Comparison of ARIMA and LSTM for prediction of hemorrhagic fever at different time scales in China

Objectives This study intends to build and compare two kinds of forecasting models at different time scales for hemorrhagic fever incidence in China. Methods Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory Neural Network (LSTM) were adopted to fit monthly, weekly and daily incidence of hemorrhagic fever in China from 2013 to 2018. The two models, combined and uncombined with rolling forecasts, were used to predict the incidence in 2019 to examine their stability and applicability. Results ARIMA (2, 1, 1) (0, 1, 1)12, ARIMA (1, 1, 3) (1, 1, 1)52 and ARIMA (5, 0, 1) were selected as the best fitting ARIMA model for monthly, weekly and daily incidence series, respectively. The LSTM model with 64 neurons and Stochastic Gradient Descent (SGDM) for monthly incidence, 8 neurons and Adaptive Moment Estimation (Adam) for weekly incidence, and 64 neurons and Root Mean Square Prop (RMSprop) for daily incidence were selected as the best fitting LSTM models. The values of root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE) of the models combined with rolling forecasts in 2019 were lower than those of the direct forecasting models for both ARIMA and LSTM. It was shown from the forecasting performance in 2019 that ARIMA was better than LSTM for monthly and weekly forecasting while the LSTM was better than ARIMA for daily forecasting in rolling forecasting models. Conclusions Both ARIMA and LSTM could be used to build a prediction model for the incidence of hemorrhagic fever. Different models might be more suitable for the incidence prediction at different time scales. The findings can provide a good reference for future selection of prediction models and establishments of early warning systems for hemorrhagic fever.

Research on Teaching Practice of Blended Higher Education Based on Deep Learning Route

This paper establishes a hybrid education teaching practice quality evaluation system in colleges and constructs a hybrid teaching quality evaluation model based on a deep belief network. Karl Pearson correlation coefficient and root mean square error (RMSE) indicators are used to measure the closeness and fluctuation between the effective online teaching quality evaluation results evaluated by this method and the actual teaching quality results. The experimental results show the following: (1) As the number of iterations increases, the fitting error of the DBN model decreases significantly. When the number of iterations reaches 20, the fitting error of the DBN model stabilizes and decreases to below 0.01. The experimental results show that the model used in this method has good learning and training performance, and the fitting error is low. (2) The evaluation correlation coefficients are all greater than 0.85, and the root mean square error of the evaluation is less than 0.45, indicating that the evaluation results of this method are similar to the actual evaluation level and have small errors, which can be effectively applied to online teaching quality evaluation in colleges and universities.

Scalar Forcing Methodology for Direct Numerical Simulations of Turbulent Stratified Mixture Combustion

Scalar forcing in the context of turbulent stratified flame simulations aims to maintain the fuel-air inhomogeneity in the unburned gas. With scalar forcing, stratified flame simulations have the potential to reach a statistically stationary state with a prescribed mixture fraction distribution and root-mean-square value in the unburned gas, irrespective of the turbulence intensity. The applicability of scalar forcing for Direct Numerical Simulations of stratified mixture combustion is assessed by considering a recently developed scalar forcing scheme, known as the reaction analogy method, applied to both passive scalar mixing and the imperfectly mixed unburned reactants of statistically planar stratified flames under low Mach number conditions. The newly developed method enables statistically symmetric scalar distributions between bell-shaped and bimodal to be maintained without any significant departure from the specified bounds of the scalar. Moreover, the performance of the newly proposed scalar forcing methodology has been assessed for a range of different velocity forcing schemes (Lundgren forcing and modified bandwidth forcing) and also without any velocity forcing. It has been found that the scalar forcing scheme has no adverse impact on flame-turbulence interaction and it only maintains the prescribed root-mean-square value of the scalar fluctuation, and its distribution. The scalar integral length scale evolution is shown to be unaffected by the scalar forcing scheme studied in this paper. Thus, the scalar forcing scheme has a high potential to provide a valuable computational tool to enable analysis of the effects of unburned mixture stratification on turbulent flame dynamics.

