information and data sharing. When a new learner asks a question, how a system recommends the answer is the problem of the learner’s cold start. In this article, our contributions are: (i) We proposed a Trust-aware Deep Neural Recommendation (TDNR) framework that addresses learner cold-start issues in informal e-learning by modeling complex nonlinear relationships. (ii) We utilized latent Dirichlet allocation for tag modeling, assigning tag categories to newly posted questions and ranking experts related to specific tags for active questioners based on hub and authority scores. (iii) We enhanced recommendation accuracy in the TDNR model by introducing a degree of trust between questioners and responders. (iv) We incorporated the questioner-responder relational graph, derived from structural preference information, into our proposed model. We evaluated the proposed model on the Stack Overflow dataset using mean absolute precision (MAP), root mean squared error (RMSE), and F-measure metrics. Our significant findings are that TDNR is a hybrid approach that provides more accurate recommendations compared to rating-based and social-trust-based approaches, the proposed model can facilitate the formation of informal e-learning communities, and experiments show that TDNR outperforms the competing methods by an improved margin. The model’s robustness, demonstrated by superior MAE, RMSE, and F-measure metrics, makes it a reliable solution for addressing information overload and user sparsity in Stack Overflow. By accurately modeling complex relationships and incorporating trust degrees, TDNR provides more relevant and personalized recommendations, even in cold-start scenarios. This enhances user experience by facilitating the formation of supportive learning communities and ensuring new learners receive accurate recommendations.

Integrating Twitter Sentiment with ML Models for Enhanced Stock Forecasting

This research investigates the integration of Twitter sentiment analysis with machine learning models, specifically Long Short-Term Memory (LSTM), Facebook Prophet, and Neural Prophet, to enhance the accuracy of stock market predictions. While these models are accurate in capturing long-term trends and seasonality, they often fail when exploring the short-term volatility in stock prices. By using real-time social media sentiments, we aim to improve short-term prediction accuracy. Sentiment data from Twitter, extracted and classified using NLP techniques, was integrated with the models to predict stock prices for Apple Inc. (AAPL), Paramount Global (PARA), and BP plc (BP) over the period 2016-2024. The study compares the performance of the models using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). Results show that Neural Prophet stood the best in terms of its accuracy and efficiency in stock predictions when compared to LSTM and Facebook Prophet, reducing RMSE by 25% and MAE by 20%. While the integration of sentiment analysis improved prediction accuracy, limitations remain in predicting the scale of sudden market changes. The findings highlight the potential of sentiment analysis in stock prediction but also call for further model enhancements to fully address short-term market volatility.

Analysis and Prediction of Path Loss for Mobile Communication Lines in Nasarawa State Using Propagation Models

The design of future-generation mobile communication systems depends critically on the path loss prediction methods and their suitability to various signal propagation regions. Even though 5G has been launched in Nigeria, its deployment still faces significant challenges especially in the suburban, and non-urban regions which still depends on 4G. Many operators are still in the process of upgrading their existing 4G networks to support network coverage and higher speed data throughput. Due to the unique mix of Nasarawa State with diverse terrains, in this study, path loss prediction for the key telecommunication networks in Nasarawa State was carried out using Signal Strength Info App. Data was collected for a period of eight weeks (August, 2nd to September, 4th 2023) focusing on 4G LTE networks operating within the 1800 MHz frequency band. Analysis of path loss using empirical models such as Hata, Egli, and COST-231 Hata models were compared with the measured path loss in different terrains of urban, sub-urban, and non-urban environments. The Root Mean Square Error (RMSE) analysis was carried out between empirical and measured path losses. Results shows that, the order of mean predicted path loss at 5km was free space model (BN>CN>DN.AN), Hata model (DN>BN>CN>AN), Cos-231 Hata model (CN>AN>BN>DN), and Egli model (AN>BN>DN>CN). Hate model predicted the highest path loss values while Egli model predicted the lowest values. Overall, Egli model is found to be the most reliable and suitable path loss prediction model for entire Nasarawa State with RMSE value 5.99dBm, 2.90dBm and 3.14dBm for Nasarawa, Keffi and Akwanga LGA respectively. However, Cost-231 Hata model with RMSE value of 5.71dBm is suitable for path loss prediction in Karu due to its irregular terrain. The findings of this study are significant for practical applications in network planning, base station coverage, frequency allocation, and signal strength management, ensuring optimal mobile network performance across various terrains.

