Efficiency Coefficient Research Articles

Downscaled‐GRACE Data Reveal Anthropogenic and Climate‐Induced Water Storage Decline Across the Indus Basin

AbstractGRACE (Gravity Recovery and Climate Experiment) has been widely used to evaluate terrestrial water storage (TWS) and groundwater storage (GWS). However, the coarse‐resolution of GRACE data has limited the ability to identify local vulnerabilities in water storage changes associated with climatic and anthropogenic stressors. This study employs high‐resolution (1 km2) GRACE data generated through machine learning (ML) based statistical downscaling to illuminate TWS and GWS dynamics across twenty sub‐regions in the Indus Basin. Monthly TWS and GWS anomalies obtained from a geographically weighted random forest (RFgw) model maintained good consistency with original GRACE data at the 25 km2 grid scale. The downscaled data at 1 km2 resolution illustrate the spatial heterogeneity of TWS and GWS depletion within each sub‐region. Comparison with in‐situ GWS from 2,200 monitoring wells shows that downscaling of GRACE data significantly improves agreement with in‐situ data, evidenced by higher Kling‐Gupta Efficiency (0.50–0.85) and correlation coefficients (0.60–0.95). Hotspots with the highest TWS and GWS decline rate between 2002 and 2023 were Dehli Doab (−442, −585 mm/year), BIST Doab (−367, −556 mm/year), Rajasthan (−242, −381 mm/year), and BARI (−188, −333 mm/year). Based on a general additive model, 47%–83% of the TWS decline was associated with anthropogenic stressors mainly due to increasing trends of crop sown area, water consumption, and human settlements. The decline rate of TWS and GWS anomalies was lower (i.e., −25 to −75 mm/year) in upstream sub‐regions (e.g., Yogo, Gilgit, Khurmong, Kabul) where climatic factors (downward shortwave radiations, air temperature, and sea surface temperature) explained 72%–91% of TWS/GWS changes. The relative influences of climatic and anthropogenic stressors varied across sub‐regions, underscoring the complex interplay of natural‐human activities in the basin. These findings inform place‐based water resource management in the Indus Basin by advancing the understanding of local vulnerabilities.

Evaluation of the Performance of Modeled and Simulated Hybrid Heat Pump Drying Systems for Industrial Applications

System configuration and design is one of the major factors that influence the performance of heat pump drying system. This study aimed to evaluate the performance and drying efficiency of two different configuration of hybrid heat pump drying system for heating process and drying of tomatoes slices at condensing temperature 15 ℃ and evaporating temperature 10 ℃ using calcium chloride-ethanol as adsorbate-adsorbent pair. The energetic coefficient of performance, and exergetic coefficient of performance and overall exergy efficiency of the system are 2.03, 15.6% and 17% respectively for simple adsorption-compression hybrid heat pump drying system while its specific moisture extraction rate is 0.234 kg/kWhr and specific energy consumption is 4.27 kg/kWhr. The energetic coefficient of performance, and exergetic coefficient of performance and overall exergy efficiency of the system are 2.18, 22.6% and 16% respectively for cascaded adsorption-compression hybrid heat pump drying system and its specific moisture extraction rate is 0.457 kg/kWhr and specific energy consumption is 3.17 kg/kWhr.

Analysis of refrigerants used in supermarket commercial equipment and the potential for increasing energy efficiency and reducing environmental impact

