Hybrid Dragonfly Research Articles

Memory Based Hybrid Dragonfly Algorithm (MHDA): a New Technique for Determining Model Parameter in Vertical Electrical Sounding (VES) Data

Vertical Electrical Sounding (VES) data inversion is a nonlinear inversion problem because several models can fit to the observed data. Therefore, a new approach based on nonlinear optimization technique is implemented which is called Memory based Hybrid Dragonfly Algorithm (MHDA). It is proposed to solve drawback of Dragonfly Algorithm (DA), i.e. low convergence rate which is caused by high exploration behaviour of DA. The drawback can lead to the local optimum solutions. MHDA successfully balances exploration and exploitation behaviours of DA to obtain global optimum solution. In this research, initially, MHDA is tested for the noise contaminated synthetic VES data to assess its performance. Subsequently, MHDA is applied for the field VES data. In both results, MHDA is able to provide Posterior Distribution Model (PDM) which is obtained from exploration process. All accepted models of PDM have lower misfit value than specified tolerance value in the inversion process. The PDM can be used to estimate solution via median value of PDM. Additionally, the uncertainty estimation of obtained solution can be determined from standard deviation value of PDM. The inversion results of synthetic and field VES data indicate that MHDA is an innovative technique to solve VES data inversion problem.

A study on performance of MHDA in training MLPs

PurposeIn recent years, the application of metaheuristics in training neural network models has gained significance due to the drawbacks of deterministic algorithms. This paper aims to propose the use of a recently developed “memory based hybrid dragonfly algorithm” (MHDA) for training multi-layer perceptron (MLP) model by finding the optimal set of weight and biases.Design/methodology/approachThe efficiency of MHDA in training MLPs is evaluated by applying it to classification and approximation benchmark data sets. Performance comparison between MHDA and other training algorithms is carried out and the significance of results is proved by statistical methods. The computational complexity of MHDA trained MLP is estimated.FindingsSimulation result shows that MHDA can effectively find the near optimum set of weight and biases at a higher convergence rate when compared to other training algorithms.Originality/valueThis paper presents MHDA as an alternative optimization algorithm for training MLP. MHDA can effectively optimize set of weight and biases and can be a potential trainer for MLPs.

RETRACTED ARTICLE: Optimal Hop Selection Based Novel Trust Based DDoS Attack Removal Framework for Reliable and Secured Transmission of Data in VANETs

This paper proposes a new Trust based DDoS Attack Removal Framework (T-DARF) for effective trust management in VANETs. Based on a newly developed data centric validation unit and intrusion detection unit, all the misbehaving nodes are removed and allows the framework to prevent DDoS attacks in a distributed and collaborative manner. In other words, our proposed framework ensures a trusted, reliable communication between vehicles and delivering reduced network overhead by handling the DDoS attacks. T-DARF is built of three elements: (1) for the detection of dishonest nodes it includes a collaborative and distributed elements, (2) to filter the malicious data it includes a data-centric validation element, and (3) a delay checking element for the detection and prevention against DoS and DDoS attacks. Also, a new trust-based routing protocol is developed that uses the idea of companions (i.e. integrating the notions of link quality and trust value of neighbors) to select the trusted vehicles and to select the optimal path we imposed a hybrid Dragonfly based particle swarm optimization (DPSO) algorithm. The simulation results proved the efficiency of our proposed trust based framework on detecting dishonest nodes, the malicious data transmitted by the honest/dishonest nodes and also detecting the colluding attacks that destroys the network resources in a short period. Also, T-DARF in a worst case scenario can sustain its performance degrees and outperforms the effectiveness of existing schemes such as TRIP, AECFV and T-CLAIDS.

An efficient clustering approach for fair semantic web content retrieval via tri-level ontology construction model with hybrid dragonfly algorithm

Web pages are heterogeneous and complex and there exists complicated associations within one web pages and linking to the others. The high interactions between terms in pages demonstrate vague and ambiguous meanings. Efficient and effective clustering methods are needed to discover latent and coherent meanings in context are necessary. This paper proposes an efficient clustering approach for fair semantic web content retrieval based on tri-level ontology construction model with hybrid dragonfly algorithm. Initially the query processing phase, by making use of systematic adaptive hierarchy method (SAHM) efficient ontology selection process is carried out by means of matching keywords retrieved form user query. Secondly, fuzzy sensitive near-neighbour influence (FSNI) based clustering approach relied on the ontology driven fuzzy linguistic measure, applied to estimate the uncertainty that may be relevant to the semantic content which belongs to the user quires. The proposed FSNI clustering approach with HDA algorithm performance is be evaluated and compared with existing clustering approaches in terms of retrieval accuracy and surfing time.

