Journal of Computational Mechanics, Power System and Control

Indian Music Classification using Neural network based Dragon fly algorithm

B. Kranthi Kiran — Sat, 03 Jul 2021 00:00:00 +0530

Generally, Indian Classical Music (ICM) is categorized into 2 Hindustani and Carnatic. Although, aforesaid music formats possess the same base, the presentation manner is different in numerous ways. The ICM basic modules are taala and raga. Fundamentally, Taala indicates rhythmic beats else patterns. From the flow of swaras, the raga is ascertained that is indicated as extensive terms. On the basis of few important factors, the raga is indicated namely aarohana-avarohna and swaras and distinctive phrases. In practice, the basic frequency is Swara that is exact via time period. In addition, there are numerous other issues with the automatic raga identification technique. Hence, raga is identified in this research without using precise note series information and vital to use an effectual classification technique. This research develops an effectual raga recognition model via the Carnatic genre music that is efficiently identified for the data mining models, which is also included. Here, the Neural Network (NN) model is proposed which is an adaptive classifier in that the feature set is exploited to learning, and the data mining models are used for the classification techniques. Moreover, a meta-heuristic approach is used for the adaptive classifier to obtain the extracted feature set knowledge. As the learning technique plays an important role in describing the accuracy of the raga identification model, it prefers to adopt the Dragonfly algorithm.

Improved Chicken Swarm Optimization based NARX neural network:Artefacts removal from ECG signal

Shivchetan Sambaragimath — Tue, 20 Jul 2021 00:00:00 +0530

One of the most important research areas is the Speech Emotion Recognition (SER) technique, which is applied in many fields such as speech processing and human-computer interaction. In general, it is mainly concentrated on using the techniques of machine learning in order to predict the precise emotional category from speech. In affective computing as well as the human-computer interaction, the developed applications of SER are very effective that are considered as the important module of the computer's next-generation system. It is due to the automatic service provisions are granted by the natural human-machine interface that requires an improved approval of user emotional conditions. Hence, this work proposes a novel SER model which integrated both emotion and gender recognition. Various features are extracted and that is fed for the emotions classifications. In this work, the Deep Belief Network (DBN) is exploited. At last, performance analysis of the developed technique is seen that better accuracy rate while comparing with the conventional models. This work proposes a novel technique for the SER model which helps both emotion as well as gender recognition. Here, the proposed Hybrid Particle Swarm Optimization (PSO) and Gravitational Search Algorithm (GSA) algorithm are introduced to identify the optimal weight of the DBN technique.

Improved Shuffled Jaya Algorithm: Advancement of LVRT Capacity of PV-Array using MPPT Algorithm

Fatema Murshid AlBalushi — Wed, 14 Jul 2021 00:00:00 +0530

Wind Turbine (WT) has become an acceptable replacement for electricity production by enormous financial and environmental compensations by fossil or nuclear power plants. Nowadays various researchers are working in this field to propose the WT performance to the Doubly Fed Induction Generator (DFIG)- Low Voltage Ride Through (LVRT) system with maximum increase as well as flexibility. In order to conquer the characteristics of non-linearity of WT, exploiting Maximum Power Point Tracking (MPPT) algorithm, the PV array is operated which is involved beside with the WT to enhance system performance. For the DFIG-LVRT system, this work proposes to experiment with the Control System (CS) which plays an important role in the experimentation of controllers to correct error signals. In this research, a new algorithm named Improved Shuffled Jaya Algorithm (IS-JA) by means of fuzzified error model to experiment optimized CS. Moreover, it differentiates the proposed model-based LVRT system by means of the conventional LVRT system as well as system by means of the least increase, utmost increase. Furthermore, the developed model is evaluated concerning the conventional models, and it also states that the results in terms of the quantitative analysis. In addition, quantitative analysis is carried out which presents the estimation of Root Mean Sqaure Error (RMSE) with altering speed. Hence, the proposed model is shown that its performance is better than conventional models.

Hybrid Particle Swarm Optimization-Gravitational Search Algorithm based Deep Belief Network: Speech Emotion Recognition

J Rajeshwar — Wed, 14 Jul 2021 00:00:00 +0530

Improved Butterfly Optimization Algorithm: PI Controller for 7-Level Inverter

Meghna Sangtani — Wed, 14 Jul 2021 00:00:00 +0530

In environmental parameters, the Photovoltaic (PV) system generates electricity which varies with deviations as well as at the Maximum Power Point (MPP) the PV network will operate. In addition, to improve the effectuality and also the Renewable Energy Sources (RES) performance, the energy storage devices can act as a possible solution. Moreover, this work tries to set up a control model using an optimization algorithm aided PI controller for the seven-level inverter. On the basis of this, the Proportional Integral (PI) controller increases are attuned dynamically using the Improved Butterfly Optimization Algorithm (IBOA). The increase must be maintained so that the error between the fault signal and reference signal must be the least. Therefore, using the developed optimized PI controller, superior dynamic performance can be attained. At last, the developed model performance is evaluated with the conventional models regarding various metrics, and the proposed method shown its dominance over the conventional models.