Parallel Direct Search Research Articles

Design and optimization of a combined supercritical CO2 recompression Brayton/Kalina cycle by using solar energy and LNG cold energy

Solar energy and LNG cold energy have broad application scenarios as two renewable sources, but their complementary potential has not been fully explored. To fill this gap, this paper constructs a novel combined cycle that utilizes solar energy and recovers LNG cold energy to generate power. This combined cycle consists of a supercritical carbon dioxide recompression Brayton cycle, an ammonia-water mixture Kalina cycle, and an LNG cryogenic power system. A systematic evaluation was conducted on the thermodynamics and exergoeconomic of the model, and the simulation results under basic operating conditions show that the first law efficiency, second law efficiency, net power output and total unit cost reach 55.73 %, 47.72 %, 13.709 MW and 27.44 $·GJ−1 respectively. The genetic algorithm and elitist non-dominated sorting genetic algorithm are used to optimize the system performance. By using the parallel direct search method, the maximum first law efficiency, as well as second law efficiency and a minimum total unit cost of 59.45 %, 52.23 % and 24.81 $·GJ−1 are obtained respectively. The optimal first law efficiency, second law efficiency, and total unit cost of the system, which are 53.52 %, 50.33 % and 26.44 $·GJ−1 respectively, are searched simultaneously from the multi-objective optimization results by LINMAP decision-making method. Furthermore, this paper analyzes the exergy flow rate of the system and heat transfer processes of the heat exchangers under optimal operating condition. A total of 23,059.35 kW of LNG cold energy is fully utilized by the system in the LNG gasification process. The exergy loss ratio and cost ratio of each equipment analysis demonstrates that the heat exchangers (HX) have the largest percentage both exergy loss ratio and cost ratio, especially the HX4.

An evolutionary hybrid Fuzzy Computationally Efficient EGARCH model for volatility prediction

Accurate modeling for forecasting of stock market volatility is a widely interesting research area both in academia as well as financial markets. This paper proposes an innovative Fuzzy Computationally Efficient EGARCH model to forecast the volatility of three stock market indexes. The proposed model represents a joint estimation of the membership function parameters of a TSK-type fuzzy inference system along with the leverage effect, asymmetric shock by leverage effect of EGARCH model in forecasting highly nonlinear and complicated financial time series model more accurately. Further unlike the conventional TSK type fuzzy neural network the proposed model uses a functional link neural network (FLANN) in the consequent part of the fuzzy rules to provide an improved mapping. Moreover, a differential evolution (DE) algorithm is suggested to solve the parameters estimation problem of Fuzzy Computationally Efficient EGARCH model. Being a parallel direct search algorithm, DE has the strength of finding global optimal solutions regardless of the initial values of its few control parameters. Furthermore, the DE based algorithm aims to achieve an optimal solution with a rapid convergence rate. The proposed model has been compared with some GARCH family models and hybrid fuzzy systems and GARCH models based on three performance metrics: MSFE, RMSFE, and MAFE. The results indicate that the proposed method offers significant improvements in volatility forecasting performance in comparison with all other specified models.

Thermodynamic optimization of the operating parameters for a combined power cycle utilizing low-temperature waste heat and LNG cold energy

This paper deals with the optimization of a novel combined power system, which can effectively recover low-temperature waste heat and fully utilize the cold energy of LNG as well, based on the first thermodynamic law and the second thermodynamic law respectively. Parametric analysis has been performed to study the effects of heat source temperature, ammonia turbine inlet pressure, LNG turbine inlet and outlet pressures, as well as ammonia mass fraction of basic solution. The simulation results show that the system performance can be improved by applying optimization techniques. The optimization is conducted under a certain set of constraints by using the differential evolution (DE) algorithm to maximize the first and the second law efficiency respectively. Through parallel direct search over the whole feasible region, it is found that a maximum first law efficiency of 39.33% can be obtained when variable vector V1 = [423.70 K, 1.8 MPa, 3.904 MPa, 0.3 MPa, 0.52]; while a maximum second law efficiency of 55.62% can be obtained when variable vector V2 = [423.93 K, 1.874 MPa, 3.493 MPa, 0.8 MPa, 0.48]. In addition, the irreversibilities in various components of the cycle under typical operating conditions and exergy efficiency optimum condition have been compared through detailed exergy analysis.

Effective Image Clustering with Differential Evolution Technique

The paper presents a novel approach of clustering image datasets with differential evolution (DE) technique. The differential evolution is a parallel direct search population based optimization method. From our simulations it is found that DE is able to optimize the quality measures of clusters of image datasets. To claim the superiority of DE based clustering we have compared the outcomes of DE with the classical K-means and popular Particle Swarm Optimization (PSO) algorithms for the same datasets. The comparisons results reveal the suitability of DE for image clustering in all image datasets.

Development of an intelligent pavement analysis toolbox

An intelligent pavement analysis toolbox was developed for backcalculating the mechanical properties of flexible pavements from non-destructive test data through an adaptive, multi-hybrid, integrated non-linear systems approach. The problem involves searching for the optimal combination of pavement layer stiffness solutions in an unsmooth, multi-modal, complex search space which makes it amenable to the application of differential evolution, a stochastic parallel direct search evolution strategy optimisation method. This innovative approach takes advantage of the combined efficiency and accuracy achieved by integrating advanced pavement numerical modelling schemes such as finite-element methods, computational intelligence-based surrogate mapping techniques such as artificial neural networks, and heuristics-based global optimisation strategies such as differential evolution, and yet provides a user-friendly automated pavement evaluation toolbox for the highway engineer to use on a real-time basis for accurate infrastructure evaluation. The main aim of the present study was to demonstrate the feasibility of using such an integrated modular systems approach to the complex non-linear parameter identification problem in pavement engineering.

