Mapping Evaluation Research Articles

Forecasting the Volatility of Tuna Fish Prices in North Sumatra using the ARCH Method in the Period January - April 2024

Tuna (Euthynnus affinis) is one of the most important fisheries commodities in Indonesia with significant economic value, especially in its contribution to fisheries export revenue. However, the price of tuna experiences significant fluctuations that can affect local and national economic stability. This study analyzes the daily price fluctuations of tuna in the North Sumatra market from January 1, 2024 to April 29, 2024 using a time series analysis approach. Daily price data were collected and analyzed to identify existing price patterns and volatility. The Autoregressive Conditional Heteroskedasticity (ARCH) model was selected to address the heteroscedasticity in the data, which suggests that the volatility of tuna prices can be well predicted based on past price behavior. The analysis steps include identifying the optimal ARCH model using the Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF), as well as testing parameter significance and normality assumptions to validate the model fit. The results show that the ARMA (1,0,0) model is the optimal one to model the price volatility of yellow tuna with the MAPE obtained of 2.382. compared to the ARMA-ARCH method with the MAPE value obtained of 2,747. Because it still contains heteroskedasticity effects, even though the results are good, the prediction results do not closely match the original data. The model is effective in improving price forecasting accuracy, which is important to support decision-making in risk management and economic planning in the fisheries sector. The findings contribute to understanding the dynamics of the yellowtail market and optimizing strategies for fisheries management.

The Use of the Normalized Difference Red Edge (NDRE) Vegetation Index from Multispectral Cameras Mounted on Unmanned Aerial Vehicle to Estimate the Nutrient Content in Oil Palm Leaves

This study aimed to develop a prediction model for the nutrient content of N, P, K, Mg, and Ca in oil palm leaves using the Normalized Difference RedEdge (NDRE) vegetation index derived from multispectral camera data. Data acquisition was carried out by an unmanned aerial vehicle (UAV), which was correlated to leaf sample analysis of the 17th frond number. Results showed that simple regression analysis successfully represented nutrient content (N, P, K, Mg, and Ca) based on NDRE values. Based on the MAPE and Correctness values, the nutrient content prediction model for N and P yields reliable results, while for K, Mg, and Ca, they are considered good, with Correctness values of 95.5%, 96.6%, 88.8%, 87.3%, and 90.0% for N, P, K, Mg, and Ca, respectively. The study found that the NDRE vegetation index can be used to predict the nutrient content of oil palm leaves with reliable results in accuracy for N and P, and good accuracy for K, Mg, and Ca. This is a promising finding, as it could lead to the development of a non-destructive and rapid method for monitoring the nutrient status of oil palm trees, with the validation models for N, P, K, Mg, and Ca are yN = 1.1089x - 0.2497, yP = 0.99x + 0.002, yK = 1.204x - 0.1576, yMg = 0.9149x + 0.0183, and yCa = 1.0418x - 0.0218. Keywords: Multispectral Cameras, Oil Palm, Leaf Nutrient Contents, Prediction.

Intelligent System Design for Oyster Mushroom Cultivation Using Mamdani Fuzzy Inference with Internet of Things (IoT)

Oyster mushroom (Pleurotus ostreatus) has great potential in agricultural development with increasing global market demand. This research develops an intelligent cultivation system based on Mamdani fuzzy and IoT to control air temperature, air humidity, and soil moisture automatically and in real-time, using DHT22 sensor, resistive soil moisture sensor, and ESP32 as microcontroller. The test results show a high level of accuracy, with an average sensor error of 1% for air temperature, 3.8% for air humidity, and 9.3% for soil moisture. The validation of watering, humidification, and fan prediction times compared to MATLAB calculations is excellent, with MAPE values of 1.52%, 1.43%, and 1.43%, respectively. This system offers a more accurate and responsive automated solution compared to conventional cultivation management. Further testing on actual oyster mushroom farms is required to determine whether the system can adapt to varying environmental conditions. The system relies heavily on a stable internet connection to optimally perform the real-time Mamdani fuzzy calculation function.

