Soft Computing Approaches for Image Segmentation

In this chapter a review is done on the image segmentation techniques by using both traditional approaches and soft computing approaches. This chapter will figure out some soft computing approaches that can be used to solve the problem of identification of objects. Some traditional approaches for the extraction of objects are discussed along with their comparison with soft computing approaches. The chapter discusses various applications of image segmentation. The soft computing approaches are then analyzed and their performance is compared with the others by identifying the advantages and disadvantages of all.

Detection of Stator and Rotor Faults in Asynchronous Motor Using Artificial Intelligence Method

This chapter deals with the implementation of a PC-based monitoring and fault identification scheme for a three-phase induction motor using artificial neural networks (ANNs). To accomplish the task, a hardware system is designed and built to acquire three phase voltages and currents from a 3.3KW squirrel-cage, three-phase induction motor. A software program is written to read the voltages and currents, which are first used to train a feed-forward neural network structure. The trained network is placed in a Lab VIEW-based program formula node that monitors the voltages and currents online and displays the fault conditions and turns the motor. The complete system is successfully tested in real time by creating different faults on the motor.

Modeling and Dynamic Surface Control of Uncertain Strict-Feedback Nonlinear Systems Using Adaptive Fuzzy Wavelet Network

This chapter proposes a new modeling and control scheme for uncertain strict-feedback nonlinear systems based on adaptive fuzzy wavelet network (FWN) and dynamic surface control (DSC) approach. It designs adaptive FWN as a nonlinear-in-parameter approximator to approximate the uncertain dynamics of the system. Then, the proposed control scheme is developed by incorporating the DSC method to the adaptive FWN-based model. Stability analysis of the proposed scheme is provided and adaptive laws are designed to learn all linear and nonlinear parameters of the network. It is proven that all the signals of the closed-loop system are uniformly ultimately bounded and the tracking error can be made arbitrary small. The proposed scheme does not require any prior knowledge about dynamics of the system and offline learning. Furthermore, it eliminates the “explosion of complexity” problems and develops accurate model of the system and simple controller. Simulation results on the numerical example and permanent magnet synchronous motor are provided to show the effectiveness of the proposed scheme.

Sizing and Placement of Battery-Sourced Solar Photovoltaic (B-SSPV) Plants in Distribution Networks

This chapter proposes a mixed-integer optimization using genetic algorithm (MIOGA) for determining the optimum sizes and placements of battery-sourced solar photovoltaic (B-SSPV) plants to reduce the total energy losses in distribution networks. Total energy loss index (TELI) is formulated as the main objective function and meanwhile bus voltage deviations and PV penetrations of B-SSPV plants are calculated. To deal the stochastic behavior of solar irradiance, 15 years of weather data is modeled by using beta probability density function (Beta-PDF). The proposed algorithm is applied on IEEE 33 bus and IEEE 69 bus test distribution networks and optimum results are acquired for different time varying voltage dependent load models. From the results, it is known that, compared to PV only, the integration of B-SSPV plants in the distribution networks resulted in higher penetration levels in distribution networks. The proposed algorithm was very effective in terms of determining the sizes of the PV plant and the battery storage, and for the charging and discharging of the battery storage.

Utility and Significance of Vague Set Theory and Advanced Optimization Mechanisms for Uncertainty Management

In this digital world, tremendous data are generated in every field. Useful information is inferred out of this data, which is valuable for effective decision making. Data mining extracts the interesting information from huge volumes of data. Association rule (AR) mining is one of the core areas of data mining where interesting information is extracted in the form of rules. Traditional AR mining is incapable of handling uncertain situations. In order to handle uncertainty, mathematical tools like vague theory can be utilized with AR mining methodologies for the development of novel vague theory based algorithms, which will be more suitable in effectively handling vague situations that helps framing effective selling strategy. Since an organization can't analyze the huge rule set obtained from these algorithms, every resultant rule should have a certain ratio of factors customized to the interest of the organization that can be achieved through optimization algorithms. This chapter explores the significance of vague theory and optimization means for effective uncertainty management.

