Search

The performance of lithium-ion battery modules significantly depends on cell-to-cell variations and connection topology. In particular, inhomogeneous distribution across the parallel battery module results in performance degradation and potential safety problems. This study evaluates the overall performance of battery modules, including parallel-connected cell groups with different system terminal positions, and examines the effect of cross-connectors. First, a battery module model with electrochemical, thermal, and aging properties is introduced. An LMN structure that allows all battery module structures was proposed for the first time. Next, modules with different topologies are simulated to analyze the cell-to-cell variations in terms of current, temperature, and aging. Finally, six performance indices are extracted from the simulation results to represent the module performance. The module with the cross-contacting system terminal reduces the effect of the connection resistance on the cell-to-cell variation, showing clear improvement over other topologies of up to 61.5%. Additionally, the use of cross-connectors reduces inconsistencies between cells and improves the safety of the battery modules in the event of potential problems. The number of cycles increases by over four times compared with the general case and no significant overdischarge state is observed in a particular cell. The proposed method can be used to solve system-level problems and assist engineers in designing battery management systems to provide longer-lasting and more efficient energy storage systems. [Display omitted] • A flexible and configurable model for Li-ion battery modules is proposed. • An electro-thermal-aging model is used to capture the dynamics of batteries. • Netlist and modified nodal analysis make large-battery system analysis easier. • The positions of system terminals are used to analyze six performance indices. • The impact of internal cross-linking on heterogeneity and safety is verified. [ABSTRACT FROM AUTHOR]

Cyanobacterial harmful algal blooms (cyanoHABs) caused by cyanobacteria negatively affect humans via river water and aquatic life. Thus, reliable cyanobacteria predictions are essential for managing cyanoHABs. With recent advancements in computer technology and big data usage, artificial intelligence (AI) technologies have gained attention in various fields, such as water resources, weather and climate, and water quality. This study evaluated the applicability of deep-learning-based AI technology for predicting cyanobacteria. A convolutional neural network (CNN)–long short-term memory (LSTM) model, a deep-learning-based AI technology advantageous for predicting time-series data and cyanobacteria features, was built. Its results were analyzed and compared with those of the existing physical Environmental Fluid Dynamics Code (EFDC)–National Institute of Environment Research (NIER) model for cyanobacteria prediction. The CNN-LSTM model performed better, with an accuracy of 69%, which is an improvement over the previous EFDC-NIER model's accuracy of 45%. In particular, there was a dramatic improvement in the prediction accuracy for low cyanobacteria cell counts in Level 1, which increased from 39% to 87%. There also was an improvement in the prediction accuracy for Levels 2 and 3. The accuracy for Level 2 increased from increased from 56% to 69%, and the accuracy for Level 3 increased from 38% to 48%. However, there was a significant decrease in prediction accuracy for high cyanobacteria cell counts in Level 4, for which the measured data were very scarce; accuracy decreased from 49% to 16.7%. The CNN-LSTM model yielded better overall prediction performance than the EFDC-NIER model, demonstrating its applicability in cyanobacteria prediction. However, it has limitations of overfitting areas with inadequate data and not accurately predicting patterns that have not occurred in the past. To address this issue, we propose an approach the combines the advantages of physics-based models and AI-based deep learning models, creating a hybrid concept. [ABSTRACT FROM AUTHOR]

The purpose of this study was to investigate the flow stress behavior and formation of ultrafine grains (UFGs) by focusing on the effects of strain-induced dynamic transformation (SIDT) and dynamic recrystallization (DRX) under hot forming near the β transus temperature. Heavy-reduction thermomechanical controlled processing (TMCP) of commercial pure (CP) titanium was performed at deformation temperatures ranging from 700 to 1000 °C, a strain rate of 1 s−1, and a 70% height reduction, followed by water cooling immediately after compression. The flow stress at 0.2 strain decreased significantly from 96 to 25 MPa when the deformation temperature increased from 800 to 900 °C. Equiaxed UFGs with a grain size of 2–4 μm formed in this temperature range. The primary mechanisms of grain refinement in the CP titanium under heavy-reduction TMCP were DRX occurring at 800 °C and a combination of DRX and SIDT (β → α) at 900 °C. Of the 700, 900, and 1000 °C TMCP-treated specimens subjected to a 90° V-bending test, the yield force of the 700 °C TMCP-treated specimen was the highest. When the TMCP temperature was increased from 900 to 1000 °C, the yield force decreased significantly, whereas the springback amount increased. The α phase grain refinement achieved by the TMCP at 700 °C increased the force, and the retained β phases within the 1000 °C TMCP-treated specimen increased the springback amount. [ABSTRACT FROM AUTHOR]

Polymer electrolyte membrane fuel cells (PEMFC) are a prevalent power source in transportation because of their ability to generate energy at low temperatures without harmful emissions. However, problems related to water management cause performance and durability degradation. The main faults are flooding, whereby the performance suffers owing to the stagnated water in gas diffusion paths and catalyst layers, and drying, which increases the ohmic loss owing to water evaporation in the electrolyte or insufficient water supply. Difficulties in recognizing these faults and normalizing operations impair the PEMFC stability. However, detecting errors in advance contributes to maintaining normal operation. Therefore, a system that diagnoses flooding and drying of the PEMFC before they occur is developed in this study using deep learning. The characteristics of flooding and drying are analyzed through preliminary experiments. Experimental data in the form of a time series are accumulated through a full-scale single-cell test. A pre-diagnosis system, developed using long short-term memory (LSTM) and a convolutional neural network (CNN), is reinforced through the bagging ensemble method. The expandability of the target future time and real-time system applicability are discussed. The detection rates achieved by the proposed system for flooding and drying that occur after 30 s are 98.52% and 95.36%, respectively. • Pre-diagnosis system that predicts the future state of PEMFC. • Flooding and drying cause performance degradation in PEMFC. • Electrical characterization of flooding and drying through PEMFC experiments. • Output voltage, relative humidity and cell temperature are selected as input features. • Bagging ensemble method based on LSTM and CNN is conducted. [ABSTRACT FROM AUTHOR]

