Symmetric Mean Absolute Percentage Error Research Articles

Deep learning algorithm development for river flow prediction: PNP algorithm

Deep learning algorithms developed in recent decades have performed well in prediction and classification using accumulated big data. However, as climate change has recently become a more serious global problem, natural disasters are occurring frequently. When analyzing natural disasters from the perspective of a data analyst, they are considered as outliers, and the ability to predict outliers (natural disasters) using deep learning algorithms based on big data acquired by computers is limited. To predict natural disasters, deep learning algorithms must be enhanced to be able to predict outliers based on information such as the correlation between the input and output. Thus, algorithms that specialize in one field must be developed, and specialized algorithms for abnormal values must be developed to predict natural disasters. Therefore, considering the correlation between the input and output, we propose a positive and negative perceptron (PNP) algorithm to predict the flow rate of rivers using climate change-sensitive precipitation. The PNP algorithm consists of a hidden deep learning layer composed of positive and negative neurons. We built deep learning models using the PNP algorithm to predict the flow of three rivers. We also built comparative deep learning models using long short-term memory (LSTM) to validate the performance of the PNP algorithm. We compared the predictive performance of each model using the root mean square error and symmetric mean absolute percentage error and demonstrated that it performed better than the LSTM algorithms .

Forecasting The Search Trends of The Keyword “Sarung Wadimor” In Indonesia on Google Trends Data Using Time Series Regression with Calender Variation and Arima Box-Jenkins

The impact of this 4.0 era is that data is growing and can be collected very easily and then reprocessed to obtain information. One of the search engines for various data and information that is often used is Google, causing a high search intensity and will further impact on increasing the amount of data generated by search engines. Google Trends is one of the official websites from Google that reflects or takes pictures of events in society based on search keywords. The search keyword that will be studied in this article is “Sarung Wadimor”. Therefore, the purpose of this research is to forecast the search trend for the keyword "Sarung Wadimor" which is interesting because the resulting time series data pattern shows a recurring pattern due to the effect of calendar variations which are thought to be related to the month of Ramadan. Forecasting modeling uses Autoregressive Integrated Moving Average (ARIMA) and Time Series Regression (TSR). The goodness of the model used in this article is the Mean Square Error (MSE), Root Mean Square Error (RMSE), and Symmetric Mean Absolute Percentage Error (SMAPE). Based on the results of the analysis, using three goodness-of-fit measures shows that the TSR model with the Calendar Variation of Ramadan + Month Periods has smaller MSE, RMSE, and SMAPE values than the other models with goodness-of-fit values of 88.602, 9.413, and 26.950, respectively. Forecasting results for the next 6 periods show that the search trend for the keyword "Sarung Wadimor" tends to decrease, this is because the month of Ramadan is still quite far in 2023.

Study of the influence of meteorological factors on HFMD and prediction based on the LSTM algorithm in Fuzhou, China

BackgroundThis study adopted complete meteorological indicators, including eight items, to explore their impact on hand, foot, and mouth disease (HFMD) in Fuzhou, and predict the incidence of HFMD through the long short-term memory (LSTM) neural network algorithm of artificial intelligence.MethodA distributed lag nonlinear model (DLNM) was used to analyse the influence of meteorological factors on HFMD in Fuzhou from 2010 to 2021. Then, the numbers of HFMD cases in 2019, 2020 and 2021 were predicted using the LSTM model through multifactor single-step and multistep rolling methods. The root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and symmetric mean absolute percentage error (SMAPE) were used to evaluate the accuracy of the model predictions.ResultsOverall, the effect of daily precipitation on HFMD was not significant. Low (4 hPa) and high (≥ 21 hPa) daily air pressure difference (PRSD) and low (< 7 °C) and high (> 12 °C) daily air temperature difference (TEMD) were risk factors for HFMD. The RMSE, MAE, MAPE and SMAPE of using the weekly multifactor data to predict the cases of HFMD on the following day, from 2019 to 2021, were lower than those of using the daily multifactor data to predict the cases of HFMD on the following day. In particular, the RMSE, MAE, MAPE and SMAPE of using weekly multifactor data to predict the following week's daily average cases of HFMD were much lower, and similar results were also found in urban and rural areas, which indicating that this approach was more accurate.ConclusionThis study’s LSTM models combined with meteorological factors (excluding PRE) can be used to accurately predict HFMD in Fuzhou, especially the method of predicting the daily average cases of HFMD in the following week using weekly multifactor data.

