Computational Intelligence Applications in Modeling and Control

Application of intelligent methods in banking becomes a challenging issue and acquiring special attention of banking supervisors and policy makers. Intelligent methods like rough set theory (RST), fuzzy set and genetic algorithm contribute significantly in multiple areas of banking and other important segments of financial sector. CAMEL is a useful tool to examine the safety and soundness of various banks and assist the banking regulators to ward off any potential risk which may lead to bank failure. RST approach may be applied for verifying authenticity and accuracy of CAMEL model and this chapter invites reader’s attention towards this relatively new and unique application of RST. The results of CAMEL model have been widely accepted by banking regulators for the purpose of assessing the financial health of banks. In this chapter we have considered ten largest public sector Indian banks on the basis of their deposit-base over a five-year period (2008–2009 to 2012–2013). The analysis of financial soundness of banks is structured under two parts. Part I is devoted to ranking of these banks on the basis of performance indices of their capital adequacy (C), asset quality (A), management efficiency (M), earnings (E) and liquidity (L). Performance analysis has been carried out in terms of two alternative approaches so as to bring implications with regard to their rank accuracy. We named these approaches as Unclassified Rank Assignment Approach and Classified Rank Assignment Approach. Part II presents analysis of accuracy of ranks obtained by CAMEL model for both the approaches that is for Unclassified Rank Assignment Approach and for Classified Rank Assignment in terms of application of Rough Set Theory (RST). The output of CAMEL model (ranking of banks for both approaches) is given as input to rough set for generating rules and for finding the reduct and core. The accuracy of the ranking generated by the CAMEL model is verified using lower and upper approximation. This chapter demonstrates the accuracy of Ranks generated by CAMEL model and decisions rules are generated by rough set method for the CAMEL model. Further, the most important attribute of CAMEL model is identified as risk-adjusted capital ratio, CRAR under capital adequacy attribute and results generated by rough set theory confirm the accuracy of the Ranks generated by CAMEL Model for various Indian public- sector banks.

Distributed computing systems are of huge importance in a number of recently established and future functions in computer science. For example, they are vital to banking applications, communication of electronic systems, air traffic control, manufacturing automation, biomedical operation works, space monitoring systems and robotics information systems. As the nature of computing comes to be increasingly directed towards intelligence and autonomy, intelligent computations will be the key for all future applications. Intelligent distributed computing will become the base for the growth of an innovative generation of intelligent distributed systems. Nowadays, research centres require the development of architectures of intelligent and collaborated systems; these systems must be capable of solving problems by themselves to save processing time and reduce costs. Building an intelligent style of distributed computing that controls the whole distributed system requires communications that must be based on a completely consistent system. The model of the ideal system to be adopted in building an intelligent distributed computing structure is the human body system, specifically the body’s cells. As an artificial and virtual simulation of the high degree of intelligence that controls the body’s cells, this chapter proposes a Cell-Oriented Computing model as a solution to accomplish the desired Intelligent Distributed Computing system.

In this chapter, the use of genetic algorithm (GA) is investigated in the field of estimating ultrasonic (US) propagation parameters. Recent works are, then, surveyed showing an ever-spread of employing GA in different applications of US. A GA is, specifically, used to estimate the propagation parameters of US waves in polycrystalline and composite materials for different applications. The objective function of the estimation is the minimization of a rational difference error between the estimated and measured transfer functions of US-wave propagation. The US propagation parameters may be the phase velocity and attenuation. Based on tentative experiments, we will demonstrate how the evolution operators and parameters of GA can be chosen for modeling of US propagation. The GA-based estimation is applied, in a test experiment, on steel alloy and Aluminum specimens with different grain sizes. Comparative results of that experiment are presented on different evolution operators for less estimation errors and complexity. The results prove the effectiveness of GA in estimating parameters for US propagation.

One of the major challenges facing requirements prioritization techniques is accuracy. The issue here is lack of robust algorithms capable of avoiding a mismatch between ranked requirements and stakeholder’s linguistic ratings. This problem has led many software developers in building systems that eventually fall short of user’s requirements. In this chapter, we propose a new approach for prioritizing software requirements that reflect high correlations between the prioritized requirements and stakeholders’ linguistic valuations. Specifically, we develop a hybridized algorithm which uses preference weights of requirements obtained from the stakeholder’s linguistic ratings. Our approach was validated with a dataset known as RALIC which comprises of requirements with relative weights of stakeholders.