Harmful algal blooms (HABs) are a serious threat to ecosystems and human health. The accurate prediction of HABs is crucial for their proactive preparation and management. While mechanism-based numerical modeling, such as the Environmental Fluid Dynamics Code (EFDC), has been widely used in the past, the recent development of machine learning technology with data-based processing capabilities has opened up new possibilities for HABs prediction. In this study, we developed and evaluated two types of machine learning-based models for HABs prediction: Gradient Boosting models (XGBoost, LightGBM, CatBoost) and attention-based CNN-LSTM models. We used Bayesian optimization techniques for hyperparameter tuning, and applied bagging and stacking ensemble techniques to obtain the final prediction results. The final prediction result was derived by applying the optimal hyperparameter and bagging and stacking ensemble techniques, and the applicability of prediction to HABs was evaluated. When predicting HABs with an ensemble technique, it is judged that the overall prediction performance can be improved by complementing the advantages of each model and averaging errors such as overfitting of individual models. Our study highlights the potential of machine learning-based models for HABs prediction and emphasizes the need to incorporate the latest technology into this important field. [ABSTRACT FROM AUTHOR]

It is extremely challenging to imitate neural networks with their high-speed parallel signal processing, low power consumption, and intelligent learning capability. In this work, we report a spike neuromorphic module composed of “synapstors” made from carbon nanotube/C60/polyimide composite and “CMOS Somas” made from complementary metal-oxide semiconductor electronic circuits. The “synapstor” emulates a biological synapse with spike signal processing, plasticity, and memory; the “CMOS Soma” emulates a Soma in a biological neuron with analog parallel signal processing and spike generation. Spikes, short potential pulses, and input to the synapstors trigger postsynaptic currents and generate output spikes from the CMOS Somas in a parallel manner with low power consumption. The module can be modified dynamically on the basis of the synapstor plasticity. Spike neuromorphic modules could potentially be scaled up to emulate biologic neural networks and their functions. [ABSTRACT FROM AUTHOR]

A simulation framework of algal bloom in a river channel with data assimilation (DA) was developed by employing two numerical models coupled to simulate a watershed and the embedded river channel. The Hydrological Simulation Program-Fortran (HSPF) model simulates flow discharge and water quality from the subwatersheds and the Environmental Fluid Dynamics Code (EFDC) model takes the subwatershed model outputs at the watershed-river confluence points as boundary forcing to simulate river hydrodynamics and water quality. The ensemble Kalman filter (EnKF) was used for assimilation of water quality variables in the framework, linking uncertainty of model simulation and observation. The simulation uncertainty of the HSPF was quantified at the confluence points as simple stochastic error models developed by comparing the model simulation and the observation. The error models reflect uncertainty of both hydrologic and water quality simulation, including uncertainty associated with point and non-point pollution sources in the watershed. The outputs of the HSPF at the confluence points were perturbed with the error models before used in the following ensemble simulation of the EFDC for the main river. DA was conducted with weekly chlorophyll- a data observed along the river to update chlorophyll- a concentrations of the EFDC model grids. The results showed that the model performance was improved by the assimilation: the root mean square error (RMSE) and the mean continuous probability rank score (CPRS) significantly decreased compared to the open-loop simulation. The updated spatial distribution of chlorophyll- a concentration along the river channel was in reasonable agreement with the observation. Although only chlorophyll- a data was involved in the assimilation, phosphate was selected among other water quality variables for update in order to evaluate the effect of chlorophyll- a assimilation on those variables. It turned out that the phosphate simulation was not much improved by the chlorophyll- a data, which was due to weak correlation between the two variables in the model ensemble. Lastly, chlorophyll-a simulation uncertainty in the river attributed to the simulation uncertainty of each variable in the watershed was evaluated. For that, two additional simulations were made, with perturbation only to flow and phosphate respectively at the confluence points. The spread of chlorophyll- a ensemble of each case became significantly narrower than the original case, indicating that the difference is attributed to the uncertainty of the other unperturbed variables. [ABSTRACT FROM AUTHOR]

• Reinforcement Learning algorithm for energy efficient speed planning of electric vehicles. • Eco-driving control policy considering road slope and safe-distance to front vehicle. • Model-based reinforcement learning is conducted using the domain knowledge. • Near optimal energy-saving performance of 93.8% of exact dynamic programming. Eco-driving is a term used to refer to a strategy for operating vehicles so as to minimize energy consumption. Without any hardware changes, eco-driving is an effective approach to improving vehicle efficiency by optimizing driving behavior, particularly for autonomous vehicles. Several approaches have been proposed for eco-driving, such as dynamic programming, Pontryagin's minimum principle, and model predictive control; however, it is difficult to control the speed of the vehicle optimally in various driving situations. This study aims to derive an eco-driving strategy for reducing the energy consumption of a vehicle in diverse driving situations, including road slopes and car-following scenarios. A reinforcement learning-based energy efficient speed planning strategy is proposed for autonomous electric vehicles, which learn an optimal control policy through a data-driven learning process. A model-based reinforcement learning algorithm is developed for the eco-driving strategy; based on domain knowledge of the vehicle powertrain, a battery energy consumption model and longitudinal dynamics model of the vehicle are approximated from the driving data and are used for reinforcement learning. The proposed algorithm is tested using a vehicle simulation, and is compared to a global optimal solution obtained using an exact dynamic programming method. The simulation results show that the reinforcement learning algorithm can adjust the speed of the vehicle by considering driving conditions such as the road slope and a safe distance from the leading vehicle while minimizing energy consumption. The reinforcement learning algorithm achieves a near-optimal performance of 93.8% relative to the dynamic programming result. [ABSTRACT FROM AUTHOR]