Refrigerants used in the commercial equipment of supermarkets are the object of the research. The Montreal Protocol calls for a complete phase-out of hydrochlorofluorocarbons (HCFCs) by 2030, and the Kigali Amendment regulates the use of hydrofluorocarbons (HFCs) from 2019. Developed countries began phasing out HFC use in 2019, while developing countries plan to freeze HFC consumption from 2024. These global efforts are aimed at reducing the depletion of the ozone layer and combating climate change. The number of supermarkets in the world varies greatly: in Europe they number from 110 thousand to 115 thousand, and in China – from 65 thousand to 70 thousand, which reflects various needs in refrigeration equipment. Stringent environmental regulations are forcing the commercial refrigeration sector to remain globally competitive. Modernization of supermarkets using natural refrigerants is important for solving emerging challenges. The results of the study show significant improvements in energy efficiency ratio (EER) and coefficient of performance (COP) when using a mixture of hydrocarbons (R290: 85 %, R600a: 15 %) compared to traditional refrigerants R404a, R449a and R502. Specifically, at the evaporation temperature of Tevap=–10 °C, EER increased by 38–44 % and COP by 26–31 % compared to R404a and R449a, respectively. At Tevap=–25 °C, EER increased by 17–34 % and COP by 2–22 % compared to R404a and R449a. Additionally, compared to R502, the hydrocarbon blend showed a 38–44 % increase in EER and 28–31 % COP at Tevap=–10 °C, and a 17–34 % increase in EER and 5–22 % COP at Tevap=–25 °C. These results highlight the advantages of the hydrocarbon mixture at different evaporation temperatures, indicating its potential to improve energy efficiency in refrigeration applications. The obtained data suggest the possibility of a wider application of the mixture of hydrocarbons in commercial refrigeration plants, offering both improved performance and compliance with safety regulations.

Analysis of the technique for experimental determination of the efficient heat transfer coefficient during drying of iron ore pellets and charge in the dense layer

Analysis of the technique for experimental determination of the efficient heat transfer coefficient during drying of iron ore pellets and charge in the dense layer

Evaluating rainfall erosivity on the Tibetan Plateau by integrating high spatiotemporal resolution gridded precipitation and gauge data

High-precision rainfall erosivity mapping is crucial for accurately evaluating regional soil erosion on the Tibetan Plateau (TP) under the backdrop of climate warming and humidification. Although high spatiotemporal resolution gridded precipitation data provides the foundation for rainfall erosivity mapping, the increasing spatial heterogeneity of rainfall with decreasing temporal granularity can lead to greater errors when directly computing rainfall erosivity from gridded precipitation data. In this study, a site-scale conversion coefficient was established so that rainfall erosivity calculated using hourly data can be converted to rainfall erosivity calculated using per-minute data. A revised model was established for calculating the rainfall erosivity based on high-resolution hourly precipitation data from the Third Pole gridded precipitation dataset (TPHiPr). The results revealed a notable underestimation in the original calculation results obtained using the TPHiPr, but strong correlation was observed between the two sets of results. There was a significant improvement in the Nash–Sutcliffe coefficient of efficiency (from −0.39 to 0.80) and the Percent Bias (from −63.95 % to 0.37 %) after model revision. The TPHiPr effectively depict the spatial characteristics of rainfall erosivity on the TP. It accurately reflected the rain shadow area on the northern flank of the Himalayas and the dry-hot valley in the Hengduan Mountains. It also showed high rainfall erosivity values in the tropical rainforest area on the southern flank of the eastern Himalayas. The overall trend of rainfall erosivity has increased on the TP during the period 1981 to 2020, with 65.91 % of the regions exhibiting an increasing trend and 22.25 % showing significant increases, indicating an intensified risk of water erosion. These findings suggest that the 40-year-high spatial resolution rainfall erosivity dataset can provide accurate data support for a quantitative understanding of soil erosion on the TP.

RISK MANAGEMENT BASED ON THE PERFORMANCE INDICATORS OF OCCUPATIONAL HEALTH AND SAFETY MEASURES AND THE SAFETY RAINBOW CLASSIFICATION