PERAMALAN PENDAPATAN BULANAN MENGGUNAKAN FUZZY TIME SERIES CHEN ORDE TINGGI

Cooperatives need consideration in the making of business strategy decisions. Forecasting can assist cooperatives in deciding on their business strategy. This study used n-orde Fuzzy Time Series Chen. n-orde Fuzzy Time Series Chen captures data patterns formed by two or more historical data in each period called fuzzy logic relation (FLR). The pattern of FLR is used to be projected in forecasting future conditions. This study used 2-orde, 3-orde, and 4-orde with 1-orde as the comparison. This study used data on the monthly revenue of the Employee Cooperative of PT. Telekomunikasi Indonesia Semarang Region for the period of January 2019 to May 2022 to predict revenue for the period of June and July 2022. This study used symmetric Mean Absolute Percentage Error (sMAPE) in calculating the forecasting error rate. 1-orde, 2-orde, 3-orde, and 4-orde of Fuzzy Time Series Chen produced different forecasting results for the period of June and July 2022. 1-orde has sMAPE value of 23.15% (good enough forecasting), 2-orde and 3-orde have sMAPE value of 10.06% (good forecasting), and 4-orde has sMAPE value of 4.52% (very good forecasting). This study showed that the larger orde used in Fuzzy Time Series Chen, the lower forecasting error rate.

Forecasting the Spread of COVID-19 Using Deep Learning and Big Data Analytics Methods.

To contain the spread of the COVID-19 pandemic, there is a need for cutting-edge approaches that make use of existing technology capabilities. Forecasting its spread in a single or multiple countries ahead of time is a common strategy in most research. There is, however, a need for all-inclusive studies that capitalize on the entire regions on the African continent. This study closes this gap by conducting a wide-ranging investigation and analysis to forecast COVID-19 cases and identify the most critical countries in terms of the COVID-19 pandemic in all five major African regions. The proposed approach leveraged both statistical and deep learning models that included the autoregressive integrated moving average (ARIMA) model with a seasonal perspective, the long-term memory (LSTM), and Prophet models. In this approach, the forecasting problem was considered as a univariate time series problem using confirmed cumulative COVID-19 cases. The model performance was evaluated using seven performance metrics that included the mean-squared error, root mean-square error, mean absolute percentage error, symmetric mean absolute percentage error, peak signal-to-noise ratio, normalized root mean-square error, and the R2 score. The best-performing model was selected and used to make future predictions for the next 61days. In this study, the long short-term memory model performed the best. Mali, Angola, Egypt, Somalia, and Gabon from the Western, Southern, Northern, Eastern, and Central African regions, with an expected increase of 22.77%, 18.97%, 11.83%, 10.72%, and 2.81%, respectively, were the most vulnerable countries with the highest expected increase in the number of cumulative positive cases.

Performance Investigation of Currently Available Reaction Mechanisms in the Estimation of NO Measurements: A Comparative Study