Utilizing VSWIR spectroscopy for macronutrient and micronutrient profiling in winter wheat

This study explores the use of leaf-level visible-to-shortwave infrared (VSWIR) reflectance observations and partial least squares regression (PLSR) to predict foliar concentrations of macronutrients (nitrogen, phosphorus, potassium, calcium, magnesium, and sulfur), micronutrients (boron, copper, iron, manganese, zinc, molybdenum, aluminum, and sodium), and moisture content in winter wheat. A total of 360 fresh wheat leaf samples were collected from a wheat breeding population over two growing seasons. These leaf samples were used to collect VSWIR reflectance observations across a spectral range spanning 350 to 2,500 nm. These samples were then processed for nutrient composition to allow for the examination of the ability of reflectance to accurately model diverse chemical components in wheat foliage. Models for each nutrient were developed using a rigorous cross-validation methodology in conjunction with three distinct component selection methods to explore the trade-offs between model complexity and performance in the final models. We examined absolute minimum predicted residual error sum of squares (PRESS), backward iteration over PRESS, and Van der Voet’s randomized t-test as component selection methods. In addition to contrasting component selection methods for each leaf trait, the importance of spectral regions through variable importance in projection scores was also examined. In general, the backward iteration method provided strong model performance while reducing model complexity relative to the other selection methods, yielding R2 [relative percent difference (RPD), root mean squared error (RMSE)] values in the validation dataset of 0.84 (2.45, 6.91), 0.75 (1.97, 18.67), 0.78 (2.13, 16.49), 0.66 (1.71, 17.13), 0.68 (1.75, 14.51), 0.66 (1.72, 12.29), and 0.84 (2.46, 2.20) for nitrogen, calcium, magnesium, sulfur, iron, zinc, and moisture content on a wet basis, respectively. These model results demonstrate that VSWIR reflectance in combination with modern statistical modeling techniques provides a powerful high throughput method for the quantification of a wide range of foliar nutrient contents in wheat crops. This work has the potential to advance rapid, precise, and nondestructive field assessments of nutrient contents and deficiencies for precision agricultural management and to advance breeding program assessments.

ANALISIS POLA DISTRIBUSI CURAH HUJAN LEBAT DOMINAN DI DAERAH ALIRAN SUNGAI (DAS) CITANDUY HULU

Rainfall plays an important role in hydrological and environmental planning. Meteorological conditions affect rainfall through its intensity, amount, and duration. However, heavy rainfall can cause significant damage and loss. Research was conducted in the Upper Citanduy watershed, West Java, using rainfall data from 8 stations over a range of 5–14 years. The study took a heavy rainfall event threshold of ≥ 50 mm in 24 hours. The research aims to analyze the dominant heavy rainfall patterns in the Upper Citandyu watershed. Empirical analysis used the Mononobe Method and Alternating Block Method (ABM), while rainfall distribution was analyzed with Inverse Distance Weighted (IDW) using Geographic Information Systems. The dominant heavy rainfall duration in the Upper Citanduy watershed occurred over 4–6 hours. Rainfall distribution analysis showed a pattern that was not influenced by the elevation of the rainfall station. The analysis results show that the Mononobe Method is closer to the dominant observed heavy rainfall distribution for 4 hours, while ABM is closer to the durations of 5 and 6 hours. The suitability of the distribution pattern was measured using Correlation (r), Root Mean Squared Error (RMSE), and Nash-Sutcliffe Efficiency (NSE) criteria, with Mononobe providing better results than ABM method. This research contributes to the understanding of rainfall patterns that can form the basis for analyzing designed rainfall intensity as a step in flood disaster mitigation in the Upper Citanduy watershed.Keywords: heavy rainfall, rainfall distribution pattern, rainfall intensity, upper Citanduy watershed, geographic information system