The article proposes the use of coefficients of efficiency and effectiveness of measures, the degree of effectiveness of measures to manage the probability of negative events and risks in general. The introduction of the first universal classifier of measures "Safety Rainbow". Design/methodology/approach. To determine the effectiveness coefficient of security measures, an expert evaluation method was used to assess the consistency of expert opinions by calculating the concordance coefficient, based on the data obtained, a formula (polynomial) of the effectiveness of measures was determined, a new calculation formula was used to determine the probability of events, the effectiveness coefficient of security measures was determined as the share of the effectiveness coefficient in the cost of the measure per year, the degree of effectiveness is the total effectiveness of all measures used, given by The effectiveness coefficient of measures can be used both to determine the level of probability of negative events, as demonstrated by the matrix method of risk assessment, and in other occupational health and safety issues. The measure effectiveness ratio allows to prioritise measures for the purpose of rational use of resources. The degree of effectiveness of measures should be used to understand the prospects for reducing the level of risk and effective hazard management. The principles of ranking measures are laid down in the universal classifier "Safety Rainbow" and allow covering all possible measures for various hazards, works and not only in the field of occupational safety and health. Research limitations/implications. The proposed methodology is universal and can be used not only in the application of the matrix method of risk assessment, but also in other methods and not only in the field of occupational safety and health. The developed methodology of risk assessment allows to perform once an expert assessment of the effectiveness of safety measures and in the future by personnel who do not have sufficient qualifications in the relevant field of activity to perform promptly and efficiently assess and manage risks, develop additional safety measures, rationally use resources, including the use of computer programs, mobile applications, etc. Originality and significance. The proposed methods are used for the first time.

Reducing convective losses in a solar cavity receiver VoCoRec by creating a controlled vortex of returned air

The efficiency of air-type solar towers influences the minimal cost of electricity/heat produced by them, which limits their use despite the widespread use of many forms of concentrated sun power plants. Nonetheless, the temperature potential of this kind of concentrated plant is the highest. To improve the efficiency of an air-type solar tower, a technique for reducing convective energy losses with return air is therefore suggested. In order to increase a cavity receiver's thermal efficiency, a little-known concept called vortex creation will be discussed in this work. The article models different ways of creating an air vortex inside the cavity receiver. Using the VoCoRec design as an example, cases are shown where the vortex increases and decreases the thermal efficiency of the receiver. The concept of Air Return Ratio (ARR) is used to determine the convective losses in a solar collector. This coefficient indicates the convective losses of the receiver due to buoyancy forces and has a direct proportional dependence on the convective efficiency coefficients of the receiver. The VoCoRec receiver, which incorporates the directional vortex inside the receiver, increased the air return coefficient by 4 % (at the same air mass flow rate). The dependence of the air return coefficient on different angles of the air outlet to the absorber plane, including in the radial direction, was also investigated. Increasing the angle of inclination of the air outlet to the main absorber increases the air return coefficient in all cases, but also increases the aerodynamic drag of the receiver (pressure drop).

Photovoltaic Capacity Management for Investment Effectiveness

The production of photovoltaic utility varies within the day/night cycle. At night, photovoltaic cells do not produce anything. However, their day-light production, unconsumed on a current basis and exported to the grid, is compensated for with supply from the grid at night. This scheme of exploitation is called net-metering and is considered herein. Solar energy produced by a prosumer and fed into the grid has to be equal to the electricity supplied from the grid at night; otherwise, a shortage or waste of photovoltaic production occurs. The above finding leads us to the proposition for the optimal solution of photovoltaic capacity. We derived a closed-form capacity solution to the maximized non-linear profit function. It solves harmonic and 2-point production functions that vary symmetrically around the mean production. To verify the solution methodology, harmonic and 2-point models from empirical production data are estimated. Then, the solution is presented together with its return rate and internal return rate. The main finding is that the unit cost of the grid electricity, photovoltaic capacity unit cost and exploitation time all affect the total profit and return rate values while not impacting the optimal capacity of the photovoltaics. The optimal capacity depends on the prosumer’s energy consumption volume and on the natural conditions of production captured here by the technology efficiency coefficient estimated from the production time series.