Elucidation of contact state on various rough surfaces via highly robust colorimetric optical interferometry

In this study, we developed and practiced colorimetric optical interferometry for the direct observation of contact states to clarify contact phenomena. We theoretically demonstrated that the effect of roughness diffuse reflection could be neglected using interferometric light intensity according to the relationship between the optical film thickness and hue. Then, we measured the static contact surfaces of spherical test pieces of different root mean square roughnesses. Results indicate that the nominal contact area is significantly larger than that obtained from the Hertzian theory of smooth contact as the surface roughness increases. The contact film thickness on the nominal contact area increases almost in proportion to the root mean square roughness. Our experiment supports the validity of the contact theory and contact simulation with very small roughnesses, which have been difficult to verify experimentally. The advantage of this measurement is that it can simultaneously capture the macroscopic contact area and microscopic film thickness distribution, which is expected to further expand the range of application.

Analysis of vibrations in a diesel engine produced by Jatropha biodiesel using heterogeneous catalyst

Extensive consumption of fossil fuel has contributed to the worldwide decline of its reserves and detrimental effect on the environment. Therefore, it is essential to explore alternative option of fuel for diesel engine. The main objective of this research article is to optimize vibrations in a single-cylinder variable compression ratio diesel engine driven by Jatropha biodiesel blend. The heterogeneous catalyst (calcium oxide) is used to manufacture of biodiesel from Jatropha curcas oil by a process of transesterification. The optimization technique (Response Surface Methodology) has been employed to optimize root mean square acceleration of vibration by taking load, compression ratio (CR), and fuel injection pressure (FIP) as engine input parameters. Experiments were designed according to central composite design. The amplitude of the frequency domain signals is determined using Fast Fourier Transform and the influence of input parameters has been investigated in the frequency domain analysis of the vibration signatures. The adequacy and significance of the models have been checked by p-value and F value tests. Regression coefficients Adj. R2, R2, Pred. R2 were also found in acceptable range. The experimental outcome reveals that biodiesel yield of 81.6% was obtained at methanol-to-oil molar ratio of 12:1, reaction temperature of 65°C, reaction time of 3 h, and catalyst concentration of 5 wt%. Simultaneously, the model obtained a series of solutions based on the desirability criteria and proposed optimum setting of engine input parameters at a load of 2.59 kg, 17.94 CR, and 268.76 bar FIP for B30 blend. B30 blend generated root mean square acceleration of 4.46 m/s2 at above optimized conditions. A validation trial was conducted and the percentage of error for root mean square acceleration was found to be 2.3356% and 1.3039%, respectively, for B0 and B30 blend.

Hourly Forecasting of Solar Photovoltaic Power in Pakistan Using Recurrent Neural Networks

The solar photovoltaic (PV) power forecast is crucial for steady grid operation, scheduling, and grid electricity management. In this work, numerous time series forecast methodologies, including the statistical and artificial intelligence-based methods, are studied and compared fastidiously to forecast PV electricity. Moreover, the impact of different environmental conditions for all of the algorithms is investigated. Hourly solar PV power forecasting is done to confirm the effectiveness of various models. Data used in this paper is of one entire year and is acquired from a 100 MW solar power plant, namely, Quaid-e-Azam Solar Park, Bahawalpur, Pakistan. This paper suggests recurrent neural networks (RNNs) as the best-performing forecasting model for PV power output. Furthermore, the bidirectional long-short-term memory RNN framework delivered high accuracy results in all weather conditions, especially under cloudy weather conditions where root mean square error (RMSE) was found lowest 0.0025, R square stands at 0.99, and coefficient of variation of root mean square error (RMSE) Cv was observed 0.0095%.