Ammonia (NH3) has been receiving the attention of researchers as an alternative promising green fuel to replace fossil sources for energy production. However, the high NOx emissions are one of the drawbacks and restrictions of using NH3 on a broad scale. The current study investigates NO production/consumption for a 70/30 (vol%) NH3/H2 mixture using kinetic reaction mechanism concepts to shed light on the essential reaction routes that promote/inhibit NO formation. Sixty-seven kinetic reaction mechanisms from the literature have been investigated and compared with recently reported measurements at a wide range of equivalence ratios (ϕ) (0.6–1.4), atmospheric pressure and temperature conditions. Both numerical simulations and experimental measurements used the same combustion reactor configuration (premixed stabilized stagnation flame). To highlight the best kinetic model for the predicting of the NO experimental measurements of NO, a symmetric mean absolute percentage error (SMAPE) has been determined as a preliminary estimation by comparing both numerical and experimental measurements. The results found that the kinetic reaction mechanism of Glarborg showed an accurate prediction with a minor error percentage of 2% at all lean and stoichiometric conditions. Meanwhile, the kinetic model of Wang accurately predicted the experimental data with 0% error at ϕ = 1.2 and underestimated the mole fraction of NO at 1.4 ϕ with an error of 10%. The sensitivity analysis and rate of production/consumption of NO mole fractions analysis have also been implemented to highlight the most important reactions that promote/inhibit NO formation. At lean and stoichiometric conditions, Glarborg kinetic model shows that the kinetic reactions of HNO + H ⇌ NO + H2, HNO + O ⇌ NO + OH, and NH + O ⇌ NO + H are the most important reaction routes with considerable effect on NO formation for 70/30 (vol%) NH3/H2 mixture. In contrast, the reactions of NH2 + NO ⇌ N2 + H2O, NH2 + NO ⇌ NNH + OH, NH + NO ⇌ N2O + H, and N + NO ⇌ N2 + O significantly consume NO to N2, NNH, and N2O. Further, Wang’s mechanism illustrated the dominant effect of each HNO + H ⇌ NO + H2, N + OH ⇌ NO + H, NH + O ⇌ NO + H in NO formation and NH + NO ⇌ N2O + H, NH2 + NO ⇌ NNH + OH, and NH2 + NO ⇌ N2 + H2O in the consumption of NO mole fractions.

A data-driven framework for medium-term electric vehicle charging demand forecasting

The rapid phase-in of electric vehicles (EV) will cause unprecedented issues with managing the supply of electricity and charging stations. It is in the interest of utility providers and everyday consumers to be able to plan for peak charging times, and related congestion. While past work has been done for localized, short-term forecasting, it has not included longer term forecasting, or considered the relationships between multiple stations. Importantly, past work has also not offered a framework for dataset construction and evaluated different dataset features. We propose a methodology to forecast demand at public EV charging stations, and use it to explore the potential of data-driven models to predict demand up to one week in advance. Our strategy includes selecting parameters for formatting a dataset given a list of charging events, a way to consider flexible prediction horizons, and deployment of deep and supervised learning-based models. To the best of our knowledge, ours is the first study to propose machine learning to forecast medium-term public EV charging demand, to exploit weather and other features at public charging stations, and to forecast demand at multiple stations and the entire network. We validated our approach using data from eleven stations over three years from Scotland, UK. Our method outperforms the benchmark time series method, and predicts network demand with a symmetric mean absolute percentage error (SMAPE) of 5.9% and a mean absolute error (MAE) of 124.7 kWh, or less than twelve percent of average daily demand.

Near real-time prediction of urgent care hospital performance metrics using scalable random forest algorithm: A multi-site development

While previous studies have shown the potential value of predictive modelling for emergency care, few models have been implemented for producing near real-time predictions across various demand, utilisation and performance metrics. In this study, 33 independent Random Forest (RF) algorithms were developed to forecast 11 urgent care metrics over a 24-hour period across three hospital sites in an Integrated Care System (ICS) in South West England. Metrics included: ambulance handover delay; emergency department occupancy; and patients awaiting admission. Mean Absolute Error (MAE), Root Mean Squared Error (RMSE) and Symmetric Mean Absolute Percentage Error (SMAPE) were used to assess the performance of RF and compare it to two alternative models: naïve baseline (NB) and Auto-Regressive Integrated Moving Average (ARIMA). Using these measures, RF outperformed NB and ARIMA in 76% (N = 25/33) of urgent care metrics according to SMAPE, 88% (N = 29/33) according to MAE and 91% (N = 30/33) according to RMSE. The RFs developed in this study have been implemented within the local ICS, providing predictions on an hourly basis that can be accessed by local healthcare planners and managers.

A Stabilized Kriging Method for Mapping Disease Rates.