Optimal refrigerant selection and parameter analysis of a wastewater heat pump using CO2 zeotropic mixtures

Heat pump technology can effectively recover waste heat from low-grade wastewater, which is crucial for energy conservation and carbon emission reduction in buildings. In this study, zeotropic mixtures were used to recover heat from low-grade wastewater and produce high-temperature water in a heat pump. The deep recovery of waste heat was realized by using a CO2 zeotropic mixture with a large glide temperature. Furthermore, this method ensures an excellent temperature matching between the heat transfer fluids and effectively reduces irreversible losses. To obtain an optimal zeotropic mixture, a method based on heuristic screening of the optimal CO2 zeotropic mixture was proposed. Furthermore, a model of a CO2 zeotropic mixture heat pump was established and experimentally validated. Based on the validated model, a sensitivity analysis of the various operating parameters of the heat pump was conducted. The results show that the zeotropic mixture CO2/R1234yf with a mass fraction of 0.17/0.83 is selected as the optimal refrigerant for the heat pump with the coefficient of performance, exergy efficiency and volumetric heating capacity of 7.46, 0.37, and 7978 kJ⋅m−3, respectively. The sensitivity analysis of operating parameters shows that the inlet temperature of hot water has more crucial influence on the coefficient of performance and exergy efficiency than various operating parameters, with sensitivity coefficients of −0.3868 and 0.2157, respectively. The volumetric heating capacity and system total exergy destruction per heating capacity unit are most sensitive to the flow rate of wastewater and hot water, with sensitivity coefficients of 0.1121 and −0.4985, respectively. These results provide a foundation for the design and optimization of a CO2 zeotropic mixture heat pump system.

Thermodynamic analysis and optimization of hydrogen liquefaction system with binary refrigerant precooling cycle

A hydrogen liquefaction process is proposed with the liquid natural gas utilization, binary refrigerant precooling and helium cryogenic cycles combined, and a general model is built for analyzing the variation of hot and cold composite curves and heat transfer temperature difference (HTTD) along with refrigerant parameters. The relationship between temperature and enthalpy of the precooling cycle and the variation of HTTD are deduced. Based on this, the composite curves and HTTD curves are constructed, and the influence of refrigerant on the HTTD and the refrigerant outlet temperature is analyzed. The optimization procedure of refrigerant is proposed based on the temperature-enthalpy relationship and HTTD considering the temperature glide. For the proposed hydrogen liquefaction system, the optimal binary refrigerant includes 60.21 % nitrogen and 39.79 % propane (mass fraction), and its mass flow rate and inlet temperature are 23,821.96 kg/h and −199.72 °C, respectively. The corresponding specific energy consumption (SEC) of the hydrogen liquefaction system is 4.44 kWh/kgLH2, reduced by 10.30 %, and the coefficient of performance (COP) and exergy efficiency are 27.10 % and 28.26 %, increased by 11.34 % and 17.80 %, respectively.

Economic and energy efficiency of resource-saving technology for switchgrass cultivation

Purpose. The study aimed to compare the efficiency of the resource-saving technology with the conventional technology for switchgrass cultivation using the developed methodology for assessing the economic and energy efficiency. Methodology / approach. The study used general and special methods, including the methodology of scientific research in agronomy, laboratory determination of dry matter content in biomass, quantitative-weight analysis to establish crop yield, and the authors’ improved methodology for assessing economic and energy efficiency. The research results were statistically processed using variance and comparative analysis. Results. The results of the research on the use of resource-saving cultivation technology in comparison with conventional technology show an increase in switchgrass biomass yield from 14.6 to 15.7 t/ha, an increase in economic efficiency with profitability growth from 73.8 to 79.0 %, and an increase in energy efficiency with a growth of the energy efficiency coefficient by 0.7 – from 4 to 4.7 (average level of energy efficiency) when applying a specific complex of agrotechnical measures. When using resource-saving technology, the average full cost of cultivating switchgrass for six years is 8305.6 UAH/ha, compared to 7952.8 UAH/ha with conventional technology. However, resource-saving technology generates an average sales revenue of 14867.5 UAH/t, which is 1045 UAH/t more than conventional technology (13822.5 UAH/t). Originality / scientific novelty. For the first time, a field experiment was conducted to compare switchgrass cultivation using resource-saving and conventional technologies. The authors have developed a methodology to assess the economic and energy efficiency of cultivating switchgrass. The results indicate that the efficiency of switchgrass biomass production is influenced by improved cultivation technology. The authors have developed a three-dimensional econometric model that demonstrates how the profitability level depends on the chosen switchgrass cultivation technology. Practical value / implications. The research results have practical significance as they have led to the development of a methodology and evaluate the economic and energy efficiency of switchgrass cultivation. These results will be useful for agrarian enterprises to save resources.