Abstract: We demonstrated a novel method for the formation of alloy nano-islands on carbon nanotube (CNT). The two metal layers (Pt, Au) were sputtered on CNTs and the intense pulsed light (IPL) was irradiated on the metal layers. The absorbed light provides enough energy for the diffusion mixing between Pt and Au, forming Pt–Au alloy phase. While the alloy is being formed by the IPL irradiation, the metal layers are broken into nano-islands on CNT due to the surface energy minimization between the metal layers and CNT. The surface characterizations of the Pt–Au/CNT were performed with X-ray diffraction, scanning electron microscope, and energy-dispersive X-ray spectroscopy. Different compositions of alloy nanoparticles were obtained by adjusting the deposition thicknesses of Pt and Au on CNT. Pt50Au50/CNT electrode showed the highest glucose oxidation current peak among Pt, Pt70Au30, Pt50Au50, Pt30Au70, and Au/CNT electrodes while the electroactive surface areas of them are kept to be similar (average surface area=7.00cm2, coefficient of variation=0.06). The amperometric response of Pt50Au50/CNT electrode to the glucose concentrations showed a wide linear range up to 24.44mM with a high detection sensitivity of 10.71μAmM−1 cm−2. Reproducibility and long-term stability of the Pt–Au/CNT electrode were also proven in the experiments. [ABSTRACT FROM AUTHOR]

Data assimilation in complex water quality modeling is inevitably multivariate because several water quality variables interact and correlate. In ensemble Kalman filter applications, determining which variables to include and the structure of the relationships among these variables is important to achieve accurate forecast results. In this study, various analysis methods with different combinations of variables and interaction structures were evaluated under two different simulation conditions: synthetic and real. In the former, a synthetic experimental setting was formulated to ensure that issues, including incorrect model error assumption problem, spurious correlation between variables, and observational data inconsistency, would not distort the analysis results. The latter did not have such considerations. Therefore, this process could demonstrate the undistorted effects of the different analysis methods on the assimilated outputs and how these effects might diminish in real applications. Under synthetic conditions, updating a single active variable was found to improve the accuracy of the other active variables, and updating multiple active variables in a multivariate manner mutually enhanced the accuracy of the variables if proper ensemble covariance and observation data consistency were ensured. The results of the real case indicated a weakened mutual enhancement effect, and the methods in which variable localization were applied yielded the best analysis results. However, the multivariate analysis methods produced more accurate forecasting results, indicating that these methods could be superior. Therefore, it is suggested that multivariate analysis methods be considered first for water quality modeling, and the application of variable localization should be considered if significant spurious correlations and data inconsistency are present. Image 1 • Multivariate DA performance in water quality modeling was investigated. • Synthetic test was used to clarify superiority among analysis methods in EnKF. • Mutual enhancement effect among state variables was presented explicitly. • Multivariate analysis may be superior to univariate analysis on forecast accuracy. • Update strategy in multivariate DA was proposed considering variable localization. [ABSTRACT FROM AUTHOR]

Parameter identification (PI) is a cost‐effective approach for estimating the parameters of an electrochemical model for lithium‐ion batteries (LIBs). However, it requires identifiability analysis (IA) of model parameters because identifiable parameters vary with reference data and electrochemical models. Therefore, we propose a PI and IA (PIIA) framework for a robust PI that can adapt to discharge data. The IA results show that the best subset with 15 parameters is determined by the Fisher information matrix and the sample‐averaged RDE criterion under various operating conditions. The identification process based on a genetic algorithm determines the optimal parameters. The identified‐parameter model predicts voltage curves with uncertainty bounds, considering the confidence intervals of identified parameters. Further, we demonstrate that the proposed PIIA framework robustly identifies the parameters of the electrochemical model from experimental data. [ABSTRACT FROM AUTHOR]

As semiconductor device structures become more complex and sophisticated, the formation of finer and deeper patterns is required. To achieve a higher yield for mass production as the number of process steps increases and process variables become more diverse, process optimization requires extensive engineering effort to meet the target process requirements, such as uniformity. In this study, we propose an efficient process design framework that can efficiently search for optimal process conditions by combining deep learning (DL) with plasma simulations. To establish the DL model, a dataset was created using a two-dimensional (2D) hybrid plasma equipment model code for an argon inductively coupled plasma system under a given process window. The DL model was implemented and trained using the dataset to learn the functional relationship between the process conditions and their consequential plasma states, which was characterized by 2D field data. The performance of the DL model was confirmed by comparison of the output with the ground truth, validating its high consistency. Moreover, the DL results provide a reasonable interpretation of the fundamental features of plasmas and show a good correlation with the experimental observations in terms of the measured etch rate characteristics. Using the designed DL, an extensive exploration of process variables was conducted to find the optimal processing condition using the multi-objective particle swarm optimization algorithm for the given objective functions of high etch rate and its uniform distribution. The obtained optimal candidates were evaluated and compared to other process conditions experimentally, demonstrating a fairly enhanced etch rate and uniformity at the same time. The proposed computational framework substantially reduced trial-and-error repetitions in tailoring process conditions from a practical perspective. Moreover, it will serve as an effective tool to narrow the processing window, particularly in the early stages of development for advanced equipment and processes. [ABSTRACT FROM AUTHOR]

The number of cyanobacterial harmful algal blooms (cyanoHABs) has increased, leading to the widespread development of prediction models for cyanoHABs. Although bacteria interact closely with cyanobacteria and directly affect cyanoHABs occurrence, related modeling studies have rarely utilized microbial community data compared to environmental data such as water quality. In this study, we built a machine learning model, the multilayer perceptron (MLP), for the prediction of Microcystis dynamics using both bacterial community and weekly water quality data from the Daechung Reservoir and Nakdong River, South Korea. The modeling performance, indicated by the R 2 value, improved to 0.97 in the model combining bacterial community data with environmental factors, compared to 0.78 in the model using only environmental factors. This underscores the importance of microbial communities in cyanoHABs prediction. Through the post-hoc analysis of the MLP models, we revealed that nitrogen sources played a more critical role than phosphorus sources in Microcystis blooms, whereas the bacterial amplicon sequence variants did not have significant differences in their contribution to each other. Similar to the MLP model results, bacterial data also had higher predictability in multiple linear regression (MLR) than environmental data. In both the MLP and MLR models, Microscillaceae showed the strongest association with Microcystis. This modeling approach provides a better understanding of the interactions between bacteria and cyanoHABs, facilitating the development of more accurate and reliable models for cyanoHABs prediction using ambient bacterial data. • MLP model better predicted Microcystis using bacterial community than water quality. • Combining bacterial data with environmental factors enhances predictive accuracy. • Microcystis blooms are more strongly influenced by nitrogen sources than phosphorus. • Robust associations between bacteria and Microcystis are evident in both MLP and MLR. [ABSTRACT FROM AUTHOR]