Research on the Markov-Chain State Interval Division Based on Predicted Data Correction

By 2022, the total length of roads in Tibet Autonomous Region reached 121,447 kilometers. Due to the unique geological conditions in Tibet, various natural disasters such as earthquakes, mudslides, landslides, avalanches, and strong winds frequently occur. Along the Sichuan-Tibet Highway alone, over 300 disasters happen each year, significantly impacting the region’s economic development. This study focuses on the complexity and randomness of natural disaster mechanisms and combines Markov chain theory to improve the accuracy of prediction data for mudslides, landslides, and earth subsidence etc. The main method is to modify the state interval of the prediction model parameter-Markov chain based on the distribution of discrete points on the number axis. The following state interval division methods are proposed: (1) If the relative error of the predicted value exceeds 50%, adjust the prediction model. (2) Obtain the lower bound of state E1 by taking the floor value downward. (3) The width of each interval does not need to be uniform. (4) Arrange continuous, dense, and close points on the number line in the same side in batches, and represent a state continuously, dividing it into one suitable interval or batches. Using this method, an improved RMSE of 0.28mm and MAPE of 0.87% were obtained for engineering examples, outperforming other models such as GM(1,1), Verhulst, DGM(2,1) with corresponding RMSE values of 0.86 mm, 0.69 mm, and 1.38 mm, and MAPE values of 2.75%, 2.53%, and 5.99%. The combined prediction results for five sets of data yielded an RMSE of 0.14 mm and MAPE of 0.56%, which are quite close to the results obtained using Markov selection correction with an RMSE of 0.37 mm and MAPE of 1.01%. Furthermore, comparing the four sets of case, the average reduction in RMSE and MAPE is 3.56mm and 1.72%, respectively, demonstrating that this method can further improve the performance of Markov chain prediction.

Advanced streamflow forecasting for Central European Rivers: The Cutting-Edge Kolmogorov-Arnold networks compared to Transformers

Accurate streamflow forecasting is crucial for effective water resource management, flood mitigation, and maintaining ecological balance, especially in Central Europe’s major rivers. This study investigates the performance of two advanced deep learning algorithms—Kolmogorov-Arnold Network (KAN) and Transformers—using long-term hydrological data from four key rivers: the Rhine, Danube, Elbe, and Oder. These rivers, characterized by varying flow regimes and significant human and climatic impacts, serve as vital arteries for both commerce and ecosystems.The dataset comprises historical daily streamflow data, alongside derived input parameters such as moving averages and flow rate changes, used to predict short-term (1 to 7 days) river discharges. The KAN model, leveraging learnable spline-based activation functions, was developed to enhance accuracy and interpretability in capturing complex hydrological patterns. In contrast, the Transformer model uses advanced attention mechanisms, which excel in handling sequential data with long-range dependencies.Results show that the KAN model significantly outperforms the Transformer in short-term forecasts (up to 3 days). For 1-day forecasts, the KAN achieved R2 values of 0.975 for the Rhine, 0.956 for the Danube, 0.992 for the Elbe, and 0.969 for the Oder, with MAPE values ranging from 3.20 % to 8.09 %. At the 3-day horizon, the KAN’s R2 values remained high, demonstrating its robustness. However, as the forecasting horizon extends to 7 days, the performance of both models converges.These findings underscore the methodological novelty of KAN, which provides enhanced short-term predictive capabilities compared to existing methods, offering valuable insights for improving water resource management strategies in complex hydrological settings.

Brownfields to Healthfields: A Retrospective Ripple Effect Mapping Evaluation in Three Rural Communities

The environments in which we live influence our health behaviors and outcomes. The redevelopment of brownfields sites to health-promoting land uses may provide an array of benefits to individuals and communities, but these impacts can be particularly difficult to assess in rural communities using traditional evaluation approaches. Using Ripple Effects Mapping, we evaluated three rural brownfields redevelopment sites across Appalachian portions of EPA Region 3. Adult members of these communities participated in guided reflection on the redevelopment and subsequent impacts. Data were constructed as digital mind maps, then coded to the Community Capitals Framework by two authors coding independently. Member checking was conducted with representative workshop participants. Commonly cited impacts were site improvements, facilitation of social and physical activity, and engaging community identity. The most discussed community capitals were social and built; the least discussed capitals were natural and political. Future directions for brownfield redevelopment evaluation are discussed.

