Normalized Root Mean Research Articles

Non-thermal atmospheric jet plasma effects on the morphological, electrical and optical properties of nanocrystalline silicon

In this study, a non-thermal atmospheric argon plasma jet (NTAPJ) was designed and used to treat the porous silicon layer, prepared by photoelectrochemical etching (PECE). The porous silicon layers were distinguished before and after the cold plasma treatment utilized by X-ray diffraction (XRD), atomic force microscopy (AFM), scanning electron microscopy (SEM), photoluminescence (PL), and dark and photo current-voltage characteristics. The results confirmed the nanostructure of the formed porous silicon. After the treatment with argon plasma, the intensity of the XRD peak increased, and the porosity decreased from 93% to 62%. Additionally, porous silicon's average roughness (Sa) and root mean square (Sq) were reduced from 60.10 to 12.31nm and 73.70 to 16.53nm, respectively. The current-voltage characteristic indicated the increase in photocurrent and the ideality factor, as well as a decrease in dynamic resistance. The photoluminescence results showed a change in PL peaks, represented by a blue shift, an increase in intensity, and a vanishing of another peak.

Improving Localization in Wireless Sensor Networks for the Internet of Things Using Data Replication-Based Deep Neural Networks.

Localization is one of the most challenging problems in wireless sensor networks (WSNs), primarily driven by the need to develop an accurate and cost-effective localization system for Internet of Things (IoT) applications. While machine learning (ML) algorithms have been widely applied in various WSN-based tasks, their effectiveness is often compromised by limited training data, leading to issues such as overfitting and reduced accuracy, especially when the number of sensor nodes is low. A key strategy to mitigate overfitting involves increasing both the quantity and diversity of the training data. To address the limitations posed by small datasets, this paper proposes an intelligent data augmentation strategy (DAS)-based deep neural network (DNN) that enhances the localization accuracy of WSNs. The proposed DAS replicates the estimated positions of unknown nodes generated by the Dv-hop algorithm and introduces Gaussian noise to these replicated positions, creating multiple modified datasets. By combining the modified datasets with the original training data, we significantly increase the dataset size, which leads to a substantial reduction in normalized root mean square error (NRMSE). The experimental results demonstrate that this data augmentation technique significantly improves the performance of DNNs compared to the traditional Dv-hop algorithm at a low number of nodes while maintaining an efficient computational cost for data augmentation. Therefore, the proposed method provides a scalable and effective solution for enhancing the localization accuracy of WSNs.

When more is more: Utilizing finite element analysis to assess chest wall injury stability after surgical stabilization of all rib fractures versus only a portion of the rib fractures.

Surgical stabilization of rib fractures (SSRF) continues to gain acceptance. Controversary exists around the number of rib fractures needing stabilization. We sought to analyze chest wall stability (CWS) after SSRF using finite element analysis (FEA) modeling in various rib fracture patterns, hypothesizing not stabilizing all fractures leaves the chest wall unstable. FEA thoracic model development was described previously. Two fracture patterns with three case scenarios each were defined for right ribs 4 to 9. Fracture Pattern 1; Case 1-all 6 ribs with lateral fractures and no stabilization; Case 2-all six fractures stabilized; Case 3-only fractures 5 to 7 were stabilized. Fracture Pattern 2; Case 4-all six ribs fractured in a flail pattern (anterior-lateral and posterior-lateral) and no stabilization; Case 5-all 12 fractures stabilized; Case 6-only six anterior-lateral fractures were stabilized. Three assessment criteria were used to quantify thoracic motion: normalized mean absolute error (NMAE), normalized root mean square error (NRMSE), and normalized interfragmentary motion (NIFM). Fracture Pattern 1: Case 1-NMAE and NRMSE analysis demonstrated significant loss of CWS up to 50% with left axial rotation; Case 2-CWS almost completely returned to nonfractured state; Case 3-CWS loss up to 37%. Fracture Pattern 2: Case 4-up to 49% of CWS lost with right axial rotation; Case 5-less than 3% CWS lost; Case 6-over 40% CWS lost. For both fracture patterns, when stabilizing all fractures, NIFM decreased by 95%. In Case 3, NIFM decreased by 56% and in Case 6, NIFM increased by 1% at the non-stabilized fracture line. Stabilizing all rib fractures significantly improves CWS. Not stabilizing both fractures of a flail segment worsens motion at the non-stabilized fractures. Therapeutic/Care Management; Level IV.