Специфика редакторской подготовки дайджеста для публикации в социальных сетях

Modern people want to keep the track of as many events as possible while saving as much time as possible and reducing the level of redundant information. Internet digests provide an opportunity to systematize online publications and structure relevant information while focusing on the main facts and ideas. The paper describes the algorithm, editing, and design of Internet digests in social media. The authors analyzed digest entries in the Russian VKontakte social network to identify the methods of collecting, processing, and structuring information at different stages of editing. The standard folding method proved to be especially efficient, considering the stylistic diversity of the source materials compiled. As a result, the editor has to standardize the style without violating the integrity of the original text, eliminating information noise, applying the principles of information literacy and multidimensionality, etc. These aspects determine the structure of the VKontate digest as title – lead – text – call to action – graphic elements. The authors developed procedures for digest editorial as a sequence of actions that provides a high-quality media product with a high coefficient of information efficiency. Editing improves the quality of the material while ensuring its relevance, reliability, and clarity, thus facilitating the reading comprehension.

Accurate Quantification of 0–30 cm Soil Organic Carbon in Croplands over the Continental United States Using Machine Learning

Increases in organic carbon within agricultural soils are widely recognized as a “negative emission” that removes CO2 from the atmosphere. Accurate quantification of soil organic carbon (SOC) to a certain depth in the spatial domain is critical for the effective implementation of improved land management practices in croplands. Currently, there is a lack of understanding regarding what depth strategy should be used to estimate SOC at 0–30 cm when sample datasets come from multiple depths. Furthermore, few studies have examined depth strategies for mapping SOC at the agricultural management level (i.e., field level), opting instead for point-based analysis. Here, three types of approaches with different depth strategies were evaluated for their ability to quantify 0–30 cm SOC content based on soil samples from 0–5 (surface), 5–30 (subsurface), and 0–30 cm (full column). These approaches involved the generalized additive model and machine learning techniques, i.e., artificial neural networks, random forest, and XGBoost. The soil samples used for the model evaluation and selection consisted of the newly collected samples in 2020–2022 and the Rapid Carbon Assessment (RaCA) legacy samples collected in 2010–2011. Environmental covariates corresponding to these SOC measurements were used in model training, including long-term physical climate, short-term weather, topographic and edaphic, and remotely sensed variables. Among the models evaluated in this study, the XGB regression model with a full column depth assignment strategy yielded the best prediction performance for 0–30 cm SOC content, with an r2 (squared Pearson correlation coefficient) of 0.48, an RMSE (root mean square error) of 0.29%, an ME (mean error) of 0.06%, an MAE of 0.25%, and an MEC (modeling efficiency coefficient) of 0.36 at the pixel level and an r2 of 0.64, an RMSE of 0.32%, an ME of −0.20%, an MAE of 0.28%, and an MEC of 0.48 at the field level. This study highlights that machine learning models with a full column depth strategy should be used to quantify 0–30 cm SOC content in agricultural soils over the continental United States (CONUS).

Analysis of the Application of Static Load Test in Bridge Bearing Capacity Testing

This article uses real engineering projects as examples to analyze how static load test technology is applied in testing the bridge-bearing capacity. The analysis covers aspects such as testing section layout, test load and efficiency coefficient, loading plan, evaluation optimization, test result modification, and result evaluation. The aim is to support the accurate detection and evaluation of bridge-bearing capacity.

Accelerated mapping and robust parallel transmit pulse design for heart and prostate imaging at 7 T.