Mapping disease rates is an important aspect of epidemiological research because it helps inform public health policy. Disease maps are often drawn according to local administrative areas (LAAs), such as counties, cities, or towns. In LAAs with small populations, disease rates are unstable and are prone to appear extremely high or low. The empirical Bayes methods consider variance differences among different LAAs, thereby stabilizing the disease rates. The methods of kriging break the constraints of geopolitical boundaries and produce a smooth curved surface in the form of contour lines, but the methods lack the stabilizing effect of the empirical Bayes methods. An easy-to-implement stabilized kriging method is proposed to map disease rates, which allows different errors in different LAAs. Monte Carlo simulations revealed that the stabilized kriging method had smaller symmetric mean absolute percentage error than three other types of methods (the original LAA-based method, empirical Bayes methods, and traditional kriging methods) in nearly all scenarios considered. Real-world data analysis of oral cancer incidence rates in men from Taiwan demonstrated that the age-standardized rates in the central mountainous sparsely-populated region of Taiwan were stabilized using our proposed method, with no more large differences in numerical values, whereas the rates in other populous regions were not over-smoothed. Additionally, the stabilized kriging map had improved resolution and helped locate several hot and cold spots in the incidence rates of oral cancer. We recommend the use of the stabilized kriging method for mapping disease rates.

Prediction of air quality index based on the SSA-BiLSTM-LightGBM model

The air quality index (AQI), as an indicator to describe the degree of air pollution and its impact on health, plays an important role in improving the quality of the atmospheric environment. Accurate prediction of the AQI can effectively serve people’s lives, reduce pollution control costs and improve the quality of the environment. In this paper, we constructed a combined prediction model based on real hourly AQI data in Beijing. First, we used singular spectrum analysis (SSA) to decompose the AQI data into different sequences, such as trend, oscillation component and noise. Then, bidirectional long short-term memory (BiLSTM) was introduced to predict the decomposed AQI data, and a light gradient boosting machine (LightGBM) was used to integrate the predicted results. The experimental results show that the prediction effect of SSA-BiLSTM-LightGBM for the AQI data set is good on the test set. The root mean squared error (RMSE) reaches 0.6897, the mean absolute error (MAE) reaches 0.4718, the symmetric mean absolute percentage error (SMAPE) reaches 1.2712%, and the adjusted R2 reaches 0.9995.

Automated Adolescence Scoliosis Detection Using Augmented U-Net With Non-square Kernels.

Purpose: Scoliosis is a deformity of the spine, and as a measure of scoliosis severity, Cobb angle is fundamental to the diagnosis of deformities that require treatment. Conventional Cobb angle measurement and assessment is usually done manually, which is inherently time-consuming, and associated with high inter- and intra-observer variability. While there exist automatic scoliosis measurement methods, they suffer from insufficient accuracy. In this work, we propose a two-step segmentation-based deep learning architecture to automate Cobb angle measurement for scoliosis assessment using X-Ray images. Methods: The proposed architecture involves two steps. In the first step, we utilize a novel Augmented U-Net architecture to generate segmentations of vertebrae. The second step includes a non-learning-based pipeline to extract landmark coordinates from the segmented vertebrae and filter undesirable landmarks. Results: Our proposed Augmented U-Net architecture achieved a Symmetric Mean Absolute Percentage Error of 9.2%, with approximately 90% of estimations having less than 10 degrees difference compared with the AASCE-MICCAI challenge 2019 dataset ground truths. We further validated the model using an internal dataset and achieved almost the same level of performance. Conclusion: The proposed architecture is robust in providing automated spinal vertebrae segmentations and Cobb angle measurement, and is potentially generalizable to real-world clinical settings.