Estimation of state of charge for polymer solid-state batteries: Ensemble learning models and temperature impact study

In addition to the research and development of solid electrolytes to improve battery performance, an efficient battery management system (BMS) is a must to ensure safe use and extend battery life, and state of charge (SOC) estimation is critical in the BMS. This paper presents a novel temperature-influenced method for estimating the SOC of polymer-based solid-state batteries in real-time, using machine learning to capture the relationship between SOC and battery characteristics at different temperatures. The results show that accurate SOC estimation results can be achieved at different temperatures, with an average root mean square error of 1.42 %. Furthermore, the method uses Shapley Additive explanation (SHAP) to reduce the degree of black box nature of the model and analyzes the electrochemical processes inside the battery at different temperatures through relaxation time distribution. Accurate estimates of SOC and guidelines for appropriate operating temperatures can serve as a basis for BMS decision-making and improve the lifetime of polymer-based solid-state batteries.

Gap filling of missing satellite data from MODIS and CMEMS for chlorophyll-a in the waters of Aceh, Indonesia

The motivation behind our study is to identify a robust method to enhance the accuracy of missing data, particularly chlorophyll-a data, which often goes undetected due to various factors. This study analyzes chlorophyll-a concentrations and sea level changes due to tides using three methods: Linear Interpolation, Fillgaps, and Modified Fillgaps. Two experiments were conducted: Experiment I involved random data removal (60% and 70%), and Experiment II combined sequential and random data removal (25% sequentially on the right, 35% and 45% randomly on the left). In Experiment I, the Modified Fillgaps method showed high correlation coefficients (up to 0.96) between original and reconstructed data, demonstrating its effectiveness in accurately filling significant data gaps. This method also exhibited low Root Mean Square Error and Mean Absolute Error values, confirming its predictive precision. In Experiment II, despite structured and realistic data loss patterns, the method maintained high correlation and low prediction errors, with low Normalized Root Mean Squared Error and Mean Absolute Percentage Error values, further validating its reliability. Additionally, the method excelled in two-dimensional chlorophyll-a maps, outperforming Linear Interpolation and Fillgaps methods in scenarios with 50% and 60% data loss, achieving higher correlation and lower prediction errors. These findings are crucial for environmental and climatological studies relying on satellite-derived data, confirming the Modified Fillgaps method as the most reliable and effective for handling significant data loss in chlorophyll-a map analyses. Future research should explore its application to other environmental data types and more complex data loss patterns.

Cross‐equalization for time‐lapse sparker seismic data

AbstractTime‐lapse seismic data processing is an important technique for observing subsurface changes over time. The conventional time‐lapse seismic exploration has been conducted using a large‐scale exploration system. However, for efficient monitoring of shallow subsurface, time‐lapse monitoring based on the small‐scale exploration system is required. Small‐scale exploration system using a sparker source offers high vertical resolution and cost efficiency, but it faces challenges, such as inconsistent waveforms of sparker sources, inaccurate positioning information and a low signal‐to‐noise ratio. Therefore, this study proposes a data processing workflow to preserve the signal and enhance the repeatability of small‐scale time‐lapse seismic data acquired using a sparker source. The proposed workflow has three stages: pre‐stack, post‐stack and machine learning–based data processing. Conventional seismic data processing methods were applied to enhance the quality of the sparker seismic data during the pre‐stack data processing stage. In the post‐stack processing stage, the positions and energy correction were performed, and the machine learning–based data processing stage attenuated random noise and applied a matched filter. The data processing was performed using only the seismic signals recorded near the seafloor, and the results confirmed the improvement in the repeatability of the entire seismic profile, including that of the target area. According to the repeatability quantification results, the predictability increased and the normalized root mean square decreased during data processing, indicating improved repeatability. In particular, the repeatability of the data was greatly improved through vertical correction, energy correction and matched filtering approaches. The processing results demonstrate that the data processing method proposed in this study can effectively enhance the repeatability of high‐resolution time‐lapse seismic data. Consequently, this approach could contribute to a more accurate understanding of temporal changes in subsurface structure and material properties.