In various watershed-based hydrological models, river topography information is generated by utilizing empirical expressions developed for large catchments. Therefore, identified the impact of channel geometry factors on hydrologic simulation in small watersheds and evaluation of reproducibility of river flow rate are required. We applied the Soil and Water Assessment Tool (SWAT) to small agricultural watersheds and compared measurements obtained from a river with DEM-generated values. The channel geometry factors determined using DEM were underestimated in comparison with measurements; however, using the measured channel geometry as the model input led to a 14.7% increase in the R2 and Nash–Sutcliffe coefficient (NSE) values in the daily average flow. In addition, the R2 and NSE values were more accurate by approximately 26% and 47% than the DEM-based optimized values obtained through SWAT-Calibration and Uncertainty Program (CUP)/Sequential Uncertainty Fitting ver.2 (SUFI-2) optimization, based on measured riverbed topography. • SWAT was applied to a small agricultural watershed considering riverbed topography. • The riverbed topography obtained from a river was compared with the DEM-generated. • Model evaluation showed high accuracy for the measured riverbed topography. • Channel geometry factors are sensitive on daily hydrological simulation for small watersheds. [ABSTRACT FROM AUTHOR]

Machine learning has shown great promise for simulating hydrological phenomena. However, the development of machine-learning-based hydrological models requires advanced skills from diverse fields, such as programming and hydrological modeling. Additionally, data pre-processing and post-processing when training and testing machine learning models are a time-intensive process. In this study, we developed a python-based framework that simplifies the process of building and training machine-learning-based hydrological models and automates the process of pre-processing hydrological data and post-processing model results. Pre-processing utilities assist in incorporating domain knowledge of hydrology in the machine learning model, such as the distribution of weather data into hydrologic response units (HRUs) based on different HRU discretization definitions. The post-processing utilities help in interpreting the model's results from a hydrological point of view. This framework will help increase the application of machine-learning-based modeling approaches in hydrological sciences. [ABSTRACT FROM AUTHOR]

Microalgae, including cyanobacteria and eukaryotic algae, are hotspots of primary production and play a critical role in global carbon cycling. However, these species often form blooms that poses a threat to aquatic ecosystems. Although the use of bacteria-derived cyanocides is regarded as an environmentally friendly method for controlling cyanobacterial blooms, only a few studies have examined their potential impact on ecosystems. This study is the first to explore the response of particle-attached (PA) and free-living (FL) bacteria to the dynamics of microalgal communities induced by the biological cyanocide paucibactin A. The microalgal community dynamics were divided into two distinct phases [phase I (days 0–2) and phase II (days 3–7)]. In phase I, paucibactin A caused a sudden decrease in the cyanobacterial concentration. Phase II was characterized by increased growth of eukaryotic microalgae (Scenedesmus , Pediastrum , Selenastrum , and Coelastrum). The stability of the bacterial community and the contribution of stochastic processes to community assembly were more pronounced in phase II than in phase I. The microalgal dynamics triggered by paucibactin A coincided with the succession of the PA and FL bacterial communities. The lysis of cyanobacteria in phase I favored the growth of microbial organic matter degraders in both the PA (e.g., Aeromonas and Rheinheimera) and FL (e.g., Vogesella) bacterial communities. In phase II, Lacibacter , Phycisphaeraceae , and Hydrogenophaga in the PA bacterial community and Lacibacter , Peredibacter , and Prosthecobacter in the FL bacterial community showed increased relative abundances. Overall, the FL bacterial community exhibited greater sensitivity to the two sequential processes compared with the PA bacterial community. These results highlight the need for studies evaluating the impact of biological cyanocides on aquatic ecosystems when used to control natural cyanobacterial blooms. [Display omitted] • Microalgal community dynamics caused by paucibactin A were divided into two phases. • Microalgal community dynamics caused the succession of bacterial community. • PA bacteria were more stable than FL bacteria in response to microalgal dynamics. • The lysis of cyanobacteria promoted eukaryotic microalgal growth. • Cyanobactericidal process enhanced bacterial interactions in the network. [ABSTRACT FROM AUTHOR]

A dual isotopes approach and the Bayesian isotope mixing model were applied to trace nitrogen pollution sources and to quantify their relative contribution to river water quality. We focused on two points to enhance the applicability of the method: 1) Direct measurement on the end-members to distinguish "sewage" and "manure" which used to be grouped in one pollution source as their isotope ranges overlap; 2) The Lagrangian sampling method was applied to consider the transport of nitrogen pollutants in a long river so that any fractionation process can be dealt with in the given Bayesian modeling framework. The results of the analysis confirmed the NO 3 − isotope composition in the river of interest to be within the range of NO 3 − with origins in "NH 4 + in fertilizer", "Soil N", and "Manure and sewage" pollution. This suggests that nitrogen pollution is mostly attributed to anthropogenic sources. The δ18O NO3 value follows the range +2.5∼+15.0‰, implying that NO 3 − in the river is mainly derived from nitrification, and possible nitrification in groundwater or waterfront other than surface water. The ratio of the concentration of δ15N NO3 to that of δ18O NO3 , and the corresponding regression equation indicates that the denitrification effect in surface water was insignificant during the study period. From the results of the contribution ratio of each source, improving the water quality of the discharge from the sewage treatment plants was proved to be the key factor to reduce nitrogen pollution in the river. [Display omitted] • Origins of nitrogen sources in the Yeongsan River in Korea were traced. • Dual isotope approach and the Bayesian isotope mixing model were used. • N transformation by Lagrangian sampling was considered. • Denitrification effect in surface water was insignificant during the study period. • Improving discharge from sewage treatment plants key to reduce nitrogen pollution. [ABSTRACT FROM AUTHOR]