Artificial Intelligence-Based Precipitation Estimation Method Using Fengyun-4B Satellite Data

This paper proposes a novel precipitation estimation method based on FY-4B meteorological satellite data (FY-4B_AI). This method facilitates the spatiotemporal matching of 125 features derived from the multi-temporal and multi-channel data of the FY-4B satellite with precipitation data at stations. Subsequently, a precipitation model was constructed using the light gradient boosting machine (LGBM) algorithm. A comparative analysis of FY-4B_AI and GPM/IMERG-L products for over 450 million station cases throughout 2023 revealed the following: (1) The results demonstrate that FY-4B_AI is more accurate than GPM/IMERG-L. Six of the eight evaluation indices exhibit superior performance for FY-4B_AI in comparison to GPM/IMERG-L. These indices include the mean absolute error (MAE), root mean square error (RMSE), relative error (RE), correlation coefficient (CC), probability of detection (POD), and critical success index (CSI). As for the MAE, the results are 1.67 (FY-4B_AI) and 1.92 (GPM/IMERG-L), respectively. The RMSEs are 3.68 and 4.07, respectively. The REs are 17.72% and 26.28%, respectively. The CCs are 0.44 and 0.36, respectively. The PODs are 61.84% and 47.31%, respectively. The CSIs are 0.30 and 0.27, respectively. However, with regard to the mean errors (MEs) and false alarm rates (FARs), FY-4B_AI (−0.88 and 62.85%, respectively) displays a slight degree of inferiority in comparison to GPM/IMERG-L (−0.80 and 62.21%, respectively). (2) An evaluation of two strong weather events to represent the spatial distribution of precipitation in different climatic zones revealed that both FY-4B_AI and GPM/IMERG-L are equally capable of accurately representing these phenomena, irrespective of whether the region in question is humid, as is the case in the southeast, or dry, as is the case in the northwest.

Estimation of Forest Growing Stock Volume with Synthetic Aperture Radar: A Comparison of Model-Fitting Methods

Satellite-based estimation of forest variables including forest biomass relies on model-based approaches since forest biomass cannot be directly measured from space. Such models require ground reference data to adapt to the local forest structure and acquired satellite data. For wide-area mapping, such reference data are too sparse to train the biomass retrieval model and approaches for calibrating that are independent from training data are sought. In this study, we compare the performance of one such calibration approach with the traditional regression modelling using reference measurements. The performance was evaluated at four sites representative of the major forest biomes in Europe focusing on growing stock volume (GSV) prediction from time series of C-band Sentinel-1 and Advanced Land Observing Satellite Phased Array L-band Synthetic Aperture Radar (ALOS-2 PALSAR-2) backscatter measurements. The retrieval model was based on a Water Cloud Model (WCM) and integrated two forest structural functions. The WCM trained with plot inventory GSV values or calibrated with the aid of auxiliary data products correctly reproduced the trend between SAR backscatter and GSV measurements across all sites. The WCM-predicted backscatter was within the range of measurements for a given GSV level with average model residuals being smaller than the range of the observations. The accuracy of the GSV estimated with the calibrated WCM was close to the accuracy obtained with the trained WCM. The difference in terms of root mean square error (RMSE) was less than 5% units. This study demonstrates that it is possible to predict biomass without providing reference measurements for model training provided that the modelling scheme is physically based and the calibration is well set and understood.