Advanced mathematical modeling for preciseestimation of CT energy spectrum using a calibration phantom

PurposeThis research aims to develop an advanced mathematical model using a CT calibration phantom to accurately estimate the CT energy spectrum in clinical settings, enhancing imaging quality and patient dose management. MethodsData were collected from a CT scanner using a CT calibration phantom at various energy levels (80, 100, 120, and 135 kVp). The data was optimized to refine the energy spectrum model, followed by cross-validation with Monte Carlo simulations. ResultsThe developed model demonstrated high precision in estimating the CT energy spectrum at all tested energy levels, with R-squared values above 0.9738 and an R-squared value of 0.9829 at 100 kVp. The model also showed low Normalized Root Mean Square Deviation (NRMSD) ranging from 0.6698 % to 1.8745 %. The Mean Energy Difference (ΔE) between the estimated and simulated spectrum consistently remained under 0.01 keV. These results were comparable to recent studies, which reported higher NRMSD and ΔE. ConclusionsThis study presents a significantly improved model for estimating the CT energy spectrum, offering greater accuracy than existing models. Its strengths include high precision and the use of standard equipment and algorithmic values. While the current use of 13 plugs is adequate, incorporating plugs with varied densities could enhance accuracy. This model has potential for improving imaging quality and optimizing patient dosing in clinical applications. Future trends may include extending energy spectrum estimation to megavoltage domains and integrating technologies like EPID and MVCT for better dose distribution prediction in high-energy photon beam therapy.

Engineering region flexibility of cellobiohydrolase Ⅰ for efficient hydrolysis of cellulose based on molecular dynamics simulation

Cellobiohydrolase, a workhorse enzyme for cellulose degradation, has the potential for application in biofuels, biochemicals, biomaterials, etc. However, highly efficient cellobiohydrolase is desirable to improve the economic feasibility of industrial applications. In this work, two variants TrCel7AT97A and TrCel7AK166A were designed to increase the flexibility of the entrance of the catalytic tunnel. We performed the molecular dynamics simulations using NAMD for 100 ns at 300 K in the NPT (isothermal–isobaric) ensemble and then we found that the region flexibility at the catalytic tunnel entrances of TrCel7AT97A and TrCel7AK166A was improved, compared to that of TrCel7A, based on the calculation of the average root mean square fluctuation. The specific activities of TrCel7AT97A and TrCel7AK166A were increased by 37 % and 68 %, respectively, compared with that of TrCel7A at 45 °C for 12 h. An effective approach named revive amino acid flexibility for enzyme activity based on amino acid flexibility was proposed for protein engineering of cellobiohydrolase TrCel7A.

Effects of water quality and nitrogen on wheat productivity: Experimental and modelling study using the SALTMED model

