Breast cancer is one of the major issues among women in developed nations. It occurs based on their lifestyles and eating habits. As per many researchers, cancer leads to death. Likewise, breast cancer is also one of the leading dangerous cancer types that may result in critical issues over a body. Hence to avoid such activities it is necessary to detect the cancer spot and provide some clinical experiments, it is termed as early detection. The detection of this type of cancer is difficult at initial stage because the cancerous tumours are rooted in the common breast tissue structures. The main objective of this research is to model a breast cancer prediction system with a novel machine learning approach based on wavelets is proposed. The prediction of breast cancer for diagnosis process is made by the proposed algorithm named as Optimally Pruned Hybrid Wavelet Kernel-based Extreme Learning Machine (HOP-WKELM). Initially, the input is pre-processed for noise reduction using kuan filter. After that, Quantum Evolutionary Algorithm (QEA) is applied to segment the cancer part in mammogram image and feature extraction using Grey-level co-occurrence matrix (GLCM), Gabor filter and Local Binary Pattern (LBP) features. The extracted features are classified using HWKELM classifier. In this HWKELW, WKELM learning algorithm utilized the Dragonfly Swarm Behaviour-based Optimization (DSBO) approach to optimize the parameters of kernel functions. The proposed strategies achieved a maximum accuracy of 98.8% and a maximum precision of 98.1% when compared with existing Adaboost systems.

Utilization of power system with management and monitoring of smart grid system is essential in global wise. Therefore, Phasor Measurement Units (PMUs) device is used for advanced monitoring of real time system, analysis and control in smart power grid system. An Optimal PMU Placement (OPP) is quite important during planning studies for both existing and future power networks. So, the problems of minimizing the number of PMUs and their optimal location for system are mainly considered. In this work, to achieve better optimal location and reduce the number of PMUs and to increase the measurement redundancy, an Enhanced Bat Algorithm (EBA) is proposed for optimization with phasing module. Phasing is mainly considered for OPP to reduce the cost. In phasing unit, PMUs are separated into three phases and those are concentrated in each phase separately. Finally, IEEE-14 bus standard systems and the Tamil Nadu state power grid (83 bus system) are utilized to test the proposed model. The simulation results are evaluated and compared with existing algorithm to show the efficient process of OPP with better optimization, less cost and minimum redundancy than the existing optimization schemes.

In the applications of Wireless Sensor Network, sensor nodes broadcast the data towards the destination. Localization is considered as the most important and interesting area in WSN as identifying the location of sensor nodes is needful for many situations. The proposed work, points on discovering the sensor node location and promotes cross layer effective routing path searching for the reliable data transmission in the network. The proposed work consists of three main phases. The initial phase includes the cross layer link establishment. The second phase includes, identifying the fuzziness of wireless sensors. It tracks the location of sensor nodes and analyses the behaviour. This can be done with the algorithm namely Fuzziness based Contiguous nodes Refining Algorithm. Since the characteristics of nodes may vary often, the algorithm finds the unsuitable communication node in the routing path and removes it. Source node controls the activities of the intermediate nodes in the routing path. After identifying and removing the unsuitable nodes, in final phase, an effective routing path is established by using the scheme called Efficient Fuzziness based Contiguous node refining with cross-layer routing Scheme (EFCRS) is proposed. This selects the routing nodes from starting node to ending node with the lesser energy usage. At last, destination node acquires the enhanced throughput and reduced consumption of energy. It also obtains minimum packet latency. Simulation is carried out and found it achieves better performance than existing algorithms in comparing metrics end to end delay, overhead, connectivity ratio, throughput and energy consumption.

In this paper a heuristic technique based optimal location and sizing of UPFC is proposed to improve the dynamic stability. The UPFC sizing is predicted and determines the favourable location to fix UPFC with constraint of power flow. Also, it generates the effects of power flow to identify the maximum power loss, voltage, real and reactive power flow. By using GA based Modified Artificial Bees Colony (MABC) algorithm the optimum location has been determined and reduce the power loss with reactive and real power flow. The violated power flow quantity is optimized by using the Improved Gravitational Search Algorithm (IGSA) and the initial condition of UPFC required quantity is recovered. IGSA is used to identify the maximum loss of power to fix the UPFC to avoid power flow affects by generator outage. The proposed work performance analyses are evaluated by implementing the proposed algorithm in the MATLAB/SIMULINK platform. The analysis results of the proposed algorithm shows better performance than the existing and also determine the pre-eminence of the proposed approach and confirm it’s prospective to solve the problem.

Wireless sensor nodes are fixed in stable environment, so more distance between nodes in network cause packet drop. During transmission period node broadcast data packets there is coverage problem. It affects network lifetime, normal sensor nodes have certain limits for packet exchange, which makes unbalanced load for routing node. It minimizes the packet delivery rate, also consume more energy. Proposed Improved Communication Confining Clustering (ICCC) Method, information is exchanged in cluster member nodes, which are organized by cluster head. Nodes confining with intra clustering, provides the stable link among nodes in cluster achieve effective packet transmission in that node. Unceasing possession Node allocation algorithm is applied to clustering, it merges many feature in cluster members nodes also reduce energy usage range. This method not allows the re-clustering of nodes that reduce time delay.