Phytoclass: A pigment‐based chemotaxonomic method to determine the biomass of phytoplankton classes

AbstractPigment‐based chemotaxonomy is a widely utilized tool to determine the biomass of phytoplankton classes from pigment biomarkers. The CHEMTAX approach is sensitive to the initial estimates of pigment‐to‐chlorophyll a (Chl a) ratios for the phytoplankton classes required, even though these are modified by the CHEMTAX process. We present an alternative chemotaxonomic method that utilizes simulated annealing with a steepest descent algorithm to derive class abundances and pigment‐to‐Chl a ratios. The simulated annealing algorithm is tested on two synthetic datasets of Southern Ocean phytoplankton communities. Each dataset is composed of 1000 inversion samples (set of phytoplankton class abundances, pigment ratios, and pigment profiles) with sizes ranging between 5 and 60 individual samples. We show that the new simulated annealing approach displays higher accuracy than two common configurations of the CHEMTAX method, with lower differences between true and estimated class abundances. Symmetric mean absolute percentage error were 4.8–11%, compared to 18–70% with CHEMTAX approaches. Proportions of variance explained (R2) between true and estimated class abundances using the simulated annealing approach were 0.98–0.99 compared to 0.71–0.89 for CHEMTAX. Overall, this new methodology is capable of determining phytoplankton class abundances at higher accuracy than CHEMTAX without sensitivity to initial estimates of pigment‐to‐Chl a ratios.

Forecasting Future Monthly Patient Volume using Deep Learning and Statistical Models.

The variety of diseases is increasing day by day, and the demand for hospitals, especially for emergency and radiology units, is also increasing. As in other units, it is necessary to prepare the radiology unit for the future, to take into account the needs and to plan for the future. Due to the radiation emitted by the devices in the radiology unit, minimizing the time spent by the patients for the radiological image is of vital importance both for the unit staff and the patient. In order to solve the aforementioned problem, in this study, it is desired to estimate the monthly number of images in the radiology unit by using deep learning models and statistical-based models, and thus to be prepared for the future in a more planned way. For prediction processes, both deep learning models such as LSTM, MLP, NNAR and ELM, as well as statistical based prediction models such as ARIMA, SES, TBATS, HOLT and THETAF were used. In order to evaluate the performance of the models, the symmetric mean absolute percentage error (sMAPE) and mean absolute scaled error (MASE) metrics, which have been in demand recently, were preferred. The results showed that the LSTM model outperformed the deep learning group in estimating the monthly number of radiological case images, while the AUTO.ARIMA model performed better in the statistical-based group. It is believed that the findings obtained will speed up the procedures of the patients who come to the hospital and are referred to the radiology unit, and will facilitate the hospital managers in managing the patient flow more efficiently, increasing both the service quality and patient satisfaction, and making important contributions to the future planning of the hospital.

Prediction of temperature distribution in a furnace using the incremental deep extreme learning machine.

In this article, a data-driven model based on the incremental deep extreme learning machine (IDELM) algorithm is proposed to predict the temperature distribution in the furnace. To this end, computational fluid dynamics (CFD) simulations are carried out first to get temperature distributions under typical working conditions. Based on the air distribution mode, the simulation results are divided into six subclasses. Then the K-means clustering method is applied to find out the benchmark working condition of each subclass. Moreover, the random sampling method is used to extract samples to reduce computational complexity. Modeling inputs are selected according to the CFD boundary conditions and combustion mechanisms, and data sets are reconstructed based on the increments of each actual working condition from the benchmark working condition. Finally, an IDBN-based prediction model is built in each subclass. The experimental results show that the IDBN-based model has a promising predictive ability with less than 11% symmetric mean absolute percentage error.

A New Regression Model for Depression Severity Prediction Based on Correlation among Audio Features Using a Graph Convolutional Neural Network.

Recent studies have revealed mutually correlated audio features in the voices of depressed patients. Thus, the voices of these patients can be characterized based on the combinatorial relationships among the audio features. To date, many deep learning-based methods have been proposed to predict the depression severity using audio data. However, existing methods have assumed that the individual audio features are independent. Hence, in this paper, we propose a new deep learning-based regression model that allows for the prediction of depression severity on the basis of the correlation among audio features. The proposed model was developed using a graph convolutional neural network. This model trains the voice characteristics using graph-structured data generated to express the correlation among audio features. We conducted prediction experiments on depression severity using the DAIC-WOZ dataset employed in several previous studies. The experimental results showed that the proposed model achieved a root mean square error (RMSE) of 2.15, a mean absolute error (MAE) of 1.25, and a symmetric mean absolute percentage error of 50.96%. Notably, RMSE and MAE significantly outperformed the existing state-of-the-art prediction methods. From these results, we conclude that the proposed model can be a promising tool for depression diagnosis.