AbstractThe effects of different water qualities and nitrogen doses need to be investigated for wheat growth to determine the optimal management strategy for sustaining wheat production potential to ensure human food security. Therefore, a 2‐year study (2020–2021 to 2021–2022) was conducted on wheat irrigated with different water qualities, including canal water (Ca), tubewell water (Tu) and mixed Ca‐Tu water (Mx), each fertilized with two nitrogen doses, that is, N75 = 75 kgN.ha−1 and N100 = 100 kgN.ha−1. Ca⨯N100 performed best, with 5.12 t.ha−1 grain yield, 11.60 t.ha−1 biomass, 102.83 cm plant height and 1.32 kg.m−3 water productivity. The best values of the root mean square error (RMSE), normalized root mean square error (NRMSE), coefficient of determination (R2) and coefficient of residual mass (CRM) were 0.11, 0.12, 0.93 and −0.004, respectively, for calibration and 0.13, 0.15, 0.87 and ±0.01, respectively, for validation of the SALTMED model. The scenario simulation was performed for additional levels of water salinity (electrical conductivity [EC] = 8 and 12 dS.m−1) and nitrogen doses (50, 125, 150 and 175 kgN.ha−1). The results revealed improvements in wheat grain yield of 107%, 16% and −6% at EC = 8 dS.m−1 and 125%, 31% and 5% at EC = 12 dS.m−1, while the improvements in biomass were 113%, 22% and −2% at EC = 8 dS.m−1 and 137%, 29% and 8% at EC = 12 dS.m−1 with increasing nitrogen doses from 50–125 kg.ha−1, 125–150 kg.ha−1 and 150–175 kg.ha−1, respectively. It is recommended that high‐quality water with the lowest possible EC and nitrogen applications of up to 150 kgN.ha−1 be adopted for wheat production in semi‐arid areas of Punjab, Pakistan.

The L1 spino-pelvic (L1SP) angle: a simplified approach for the assessment of the PI-LL mismatch in hip surgery

Introduction: Pelvic incidence - lumbar lordosis (PI-LL) mismatch is often considered when assessing spinopelvic alignment in the sagittal plane. The mismatch is conventionally obtained by measuring 2 separate angles on lateral spinopelvic radiographs. This study describes a simplified approach for assessing spinopelvic mobility and measuring the PI-LL mismatch through the evaluation of the L1-spinopelvis angle (L1SP). Methods: 96 standing lateral radiographs were obtained from consecutive patients presenting for total hip arthroplasty between November 2020 and July 2021. 3 operators were recruited to annotate landmarks on digital radiographs. Correlation analysis and error analysis were applied. Measurement reproducibility was assessed using intraclass correlation coefficient (ICC). Results: The correlation coefficients of the 3 variables were respectively 0.87 for PI, 0.94 for LL, and 0.96 for L1SP. The normalised root mean square error between the 2 measurement sets was 9.96% for PI, 5.97% for LL, and 4.41% for L1SP. The absolute error was 3.49° ± 4.63° for PI, 3.23° ± 3.78° for LL, 2.68° ± 3.19° for PI-LL conventional, and 2.35° ± 2.88° for PI-LL via L1SP, respectively. In terms of reproducibility, measurement of L1SP outperformed that of PI and LL (ICC = 0.97 versus 0.83 and 0.93, respectively). Conclusion: The simplified L1SP method, through the measurement of a single angle, produced similar measurements to the conventional PI-LL method. The measurement repeatability between operators was improved using the L1SP method. From a clinical practice perspective, both methods are equivalent. The new method is readily reproducible using commercially available PACS software during preoperative templating.

Implementation of a multistage predictive energy management strategy considering electric vehicles using a novel hybrid optimization technique

In this paper, a novel artificial intelligence (AI)-based energy management strategy across three levels is designed for isolated microgrids. During the initial phase, AI is not employed. A precise and rapid-response microcontroller, namely the FPGA, is utilized. The performance of the FPGA is experimentally investigated under various operation conditions. In the second phase, AI plays a crucial role in enhancing both the reliability and economic effectiveness of the system. The multi-objective snake optimization (SO) algorithm is employed to attain high technical and economic performance within predefined constraints. Three objective functions are involved in this phase, focusing on the minimization of operational costs, loss of power supply probability (LPSP), and electrical energy losses in the dummy load. In the third level, a novel control strategy-based coordinated model predictive control is presented as part of the proposed AI-embedded energy management strategy. A hybrid optimization algorithm combining the multilayer feed-forward neural networks (MFFNN) and SO algorithms is proposed to predict the output power of backup sources. The microgrid under consideration is supported by a hybrid backup system comprising battery energy storage systems (BESS), electric vehicle (EV) batteries, and fuel cells (FCs). The effectiveness of this hybrid algorithm is evaluated through comparison with many other algorithms, aiming to assess its performance in accurately predicting the output power of backup sources. The results show the feasibility of using AI to achieve efficient operation and energy management in microgrids. The proposed MFFNN-SO algorithm achieves the best performance, as the normalized root mean square error (NRMSE) for FCs, BESS, and EV is about 0.1296 %, 0.0094 %, and 0.1304 %, respectively. The SO algorithm achieves a notable 6.3361% reduction in operating costs, resulting in a final operational cost of 166.4811 $.