• Data nexus was prepared from in-situ , synthetic measurement, and remote sensing data. • Convolutional attention model was designed to predict cyanobacteria cells. • Attention network delivered the importance of Chlorophyll-a map for the prediction. • Convolutional attention model was feasible to harmful algae with data assemblage. Massive cyanobacterial blooms in river water causes adverse impacts on aquatic ecosystems and water quality. Complex and diverse data sources are available to investigate the cyanobacteria phenomena, including in situ data, synthetic measurements, and remote sensing images. Deep learning attention models can process these intricate sources to forecast cyanobacteria by identifying important variables in the data sources. However, deep learning attention models for predicting cyanobacteria have rarely been studied using an assemblage of various datasets. Thus, in this study, a convolutional neural network (CNN) model with a convolutional block attention module (CNN an) was developed to predict cyanobacterial cell concentrations by using the observed cell data from field monitoring, chlorophyll-a distribution map from hyperspectral image sensing, and simulated water quality outputs from a hydrodynamic model. Then, the prediction performance of the CNN an model was compared to an environmental fluid dynamics code (EFDC) simulation and a CNN model without an attention network. The seasonal variations of the predicted cyanobacteria that was obtained from CNN an showed the best agreement with the observed variations with Nash–Sutcliffe efficiency values higher than 0.76 when compared to the EFDC and CNN predictions. The daily hydrodynamic outputs allowed the prediction of cyanobacteria cells, while the rich information of the chlorophyll-a map contributed to the improvement of the prediction performance at certain periods. Moreover, the attention network visualized the importance of the additional chlorophyll-a map and improved the CNN an model prediction performance by refining the input features. Therefore, this study demonstrated that a deep learning model with data assemblage is practically feasible for predicting the presence of harmful algae in inland water. [Display omitted] [ABSTRACT FROM AUTHOR]

A hybrid electric vehicle (HEV) is defined as a vehicle that has two or more power sources, the hybrid electric vehicle is a representative eco-friendly vehicle because it can operate efficiently with each power source and requires only a small sized electric power source. However, it is not possible to develop high efficiency HEVs without an effective energy management system (EMS), a well-designed EMS is vital in HEVs because they need to manage two power sources. Motivated by this, there are continuing efforts being made to research and establish suitable energy management strategies in order to develop high efficiency HEVs. In the past, many energy management strategies for HEVs were developed based on optimal control theory. Recently, various kinds of machine learning technologies have been applied to HEV EMS development based on breakthroughs in the fields of machine learning and artificial intelligence (AI). Machine learning is a field of research that allows computers to perform arbitrary tasks guided by data rather than explicit programming. Machine learning can be classified into supervised learning, reinforcement learning (semi-supervised learning), and unsupervised learning depending on how the training data is structured. In this study, we look at cases and studies in which machine learning techniques from each category were used to develop HEV energy management strategies. [ABSTRACT FROM AUTHOR]

Machine learning has shown great promise for simulating hydrological phenomena. However, the development of machine learning-based hydrological models requires advanced skills from diverse fields, such as programming and hydrological modeling. Additionally, data pre-processing and post-processing when training and testing machine learning models is a time-intensive process. In this study, we developed a python-based framework that simplifies the process of building and training machine learning-based hydrological models and automates the process of pre-processing of hydrological data and post-processing of model results. Pre-processing utilities assist in incorporating domain knowledge of hydrology in the machine learning model, such as the distribution of weather data into hydrologic response units (HRUs) based on different HRU discretization definitions. The post-processing utilities help in interpreting the model's results from a hydrological point of view. This framework will help increase the application of machine learning-based modeling approaches in hydrological sciences. [ABSTRACT FROM AUTHOR]

• Suspect and non-target screening were applied to identify new micropollutants. • Twenty-six substances were newly identified in domestic effluents. • Quantitative analysis measured concentrations for 70 substances. • Telmisartan, mefenamic acid, PFOA, and cimetidine were prioritized as major concern. The present study was designed to identify recently (or rarely) recognized or unreported substances (RRS or URS) contained in the effluents from water treatment plants in two industrialized urban areas, Gumi and Daegu, in Korea. In addition to 30 initial targets, 72 substances were identified through suspect and non-target screening (SNTS). Among them were 4 RRSs and 22 URSs, respectively. The quantitative analyses were applied to 35 pharmaceuticals, 15 pesticides, 13 poly-/perfluorinated alkyl substances (PFASs), 2 organophosphate flame retardants (OPFRs), 2 corrosion inhibitors, and 3 metabolites. The highest average concentration was observed for benzotriazole, followed by those for niflumic acid, and metformin. Effluents from Gumi mainly contained benzotriazole and metformin whereas niflumic acid and tramadol were the major components in effluents from Daegu. According to a scoring system based on risk relevant parameters, higher priorities were given to telmisartan, PFOA, and cimetidine. Yet, priorities for some substances were area specific (e.g., benzotriazole from Gumi, PFASs from Daegu), reflecting differences in industry profiles and populations. Many of the RRSs and URSs were recognized as potential hazards. The new identifications and evaluations should be taken into consideration for constant monitoring and management, as do the previously recognized contaminants. [ABSTRACT FROM AUTHOR]

