A new methodology for evolutionary optimization of energy systems

This paper presents a novel technique to significantly reduce the compute time for evolutionary optimization of systems modeled using CFD. In this scheme the typical roulette selection process is modified with a process in which competing members are represented by a Gaussian fitness distribution obtained from an artificial neural network with a feature weighted general regression neural network to create a universal approximator. This approximator develops a real-time estimate of the final fitness and error bounds during each iteration of the CFD solver. The iteration process continues until the estimated fitness and error bounds indicate that additional iterations will have a small effect on the outcome of the roulette selection process. This reduces the time required for each system call and hence reduces the overall computational time required.     

Effects of a proper feature selection on prediction and optimization of drilling rate using intelligent techniques

Engineering with Computers - One of the important factors during drilling times is the rate of penetration (ROP), which is controlled based on different variables. Factors affecting different...     

A systematic methodology for firewall penetration testing

Firewall testing is one of the most useful of a set of alternatives for evaluating the security effectiveness of a firewall. A major advantage of firewall testing is being able to empirically determine how secure a firewall is against attacks that are likely to be launched by network intruders. This article advances the view that firewall testing should examine not only the ability of a firewall to resist attacks from external sources, but also the defences of the entire network that the firewall protects against external threats. Accordingly, testing should follow a systematic methodology to ensure that it is complete and appropriate, and to reduce the risk of damage and/or disruption to networks and hosts within.     

Application of sensitivity analysis to the optimization of petroleum drilling operations

Optimized drilling models which are concerned with determining the mimimum drilling costs of petroleum operations are analyzed from the point of view of sensitivity of the optimal solution to changes in model parameters. These models are nonlinear and the parameters are typically derived from field or laboratory data. The sensitivity analysis procedures of geometric programming are used to study the impact of parameter changes such as rock hardness and abrasiveness, rate of drill wear and bearing wear among others. The procedures are illustrated with a numerical example which is used to determine an optimal drilling rig configuration as well as determine the impact on that configuration of changing drilling parameters.     

A new hybrid methodology for cooperative-coevolutionary optimization of radial basis function networks

This paper presents a new multiobjective cooperative–coevolutive hybrid algorithm for the design of a Radial Basis Function Network (RBFN). This approach codifies a population of Radial Basis Functions (RBFs) (hidden neurons), which evolve by means of cooperation and competition to obtain a compact and accurate RBFN. To evaluate the significance of a given RBF in the whole network, three factors have been proposed: the basis function’s contribution to the network’s output, the error produced in the basis function radius, and the overlapping among RBFs. To achieve an RBFN composed of RBFs with proper values for these quality factors our algorithm follows a multiobjective approach in the selection process. In the design process, a Fuzzy Rule Based System (FRBS) is used to determine the possibility of applying operators to a certain RBF. As the time required by our evolutionary algorithm to converge is relatively small, it is possible to get a further improvement of the solution found by using a local minimization algorithm (for example, the Levenberg–Marquardt method). In this paper the results of applying our methodology to function approximation and time series prediction problems are also presented and compared with other alternatives proposed in the bibliography.     

Two-level intelligent modeling method for the rate of penetration in complex geological drilling process

The rate of penetration (ROP) model is of great importance in achieving a high efficiency in the complex geological drilling process. In this paper, a novel two-level intelligent modeling method is proposed for the ROP considering the drilling characteristics of data incompleteness, couplings, and strong nonlinearities. Firstly, a piecewise cubic Hermite interpolation method is introduced to complete the lost drilling data. Then, a formation drillability (FD) fusion submodel is established by using Nadaboost extreme learning machine (Nadaboost-ELM) algorithm, and the mutual information method is used to obtain the parameters, strongly correlated with the ROP. Finally, a ROP submodel is established by a neural network with radial basis function optimized by the improved particle swarm optimization (RBFNN-IPSO). This two-level ROP model is applied to a real drilling process and the proposed method shows the best performance in ROP prediction as compared with conventional methods. The proposed ROP model provides the basis for intelligent optimization and control in the complex geological drilling process.     

A simulation modelling methodology for evaluating flat-shunted yard operations

Provided, in this paper, is a simulation modelling methodology for analysing and evaluating flat-shunted yard operations. Created and implemented is a yard simulation model using a computer package for event-based simulation, SIMUL’8. The main idea behind the simulation modelling approach is to simulate the flat-shunted yard operations dividing the yard into segments so that the behaviour of each segment can be described and analyzed separately. The simulation model takes the shape of queuing network. The components of the queuing network are interconnected queuing systems that interact and influence one another, so that the global impact of freight train operations is captured. The queuing systems replicate the preliminary specified segments that consist of a set of work centres (i.e., servers) and/or storage areas (i.e., capacity limitations). This modelling approach allows us to study the processing capabilities of the yards. In the shape of “Case Study”, we demonstrate how the proposed approach is implemented in terms of “real world” case.     