Electric load forecasting is a real-life problem in industry. Electricity supplier’s use forecasting models to predict the load demand of their customers to increase/decrease the power generated and to minimize the operating costs of producing electricity. This paper presents the development and the implementation of three new electricity demand-forecasting models using the adaptive neuro-fuzzy inference system (ANFIS) approach in parallel load series. The input-output data pairs used are the real-time quart-hourly metropolitan France electricity load obtained from the RTE website and forecasts are done for lead-time of a 1 h ahead. Results and forecasting performance obtained reveal the effectiveness of the third proposed approach and shows that 56 % of the forecasted loads have an APE (absolute percentage error) under 0.5, and an APE under one was achieved for about 80 % of cases. Which mean that it is possible to build a high accuracy model with less historical data using a combination of neural network and fuzzy logic.

Classical optimization methods often face great difficulties while dealing with several engineering applications. Under such conditions, the use of computational intelligence approaches has been recently extended to address challenging real-world optimization problems. On the other hand, the interesting and exotic collective behavior of social insects have fascinated and attracted researchers for many years. The collaborative swarming behavior observed in these groups provides survival advantages, where insect aggregations of relatively simple and “unintelligent” individuals can accomplish very complex tasks using only limited local information and simple rules of behavior. Swarm intelligence, as a computational intelligence paradigm, models the collective behavior in swarms of insects or animals. Several algorithms arising from such models have been proposed to solve a wide range of complex optimization problems. In this chapter, a novel swarm algorithm called the Social Spider Optimization (SSO) is proposed for solving optimization tasks. The SSO algorithm is based on the simulation of cooperative behavior of social-spiders. In the proposed algorithm, individuals emulate a group of spiders which interact to each other based on the biological laws of the cooperative colony. The algorithm considers two different search agents (spiders): males and females. Depending on gender, each individual is conducted by a set of different evolutionary operators which mimic different cooperative behaviors that are typically found in the colony. In order to illustrate the proficiency and robustness of the proposed approach, it is compared to other well-known evolutionary methods. The comparison examines several standard benchmark functions that are commonly considered within the literature of evolutionary algorithms. The outcome shows a high performance of the proposed method for searching a global optimum with several benchmark functions.

Bio-inspired computation is a field of study that connects together numerous subfields of connectionism (neural network), social behavior, emergence field of artificial intelligence and machine learning algorithms for complex problem optimization. Bio-inspired computation is motivated by nature and over the last few years, it has encouraged numerous advance algorithms and set of computational tools for dealing with complex combinatorial optimization problems. Black Hole is a new bio-inspired metaheuristic approach based on observable fact of black hole phenomena. It is a population based algorithmic approach like genetic algorithm (GAs), ant colony optimization (ACO) algorithm, particle swarm optimization (PSO), firefly and other bio-inspired computation algorithms. The objective of this book chapter is to provide a comprehensive study of black hole approach and its applications in different research fields like data clustering problem, image processing, data mining, computer vision, science and engineering. This chapter provides with the stepping stone for future researches to unveil how metaheuristic and bio-inspired commutating algorithms can improve the solutions of hard or complex problem of optimization.

This chapter presents how multiobjective bilevel programming (MOBLP) in a hierarchical structure can be efficiently used for modeling and solving optimal power generation and dispatch problems via genetic algorithm (GA) based Fuzzy Goal Programming (FGP) method in a power system operation and planning horizon. In MOBLP formulation of the proposed problem, first the objectives of real and reactive power (P-Q) optimization are considered as two optimization problems at two individual levels (top level and bottom level) with the control of more than one objective at each level. Then the hierarchically ordered problem is fuzzily described to accommodate the impression in P-Q optimization simultaneously in the decision making context. In the model formulation, the concept of membership functions in fuzzy sets for measuring the achievement of highest membership value (unity) of the defined fuzzy goals to the extent possible by minimising their under-deviational variables on the basis of their weights of importance is considered. The aspects of FGP are used to incorporate the various uncertainties in power generation and dispatch. In the solution process, the GA is used in the framework of FGP model in an iterative manner to reach a satisfactory decision on the basis of needs and desires of the decision making environment. The GA scheme is employed at two different stages. At the first stage, individual optimal decisions of the objectives are determined for fuzzy goal description of them. At the second stage, evaluation of goal achievement function for minimization of the weighted under-deviational variables of the membership goals associated with the defined fuzzy goals is considered for achieving the highest membership value (unity) of the defined fuzzy goals on the basis of hierarchical order of optimizing them in the decision situation. The proposed approach is tested on the standard IEEE 6-Generator 30-Bus System to illustrate the potential use of the approach.