Mechanical properties and JC constitutive modification of tungsten alloys under high strain rates and high/low temperatures

The original Johnson-Cook model's poor prediction of W90 alloy stress at high strain rate and high and low temperatures restricts research and utilization of this alloy. This study used a Split Hopkinson Pressure Bar (SHPB) device to conduct dynamic compression tests on W90 alloy over a temperature range of [-90℃,700℃] and analyze its mechanical behavior. We examined the material's different plastic phase characteristics at high and low temperatures and conducted microstructure research on impact test specimens. Then, we analyzed the stress-strain data from SHPB experiments and quasi-static compression tests on the classical Johnson-Cook model. The original Johnson-Cook model was calibrated using the stress-strain data from SHPB and quasi-static compression tests, and the modified Johnson-Cook model was established and verified. The results show that: low temperature significantly strengthens the material's stress, while increasing temperature thermal softening reduces stress; increasing strain rate increases the strain generated by the workpiece, causing faster strain hardening and thermal softening effects, resulting in higher and more stable stress; low temperature makes thicker grain boundaries, increasing plastic stress, and increasing strain rate also increases plastic stress, as well as making the material's grain boundaries thicker. The increase of strain rate makes the grain more dense, which will also show up in the increase of stress on the macroscopic level; the modified Johnson-Cook model in this paper has high agreement with experimental data, with a mean value of MAPE of only 1.85 %,which is 90.20 % lower than the traditional model, and the average value of the RMSE is 85.56 % lower than the traditional model. The modified model significantly outperforms the traditional model in predicting stress and better describes stress changes in W90 at different temperatures.

Analysis, modelling, and optimization of force in ultra-precision hard turning of cold work hardened steel using the CBN tool

The machinability of high-performance materials such as superalloys, composites, and hardened steel has been a big challenge due to their mechanical, physical, and chemical properties, which give them inherent complex machining characteristics. Additionally, majority of machinability tests conducted on these materials have been carried out on conventional and less precise lathes based on Taguchi, composite, and other designs of experiments that do not exploit all the possible combinations of cutting parameters. This work reports an investigation on ultra-precision hard turning (UHT) of cold work hardened AISI D2 steel of HRC 62, based on the full factorial design of experiment, carried out on an ultra-precision lathe. A theoretical analysis of the force components generated is reported. Modelling of the process, based on the resultant force, is also reported through a machine learning model. The model was developed from the experimental data and statistically evaluated with validation data. Its average MAPE values of 1.47%, 4.81%, and 10.66% for training, testing, and validation, respectively, attest to its robustness. The excellent coefficient of determination values, R2, also justify the model’s robustness. Multi-objective optimization was also conducted to optimize material removal rate (MRR), resultant force, and vibration simultaneously. For sustainable and efficient UHT, optimal cutting velocity (158.8 m/min), feed (0.125 mm/rev), and depth of cut (0.074 mm) were proposed to generate optimal resultant force (224.8 N), MRR (2603.6 mm3/min), and vibration (0.03 m/s^2) simultaneously. These results can be beneficial in planning UHT processes for high-performance materials.

Artificial neural network based forecasting of diesel engine performance and emissions utilizing waste cooking biodiesel