A new method for overhead drilling

In the construction sector, overhead drilling into concrete or metal ceilings is a strenuous task associated with shoulder, neck and back musculoskeletal disorders due to the large applied forces and awkward arm postures. Two intervention devices, an inverted drill press and a foot lever design, were developed then compared to the usual method by construction workers performing their normal overhead drilling activities (n = 14). While the intervention devices were rated as less fatiguing than the usual method, their ratings on usability measures were worse than the usual method. The study demonstrates that the intervention devices can reduce fatigue; however, additional modifications are necessary in order to improve usability and productivity. Devices designed to improve workplace safety may need to undergo several rounds of field testing and modification prior to implementation.     

A new methodology of extraction, optimization and application ofcrisp and fuzzy logical rules

A new methodology of extraction, optimization, and application of sets of logical rules is described. Neural networks are used for initial rule extraction, local or global minimization procedures for optimization, and Gaussian uncertainties of measurements are assumed during application of logical rules. Algorithms for extraction of logical rules from data with real-valued features require determination of linguistic variables or membership functions. Contest-dependent membership functions for crisp and fuzzy linguistic variables are introduced and methods of their determination described. Several neural and machine learning methods of logical rule extraction generating initial rules are described, based on constrained multilayer perceptron, networks with localized transfer functions or on separability criteria for determination of linguistic variables. A tradeoff between accurary/simplicity is explored at the rule extraction stage and between rejection/error level at the optimization stage. Gaussian uncertainties of measurements are assumed during application of crisp logical rules, leading to "soft trapezoidal" membership functions and allowing to optimize the linguistic variables using gradient procedures. Numerous applications of this methodology to benchmark and real-life problems are reported and very simple crisp logical rules for many datasets provided.     

A Bayes-SLIM based methodology for human reliability analysis of lifting operations

Investigations have shown that human error (HE) is the most common cause of accidents involving lifting operations. So it is essential to conduct human reliability analysis (HRA) in lifting operations. The estimation of human error probabilities (HEPs) is a key to HRA. In this paper, five useful performance shaping factors (PSFs) of lifting operations were introduced and an approach to combine Bayesian methodology and the success likelihood index method (SLIM) was applied to determine the HEPs of various lifting operation errors. A numerical example was used to illustrate the application of the proposed methodology.Relevance to industryReliability analysis of crane lifting operation has traditionally been difficult without considering human error. The present study suggests that managers could use the Bayesian-SLIM methodology to conduct human reliability analysis and minimize human error in crane lifting operation.     

A methodology for missile countermeasures optimization under uncertainty

The missile countermeasures optimization problem is a complex strategy optimization problem that combines aircraft maneuvers with additional countermeasures in an attempt to survive attack from a single surface-launched, anti-aircraft missile. Classic solutions require the evading aircraft to execute specific sequences of maneuvers at precise distances from the pursuing missile and do not effectively account for uncertainty about the type and/or current state of the missile. This paper defines a new methodology for solving the missile countermeasures optimization problem under conditions of uncertainty. The resulting genetic programming system evolves programs that combine maneuvers with such countermeasures as chaff, flares, and jamming to optimize aircraft survivability. This methodology may be generalized to solve strategy optimization problems for intelligent, autonomous agents operating under conditions of uncertainty.     

Online tool wear prediction in drilling operations using selective artificial neural network ensemble model

Online tool wear prediction plays a key role in industry automation for higher productivity and product quality. In recent past, several artificial neural network (ANN) models using multiple sensor signals as inputs for prediction as well as classification of tool wear have been proposed. However, a single ANN used in these models is often tries, which could limit their wide applications due to the complicated procedure of constructing a single ANN model. This study proposed a selective ANN ensemble approach DPSOEN, where several selected component ANNs are jointly used to online predict flank wear in drilling operation. DPSOEN provides more simple training and better generalization performance than using single ANN and hence is easier to be used by operators who often are not good at ANN techniques. Two benchmark cases were used to evaluate the performance of DPSOEN in predicting flank wear. It shows improved generalization performance that outperforms those of single ANN and Ensemble ALL approach. The investigation proposed a heuristic approach for applying the DPSOEN-based model as an effective and useful tool to predict tool wear online with potential applications for tool condition monitoring in general. Analysis from this study provides guidelines in developing ANN ensemble-based tool wear prediction systems.     

Evolutionary optimization for disaster relief operations: A survey

Effective planning and scheduling of relief operations play a key role in saving lives and reducing damage in disasters. These emergency operations involve a variety of challenging optimization problems, for which evolutionary computation methods are well suited. In this paper we survey the research advances in evolutionary algorithms (EAs) applied to disaster relief operations. The operational problems are classified into five typical categories, and representative works on EAs for solving the problems are summarized, in order to give readers a general overview of the state-of-the-arts and facilitate them to find suitable methods in practical applications. Several state-of-art methods are compared on a set of real-world emergency transportation problem instances, and some lessons are drawn from the experimental analysis. Finally, the strengths, limitations and future directions in the area are discussed.     