AN EFFICIENT CLUSTERING APPROACH FOR FAIR SEMANTIC WEB CONTENT RETRIEVAL VIA TRI-LEVEL ONTOLOGY CONSTRUCTION MODEL WITH HYBRID DRAGONFLY ALGORITHM

Web pages are heterogeneous and complex and there exists complicated associations within one web pages and linking to the others. The high interactions between terms in pages demonstrate vague and ambiguous meanings. Efficient and effective clustering methods are needed to discover latent and coherent meanings in context are necessary. This paper proposes an efficient clustering approach for fair semantic web content retrieval based on tri-level ontology construction model with hybrid dragonfly algorithm. Initially the query processing phase, by making use of systematic adaptive hierarchy method (SAHM) efficient ontology selection process is carried out by means of matching keywords retrieved form user query. Secondly, fuzzy sensitive near-neighbour influence (FSNI) based clustering approach relied on the ontology driven fuzzy linguistic measure, applied to estimate the uncertainty that may be relevant to the semantic content which belongs to the user quires. The proposed FSNI clustering approach with HDA algorithm performance is be evaluated and compared with existing clustering approaches in terms of retrieval accuracy and surfing time.

Hybrid Dragonfly Optimization-Based Artificial Neural Network for the Recognition of Epilepsy

Epilepsy can well be stated as a disorder of the central nervous systems (CNS) that brought about recurring seizures owing to chronic abnormal blasts of electrical discharge on the brain. Knowing if an individual is having a seizure and diagnosing the seizure type or epilepsy syndrome could be hard. Many methods were developed to recognize this disease. But the existing techniques for detection of epilepsy are not satisfied with accuracy, and cannot identify the diseases effectively. To trounce these drawbacks, this paper proposes an approach for the recognition of Epilepsy as of the electroencephalography (EEG) signals. This is implemented as follows. Primarily, the Kalman filter (KF)is utilized for pre-processing to eradicate the impulse noise present in the EEG signals. This filtered signal is then decomposed utilizing variable modes decomposition (VMD). Feature extraction (FE) is performed by computing 7 features. The dimensionality of this signal is then lessened using Modified-Principal Components Analysis (M-PCA). Finally, classification is conducted utilizing the artificial neural networks (ANN) that is optimized using the hybrid dragonfly algorithm (HDA). Disparate performance metrics such as sensitivity, accuracy, and false discovery rates (FDR) are ascertained and as well weighted against with the existent works.

Design and Analysis of Tilt Integral Derivative Controller for Frequency Control in an Islanded Microgrid: A Novel Hybrid Dragonfly and Pattern Search Algorithm Approach

In this research work, a maiden approach is made for the frequency control in an islanded AC microgrid (MG). A MG can be formed by combining the different sources like renewable energy source, wind power generation and the solar energy generation. Variation in any of the source influences the MG frequency, and thus, the frequency control issue for MG is always a challenge for the researcher industry. In light of these difficulties, this paper considers a tilt integral derivative (TID) controller for the secondary frequency control of the islanded MGs and a novel hybrid dragonfly algorithm and pattern search (hDF-PS) algorithm is used to tune the controller parameters. In the proposed control conspire, some sources like microturbine, diesel engine generator and fuel cell are used which balance the load and power can demand of the MG. The novel hybrid controller is inspected on a MG test system, and the robustness and execution are assessed by considering different disturbances and parametric variations. In order to show the effectiveness of the proposed hybrid algorithm-based TID controller, it is being compared with some conventional controllers like integral, proportional integral and proportional integral derivative-based controller. It can be demonstrated that the proposed hDF-PS-based TID controller approach (because of considering organized/parametric instabilities) gives preferably better execution over the other control techniques.

Memory based Hybrid Dragonfly Algorithm for numerical optimization problems

A novel hybrid algorithm (MHDA) based on Dragon Fly and PSO is proposed.Performance is tested using standard benchmark problems.Proposed algorithm is compared with well-known optimization algorithms.Statistical analysis is done using Friedmans test and Wilcoxon signed ranksum test.Superiority of MHDA is also proved by applying on engineering design problems. Dragonfly algorithm (DA) is a recently proposed optimization algorithm based on the static and dynamic swarming behaviour of dragonflies. Due to its simplicity and efficiency, DA has received interest of researchers from different fields. However, it lacks internal memory which may lead to its premature convergence to local optima. To overcome this drawback, we propose a novel Memory based Hybrid Dragonfly Algorithm (MHDA) for solving numerical optimization problems. The pbest and gbest concept of Particle Swarm optimization (PSO) is added to conventional DA to guide the search process for potential candidate solutions and PSO is then initialized with pbest of DA to further exploit the search space. The proposed method combines the exploration capability of DA and exploitation capability of PSO to achieve global optimal solutions. The efficiency of the MHDA is validated by testing on basic unconstrained benchmark functions and CEC 2014 test functions. A comparative performance analysis between MHDA and other powerful optimization algorithms have been carried out and significance of the results is proved by statistical methods. The results show that MHDA gives better performance than conventional DA and PSO. Moreover, it gives competitive results in terms of convergence, accuracy and search-ability when compared with the state-of-the-art algorithms. The efficacy of MHDA in solving real world problems is also explained with three engineering design problems.