Maintenance and its integration with control and monitoring systems enable the improvement of systems functioning, regarding availability, efficiency, productivity and quality. This paper proposes a monitoring-maintenance approach based on fuzzy Petri Nets (PN’s) for manufacturing job-shops with time constraints. In such systems, operation times are included between a minimum and a maximum value. In this context, we propose a new fuzzy Petri net called Fuzzy Petri Net for maintenance (FPNM). This tool is able to identify and select maintenance activities of a discrete event system with time constraints, using a temporal fuzzy approach. The maintenance module is consists of P-time PNs and fault tree. The first is used for modelling of normal behaviour of the system by temporal spectrum of the marking. The second model corresponds to diagnosis activities. Finally, to illustrate the effectiveness and accuracy of proposed maintenance approach, two industrial examples are depicted.

In this work, we use radial basis function neural network for modeling nonlinear systems. Generally, the main problem in artificial neural network is often to find a better structure. The choice of the architecture of artificial neural network for a given problem has long been a problem. Developments show that it is often possible to find architecture of artificial neural network that greatly improves the results obtained with conventional methods. We propose in this work a method based on No Sorting Genetic Algorithm II (NSGA II) to determine the best parameters of a radial basis function neural network. The NSGAII should provide the best connection weights between the hidden layer and output layer, find the parameters of the radial function of neurons in the hidden layer and the optimal number of neurons in the hidden layers and thus ensure learning necessary. Two functions are optimized by NSGAII: the number of neurons in the hidden layer of the radial basis function neural network, and the error which is the difference between desired input and the output of the radial basis function neural network. This method is applied to modeling Box and Jenkins system. The obtained results are very satisfactory.

Determination of gender is the foremost and important step of forensic anthropology in determining a positive identification from unidentified skeletal remains. Gender determination is the classification of an individual into one of two groups, male or female. The classification technique most used by anthropologists or researchers is traditional gender determination with applied linear approach, such as Discriminant Function Analysis (DFA). This paper proposed non-linear approach specific Back-Propagation Neural Network (BPNN) to determine gender from sacrum bone. Sacrum bone is one part of the body that is usually regarded as the most reliable indicator of sex. The data used in the experiment were taken from previous research, a total of 91 sacrum bones consisting of 34 females and 57 males. Method of measurement used is metric method which is measured based on six variables; real height, anterior length, anterior superior breadth, mid-ventral breadth, anterior posterior diameter of the base, and max-transverse diameter of the base. The objective of this paper is to examine and compare the degree of accuracy between previous research (DFA) and BPNN. There are two architectures of BPNN built for this case, namely [6; 6; 2] and [6; 12; 2]. The best average accuracy obtained by BPNN is model [6; 12; 2] with accuracy 99.030 % for training and 97.379 % for testing on experiment lr = 0.5 and mc = 0.9, then obtained Mean Squared Error (MSE) training is 0.01 and MSE testing is 1.660. Previous research using DFA only obtained accuracy as high as 87 %. Hence, it can be concluded that BPNN provide classification accuracy higher than DFA for gender determination in forensic anthropology.

Fault location and distance protection in transmission lines are essential smart grid technologies ensuring reliability of the power system and achieve the continuity of service. The objective of this chapter is to presents an accurate algorithm for estimating fault location in Extra High Voltage (EHV) transmission lines using Artificial Neural Networks (ANNs) for high speed protection. The development of this algorithm is based on disturbed transmission line models. The proposed fault protection (fault detection/classification and location) uses only the three phase currents signals at the one end of the line. The proposed technique uses five ANNs networks and consists of two steps, including fault detection/classification and fault location. For fault detection/classification, one ANN network is used in order to identify the fault type; the fault detection/classification procedure uses the fundamental components of pre-fault and post-fault sequence samples of three phase currents and zero sequence current. For fault location, four ANNs networks are used in order to estimate the exact fault location in transmission line. Magnitudes of pre-fault and post-fault of three phase currents are used. The ANNs are trained with data under a wide variety of fault conditions and used for the fault classification and fault location on the transmission line. The proposed fault detection/classification and location approaches are tested under different fault conditions such as different fault locations, different fault resistances and different fault inception angles via digital simulation using MATLAB software in order to verify the performances of the proposed methods. The ANN-based fault classifier and locator gives high accuracy for all tests under different fault conditions. The simulations results show that the proposed scheme based on ANNs can be used for on-line fault protection in transmission line.

Motivated by the demand for formalized representation of outcomes of data mining investigations and the successful results of using Formal Concept Analysis (FCA) and Ontology, this chapter addresses the task of constructing an ontology of data mining in order to support flexible query in large Database using FCA and Fuzzy Ontology. A new approach for automatic generation of Fuzzy Ontology of Data Mining (FODM), through the combination of conceptual clustering, fuzzy logic and FCA will be presented. Then, a new algorithm to support database flexible querying using the generated fuzzy ontology will be defined. The approach starts with the organization of the data in homogeneous clusters having common properties which allows to deduce the data’s semantic. Then, it models these clusters by an extension of the FCA. This lattice will be used to build a core of ontology. This ontology will be represented, then, as a set of fuzzy rules as an efficient answers to flexible queries. We show that this approach is optimum because the evaluation of the query is not done on the set of starting data which is huge but rather by using the generated fuzzy ontology.