A new methodology for observer design and implementation

Observers are usually formulated as explicit systems of differential equations and implemented using standard ODE solvers. In this paper, we show that there can be advantages in formulating the observer as a differential-algebraic equation (DAE). We propose a canonical DAE observer and show that it can be implemented using standard DAE solvers. The DAE implementation of the observer can be considered regardless of the design philosophy used. We present a simple design strategy for this canonical DAE observer based on the extended linearization method     

A new class of hybrid global optimization algorithms for peptide structure prediction: integrated hybrids

A novel class of hybrid global optimization methods for application to the structure prediction in protein-folding problem is introduced. These optimization methods take the form of a hybrid between a deterministic global optimization algorithm, the αBB, and a stochastically based method, conformational space annealing (CSA), and attempt to combine the beneficial features of these two algorithms. The αBB method as previously extant exhibits consistency, as it guarantees convergence to the global minimum for twice-continuously differentiable constrained nonlinear programming problems, but can benefit from improvements in the computational front. Computational studies for met-enkephalin demonstrate the promise for the proposed hybrid global optimization method.     

A new methodology for measuring land fragmentation

The presence of land fragmentation can indicate that an existing land tenure structure is problematic. It can be a major problem in many regions because it restricts rational agricultural development and reduces the opportunities for sustainable rural development although in some cases, it can prove beneficial and desirable for social and environmental reasons. Whilst policies to counter land fragmentation require reliable measurement of the situation, current fragmentation indices have significant weaknesses. In particular, they ignore critical spatial variables such as the shape of parcels as well as non-spatial variables such as ownership type and the existence or absence of road access for each land parcel. Furthermore, there is no flexibility for users to select the variables that they think appropriate for inclusion in the fragmentation index, and no variable weighting mechanism is available. The aim of this paper is to introduce a new ‘global land fragmentation index’ that combines a multi-attribute decision-making method with a geographic information system. When applied to a case study area in Cyprus, the new index outperforms the existing indices in terms of reliability as it is comprehensive, flexible, problem specific and knowledge-based. The methodology can be easily applied to assess the quality of any existing system for which evaluation criteria can be defined with values ranging from the worst to best conditions.     

A Genetic Algorithm-based optimization model for supporting green transportation operations

Green Logistics (GL) has emerged as a trend in the management of the distribution of goods and the collection of end-of-life products. With its focus on maximizing the economic and environmental value by means of recycling and emission control, GL contributes to the sustainable development of industry but also requires a more comprehensive transportation scheme when conducting logistics services. This study is motivated by the practice of delivering and collecting water carboys. In this paper, a Genetic Algorithm-based optimization model (GOM) is proposed for designing a green transportation scheme of economic and environmental cost efficiency in forward and reverse logistics. Two vehicle routing models with simultaneous delivery and pickup (full or partial pickup) are formulated and solved by a Genetic Algorithm. A cost generation engine is designed to perform a comprehensive cost comparison and analysis based on a set of economic and environmental cost factors, so as to examine the impact of the two models and to suggest optimal transportation schemes. The computational experiments show that the overall cost is evidently lower in the full pickup model. Notably, the impact of product cost after recycling and reusing empty carboys on total cost is more significant than the impact of transportation cost and CO₂ emission cost. In summary, the proposed GOM is capable of suggesting a guidance for the logistics service providers, who deal with green operations, to adopt a beneficial transportation scheme so as to eventually achieve a low economic and environmental cost.     

A novel methodology for prediction of spatial-temporal activities using latent features

In today's era of big data, huge amounts of spatial-temporal data are generated daily from all kinds of citywide infrastructures. Understanding and predicting accurately such a large amount of data could benefit many real-world applications. In this paper, we propose a novel methodology for prediction of spatial-temporal activities such as human mobility, especially the inflow and outflow of people in urban environments based on existing large-scale mobility datasets. Our methodology first identifies and quantifies the latent characteristics of different spatial environments and temporal factors through tensor factorization. Our hypothesis is that the patterns of spatial-temporal activities are highly dependent on or caused by these latent spatial-temporal features. We model this hidden dependent relationship as a Gaussian process, which can be viewed as a distribution over the possible functions to predict human mobility. We tested our proposed methodology through experiments conducted on a case study of New York City's taxi trips and focused on the mobility patterns of spatial-temporal inflow and outflow across different spatial areas and temporal time periods. The results of the experiments verify our hypothesis and show that our prediction methodology achieves a much higher accuracy than other existing methodologies.