Ecological and environmental problems resulting from fossil fuels are due to the harmful emissions released into the atmosphere. The rising interest in searching for alternative fuels like biodiesel is growing to solve these problems. Waste cooking oil (WCO) is transformed into methyl ester and combined with biodiesel in percentages of 25, 50, 75, and 100%. Research is done on the impacts of methyl ester blends on engine performance and emissions. Compared to diesel, the methyl ester combination showed 25% lower brake power and 24% loss in thermal efficiency at maximum load and 1500 rpm. However, diesel fuel showed 23% lower specific fuel consumption increase than biodiesel. Compared to diesel, methyl ester exhibits 15% lower air-fuel ratio and 4% volumetric efficiency. Biodiesel lowers CO, HC, and smoke concentrations by 12, 44, and 48%, respectively, compared to diesel. Biodiesel emits 23% higher NOx at 1500 rpm and 100% engine load. To predict the emissions and performance of different percentages of biodiesel at engine speed variation, an artificial neural network (ANN) model is presented. ANN modeling minimizes labor, time, and finances and uses nonlinear data. Predictions were produced about the brake output power, specific fuel consumption, thermal efficiency, air-fuel ratio, volumetric efficiency, and emissions of smoke, CO, HC, and NOx as a function of engine speed and blend ratio. All correlation coefficients (r) over 0.99 and R2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$R^{2}$$\\end{document} values were beyond 0.98 for all variables. There were low values of MSE, MAPE, and MSLE with significant predictive ability. WCO’s biodiesel is a viable diesel engine replacement fuel.

Data-driven energy consumption prediction of a university office building using machine learning algorithms

Redundant consumption of energy in buildings is an important issue that causes increasing problems of climate change and global warming in the world. Therefore, it is necessary to develop efficient energy management approaches in buildings. Accurate prediction of energy consumption plays an important role to obtain energy-efficient buildings. Data-driven methods gained attention for estimation of energy consumption in buildings which would provide more accurate prediction results. In this study, hourly energy consumption prediction is performed on a university office building to increase energy efficiency in the building using machine learning algorithms. A new parameter is proposed, air conditioning demand, to improve accuracy of the algorithms. Moreover, temporal parameters, i.e. day of week, month of year, and hour of day, were used along with meteorological parameters to improve prediction performance of the algorithms. Experimental results show that hourly energy consumption of the building could be predicted using machine learning algorithms with high performance. When the results were analysed, Deep Neural Network (DNN) achieved better performance among other alternative algorithms. The average values of R2, RMSE and MAPE for DNN were 0.959, 4.796 kWh, and 5.738 %, respectively. Also, the addition of proposed air conditioning demand parameter provided improved performance to the algorithms.

Sustainable Textile Manufacturing with Revolutionizing Textile Dyeing: Deep Learning-Based, for Energy Efficiency and Environmental-Impact Reduction, Pioneering Green Practices for a Sustainable Future

The textile industry, a substantial component of the global economy, holds significant importance due to its environmental impacts. Particularly, the use of water and chemicals during dyeing processes raises concerns in the context of climate change and environmental sustainability. Hence, it is crucial from both environmental and economic standpoints for textile factories to adopt green industry standards, particularly in their dyeing operations. Adapting to the green industry aims to reduce water and energy consumption in textile dyeing processes, minimize waste, and decrease the carbon footprint. This approach has become crucial in achieving sustainability in textiles following the signing of the Paris Climate Agreement. Important elements of this transformation include the reuse of washing waters used in the dyeing process, the recycling of wastewater, and the enhancement of energy efficiency through necessary methodological and equipment changes. This study analyzes the energy, labor, production, and consumption data since 2011 for a textile factories with four branches located in the Adana Organized Industrial Zone. Among these factories, the one designated as UT1, which has the highest average energy and water consumption compared to the other three branches, is selected. In recent years, the use of artificial intelligence and machine learning technologies in predicting industrial processes has been increasingly observed. The data are analyzed using LSTM (Long Short-Term Memory) and ANN (Artificial Neural Networks) forecasting methods. Particularly, the LSTM algorithms, which provided the most accurate results, have enabled advanced forecasting of electricity consumption in dyeing processes for future years. In 2020, electricity consumption was recorded as 3,717,224 kWh and this consumption was reflected in the total energy cost as TRY 1,916,032. Electricity consumption accounts for 22.34% of total energy consumption, while the share of this energy type in the cost is 43.25%. In the light of these data, the MAPE value for energy consumption forecasts using the LSTM model was 0.45%, which shows that the model is able to forecast with high accuracy. As a result, a solar power plant was installed to optimize energy consumption, and in 2023 60% energy savings were achieved in summer and 25% in winter. The electricity consumption forecasting results have been an essential guide in planning strategic initiatives to enhance factory efficiency. Following improvement efforts aimed at reducing energy consumption and lowering the carbon footprint, significant optimizations in processes and layouts have been made at specific bottleneck points within the facility. These improvements have led to savings in labor, time, and space, and have reduced unit production costs.