To investigate the impact of reduced k-space sampling on mapping and the resulting impact on phase shimming and dynamic/universal parallel transmit (pTx) RF pulse design. Channel-wise 3D maps were measured at 7 T in 35 and 23 healthy subjects for the heart and prostate region, respectively. With these maps, universal phase shims optimizing homogeneity and efficiency were designed for heart and prostate imaging. In addition, universal 4kT-point pulses were designed for the heart. Subsequently, individual phase shims and individual 4kT-pulses were designed based on maps with different acceleration factors and tested on the original maps. The performance of the pulses was compared by evaluating their coefficients of variation (CoV), efficiencies and specific energy doses (SED). Furthermore, validation measurements were carried out for one heart and one prostate subject. For both organs, the universal phase shims showed significantly higher efficiencies and lower CoVs compared to the vendor provided default shim, but could still be improved with individual phase shims based on accelerated maps (acquisition time = 30 s). In the heart, the universal 4kT-pulse achieved significantly lower CoVs than tailored phase shims. Tailored 4kT-pulses based on accelerated maps resulted in even further reduced CoVs or a 2.5-fold reduction in SED at the same CoVs as the universal 4kT-pulse. Accelerated maps can be used for the design of tailored pTx pulses for prostate and cardiac imaging at 7 T, which further improve homogeneity, efficiency, or SED compared to universal pulses.

Modeling of Filtration Phenomenon in Hydrostatic Drives

Some users consider modern hydrostatic drives and controls to be unreliable and difficult to maintain. This view is often due to operational problems caused by issues with obtaining and then maintaining the appropriate cleanliness class of the working fluid. Recommendations on the selection of appropriate filtration system elements can be found in the literature, but there is no numerical model that could be helpful in a detailed analysis of the phenomenon. In the article, the authors tried to fill the research gap regarding the lack of a filtration model based on the filtration efficiency coefficient of filter elements used in hydraulic drives and controls. The developed model allows users to determine the influence of selected filtration system parameters on the separation of contaminants by filter elements. The model is intended to help designers and users of hydraulic drives and controls in optimizing the filtration system in order to obtain and then maintain the required cleanliness class of the hydraulic fluid. This paper also includes the results of the sensitivity analysis of selected filtration-system operating parameters in terms of the highest efficiency. In order to verify the developed model, experimental tests were also carried out, with the results presented in this paper. Based on the numerical analyses and experimental studies, recommendations that may be helpful in the selection or development of filtration systems used in hydrostatic drives and controls were developed.

SPECIFIC ENERGY CAPACITY OF PROCESSING AND ENERGY EFFICIENCY FOR PROCESS OF GRINDING WITH WHEELS FROM SUPERHARD MATERIALS

The analysis of modern research shows that when evaluating the specific energy intensity of the abrasive processing process, one should pay attention not only to the indicators of grinding power and material removal rate, but also to the indicator of the wear of the abrasive tool, which we will show below for the grinding tool made of superhard materials. It is shown that the traditional method of estimating the specific energy capacity based on the ratio of the grinding power to the processing productivity does not provide an adequate solution, since with it the specific heat capacity of processing exceeds the specific heat capacity of melting of the processing material by almost an order of magnitude. Therefore, it is the application of a new approach to the assessment of the specific energy intensity of diamond grinding, taking into account the wear of the working layer of the diamond wheel, and makes it possible to estimate the indicators of the specific energy intensity of grinding and the energy efficiency coefficient. It has been proven that when estimating the specific energy capacity of grinding metal-ceramic composite materials consisting of a low-melting and refractory component, the latent heat capacity of melting of the low-melting component should be taken as the basis. It is shown that the plastic mode of grinding occurs precisely when the specific energy capacity of grinding, taking into account the wear of the wheel, becomes close to the specific heat capacity of melting of a brittle material.

Dimensionless resolutions for heat flux decrement factor and time lag of the wall during cyclic variations in outdoor air temperature

The heat flux decrement factor (DF) and time lag (TL) are two vital indicators for assessing wall dynamic thermal performance. However, computing these indicators involves complex variables such as material thermal properties, wall thickness, and others. Therefore, exploring the inherent relation between these two indicators and the influencing parameters is an urgent problem for reducing building energy consumption. To this end, an analytical method based on the auxiliary function was employed to calculate the heat flux DF and TL of a single-layer wall. Then, the impact of the influencing factors on heat flux DF and TL were evaluated. The results showed that heat flux DF and TL are only determined by two dimensionless parameters: diffusion length and thermal efficiency coefficient. The heat flux DF decreases by at least 5 times as the diffusion length varies from 0 to 2. Besides, it is also affected by the diffusion length when the thermal efficiency coefficient (ε) is less than 5, and the smaller the ε, the smaller the heat flux DF. On the other hand, the heat flux TL increases linearly with the diffusion length initially but declines precipitously as the diffusion length approaches 6–7. When the diffusion length is the same, the heat flux TL slightly decreases as the thermal efficiency coefficient increases.