Monitoring river discharge from space: An optimization approach with uncertainty quantification for small ungauged rivers

The number of in-situ stations measuring river discharge, one of the Essential Climate Variables (ECV), is declining steadily, and numerous basins have never been gauged. With the aim of improving data availability worldwide, we propose an easily applicable and transferable approach to estimate reach-scale discharge solely using remote sensing data that is suitable for filling gaps in the in-situ network. We combine 20 years of satellite altimetry observations with high-resolution satellite imagery via a hypsometric function to observe large portions of the reach-scale bathymetry. The high-resolution satellite images, which are classified using deep learning image segmentation, allow for detecting small rivers (narrower than 100m) and can capture small width variations. The unobserved part of the bathymetry is estimated using an empirical width-to-depth function. Combined with precise satellite-derived slope measurements, river discharge is calculated at multiple consecutive cross-sections within the reach. The unknown roughness coefficient is optimized by minimizing the discharge differences between the cross-sections. The approach requires minimal input and approximate boundary conditions based on expert knowledge but is not dependent on calibration. We provide realistic uncertainties, which are crucial for data assimilation, by accounting for errors and uncertainties in the different input quantities. The approach is applied globally to 27 river sections with a median normalized root mean square error of 12% and a Nash–Sutcliffe model efficiency of 0.560. On average, the 90% uncertainty range includes 91% of the in-situ measurements.

Industrial activated sludge model identification using hyperparameter-tuned metaheuristics

This study focuses on the parameter estimation of an industrial activated sludge model using hyperparameter-tuned metaheuristic techniques. The data used in this study were collected on-site from a textile industry wastewater treatment plant. A Modified Activated Sludge Model (M-ASM) was the 'first-principle model’ selected and implemented with suitable assumptions. Advanced metaheuristic techniques, as Adaptive Tunicate Swarm Optimization (ATSO), Whale Optimization Algorithm (WOA), Rao-3 Optimization (Rao-3) and Driving Training Based Optimization (DTBO) were implemented. The hyperparameter tuning was performed with Bayesian Optimization (BO). Optimized metaheuristic algorithms were implemented for model-parameter identification. The Bayesian optimized Rao-3(BO-Rao-3) algorithm provided the best validation results, with a Mean Absolute Percentage Error (MAPE) value of 7.0141 and Normalized Root Mean Square Error (NRMSE) value of 0.2629. It also had the least execution time. BO-Rao-3 is 0.93% to 4.7% better than the other implemented hyperparameter-tuned metaheuristic techniques.

Implementasi Data Mining Prediksi Penjualan Produk Semen Menggunakan Metode Linear Regression (Studi Kasus PT. Toyo Mortar Indonesia)

The cement industry in Indonesia plays a vital role in infrastructure development and the construction sector. PT. Toyo Mortar Indonesia, established in 2015 in the Tangerang industrial area, focuses on the production of instant mortar cement. However, the national cement industry faces the challenge of declining sales due to the pandemic, which has affected people's purchasing power and slowed down property and infrastructure projects. To achieve future growth potential, the right strategy is needed in sales planning. This study implements data mining using the linear regression method with the RapidMiner application to predict cement product sales. The data used includes historical sales data and product information from PT. Toyo Mortar Indonesia. In the tests conducted, the data was processed using 12 variables, of which 10 variables influenced the prediction results: TM-168, TM-178, TM-188, TM-189, TM-198, UM-100, UM-101, UM BIRU, raw materials, and other sales. The test results show that the Linear Regression method can predict cement product sales with an average Root Mean Square Error (RMSE) value of 493,125.701 with a standard deviation of 126,330.525. The average Absolute Error is 303,270.965 with a standard deviation of 46,207.586, and the average relative error is 1.89% with a standard deviation of 0.36%, indicating very good prediction accuracy. This can help the company in making decisions related to sales planning using the available sales data.