Factors Affecting The Number Of Domestic Flights In Indonesia During Covid-19 Pandemic Using SARIMAX Method

Indonesia, which consists of thousands of large and small islands, relies heavily on-air transportation to support mobility between regions. As many as 80% of Indonesia's total air transportation passengers are domestic flight passengers. This shows how vital domestic flights are in Indonesia's air transportation system. However, in 2020, the COVID-19 pandemic had an impact that resulted in a decrease in the number of domestic flights in Indonesia. Therefore, an analysis is needed to determine the factors that affect the number of domestic flights in Indonesia. This study uses the SARIMAX method, a time series regression with the addition of seasonal factors and other variables or exogenous factors that significantly affect the model to improve the model's accuracy. Several exogenous variables are considered, including the number of operating civil aviation airports, positive daily cases of COVID-19, calendar effects during Eid al-Fitr and New Year's Day, and social restriction policies. The results showed that the number of operating airports one week before Eid al-Fitr, one week during Eid al-Fitr, one week before New Year, and Emergency PPKM significantly influenced the number of domestic flights. These variables offer pivotal insights into the influence of external factors on domestic flight patterns, exerting significant impacts on passenger travel behavior and subsequently influencing domestic flight volume. The integration of these variables in the SARIMAX model allows for a comprehensive analysis of the complex dynamics influencing domestic air travel in Indonesia. The best SARIMAX model obtained is SARIMAX (1,1,1)(4,1,1)7 with a MAPE value of 5.35% and a coefficient of determination is 97%.

Nonlinear Ensemble Deep Learning Model for Energy Consumption Prediction with Bayesian Optimization

Accurate prediction of electric energy consumption is crucial for efficient load dispatching, energy utilization, and grid operation. Traditional statistical and classical machine learning methods struggle with the nonlinear nature of energy consumption data, often leading to higher prediction errors. Additionally, deep learning models using a single approach face challenges such as convergence to local minima and poor generalization. This paper proposes a nonlinear ensemble deep learning model for residential energy consumption prediction, incorporating Bayesian optimization for hyperparameter tuning. The model combines Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and 1D Convolutional Neural Networks (1D-CNN), leveraging their powerful nonlinear feature learning capabilities. A k-means clustering approach is used to preprocess and reduce variability in the data, enhancing the ensemble model's performance. The ensemble model was tested on real energy consumption data from two districts in Addis Ababa, showing significant improvements in prediction accuracy with lower MAE, RMSE, and MAPE values compared to single models and un-clustered data. The integration of clustering and Bayesian optimization further enhanced model generalizability and minimized overfitting, demonstrating the effectiveness of a nonlinear approach in capturing complex energy consumption patterns.

Unravelling flood risk in the Rel River watershed, Gujarat using coupled earth observations, multi criteria decision making and Google Earth Engine

Socioeconomic developments, ineffective drainage systems, and insufficient river control, all contribute to significant loss of property and life due to the constant threat of floods. Therefore, controlling flood threats across Rel River, Dhanera, Gujarat has become even more crucial due to floods causing strain throughout the area during monsoon season. The different 52 micro-watersheds were formed across the study region using earth observations for the estimation of flood hazards, vulnerability, and risk. The AHP-MCDM was employed to assign priority rank, weightage, and risk category for each micro watershed. The flood hazard zone was mapped and its vulnerability was characterized in different categories varying from very low to very high. The normalized weights of each factor i.e, hazard indicator (soil, elevation, slope, flow accumulation, rainfall) and vulnerability indicator (LULC, distance from the hospital, population density map) were estimated employing the AHP-MCDM technique whereas LULC along with most of other factors were derived from GEE. The integration of vulnerability and hazard indicators, provides insights into understanding the flood sensitivity, facilitating the preparation of the flood risk map. 20 micro-watersheds were susceptible to high to very high risk and covered an area of 213.15 km2 whereas 32 micro-watersheds were in the range of very low to moderate category which covers the area of 228.41 km2. Therefore, the integration of GEE and spectral indices for obtaining various hazard & vulnerability indicators, which were prioritized and ranked using AHP is a unique methodology, facilitating a robust evaluation of flood risk mapping.