In this study, we investigated the activity concentrations of natural (226Ra, 232Th, 40K) and artificial (137Cs) radionuclides of soils in Jeju Island, and radiation hazard indices arising from the activity concentrations. The activity concentrations of 232Th and 40K based on soil color were in the order of black volcanic ash soils (BVAS) < very dark brown volcanic ash soils (VDBAS) < dark brown soils, but 137Cs was highest in BVAS, and 226Ra was lowest in VDBAS. The radiation hazard indices calculated from the activity concentrations were negligible. In terms of annual outdoor effective dose rate (AEDRout), the contribution of radionuclides to the soils was 137Cs (5.1%) < 40K (24.7%) < 226Ra (29.0%) < 232Th (41.3%), i.e., dominated by natural radionuclides, but contributions were dependent on soil properties. [ABSTRACT FROM AUTHOR]

[Display omitted] • Increased drought frequency & nonpoint source pollution due to high carbon emissions. • Integration of SSP climate change scenarios with multiple indicators and models. • Assessing environmental impact using heat index, SPI, EMC, and environmental flow. • Identified vulnerable watersheds requiring non-point pollution management. • Practical implications for water resource management and environmental preservation. Abnormal climate phenomena that exceed conventional weather observations occur worldwide, such as severe floods, droughts, and environmental issues, and have attracted increasing attention. To avoid indiscriminate industrialization and achieve sustainable development, environmental experts have presented various opinions based on climate change scenario data. In this study, shared socioeconomic pathway (SSP) climate change scenarios that consider socioeconomic conditions were used to assess the environmental impact of extreme climate conditions on watersheds. Using the UK Earth System Modelling (UKESM1) SSP scenarios, we analyzed the following: 1) Heat index for each SSP climate change scenario using the H-Index; 2) extreme climate for each SSP climate change scenario using the standardized precipitation index (SPI); 3) the impact of extreme climate on non-point pollution using event mean concentration (EMC); and 4) the environmental flow caused by extreme climate by combining the K-water Distributed RUnoff Model (K-DRUM) and global environmental flow calculator (GEFC) models. According to the heat index analysis regarding SSPs climate change scenarios, the heatwave index will continue to rise if high carbon emissions persist. Temperature serves as an important indicator that has the most meaningful impact on ecosystems. The SPI analysis showed an increase in "extremely dry" conditions, and overall, more severe droughts are likely to occur due to high carbon emission-induced climate change. The nonpoint source pollution is also higher in climate change scenarios with high carbon emissions. The extreme drought event assessment using SPI and environmental flow revealed a shift from the "moist conditions" category to grade C. Our assessments in this study conclusively indicate that drought frequency will increase along with nonpoint source pollution. Furthermore, environmental flows will shift to grade C, resulting in the disappearance of some sensitive ecological species and an increase in invasive species. These analyses determined that high carbon emission-induced climate change caused leads to severe droughts and significant alterations in the overall water circulation, thereby complicating water resource management. Furthermore, we identified watersheds that are highly vulnerable to climate change and designated these as "mid-watersheds" that first require non-point pollution management. Here, the aquatic ecosystem environment can be affected by climate change without any artificial environmental influence. Various research-based analysis methods based on modeling have been presented using SSP climate change scenarios, and are highly beneficial for establishing policies and coping strategies for environmental preservation and sustainable development. [ABSTRACT FROM AUTHOR]

Pharmaceuticals and personal care products (PPCPs) in the Yeongsan River, Korea were prioritized via suspect and non-target analysis using LC-HRMS (QExactive plus Orbitrap) followed by semi-quantitative analysis to confirm the priority of PPCPs. A scoring and ranking system for prioritization was suggested based on occurrence frequency and chromatographic peak area or concentration. Through suspect and non-target screening, more than 50 PPCPs were tentatively identified and ranked by the scoring system. Among them, 28 substances were finally confirmed using reference standards. For estimating concentration, 26 confirmed PPCPs and 12 additional substances not included in the first ranking were semi-quantitatively analyzed. We found that carbamazepine, metformin, paraxanthine, naproxen, and fluconazole occurred 100% of the time above the limit of quantification in 14 samples, whereas carbamazepine, metformin, paraxanthine, caffeine, and cimetidine showed maximum concentrations above 1000 ng/L. Thus, in the final prioritization list, carbamazepine, metformin, and paraxanthine shared first place, followed by caffeine, cimetidine, lidocaine, naproxen, cetirizine, climbazole, fexofenadine, tramadol, and fluconazole, with scores of 100 or above. We suggest that these 12 PPCPs are the most highly exposable substances, and thus must be considered in future water monitoring in the Yeongsan River. [ABSTRACT FROM AUTHOR]

In this paper, we report a miniaturized low-power wireless remote environmental monitoring system. This system has been developed for on-site monitoring of water pollution by heavy-metal ions. The system is composed of three parts: an electrochemical sensor module using microfabricated electrodes for detecting heavy-metal contamination in sample water; a custom potentiostat module including readout circuitry, analog-to-digital converter and microcontroller; and a radio frequency (RF) module for sending detected signals to a base station through wireless communication. The electrochemical sensor module is implemented using microfabricated mercury working electrodes (WEs), solid-state reference electrode (SSRE), and platinum counter electrode (CE). For the low-power operation, direct frequency-shift keying (FSK) modulation and simple binary FSK demodulation methods are used for RF module which is realized using 0.18 μm CMOS technology. All the modules are hybrid integrated in a printed circuit board (PCB) and low-power consumption below 1 mW has been achieved. [Copyright &y& Elsevier]

This study introduces a semi-empirical algorithm to estimate the extent of the phycocyanin (PC) concentration in eutrophic freshwater bodies; this is achieved by studying the reflectance characteristics of the red and near-red spectral regions, especially the shifting of the peak near 700 nm towards longer wavelengths. Spectral measurements in a darkroom environment over the pure-cultured cyanobacteria Microcystis showed that the shift is proportional to the algal biomass. A similar proportional trend was found from extensive field measurement data. The data also showed that the correlation of the magnitude of the shift with the PC concentration was greater than that with chlorophyll-a. This indicates that the characteristic can be a useful index to quantify cyanobacterial biomass. Based on these observations, a new PC algorithm was proposed that uses the remote sensing reflectance of the peak band around 700 nm and the trough band around 620 nm, and the magnitude of the peak shift near 700 nm. The efficacy of the algorithm was tested with 300 sets of field data, and the results were compared to select algorithms for the PC concentration prediction. The new algorithm performed better than the other algorithms with respect to most error indices, especially the mean relative error, indicating that the algorithm can reduce errors when PC concentrations are low. The algorithm was also applied to a hyperspectral dataset obtained through aerial imaging, in order to predict the spatial distribution of the PC concentration in an approximately 86 km long reach of the Nakdong River. [ABSTRACT FROM AUTHOR]