Rapid distance estimation of odor sources by electronic nose with multi-sensor fusion based on spiking neural network

The inference of odor source locations holds great potential value in chemical leak detection, explosive searches and flammable substance detection. Sensor response and derivative changes are common indicators used to estimate source distance. However, most studies often rely on single-feature methods using a single sensor. In this study, a novel distance indicator based on a spiking neural network (SNN) that integrates multiple sensors is proposed. SNN was introduced for effective multi-sensor data fusion, which can provide accurate and rapid source distance estimation. The indicator's accuracy and quick detection capabilities were tested using a publicly available wind tunnel dataset and compared with single-sensor indicators. Additionally, an olfactory robot platform was constructed to test the method's real-time performance. Results demonstrate that this method effectively utilizes sensor arrays to estimate source distance, with an average root mean square error (RMSE) of less than 0.1 m across different wind speeds, outperforming single-sensor-based indicators. Additionally, the model exhibits good distance estimation performance within a relatively short detection time, with an average RMSE of 0.118 m. Practical testing on a mobile robot platform confirms the applicability of this distance indicator for real-time distance monitoring. In conclusion, the proposed distance indicator enables effective distance estimation for odor sources, providing a promising method for odor source localization and olfactory navigation.

A centralized power prediction method for large-scale wind power clusters based on dynamic graph neural network

The uncertainty of the wind process leads to the randomness of its regional propagation, and the spatial correlation between nearby wind farms also shows a dynamic change tendency under the influence of wind direction. To consider the influence of dynamic spatial correlation on wind power prediction modeling, a short-term wind power prediction method based on a dynamic graph neural network is proposed. First, the topology graph of the wind farm cluster was established to represent the correlation of wind farms based on graph theory. Then, a dynamic spatiotemporal graph neural network was constructed to adapt graph topology by node embedding, which can explore the changing characteristics of spatial correlation among wind farms. Finally, we proposed a decoupling error model that can quantify the proportion of errors caused by the modeling process, which can assist in evaluating the performance of predictive models. We conducted experiments using data provided by the wind farm cluster in Inner Mongolia, and the average normalized Root Mean Square Error for the 12 wind farms was 0.1424, which verified the effectiveness of the proposed model.

Machine learning and deep learning models based grid search cross validation for short-term solar irradiance forecasting

In late 2023, the United Nations conference on climate change (COP28), which was held in Dubai, encouraged a quick move from fossil fuels to renewable energy. Solar energy is one of the most promising forms of energy that is both sustainable and renewable. Generally, photovoltaic systems transform solar irradiance into electricity. Unfortunately, instability and intermittency in solar radiation can lead to interruptions in electricity production. The accurate forecasting of solar irradiance guarantees sustainable power production even when solar irradiance is not present. Batteries can store solar energy to be used during periods of solar absence. Additionally, deterministic models take into account the specification of technical PV systems and may be not accurate for low solar irradiance. This paper presents a comparative study for the most common Deep Learning (DL) and Machine Learning (ML) algorithms employed for short-term solar irradiance forecasting. The dataset was gathered in Islamabad during a five-year period, from 2015 to 2019, at hourly intervals with accurate meteorological sensors. Furthermore, the Grid Search Cross Validation (GSCV) with five folds is introduced to ML and DL models for optimizing the hyperparameters of these models. Several performance metrics are used to assess the algorithms, such as the Adjusted R2 score, Normalized Root Mean Square Error (NRMSE), Mean Absolute Deviation (MAD), Mean Absolute Error (MAE) and Mean Square Error (MSE). The statistical analysis shows that CNN-LSTM outperforms its counterparts of nine well-known DL models with Adjusted R2 score value of 0.984. For ML algorithms, gradient boosting regression is an effective forecasting method with Adjusted R2 score value of 0.962, beating its rivals of six ML models. Furthermore, SHAP and LIME are examples of explainable Artificial Intelligence (XAI) utilized for understanding the reasons behind the obtained results.