A Hybrid Neural Network Model for Short-Term Wind Speed Forecasting

This study proposes an effective wind speed forecasting model combining a data processing strategy, neural network predictor, and parameter optimization method. (a) Variational mode decomposition (VMD) is adopted to decompose the wind speed data into multiple subseries where each subseries contains unique local characteristics, and all the subseries are converted into two-dimensional samples. (b) A gated recurrent unit (GRU) is sequentially modeled based on the obtained samples and makes the predictions for future wind speed. (c) The grid search with rolling cross-validation (GSRCV) is designed to simultaneously optimize the key parameters of VMD and GRU. To evaluate the effectiveness of the proposed VMD-GRU-GSRCV model, comparative experiments based on hourly wind speed data collected from the National Renewable Energy Laboratory are implemented. Numerical results show that the root mean square error, mean absolute error, mean absolute percentage error, and symmetric mean absolute percentage error of this proposed model reach 0.2047, 0.1435, 3.77%, and 3.74%, respectively, which outperform the benchmark predictions using popular parameter optimization methods, data processing techniques, and hybrid neural network forecasting models.

Machine Learning Predictive Models for Evaluating Risk Factors Affecting Sperm Count: Predictions Based on Health Screening Indicators.

In many countries, especially developed nations, the fertility rate and birth rate have continually declined. Taiwan's fertility rate has paralleled this trend and reached its nadir in 2022. Therefore, the government uses many strategies to encourage more married couples to have children. However, couples marrying at an older age may have declining physical status, as well as hypertension and other metabolic syndrome symptoms, in addition to possibly being overweight, which have been the focus of the studies for their influences on male and female gamete quality. Many previous studies based on infertile people are not truly representative of the general population. This study proposed a framework using five machine learning (ML) predictive algorithms-random forest, stochastic gradient boosting, least absolute shrinkage and selection operator regression, ridge regression, and extreme gradient boosting-to identify the major risk factors affecting male sperm count based on a major health screening database in Taiwan. Unlike traditional multiple linear regression, ML algorithms do not need statistical assumptions and can capture non-linear relationships or complex interactions between dependent and independent variables to generate promising performance. We analyzed annual health screening data of 1375 males from 2010 to 2017, including data on health screening indicators, sourced from the MJ Group, a major health screening center in Taiwan. The symmetric mean absolute percentage error, relative absolute error, root relative squared error, and root mean squared error were used as performance evaluation metrics. Our results show that sleep time (ST), alpha-fetoprotein (AFP), body fat (BF), systolic blood pressure (SBP), and blood urea nitrogen (BUN) are the top five risk factors associated with sperm count. ST is a known risk factor influencing reproductive hormone balance, which can affect spermatogenesis and final sperm count. BF and SBP are risk factors associated with metabolic syndrome, another known risk factor of altered male reproductive hormone systems. However, AFP has not been the focus of previous studies on male fertility or semen quality. BUN, the index for kidney function, is also identified as a risk factor by our established ML model. Our results support previous findings that metabolic syndrome has negative impacts on sperm count and semen quality. Sleep duration also has an impact on sperm generation in the testes. AFP and BUN are two novel risk factors linked to sperm counts. These findings could help healthcare personnel and law makers create strategies for creating environments to increase the country's fertility rate. This study should also be of value to follow-up research.