In the area of image processing, segmentation of an image into multiple regions is very important for classification and recognition steps. It has been widely used in many application fields such as medical image analysis to characterize and detect anatomical structures, robotics features extraction for mobile robot localization and detection and map procession for lines and legends finding. Many techniques have been developed in the field of image segmentation. Methods based on intelligent techniques are the most used such as Genetic Algorithm (GA), Ant Colony Optimization (ACO), Artificial Bee Colony (ABC), and Particle Swarm Optimization (PSO) called metaheuristics algorithms. In this paper, we describe a novel method for segmentation of images based on one of the most popular and efficient metaheuristic algorithm called Particle Swarm optimization (PSO) for determining multilevel threshold for a given image. The proposed method takes advantage of the characteristics of the particle swarm optimization and improves the objective function value to updating the velocity and the position of particles. This method is compared to the basic PSO method, also, it is compared with other known multilevel segmentation methods to demonstrate its efficiency. Experimental results show that this method can reliably segment and give threshold values than other methods considering different measures.

In this paper, a heuristic method based on Firefly Algorithm is proposed for inverse kinematics problems in articulated robotics. The proposal is called, IK-FA. Solving inverse kinematics, IK, consists in finding a set of joint-positions allowing a specific point of the system to achieve a target position. In IK-FA, the Fireflies positions are assumed to be a possible solution for joints elementary motions. For a robotic system with a known forward kinematic model, IK-Fireflies, is used to generate iteratively a set of joint motions, then the forward kinematic model of the system is used to compute the relative Cartesian positions of a specific end-segment, and to compare it to the needed target position. This is a heuristic approach for solving inverse kinematics without computing the inverse model. IK-FA tends to minimize the distance to a target position, the fitness function could be established as the distance between the obtained forward positions and the desired one, it is subject to minimization. In this paper IK-FA is tested over a 3 links articulated planar system, the evaluation is based on statistical analysis of the convergence and the solution quality for 100 tests. The impact of key FA parameters is also investigated with a focus on the impact of the number of fireflies, the impact of the maximum iteration number and also the impact of (α, β, γ, δ) parameters. For a given set of valuable parameters, the heuristic converges to a static fitness value within a fix maximum number of iterations. IK-FA has a fair convergence time, for the tested configuration, the average was about 2.3394 × 10⁻³ seconds with a position error fitness around 3.116 × 10⁻⁸ for 100 tests. The algorithm showed also evidence of robustness over the target position, since for all conducted tests with a random target position IK-FA achieved a solution with a position error lower or equal to 5.4722 × 10⁻⁹.

Breast cancer is the common form of cancer and leading cause of mortality among woman, especially in developed countries. In western countries about 53–92 % of the population has this disease. As with any form of cancer, early detection and diagnosis of breast cancer can increase the survival rate. Mammography is the current diagnostic method for early detection of breast cancer. Breast parenchymal patterns are not stable between patients, between left and right breasts, and even within the same breast from year to year in the same patient. Breast cancer has a varied appearance on mammograms, from the obvious spiculated masses, to very subtle asymmetries noted on only one view, to faint calcifications seen only with full digital resolution or a magnifying glass. The large volume of cases requiring interpretation in many practices is also daunting, given the number of women in the population for whom yearly screening mammography is recommended. It seems obvious that this difficult task could likely be made less error prone with the help of computer algorithms. Computer-aided detection (CAD) systems have been shown to be capable of reducing false-negative rates in the detection of breast cancer by highlighting suspicious masses and microcalcifications on mammograms. These systems aid the radiologist as a ‘second opinion’ in detecting cancers and the final decision is taken by the radiologist. A supervised machine learning algorithm is investigated—Differential Evolution Optimized Wavelet Neural Network (DEOWNN) for detection of abnormalities in mammograms. Differential Evolution (DE) is a population based optimization algorithm based on the principle of natural evolution, which optimizes real parameters and real valued functions. By utilizing the DE algorithm, the parameters of the Wavelet Neural Network (WNN) are optimized. To increase the detection accuracy a feature extraction methodology is used to extract the texture based features of the abnormal breast tissues prior to classification. Then differential evolution optimized wavelet neural network classifier is applied at the end to determine whether the given input data is normal or abnormal. The performance of the computerized decision support system is evaluated using a mini database from Mammographic Image Analysis Society (MIAS) and images collected from mammogram screening centres.