Spatiotemporal wind speed forecasting using conditional local convolution and multidimensional meteorology features

Wind speed prediction is crucial for precisely wind power forecasting and reduced maintenance costs. Highland regions, which possess a considerable wind potential, present complex meteorological conditions, making wind speed prediction challenging. Traditional weather forecasting relies on complex statistical methods and extensive prior knowledge. While recent deep learning models have improved prediction accuracy, they often assume uniform influence weight structure, limiting model effectiveness. This study introduces an enhanced Conditional Local Convolution Recurrent Network (CLCRN) model to improve spatiotemporal wind speed forecasting using multidimensional meteorological inputs such as temperature, pressure, and dew point, alongside wind components. This model addresses uniform influence model weight issue by redesigning convolution kernels to better capture local meteorological features and integrating multiple influencing factors. Our model consistently achieves lower Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) values across various prediction intervals (3, 6, 9, and 12 h) compared to other models, supported by the meteorological station data from 2019 to 2021. Furthermore, the spatial distribution of the local convolution weights aligns with local wind velocity patterns in Inner Mongolia, enhancing model interpretability. These results demonstrate potential for practical applications in renewable energy planning and wind dynamics simulation.

Evaluation of Machine Learning Application on the Prediction of Particulate Matter Concentrations in Small/Medium-Sized City

Objectives:In this study, Machine Learning (ML) algorithms were evaluated to predict the concentration of particulate matter (PM10 and PM2.5) using air quality and meteorological data in small/medium-sized city.Methods:ML models, including Multiple Linear Regression (MLR), Decision Tree Regression (DTR), Random Forest (RF), Extreme Gradient Boosting (XGB), Light Gradient Boosting Machine (LGB), were used to predict PM10 and PM2.5 concentrations. Five air quality variables, including NO2, SO2, CO, PM10 and PM2.5, and seven meteorological variables, including temperature, humidity, vapor pressure, wind speed, precipitation, local atmospheric pressure, and sea-level atmosphere pressure, were collected from three air quality monitoring stations and one meteorological observatory from 2017 to 2022. A total of 52,583 sets of data were used for ML. The prediction accuracies of the applied ML models were evaluated using the Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Coefficient of determination (R2).Results and Discussion:Higher ML performance was obtained when using the data including PM10 and PM2.5 compared to the data excluding these variables. Among five different ML models, the XGB model showed the highest accuracy in predicting PM10 and PM2.5 one hour in the future. However, poorer performance was obtained as the predicted period increased from one hour to 72 hours.Conclusion:The application of ML algorithms for the short-term prediction of PM10 and PM2.5 was successful in this study. However, more input variables and deep learning algorithms are needed for long-term prediction of PM10 and PM2.5.

Deep learning models for hepatitis E incidence prediction leveraging Baidu index.

Infectious diseases are major medical and social challenges of the 21st century. Accurately predicting incidence is of great significance for public health organizations to prevent the spread of diseases. Internet search engine data, like Baidu search index, may be useful for analyzing epidemics and improving prediction. We collected data on hepatitis E incidence and cases in Shandong province from January 2009 to December 2022 are extracted. Baidu index is available from January 2009 to December 2022. Employing Pearson correlation analysis, we validated the relationship between the Baidu index and hepatitis E incidence. We utilized various LSTM architectures, including LSTM, stacked LSTM, attention-based LSTM, and attention-based stacked LSTM, to forecast hepatitis E incidence both with and without incorporating the Baidu index. Meanwhile, we introduce KAN to LSTM models for improving nonlinear learning capability. The performance of models are evaluated by three standard quality metrics, including root mean square error(RMSE), mean absolute percentage error(MAPE) and mean absolute error(MAE). Adjusting for the Baidu index altered the correlation between hepatitis E incidence and the Baidu index from -0.1654 to 0.1733. Without Baidu index, we obtained 17.04±0.13%, 17.19±0.57%, in terms of MAPE, by LSTM and attention based stacked LSTM, respectively. With the Baidu index, we obtained 15.36±0.16%, 15.15±0.07%, in term of MAPE, by the same methods. The prediction accuracy increased by 2%. The methods with KAN can improve the performance by 0.3%. More detailed results are shown in results section of this paper. Our experiments reveal a weak correlation and similar trends between the Baidu index and hepatitis E incidence. Baidu index proves to be valuable for predicting hepatitis E incidence. Furthermore, stack layers and KAN can also improve the representational ability of LSTM models.