DEVELOPMENT OF RAINFALL FORECASTING MODEL USING MACHINE LEARNING WITH SINGULAR SPECTRUM ANALYSIS

Agriculture is the key point for survival for developing nations like India. For farming, rainfall is generally significant. Rainfall updates are help for evaluate water assets, farming, ecosystems and hydrology. Nowadays rainfall anticipation has become a foremost issue. Forecast of rainfall offers attention to individuals and knows in advance about rainfall to avoid potential risk to shield their crop yields from severe rainfall. This study intends to investigate the dependability of integrating a data pre-processing technique called singular-spectrum-analysis (SSA) with supervised learning models called least-squares support vector regression (LS-SVR), and Random-Forest (RF), for rainfall prediction. Integrating SSA with LS-SVR and RF, the combined framework is designed and contrasted with the customary approaches (LS-SVR and RF). The presented frameworks were trained and tested utilizing a monthly climate dataset which is separated into 80:20 ratios for training and testing respectively. Performance of the model was assessed using Root Mean Square Error (RMSE) and Nash–Sutcliffe Efficiency (NSE) and the proposed model produces the values as 71.6 %, 90.2 % respectively. Experimental outcomes illustrate that the proposed model can productively predict the rainfall. ABSTRAK:Pertanian adalah titik utama kelangsungan hidup negara-negara membangun seperti India. Untuk pertanian, curah hujan pada amnya ketara. Kemas kini hujan adalah bantuan untuk menilai aset air, pertanian, ekosistem dan hidrologi. Kini, jangkaan hujan telah menjadi isu utama. Ramalan hujan memberikan perhatian kepada individu dan mengetahui terlebih dahulu mengenai hujan untuk menghindari potensi risiko untuk melindungi hasil tanaman mereka dari hujan lebat. Kajian ini bertujuan untuk menyelidiki kebolehpercayaan mengintegrasikan teknik pra-pemprosesan data yang disebut analisis-spektrum tunggal (SSA) dengan model pembelajaran yang diawasi yang disebut regresi vektor sokongan paling rendah (LS-SVR), dan Random-Forest (RF), ramalan hujan. Menggabungkan SSA dengan LS-SVR dan RF, kerangka gabungan dirancang dan dibeza-bezakan dengan pendekatan biasa (LS-SVR dan RF). Kerangka kerja yang disajikan dilatih dan diuji dengan menggunakan set data iklim bulanan yang masing-masing dipisahkan menjadi nisbah 80:20 untuk latihan dan ujian. Prestasi model dinilai menggunakan Root Mean Square Error (RMSE) dan Nash – Sutcliffe Efficiency (NSE) dan model yang dicadangkan menghasilkan nilai masing-masing sebanyak 71.6%, 90.2%. Hasil eksperimen menggambarkan bahawa model yang dicadangkan dapat meramalkan hujan secara produktif.

Effect of the kinesio taping application method for the activation and strength of gastrocnemius muscles in healthy adults

The objective of this study was to verify the effect of kinesio-taping (KT) application of the origin for muscle insertion (O>I) and insertion for muscle origin (I>O) in healthy participants, through surface electromyography and peak torque of gastrocnemius muscles. A total of 69 participants with an average age of 22.9±5.2 years were evaluated, 41 women with an average age of 22.0±5.1 years, BMI 25.1±4.6 kg/m2, and 28 men with an average age of 22.0±5.1 years, BMI 23.1±3.3 kg/m2, randomized under three conditions: O>I, I>O and no KT (control), repeated three times with five-minute rest between sessions. The peak torque of the gastrocnemius lateral (GML) and medial (GMM) muscles was evaluated at speeds 30º/s 60º/s 120º/s and muscle activity through surface electromyography. Repeated-measurement ANOVA showed effect only on the variable speed (F=767,1; p<0.001) and the variables condition (F=0.010; p=0.990) and interaction (F=0.199; p=0.892) were not significant. In electromyography, Root mean Square (RMS) did not differ in the conditions evaluated, presenting standard behavior without significant differences. The KT application regardless of being O>I or I>O muscular, did not alter the muscle recruitment or contribute to the increase in peak torque performance during the three speeds.

Fruit Bromelain-Derived Peptide Potentially Restrains the Attachment of SARS-CoV-2 Variants to hACE2: A Pharmacoinformatics Approach

Before entering the cell, the SARS-CoV-2 spike glycoprotein receptor-binding domain (RBD) binds to the human angiotensin-converting enzyme 2 (hACE2) receptor. Hence, this RBD is a critical target for the development of antiviral agents. Recent studies have discovered that SARS-CoV-2 variants with mutations in the RBD have spread globally. The purpose of this in silico study was to determine the potential of a fruit bromelain-derived peptide. DYGAVNEVK. to inhibit the entry of various SARS-CoV-2 variants into human cells by targeting the hACE binding site within the RBD. Molecular docking analysis revealed that DYGAVNEVK interacts with several critical RBD binding residues responsible for the adhesion of the RBD to hACE2. Moreover, 100 ns MD simulations revealed stable interactions between DYGAVNEVK and RBD variants derived from the trajectory of root-mean-square deviation (RMSD), radius of gyration (Rg), and root-mean-square fluctuation (RMSF) analysis, as well as free binding energy calculations. Overall, our computational results indicate that DYGAVNEVK warrants further investigation as a candidate for preventing SARS-CoV-2 due to its interaction with the RBD of SARS-CoV-2 variants.