Enhancing Stock Price Prediction in the Indonesian Market: a Concave LSTM Approach with RunReLU

This study addresses the pressing need for improved stock price prediction models in the financial markets, focusing on the Indonesian stock market. It introduces an innovative approach that utilizes the custom activation function RunReLU within a Concave Long Short Term Memory (LSTM) framework. The primary objective is to enhance prediction accuracy, ultimately assisting investors and market participants in making more informed decisions. The research methodology used historical stock price data from ten prominent companies listed on the Indonesia Stock Exchange, covering the period from July 6, 2015, to October 14, 2021. Evaluation metrics such as RMSE, MAE, MAPE, and R2 were employed to assess model performance. The results consistently favored the RunReLU-based model over the ReLU-based model, showcasing lower RMSE and MAE values, higher R2 values, and notably reduced MAPE values. These findings underscore the practical applicability of custom activation functions for financial time series data, providing valuable tools for enhancing prediction precision in the dynamic landscape of the Indonesian stock market.

Multi-step carbon emissions forecasting using an interpretable framework of new data preprocessing techniques and improved grey multivariable convolution model

As China stands at a critical juncture in its transition towards a low-carbon future, the precise prediction and analysis of provincial carbon emissions have emerged as paramount tasks. Focusing on forecasting in this intricate and vital domain, an updated grey multivariable convolution model is designed by employing the unified new-information-oriented accumulating generation operator (UNAGO) technique to process raw sequences. Equipped with UNAGO's hyper-parameters that enable the independent scaling effects and prioritize new information, the newly-designed model offers high flexibility and adaptability for handling complex provincial carbon emissions. Subsequently, four different restricted carbon emission sequences are utilized as case studies for validation purposes, and the new model's performance is scrutinized against five contrasting methods across three prediction forecasting horizons. Comparative experimental results reveal the new model's superior level of accuracy in predicting carbon emissions across four different provinces, with MAPE values of <4 % and 10 % in the in-sample and out-of-sample periods, respectively. Furthermore, rigorous evaluations with Diebold-Mariano (DM) and Probability Density Analysis (PDA) tests confirm the model's robust and general forecasting capabilities.

Applicability of energy-theta mapping in resonant soft X-ray reflectometry to probe depth dependence of oxidation state and crystallographic environment in iron oxide multilayers

In the present paper we discuss applicability of the synchrotron method of resonant x-ray reflectometry to non-destructive depth profiling of multilayer heterostructures with low optical contrast between the sublayers. The two-dimensional mapping approach comprising evaluation and measurement of reflectance as a function of photon energy and grazing angle is shown to provide a convenient way to effectively utilize the enhancement of optical contrast between the sublayers that exists at the absorption edges of chemical elements due to particular spectral shape peculiarities related to oxidation state, crystallographic environment and magnetization. In the present study the evaluation of the resonant X-ray reflectivity maps is performed using a specially developed modeling and fitting software. Estimation of the photon energy / grazing angle combinations at which the reflectance is most sensitive to the minor changes in the physical properties of the sublayers is illustrated by the example of ferroic-on-semiconductor nanoscale heterostructures composed of nanoscale film of metallic iron oxidized from either top or bottom side. The subtle differences in resonant soft x-ray reflectance caused by variation of the oxide film thickness, oxidation state and crystallographic environment are discussed. The presented approach is applicable to a large class of heterostructures composed of sublayers that can be distinguished only by the differences in their near edge X-ray absorption fine structure.