Temperature modeling of wave rotor refrigeration process based on variable time-delay compensation and SDAE-GRU

Abstract Aiming at the problem of noise and time-delay in temperature prediction modeling of large wave rotor refrigeration process, this paper proposes a novel modeling algorithm that combines the advantages of stacked denoising autoencoder (SDAE) network and Gate Recurrent Unit (GRU) algorithm and utilizes the Maximal Information Coefficient (MIC) algorithm to solve the targeting problem. The MIC algorithm is utilized to calculate the delay time of each input variable and output variable, and the input matrix is reconstructed to compensate for the time-delay, thus solving the problem afterwards. The SDAE network is utilized for denoising the input data and feature extraction to solve the problem of presence of noise in the data. The simulation results show that the proposed MIC-SDAE-GRU algorithm achieves a Root Mean Squared Error (RMSE) of 0.6432 and an R-squared (R²) value of 0.9638, outperforming traditional machine learning methods and other deep learning approaches. Specifically, compared to the standard GRU algorithm, it improves the accuracy of temperature prediction for the wave rotor refrigeration process by 24.7% and exhibits strong generalization ability across various operating conditions.

A Multi-Spatial Scale Ocean Sound Speed Prediction Method Based on Deep Learning

As sound speed is a fundamental parameter of ocean acoustic characteristics, its prediction is a central focus of underwater acoustics research. Traditional numerical and statistical forecasting methods often exhibit suboptimal performance under complex conditions, whereas deep learning approaches demonstrate promising results. However, these methodologies fall short in adequately addressing multi-spatial coupling effects and spatiotemporal weighting, particularly in scenarios characterized by limited data availability. To investigate the interactions across multiple spatial scales and to achieve accurate predictions, we propose the STA-ConvLSTM framework that integrates spatiotemporal attention mechanisms with convolutional long short-term memory neural networks (ConvLSTM). The core concept involves accounting for the coupling effects among various spatial scales while extracting temporal and spatial information from the data and assigning appropriate weights to different spatiotemporal entities. Furthermore, we introduce an interpolation method for ocean temperature and salinity data based on the KNN algorithm to enhance dataset resolution. Experimental results indicate that STA-ConvLSTM provides precise predictions of sound speed. Specifically, relative to the measured data, it achieved a root mean square error (RMSE) of approximately 0.57 m/s and a mean absolute error (MAE) of about 0.29 m/s. Additionally, when compared to single-dimensional spatial analysis, incorporating multi-spatial scale considerations yielded superior predictive performance.

Performance Assessment of Satellite-Based Precipitation Products in the 2023 Summer Extreme Precipitation Events over North China