An improved Wavenet network for multi-step-ahead wind energy forecasting

Accurate multi-step-ahead wind speed (WS) and wind power (WP) forecasting are critical to the scheduling, planning, and maintenance of wind farms. Previous forecasting methods tend to focus on improving forecast accuracy by integrating different models and disaggregating data while neglecting the forecasting ability of basic models. In addition, traditional multi-step-ahead output strategies have limitations that constrain the forecasting capability of models. To overcome the above challenges, this study proposes a novel forecasting model called ED-Wavenet-TF. It adopts two Wavenet networks as Encoder and Decoder connected by the multi-head self-attention mechanism. And, teacher forcing is used as the multi-step-ahead output strategy for WS and WP forecasting. In the training phase, ED-Wavenet-TF uses a portion of the actual data to correct the errors at the intermediate forecasting steps, while in the forecasting phase, it runs through an inference loop to make forecasts. In this study, two WS datasets and two WP datasets are used to validate the performance of ED-Wavenet-TF with univariate input. The results show that compared with Wavenet, the symmetric mean absolute percentage error of ED-Wavenet-TF at four forecasting steps is lower by at least 4.8577% on average for the WS datasets and 8.9463% on average for the WP datasets. The advantages of ED-Wavenert-TF over ten comparable models are confirmed by four evaluation indicators and the Harvey, Leybourne, and Newbold statistical hypothesis test. Moreover, ED-Wavenet-TF is extended to make multi-step-ahead forecasts with multivariate inputs, whose effectiveness is demonstrated on another open WS dataset.

Forecasting energy consumption demand of customers in smart grid using Temporal Fusion Transformer (TFT)

Energy consumption prediction has always remained a concern for researchers because of the rapid growth of the human population and customers joining smart grids network for smart home facilities. Recently, the spread of COVID-19 has dramatically increased energy consumption in the residential sector. Hence, it is essential to produce energy per the residential customers' requirements, improve economic efficiency, and reduce production costs. The previously published papers in the literature have considered the overall energy consumption prediction, making it difficult for production companies to produce energy per customers' future demand. Using the proposed study, production companies can accurately have energy per their customers' needs by forecasting future energy consumption demands.Scientists and researchers are trying to minimize energy consumption by applying different optimization and prediction techniques; hence this study proposed a daily, weekly, and monthly energy consumption prediction model using Temporal Fusion Transformer (TFT). This study relies on a TFT model for energy forecasting, which considers both primary and valuable data sources and batch training techniques. The model's performance has been related to the Long Short-Term Memory (LSTM), LSTM interpretable, and Temporal Convolutional Network (TCN) models. The model's performance has remained better than the other algorithms, with mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) of 4.09, 2.02, and 1.50. Further, the overall symmetric mean absolute percentage error (sMAPE) of LSTM, LSTM interpretable, TCN, and proposed TFT remained at 29.78%, 31.10%, 36.42%, and 26.46%, respectively. The sMAPE of the TFT has proved that the model has performed better than the other deep learning models.

Quantification of Physiological Parameters of Rice Varieties Based on Multi-Spectral Remote Sensing and Machine Learning Models

Estimating plant physiological indicators with remote sensing technology is critical for ensuring precise field management. Compared with other remote sensing platforms, low-altitude unmanned aerial vehicles (UAVs) produce images with high spatial resolution that can be used to clearly identify vegetation. However, the information of UAV image data is relatively complex and difficult to analyze, which is the main problem limiting its large-scale use at present. In order to monitor plant physiological indexes from the multi-spectral data, a new method based on machine learning is studied in this paper. Using UAV for deriving the absorption coefficients of plant canopies and whole leaf area, this paper quantifies the effects of plant physiological indicators such as the soil and plant analyzer development (SPAD) value, whole leaf area, and dry matter accumulation on the relationship between the reflectance spectra. Nine vegetation indexes were then extracted as the sensitive vegetation indexes of the rice physiological indicators. Using the SVM model to predict the SPAD value of the plant, the mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and symmetric mean absolute percentage error (SMAPE) values of the model were 1.90, 1.38, 0.13, 0.86, and 4.13, respectively. The results demonstrate that the rice plants display a considerable biochemical and spectral correlation. Using SVM to predict the SPAD value has a better effect because of a better adaptation and a higher accuracy than other models. This study suggests that the multi-spectral data acquired using UAV can quickly estimate field physiological indicators, which has potential in the pre-visual detection of SPAD value information in the field. At the same time, it can also be extended to the detection and inversion of other key variables of crops.