Lyapunov-Based Predictive Control Methodologies for Networked Control Systems

The problem considered in this chapter is to control a vehicle drivetrain in order to minimize its oscillations while coping with the time-varying delays introduced by the CAN communication network and the strict timing limitations. As such, two Lyapunov-based model predictive control design methodologies are presented: one based on modeling the network-induced time-varying delays using a polytopic approximation technique and the second one based on modeling the delays as disturbances. Several tests performed using an industry validated drivetrain model indicate that the proposed design methodologies can handle both the performance/physical constraints and the strict limitations on the computational complexity, while effectively coping with the time-varying delays. Moreover, a comparative analysis between the two Lyapunov-based model predictive control design methodologies in terms of computational complexity, number of optimization variables, and obtained performances is carried out.

Fast Chaotic Encryption Using Circuits for Mobile and Cloud Computing

In last few decades, a lot of work has been done in the field of cryptography; it is being considered one of the safe methods to protect data. It was first used to protect communication by individuals, armies, and organizational companies. With the help encryption method, anyone can protect their data from a third-party attack. Images are used in various areas like biometric authentication, medical science, military, etc., where they are being stored or transferred over the network and the safety of such images are very important. The newest movement in encryption is chaos-based, which is a better encryption technique than AES, DES, RSA, etc. It consists of different property such as sensitive independence on original situation, non-periodicity, non-convergence, etc. In recent times, many chaos-based image encryption algorithms have been proposed, but most of them are not sufficient to provide full protection to data. In this chapter, a survey of different chaos-based image encryption techniques is discussed.

Classification Techniques for Thunderstorms and Lightning Prediction

Environmental disasters affect the economy, biodiversity, human life, and living organisms. Thunderstorms are one of such environmental disaster. By using proper methodology of forecasting thunderstorms, the adverse effects can be reduced. The prediction of thunderstorms is the most difficult task in weather forecasting due to their temporal and spatial extension either physically or dynamically. Lightning is associated with thunderstorms, which causes wildfires, kills people and other living organisms. Heavy rain from thunderstorms causes flash flooding. In this regard, several researchers have proposed different methodology such as statistical, numerical mode, data mining, soft computing, and machine learning for forecasting of severe weather to reduce the damages. This chapter focuses existing classification methods on thunderstorms and lightning prediction. This chapter includes suggestions on the future research directions.

An Innovative Design of RF Energy Harvester for Wireless Sensor Devices

Due to growing demand of energy, green technologies are highly attractive among researchers because of their non-conventional nature. Energy harvesting is one of their best parts. Very low cost of maintenance and non-polluting nature are major reasons behind their growing demand. However, for ultra-low power applications, such as in wireless sensor devices, the energy scavenging from RF signal is another alternative. In the last few years, a great interest has been seen in microwave power scavenging for charging wireless devices. This chapter presents a RF energy harvesting circuit with tuned π-matching network that resonates at desired incident RF frequency to boost these signals. Various computer intelligent techniques have been used to optimize parameters value of matching circuit. The designed circuit has been analyzed for input power range from -30 dBm to 0 dBm. Approximately 80% maximum PCE is achieved at RF input of 0 dBm with 4 KΩ load. It is also demonstrated that better output power is produced for power range -15 dBm to 0 dBm at higher load values.

Routing in Wireless Sensor Networks Using Soft Computing

Wireless sensor networks are the evolutionary self-organizing multi-node networks. Due to dynamic network conditions and stochastically varying network environments, routing in WSNs is critically affected and needs to be optimized. The routing strategies developed for WSNs must be efficient to make it an operationally self-configurable network. For this we need to resort to near shortest path evaluation. Therefore, some soft computing approaches that can calculate the near shortest path available in an affordable computing time are required. WSNs have a high computational environment with limited and precise transmission range, processing, and limited energy sources. The sever power constraints strongly affect the existence of active nodes and hence the network lifetime. So, here, the authors use the power of soft computing because the potential features of soft-computing (SC) approach highly address their adaptability and compatibility to overwhelm the complex challenges in WSNs.