PERAMALAN JUMLAH TITIK PANAS PROVINSI KALIMANTAN TIMUR MENGGUNAKAN METODE RADIAL BASIS FUNCTION NEURAL NETWORK

Radial Basis Function Neural Network (RBFNN) is a neural that uses a radial base function in hidden layers for classification and forecasting purposes. Neural Network is developed into a radial function base with an information processing system that has characteristics similar to biological neural networks, consisting of input layers, hidden layers, and output layers. The data used in this study is data on the number of hotspots in East Kalimantan Province obtained from the official website of the National Aeronautics and Space Administration (NASA). The purpose of this research is to obtain the RBFNN model and the results of forecasting the number of hotspots for the period January 2020 to March 2020. The radial basis function used is the local Gaussian function and the linear activation function. In this study using the proportion of training data and testing data 70: 30; 80:20; and 90:10. The results showed that the input network using significant Partial Autocorrelation Function (PACF) at lag 1 and lag 2, so that the RBFNN model that was formed involved Xt-1 and Xt-2. The best Mean Absolute Percentage Error (MAPE) minimum obtained the 80:20 data proportion with 2 hidden networks. The RBFNN architecture that is formed is 2 input layers, 2 hidden layers and 1 output layer. Data from forecasting the number of hotspots in East Kalimantan Province shows that from January 2020 to February 2020 there was a decline and March 2020 an increase.

SHIFTED LEGENDRE FRACTIONAL PSEUDOSPECTRAL DIFFERENTIATION MATRICES FOR SOLVING FRACTIONAL DIFFERENTIAL PROBLEMS

A new differentiation technique, fractional pseudospectral shifted Legendre differentiation matrices (FSL D-matrices), was introduced. It depends on shifted Legendre polynomials (SLPs) as a base function. We take into consideration its extreme points and inner product. The technique was used to solve fractional ordinary differential equations (FODEs). Moreover, it extended to approximate fractional integro-differential equations (FIDEs) and fractional optimal control problems (FOCPs). The novel FSL D-matrices transformed these fractional differential problems (FDPs) into an algebraic system of equations. Also, an error and a convergence analysis for that technique were investigated. Finally, the correctness and efficiency of this technique were examined with test functions and several examples. All the results were compared with the results of other methods to ensure the investigated error analysis.

Automatic Differentiation in Automatic Generation of the Linearized Equations of Motion

In the ongoing search for mathematically efficient methods of predicting the motion of vehicle and other multibody systems, and presenting the associated results, one of the avenues of continued interest is the linearization of the equations of motion. While linearization can potentially result in reduced fidelity in the model, the benefits in computational speed often make it the pragmatic choice. Linearization techniques are also useful in modal and stability analysis, model order reduction, and state and input estimation. This paper explores the application of automatic differentiation to the generation of the linearized equations of motion. Automatic differentiation allows one to numerically evaluate the derivative of any function, with no prior knowledge of the differential relationship to other functions. It exploits the fact that every computer program must evaluate every function using only elementary arithmetic operations. Using automatic differentiation, derivatives of arbitrary order can be computed, accurately to working precision, with minimal additional computational cost over the evaluation of the base function. There are several freely available software libraries that implement automatic differentiation in modern computing languages. In the paper, several example multibody systems are analyzed, and the computation times of the stiffness matrix are compared using direct evaluation and automatic differentiation. The results show that automatic differentiation can be surprisingly competitive in terms of computational efficiency.

Molecular Docking Study of Alzheimer Disease Responsible Protein Inhibition Activity by Bioactive Ginkgolides

Terpenoids are major components present in herbal formulations of Ginkgo biloba which are considered to slow down progression of Alzheimer disease. Ginkgolide A, Ginkgolide B, Ginkgolide C, Ginkgolide M, Ginkgolide J, Ginkgolide K and Bilobalide are some of the terpenoids selected for computational theoretical calculations using DFT theory at B3LYP/6-311+G*(d,p) basic set level using Gaussian 16W. To study the interaction between selected terpenoids and selected proteins, molecular docking analysis is carried out using Argus Lab (4.0.1) and Auto Dock (4.2). Calculations are carried out on efficient shape-based search algorithm principle and a score base function to calculate the binding energies between them. ADMET analysis provide properties insight of terpenoids compounds. Results from calculated data reveal that there are possible interactions. This data can help in development of potent protein kinase inhibitor for the treatment of Alzheimer.

