An Enhanced Version of Cat Swarm Optimization Algorithm for Cluster Analysis

Clustering is an unsupervised machine learning technique that optimally organizes the data objects in a group of clusters. In present work, a meta-heuristic algorithm based on cat intelligence is adopted for optimizing clustering problems. Further, to make the cat swarm algorithm (CSO) more robust for partitional clustering, some modifications are incorporated in it. These modifications include an improved solution search equation for balancing global and local searches, accelerated velocity equation for addressing diversity, especially in tracing mode. Furthermore, a neighborhood-based search strategy is introduced to handle the local optima and premature convergence problems. The performance of enhanced cat swarm optimization (ECSO) algorithm is tested on eight real-life datasets and compared with the well-known clustering algorithms. The simulation results confirm that the proposed algorithm attains the optimal results than other clustering algorithms.

Performance Analysis of CAT Swarm Optimization Algorithm in Disruption Tolerant Networks

Disruption tolerant networks (DTN) are networks that provide unguided technologies. Solar technologies and Radio frequencies are used to operate the network. In DTN networks the connectivity does not last for a long time and they do not provide end to end connectivity. Therefore it uses store and forward technique to forward the packets to the destination nodes. When N number of nodes is participating in the network, each node receives packets from the previous node and sends acknowledgment to the sender node. Time delay occurs on receiving and sending acknowledgment continuously. Collision occurs due to the congestion in the network. Due to the irregular connectivity in the network, the compromised nodes try to drop the whole packet or part of the packet. The Blackhole and Greyhole attacks occur due to the packet loss. Optimized algorithm can be used to solve the above attacks. By using CAT Swarm optimization algorithm, the attacks can be prevented and it minimizes the Time delay in delivering the packets.

Global Maximum Power Point Tracking of Solar Photovoltaic Strings under Partial Shading Conditions Using Cat Swarm Optimization Technique

The power versus voltage curves of solar photovoltaic panels form several peaks under fractional (partial) shading conditions. Traditional maximum output power tracking (MPPT) techniques fail to achieve global peak power at the output terminals. The proposed Cat Swarm Optimization (CSO) method intends to apply MPPT techniques to extract the global maxima from the shaded photovoltaic systems. CSO is a robust and powerful metaheuristic swarm-based optimization technique that has received very positive feedback since its emergence. It has been used to solve a variety of optimization issues, and several variations have been developed. The CSO-based maximum power tracking technique can successfully tackle two major issues of the PV system during shading conditions, including random oscillations caused by conventional tracking techniques and power loss. The proposed techniques have been extensively used in comparison to conventional algorithms like the Perturb and the Observe (P and O) technique. The main objective is to achieve a tracking speed for extracting the Maximum Power Point (MPP) from the solar Photovoltaic (PV) system under fractional shading conditions by using CSO. Modeling of the solar photovoltaic array in the MATLAB/Simulink platform comprises a photovoltaic module, a switching converter (Boost Converter), and the load. The PSO and CSO techniques are applied to the PV module under different weather conditions. The PSO algorithm is compared to the CSO algorithm according to simulation results, revealing that the CSO algorithm can provide better accuracy and a faster tracking speed.