Solid fuelization technology can increase the heating value of sewage sludge such that it can be utilised as a fossil fuel substitutes. Reducing landfilling of bottom and fly ash resulting from heavy metals contained in sewage sludge is challenging. Hence, combustion melting technology (CMT), which can discharge bottom ash in the form of slag, has been proposed herein as an alternative to the conventional incineration technology. However, further research is required to improve the flowability of slag. Applicability of CMT for the stable treatment of heavy metals in the ash generated during the energisation of sewage sludge solid fuel has been reviewed. The change in the degree of fluidity was identified via a laboratory-scale fluidity measurement experiment following changes in melting temperature, mixing ratio of sewage sludge and sawdust, and basicity. The pouring index (PI) of sewage sludge solid fuel (pellet) was maintained at a level of about 60% at a basicity index of 0.8. Based on the results, the slagging rates and volume reduction rates, exhaust gas analysis, and heavy metal elution characteristics under oxygen enrichment were derived from a 2 ton/day combustion melting pilot plant experiment; thereafter, the feasibility of combustion melting of sewage sludge solid fuel was determined. [ABSTRACT FROM AUTHOR]

Composted livestock manures, in both solid and liquid form, are used as fertilizers in cropland. However, excess solid and liquid manures in agricultural watersheds are considered as nonpoint pollution sources because of their high nutrient and heavy metal contents of, as well as their antibiotic contents, especially veterinary antibiotics (VAs). In this study, 21 VAs under nine classes (i.e., cephems, ionophores, lincosamides, penicillins, pleuromutilins, quinolones, streptogramins, sulfonamides, and tetracyclines) found in agricultural watersheds were simultaneously analyzed via UHPLC-q-orbitrap high-resolution mass spectrometry using an on-line solid-phase extraction system. The residues of VAs in the surface water of two intensive livestock rearing watersheds (Cheongmi and Gwangcheon streams) in Korea were successfully quantified, and the values were found to range from 1.84 ± 0.42 ng L−1 to 835.6 ± 31.9 ng L−1. Time lags of 2–3 months were observed between the periods of liquid manure application and the periods with the maximum concentrations of VAs. In both watersheds, samples from points close to areas with extensive application of liquid manure exhibited high concentrations of most of the 21 VAs. Between the watersheds, the one with heavier application of liquid manure showed higher concentrations of the target VAs. To the best of our knowledge, this study represents the first attempt at evaluating the correlation between liquid manure application and environmental occurrence of VAs in surface water. The findings reveal that liquid manure application plays an important role in introducing VAs into aquatic environments. Unlabelled Image • Spatiotemporal trends of veterinary antibiotics in farming watersheds were analyzed. • Agricultural activities affected seasonal variations of antibiotic concentrations. • Areas with heavier liquid manure application showed higher antibiotic concentrations. • Liquid manure application can affect residual antibiotic concentrations. [ABSTRACT FROM AUTHOR]

• The natural wetland is polluted by more than 30 synthetic organic substances. • Fungicides are dominant pollutants, followed by insecticide and herbicides. • Suspect and non-target screening revealed 48 TPs of pesticides in the wetland. • QSAR estimations indicate that most TPs are less risky than their parents. • The natural wetland is capable of mitigating the risk of pollutants by forming various TPs. Natural wetlands have been recognized as a natural reactor for degradation and elimination of environmental pollutants. The Upo Wetland, the largest inland wetland in Korea, is mainly surrounded by agricultural lands and it is susceptible to contamination from excess nutrient loads and synthetic organic contaminants (SOCs) (e.g., pesticides). The aim of this study was to identify major SOCs in the wetland and evaluate their degradation. We used high resolution mass spectrometry (HRMS) with a two-step analysis approach (i.e., 1st analysis for target measurement along with suspect and non-target screening (SNTS) and 2nd analysis for complimentary suspect screening) to identify and quantify the transformation products (TPs) of the identified parent SOCs. Quantitative analysis of 30 targets, mainly including pesticides, showed that fungicides were the major SOCs detected in the wetland, accounting for about 50% of the composition ratio of the total SOCs quantified. Orysastrobin occurred at the highest mean concentration (>700 ng/L), followed by two other fungicides, carbendazim and tricyclazole. The first analysis (SNTS) tentatively identified 39 TPs (30 by suspect, 9 by non-target screening) of 14 parent pesticides. Additionally, the second analysis (complimentary suspect screening) identified 9 more TPs. Among the 48 total TPs identified, 7 were confirmed with reference standards. The identification of the remaining TPs had a high confidence level (e.g., level 2 or 3). Regarding transport though the wetland, most TPs showed greater peak area ratios (i.e., the relative portion of chromatographic area of the TPs to the parent compound) at the outlet point of the wetland compared to the inlet point. The risk quotient, which was calculated using the concentrations of parent compounds, decreased toward the outlet, demonstrating the degradation capacity of the wetland. The estimates for biodegradability, hydrophobicity, and toxicity by an in-silico quantitative structure-activity relationship (QSAR) model indicated a lower half-life, lower logD OW , and greater effect concentration for most TPs compared to the parent compounds. Based on these results, we conclude that natural wetlands play a role as an eco-friendly reactor for degrading SOCs to form numerous TPs that are lower risk than the parent compounds. Image, graphical abstract [ABSTRACT FROM AUTHOR]