Enhanced 2D Hand Pose Estimation for Gloved Medical Applications: A Preliminary Model.

(1) Background: As digital health technology evolves, the role of accurate medical-gloved hand tracking is becoming more important for the assessment and training of practitioners to reduce procedural errors in clinical settings. (2) Method: This study utilized computer vision for hand pose estimation to model skeletal hand movements during in situ aseptic drug compounding procedures. High-definition video cameras recorded hand movements while practitioners wore medical gloves of different colors. Hand poses were manually annotated, and machine learning models were developed and trained using the DeepLabCut interface via an 80/20 training/testing split. (3) Results: The developed model achieved an average root mean square error (RMSE) of 5.89 pixels across the training data set and 10.06 pixels across the test set. When excluding keypoints with a confidence value below 60%, the test set RMSE improved to 7.48 pixels, reflecting high accuracy in hand pose tracking. (4) Conclusions: The developed hand pose estimation model effectively tracks hand movements across both controlled and in situ drug compounding contexts, offering a first-of-its-kind medical glove hand tracking method. This model holds potential for enhancing clinical training and ensuring procedural safety, particularly in tasks requiring high precision such as drug compounding.

A new lightweight framework based on knowledge distillation for reducing the complexity of multi-modal solar irradiance prediction model

The inherent uncertainty of solar energy brings great difficulties to the grid connection and short-term energy planning and dispatching. Deep learning method makes it possible to predict the short-term solar energy with its powerful learning ability, but its complex model structure and huge trainable parameters bring great difficulties to the practical deployment. Therefore, this paper proposes a lightweight framework based on knowledge distillation strategy, which greatly reduces the complexity of multi-modal solar irradiance prediction model meanwhile ensuring an acceptable accuracy, facilitating the practical deployment. Firstly, a teacher model with multi-modal structure and good accuracy is built based on ResNet18-Informer. Then, the lightweight model is obtained by the proposed lightweight framework depending on the knowledge of teacher model. The comparisons of various models and the optimal settings of knowledge distillation are analyzed. Results show that the lightweight model can reduce the trainable parameters, inference time, and GPU memory by 97.7%, 52.5%, and 36.3%, respectively. The normalized root mean square error is reduced by 24.87% compared with the same structure model but without knowledge distillation, verifying the superiority of the proposed framework. The soft loss using the light loss with the ratio of 0.3 can obtain the best training results for the lightweight model. The structure with 3 residual blocks and 3 LSTM layers is proved to be the best for the lightweight model in the solar irradiance prediction task.

A Data-Driven Multi-Step Flood Inundation Forecast System

Inundation maps that show water depths that occur in the event of a flood are essential for protection. Especially information on timings is crucial. Creating a dynamic inundation map with depth data in temporal resolution is a major challenge and is not possible with physical models, as these are too slow for real-time predictions. To provide a dynamic inundation map in real-time, we developed a data-driven multi-step inundation forecast system for fluvial flood events. The forecast system is based on a convolutional neural network (CNN), feature-informed dense layers, and a recursive connection from the predicted inundation at timestep t as a new input for timestep t + 1. The forecast system takes a hydrograph as input, cuts it at desired timesteps (t), and outputs the respective inundation for each timestep, concluding in a dynamic inundation map with a temporal resolution (t). The prediction shows a Critical Success Index (CSI) of over 90%, an average Root Mean Square Error (RMSE) of 0.07, 0.12, and 0.15 for the next 6 h, 12 h, and 24 h, respectively, and an individual RMSE value below 0.3 m, for all test datasets when compared with the results from a physically based model.