In the summer of 2023, North China experienced a rare extreme precipitation storm due to Typhoons Doksuri and Khanun, leading to significant secondary disasters and highlighting the urgent need for accurate rainfall forecasting. Satellite-based quantitative precipitation estimation (QPE) products like Integrated Multi-Satellite Retrievals for GPM (IMERG) and Global Satellite Mapping of Precipitation (GSMaP) from the Global Precipitation Measurement (GPM) Mission have great potential for enhancing forecasts, necessitating a quantitative evaluation before deployment. This study uses a dense rain gauge as a benchmark to assess the accuracy and capability of the latest version 7B IMERG and version 8 GSMaP satellite-based QPE products for the 2023 summer extreme precipitation in North China. These satellite-based QPE products include four satellite-only products, namely IMERG early run (IMERG_ER) and IMERG late run (IMERG_LR), GSMaP near-real-time (GSMaP_NRT), and GSMaP microwave-infrared reanalyzed (GSMaP_MVK), along with two gauge-corrected products, namely IMERG final run (IMERG_FR) and GSMaP gauge adjusted (GSMaP_Gauge). The results show that (1) GSMaP_MVK, IMERG_LR, and IMERG_FR effectively capture the space distribution of the extreme rainfall, with relatively high correlation coefficients (CCs) of approximately 0.77, 0.75, and 0.79. The IMERG_ER, GSMaP_NRT, and GSMaP_Gauge products exhibit a less accurate spatial pattern capture (CCs about 0.66, 0.73, and 0.67, respectively). Each of the six QPE products tends to underestimate rainfall (RBs < 0%). (2) The IMERG products surpass the corresponding GSMaP products in serial rainfall measurement. IMERG_LR demonstrates superior performance with the lowest root-mean-square error (RMSE) (about 0.38 mm), the highest CC (0.97), and less underestimation (RB about −6.37%). (3) The IMERG products at rainfall rates ≥ 30 mm/h, GSMaP_NRT and GSMaP_MVK products at rainfall rates ≥ 55 mm/h, and GSMaP_Gauge products at ≥ 40 mm/h showed marked limitations in event detection, with a near-zero probability of detection (POD) and a nearly 100% false alarm ratio (FAR). In this extreme precipitation event, caution is needed when using the IMERG and GSMaP products.

Machine Learning based Pointing Corrections for Radio/Sub-millimeter Telescopes

Radio, sub-millimeter and millimeter ground-based telescopes are powerful instruments for studying the gas and dust-rich regions of the Universe that are invisible at optical wavelengths, but the pointing accuracy is crucial for obtaining high-quality data. Pointing errors are small deviations of the telescope’s orientation from its desired direction. The telescopes use linear regression pointing models to correct for these errors, taking into account various factors such as weather conditions, telescope mechanical structure, and the target’s position in the sky. However, residual pointing errors can still occur due to factors that are hard to model accurately, such as thermal and gravitational deformation and environmental conditions like humidity and wind. Here we present a proof-of-concept for reducing pointing error for the Atacama Pathfinder EXperiment (APEX) telescope in the high-altitude Atacama Desert in Chile based on machine learning. Using historic pointing data from 2022, we trained eXtreme Gradient Boosting (XGBoost) models that reduced the root-mean-square errors (RMSE) for azimuth and elevation (horizontal and vertical angle) pointing corrections by 4.3% and 9.5%, respectively, on hold-out test data. Our results will inform operations of current and future facilities such as the next-generation Atacama Large Aperture Submillimeter Telescope (AtLAST)

Development and Applicability Assessment of an Hourly Water Level Prediction Model for Agricultural Reservoirs Using Automated Machine Learning

Objectives:This study aims to develop and evaluate hourly water level prediction models for agricultural reservoirs using an automated machine learning (AutoML) approach, specifically, employing TPOT.Methods:The study focuses on the Baekrok and Baekryeon reservoirs using rainfall forecast data from the Korea Meteorological Administration, along with observed rainfall and reservoir water level data. TPOT, which utilizes genetic algorithms to automate the generation of optimal machine learning pipelines, was used to build models. Additionally, Random Forest (RF) was implemented as a benchmark to evaluate TPOT’s applicability. Predictions were generated with lead times of 1, 3, 6, and 12 hours, and model accuracy was evaluated using the Nash-Sutcliffe Efficiency (NSE), coefficient of determination (R2), and root-mean-square error (RMSE).Results and Discussion:The predictions from both TPOT and RF showed high accuracy for shorter lead times, with NSE values exceeding 0.99 for the 1-hour predictions. However, predictive accuracy decreased as lead time increased, likely due to greater uncertainty in rainfall forecast data, particularly for the 12-hour predictions. Despite this trend, TPOT-derived models maintained more stable and accurate performance compared to RF models across all lead times.Conclusion:This study demonstrates the applicability of TPOT-based AutoML techniques in predicting hourly water levels in agricultural reservoirs. Future work should explore strategies to mitigate the decline in accuracy for longer lead times.