• Synthetic water quality data were generated by EFDC–NIER model. • Convolutional neural network model was developed to predict Microcystis biomass. • Convolutional neural network provided a good short-term prediction of Microcystis. • Convolutional neural network predictions were sensitive to temporal density of data. Machine learning modeling techniques have emerged as a potential means for predicting algal blooms. In this study, synthetic spatio-temporal water quality data for a river section were generated with a 3D water quality model and used to investigate the capability of a convolutional neural network (CNN) for predicting harmful cyanobacterial blooms. The CNN model displayed a reasonable capacity for short-term predictions of cyanobacteria (Microcystis) biomass. In the nowcasting of Microcystis, the CNN performance had a Nash-Sutcliffe Efficiency (NSE) of 0.87. An increase in the forecast lead time resulted in a decrease in the prediction accuracy, reducing the NSE from 0.87 to 0.58. As the spatial observation density increased from 20% to 100% of the input image grids, the CNN prediction NSE had improved from 0.70 to 0.84. Adding noise to the data resulted in accuracy deterioration, but even at the noise amplitude of 10%, the accuracy was acceptable for some applications, with NSE = 0.76. Visualization of the CNN results characterized its performance variations across the studied river reach. Overall, this study successfully demonstrated the capability of the CNN model for cyanobacterial bloom prediction using high temporal frequency images. Image, graphical abstract [ABSTRACT FROM AUTHOR]

Hyperspectral image sensing can be used to effectively detect the distribution of harmful cyanobacteria. To accomplish this, physical- and/or model-based simulations have been conducted to perform an atmospheric correction (AC) and an estimation of pigments, including phycocyanin (PC) and chlorophyll-a (Chl-a), in cyanobacteria. However, such simulations were undesirable in certain cases, due to the difficulty of representing dynamically changing aerosol and water vapor in the atmosphere and the optical complexity of inland water. Thus, this study was focused on the development of a deep neural network model for AC and cyanobacteria estimation, without considering the physical formulation. The stacked autoencoder (SAE) network was adopted for the feature extraction and dimensionality reduction of hyperspectral imagery. The artificial neural network (ANN) and support vector regression (SVR) were sequentially applied to achieve AC and estimate cyanobacteria concentrations (i.e., SAE-ANN and SAE-SVR). Further, the ANN and SVR models without SAE were compared with SAE-ANN and SAE-SVR models for the performance evaluations. In terms of AC performance, both SAE-ANN and SAE-SVR displayed reasonable accuracy with the Nash–Sutcliffe efficiency (NSE) > 0.7. For PC and Chl-a estimation, the SAE-ANN model showed the best performance, by yielding NSE values > 0.79 and > 0.77, respectively. SAE, with fine tuning operators, improved the accuracy of the original ANN and SVR estimations, in terms of both AC and cyanobacteria estimation. This is primarily attributed to the high-level feature extraction of SAE, which can represent the spatial features of cyanobacteria. Therefore, this study demonstrated that the deep neural network has a strong potential to realize an integrative remote sensing application. [ABSTRACT FROM AUTHOR]

River water quality is one of the main challenges that societies face during the 21st century. Accurate and reliable real-time prediction of water quality is an effective adaptation measure to counteract water quality issues such as accidental spill and harmful algae blooms. To improve accuracy and skill of water quality forecasts along the Yeongsan River in South Korea three different ensemble data assimilation (DA) methods have been investigated: the traditional Ensemble Kalman Filter (EnKF) and two related algorithms (Dud-EnKF and EnKF-GS) that offer either possibilities to improve initial conditions for non-linear models or reduce computation time (important for real-time forecasting) by using a (smaller) time-lagged ensemble to estimate the Kalman gain. Twin experiments, assimilating synthetic observations of three algae species and phosphate concentrations, with relatively small ensemble sizes showed that all three DA methods improved forecast accuracy and skill with only subtle difference between the methods. They all improved the model accuracy at downstream locations with very similar performances but due to spurious correlation, the accuracy at upstream locations was somewhat deteriorated. The experiments also showed no clear trend of improvement by increasing the ensemble size from 8 to 64. The real world experiments, assimilating real observations of three algae species and phosphate concentrations, showed that less improvement was achieved compared to the twin experiments. Further improvement of the model accuracy may be achieved with different state variable definitions, use of different perturbation and error modelling settings and/or better calibration of the deterministic water quality model. Image 1 • State updating has been tested using an operational water quality model in Korea. • A small EnKF ensemble, practical in operational use, shows improved forecast accuracy. • Twin experiments provide insight in the performance of data assimilation methods. • All ensemble DA methods improve model results downstream of the assimilated location. • At upstream locations the model state updates may suffer from spurious correlation. [ABSTRACT FROM AUTHOR]

Hyperspectral imagery (HSI) provides substantial information on optical features of water bodies that is usually applicable to water quality monitoring. However, it generates considerable uncertainties in assessments of spatial and temporal variation in water quality. Thus, this study explored the influence of different optical methods on the spatial distribution and concentration of phycocyanin (PC), chlorophyll-a (Chl-a), and total suspended solids (TSSs) and evaluated the dependence of algal distribution on flow velocity. Four ground-based and airborne monitoring campaigns were conducted to measure water surface reflectance. The actual concentrations of PC, Chl-a, and TSSs were also determined, while four bio-optical algorithms were calibrated to estimate the PC and Chl-a concentrations. Artificial neural network atmospheric correction achieved Nash-Sutcliffe Efficiency (NSE) values of 0.80 and 0.76 for the training and validation steps, respectively. Moderate resolution atmospheric transmission 6 (MODTRAN 6) showed an NSE value >0.8; whereas, atmospheric and topographic correction 4 (ATCOR 4) yielded a negative NSE value. The MODTRAN 6 correction led to the highest R2 values and lowest root mean square error values for all algorithms in terms of PC and Chl-a. The PC:Chl-a distribution generated using HSI proved to be negatively dependent on flow velocity (p-value = 0.003) and successfully indicated cyanobacteria risk regions in the study area. [ABSTRACT FROM AUTHOR]