Optimized Hybrid Deep Learning for Real-Time Pandemic Data Forecasting: Long and Short-Term Perspectives

Background: With new variants of COVID-19 causing challenges, we need to focus on integrating multiple deep-learning frameworks to develop intelligent healthcare systems for early detection and diagnosis. Objective: This article suggests three hybrid deep learning models, namely CNN-LSTM, CNN-Bi- LSTM, and CNN-GRU, to address the pressing need for an intelligent healthcare system. These models are designed to capture spatial and temporal patterns in COVID-19 data, thereby improving the accuracy and timeliness of predictions. An output forecasting framework integrates these models, and an optimization algorithm automatically selects the hyperparameters for the 13 baselines and the three proposed hybrid models. Methods: Real-time time series data from the five most affected countries were used to test the effectiveness of the proposed models. Baseline models were compared, and optimization algorithms were employed to improve forecasting capabilities. Results: CNN-GRU and CNN-LSTM are the top short- and long-term forecasting models. CNNGRU had the best performance with the lowest SMAPE and MAPE values for long-term forecasting in India at 3.07% and 3.17%, respectively, and impressive results for short-term forecasting with SMAPE and MAPE values of 1.46% and 1.47%. Conclusion: Hybrid deep learning models, like CNN-GRU, can aid in early COVID-19 assessment and diagnosis. They detect patterns in data for effective governmental strategies and forecasting. This helps manage and mitigate the pandemic faster and more accurately.

An Improved Type-1 Fuzzy Regression Function for COVID-19 Cases Prediction

The deficiency in data collection and reporting has led to the emergence of uncertainty in the data in the pandemic process. These matters cause that traditional statistical and mathematical models have been functionless and unreliable in this issue. In this respect, this study presents an ensemble prediction model with innovative and contemporary properties to model COVID-19 cases in the UK and USA inclusively throughout the whole pandemic process. The proposed ensemble prediction model is composed of an assembly of Type-1 fuzzy regression functions with elastic net regularization (E-T1FRF) and radial basis function neural networks (RBFNNs). Thus, the proposed ensemble model can successfully model the uncertainty in the data with the fuzzy modelling perspective of T1FRF and also thoroughly adapt to the patterns in the data thanks to the flexible modelling ability of RBFNN based on data. With the proposed ensemble prediction model, the cases in the entire pandemic process from the beginning of March 2020 to the end of June 2022 were modelled and predicted for 23 different periods in one-month prediction steps. The proposed model produced predictions with MAPE values below 3% in all but periods except for three periods. Also, the average MAPE values for all periods were obtained at around 2% for the UK and only 1.5% for the US. These results, at a reasonable level of error, demonstrated the practicality of the usage of the proposed community model for other countries and provided valuable information for future action.

Scientific preparation for JRT: Wind pressure prediction model for large radio telescope based on real data from multi-sensors

Jingdong 120-meter radio telescope (JRT) is poised to become the world's largest single-aperture fully steerable medium-low frequency radio telescope. However, like other large-aperture radio telescopes, the JRT is vulnerable to wind loads, which can cause structural deformation and pointing errors. Addressing this challenge requires the ability to predict dynamic winds in real-time. This study developed a wind pressure preprocessing and prediction model using sensor data collected from the Kunming 40-meter radio telescope (KRT), enabling real-time prediction of wind pressure on the telescope. The model employs adaptive noise and Variational Mode Decomposition (VMD) techniques to eliminate random noise from the original wind pressure data. Subsequently, wind pressure predictions are made using a Bidirectional Long Short-term Memory (BiLSTM) model. By conducting predictions under various stabilization conditions and conducting a thorough analysis of measurement data from five sensors, the study has achieved impressive results in predicting wind pressure on the KRT reflector surface. The proposed model demonstrates the lowest MAE, RMSE, and MAPE, while achieving the highest R2 across various data sets. Where the average R2 of the proposed model is 0.9392 at 45° pitch angle attitude and the RMSE, MAE and MAPE values are 1.4923, 1.2377 and 1.82% respectively. This model helps wind load monitoring of real-time wind pressure monitoring of the telescope surface, to study the effects of wind load on pointing accuracy. By adjusting the control parameters to reduce wind load interference, to ensure the high-precision work of a large radio telescope, such as JRT.