A novel strategy for flood flow Prediction: Integrating Spatio-Temporal information through a Two-Dimensional hidden layer structure

In recent years, neural network models have been extensively applied in flood prediction due to their superior performance. However, most studies aimed at enhancing models have simply stacked basic models, overlooking the need for designs specific to the physical attributes of hydrological data. While stacking basic models can enhance flood prediction to a degree, they often introduce significant unnecessary complexity redundancy. Moreover, improvement strategies are challenging to generalize across different models. To address this issue, a widely applicable Two-Dimensional Hidden Layer (Td) architecture is proposed. Unlike single hidden layer architectures, this design specifically tackles hydrological data’s spatiotemporal characteristics. Discretizing spatiotemporal information in flood and rainfall data yields an abstract, two-dimensional representation of the hydrological data to enhance the model’s ability to utilize hydrological information. Using a Td structure instead of the single hidden layer in Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (BiLSTM), and Convolutional Neural Network (CNN), the TdRNN, TdLSTM, TdBiLSTM, and TdCNN models are developed to assess the novel structure’s impact on foundational models. Experiments were conducted using empirical flood process data from the LouDe Station in Shandong, China, and the NingXiang Station in Hunan, China. The results indicate that, compared to benchmark models, network models incorporating a Td structure exhibit superior adaptability across various lead times. For instance, at the LouDe Station, with a one-hour lead time, the TdCNN model showed improvements in the Nash-Sutcliffe Efficiency Coefficient (NSE), Kling-Gupta Efficiency Coefficient (KGE), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE) performance metrics by 26.24%, 33.48%, 62.51%, and 56.88%, respectively, compared to the CNN model. The new models also demonstrated significant performance enhancements under other conditions. The experimental results demonstrate that the Td structure is an effective and versatile model improvement strategy. It enhances the prediction accuracy by strengthening the model’s ability to utilize information. This study may contribute to the development of a flood prediction system and provide data to support subsequent reservoir scheduling and disaster mitigation planning.

Sea surface temperature prediction by stacked generalization ensemble of deep learning

Accurate sea surface temperature (SST) prediction is of great significance for fishery farming, marine ecological protection, and planning of maritime activities. In this paper, the stacked generalization ensemble is demonstrated for SST prediction to improve every single deep-learning model. Long-term high-resolution satellite-derived SST is used with the sub-regions of the Taiwan Strait and East China Sea taken as the study area. We select the Multilayer Perceptron, Long short-term memory (LSTM), Convolutional Neural Networks (CNN), CNN-LSTM as individual learners, and Convolutional LSTM as the meta-learner. The individual learners are trained and validated on the retained data subset I, while the meta-learner is trained and validated by constructing the samples with the predictions of validated individual learners on the retained data subset II. The two types of models are evaluated on the same test dataset with root-mean-square error and coefficient of efficiency as the scoring criteria. We find that the meta-model outperforms any individual model and other baselines for the one-day-/three-day-ahead forecasts in the Taiwan Strait and one-day-/three-day-/five-day-ahead predictions in the East China Sea. Furthermore, when the lead time is 1 day and 3 days, the meta-model has a better spatial distribution of prediction metrics across all grid points in the Taiwan Strait sub-area. For the East China Sea sub-region, the meta-model advantage is extended to the lead time of 5 days. Probably due to the higher quality of offshore satellite data, the prediction ability enhancement of the stacked ensemble applied in the East China Sea is better than that in the Taiwan Strait. The better-performing meta-model prediction suggests that the stacked generalization ensemble is encouraging and promising for improving the short-term prediction of the daily SST field.