Sliding Mode Control of a 2DOF Robot Manipulator: A Simulation Study Using Artificial Neural Networks with Minimum Parameter Learning

Background: In this paper, we have developed an intelligent control law for the control of mobile manipulator robots by investigating the various techniques proposed in the literature. Thus, we have adopted a hybrid approach that integrates a part of classical and advanced automation in order to create an efficient control structure that can cope with a certain level of complexity. Our research logic is based on the process of keeping in mind that the control system must comply with the constraints imposed during the implementation of the control architecture. Objective: This paper aims to develop a control law in order to guarantee a certain level of performance, more precisely, during a trajectory tracking application for mobile handling missions. The developed control law guarantees robustness with respect to external disturbances and parametric uncertainties due to the modelling of the system. Methods: In this paper, a study of the basic concepts of robotics and robot modelling is presented in order to set up the dynamic model used for the elaboration of the command. A sliding-mode controller based on a radial base function neural network with minimum parameter learning is developed for the Pelican robot as a two-link robot manipulator. This approach, which combines a radial base function neuronal network (RBFNN) and a sliding mode control (SMC), is presented for the tracking control of this class of systems with unknown non-linearities. The centre and output weights of the RBFNN are updated via online learning in accordance with the adaptive laws, allowing the control output of the neural network to approach the equivalent control in the sliding mode in the predetermined direction. The Lyapunov function is used to develop the adaptive control algorithm based on the RBFNN model. For reducing the computational load and increasing real-time arm performance, an RBFNN-based on the SMC with the Minimum Parameter Learning (MPL) method is designed. Results: Neural network sliding mode control is designed to underline the effectiveness of the approach to control the manipulator;—this method of control is used to ensure the tracking trajectories. Conclusion: The results of the simulation for the manipulator's arm demonstrated the effectiveness of the modelling strategy, the correction, and the robustness of the control approach.

Least Square-Support Vector Machine Based Modeling for Electric Arc Furnaces

The process of making steel from the scrap metal by means of electric arc furnaces (EAF) has been used extensively in the industry. Accurate modelling of EAFs is, therefore, desired to assess their operations and their impacts on the electrical network. A number of approaches have already been used to model the v-i behavior of electric arc furnaces including mathematical methods and data-driven models. The objective of this thesis is to investigate the data-driven modelling methodologies, in particular, least square support vector machine (LS-SVM). The results obtained show that the proposed method with radial base function kernel provides the model to predict both arc current and arc voltage of EAFs.

Least Square-Support Vector Machine Based Modeling for Electric Arc Furnaces

The process of making steel from the scrap metal by means of electric arc furnaces (EAF) has been used extensively in the industry. Accurate modelling of EAFs is, therefore, desired to assess their operations and their impacts on the electrical network. A number of approaches have already been used to model the v-i behavior of electric arc furnaces including mathematical methods and data-driven models. The objective of this thesis is to investigate the data-driven modelling methodologies, in particular, least square support vector machine (LS-SVM). The results obtained show that the proposed method with radial base function kernel provides the model to predict both arc current and arc voltage of EAFs.

Memprediksi Jumlah Siswa Baru Menggunakan Metode Backpropagation (STUDI KASUS: SMK HARAPAN BANGSA KUALA)

Artificial Neural Network (ANN) and time series data can be used for forecasting methods well. Artificial Neural Network is a method whose working principle is adapted from a mathematical model in humans or biological nerves. Neural networks are characterized by; (1) the pattern of connections between neurons (called architecture), (2) determining the weight of the connection (called training or learning), and (3) the activation function. The research objective was to obtain the best artificial neural network architecture, comparing the two methods of Backpropogation Neural Networks with the Radial Base Function Artificial Neural Network (RBF) method. This research is a research using real data (true experimental). This research was conducted at SMK Harapan Bangsa Kuala, which was obtained from 2015 to 2019. The results showed that for one iteration using the backpropagation method the result was 0,378197657 with a squared error 0.143033468, then the results achieved were not in accordance with the target.

Least square-support vector machine based modeling for electric arc furnaces

The process of making steel from the scrap metal by means of electric arc furnaces (EAF) has been used extensively in the industry. Accurate modelling of EAFs is, therefore, desired to assess their operations and their impacts on the electrical network. A number of approaches have already been used to model the v-i behavior of electric arc furnaces including mathematical methods and data-driven models. The objective of this thesis is to investigate the data-driven modelling methodologies, in particular, least square support vector machine (LS-SVM). The results obtained show that the proposed method with radial base function kernel provides the model to predict both arc current and arc voltage of EAFs.