Down-hill Simplex Method Based Differential Evolution

Differential Evolution (DE) is based on both an evolutionary strategy and a parallel direct search method employing a population. DE is an effective optimization method available for solving global optimization problem over continuous space. DE has a few control parameters that have to be set by users. This paper describes a new DE using Down-hill Simplex Method. Then we consider average distance of a new DE and we examine DE to choice a proposal method or the other proposal method according to the average distance. The feasibility and advantage of the proposed DE are demonstrated through some numerical simulations using four different typical global optimization test problems.

Differential evolution tuned fuzzy supervisor adapted extended Kalman filtering for SLAM problems in mobile robots

SUMMARYThe present paper proposes a successful application of differential evolution (DE) optimized fuzzy logic supervisors (FLS) to improve the quality of solutions that extended Kalman filters (EKFs) can offer to solve simultaneous localization and mapping (SLAM) problems for mobile robots and autonomous vehicles. The utility of the proposed system can be readily appreciated in those situations where an incorrect knowledge of Q and R matrices of EKF can significantly degrade the SLAM performance. A fuzzy supervisor has been implemented to adapt the R matrix of the EKF online, in order to improve its performance. The free parameters of the fuzzy supervisor are suitably optimized by employing the DE algorithm, a comparatively recent method, popularly employed now-a-days for high-dimensional parallel direct search problems. The utility of the proposed system is aptly demonstrated by solving the SLAM problem for a mobile robot with several landmarks and with wrong knowledge of sensor statistics. The system could successfully demonstrate enhanced performance in comparison with usual EKF-based solutions for identical environment situations.

Deterministic parallel global parameter estimation for a model of the budding yeast cell cycle

Two parallel deterministic direct search algorithms are combined to find improved parameters for a system of differential equations designed to simulate the cell cycle of budding yeast. Comparing the model simulation results to experimental data is difficult because most of the experimental data is qualitative rather than quantitative. An algorithm to convert simulation results to mutant phenotypes is presented. Vectors of the 143 parameters defining the differential equation model are rated by a discontinuous objective function. Parallel results on a 2200 processor supercomputer are presented for a global optimization algorithm, DIRECT, a local optimization algorithm, MADS, and a hybrid of the two.

Concurrent Implementation of the Optimal Incremental Approximation Method for the Adaptive and Meshless Solution of Differential Equations

The optimal incremental function approximation method is implemented for the adaptive and meshless solution of differential equations. The basis functions and associated coefficients of a series expansion representing the solution are selected optimally at each step of the algorithm according to appropriate error minimization criteria. Thus, the solution is built incrementally. In this manner, the computational technique is adaptive in nature, although a grid is neither built nor adapted in the traditional sense using a posteriori error estimates. Since the basis functions are associated with the nodes only, the method can be viewed as a meshless method. Variational principles are utilized for the definition of the objective function to be extremized in the associated optimization problems. Complicated data structures, expensive remeshing algorithms, and systems solvers are avoided. Computational efficiency is increased by using low-order local basis functions and the parallel direct search (PDS) optimization algorithm. Numerical results are reported for both a linear and a nonlinear problem associated with fluid dynamics. Challenges and opportunities regarding the use of this method are discussed.

A Class of Trust-Region Methods for Parallel Optimization

We present a new class of optimization methods that incorporates a parallel direct search (PDS) method within a trust-region Newton framework. This approach combines the inherent parallelism of PDS with the rapid and robust convergence properties of Newton methods. Numerical tests have yielded favorable results for both standard test problems and engineering applications. In addition, the new method appears to be more robust in the presence of noisy functions, which are inherent in many engineering simulations.

Sequential function approximation for the solution of differential equations

A computational method for the solution of differential equations is proposed. With this method an accurate approximation is built by incremental additions of optimal local basis functions. The parallel direct search software package (PDS), that supports parallel objective function evaluations, is used to solve the associated optimization problem efficiently. The advantage of the method is that, although it resembles adaptive methods in computational mechanics, an a priori grid is not necessary. Moreover, the traditional matrix construction and evaluations are avoided. Computational cost is reduced while efficiency is enhanced by the low-dimensional parallel-executed optimization and parallel function evaluations. In addition, the method should be applicable to a broad class of interpolation functions. Results and global convergence rates obtained for one- and two-dimensional boundary value problems are satisfactorily compared to those obtained by the conventional Galerkin finite element method. © 1997 John Wiley & Sons, Ltd.

Electrical defibrillation optimization: an automated, iterative parallel finite-element approach.

To date, optimization of electrode systems for electrical defibrillation has been limited to hand-selected electrode configurations. In this paper we present an automated approach which combines detailed, three-dimensional (3-D) finite-element torso models with optimization techniques to provide a flexible analysis and design tool for electrical defibrillation optimization. Specifically, a parallel direct search (PDS) optimization technique is used with a representative objective function to find an electrode configuration which corresponds to the satisfaction of a postulated defibrillation criterion with a minimum amount of power and a low possibility of myocardium damage. For adequate representation of the thoracic inhomogeneities, 3-D finite-element torso models are used in the objective function computations. The CPU-intensive finite-element calculations required for the objective function evaluation have been implemented on a message-passing parallel computer in order to complete the optimization calculations in a timely manner. To illustrate the optimization procedure, it has been applied to a representative electrode configuration for transmyocardial defibrillation, namely the subcutaneous patch-right ventricular catheter (SP-RVC) system. Sensitivity of the optimal solutions to various tissue conductivities has been studied. Results for the optimization of defibrillation systems are presented which demonstrate the feasibility of the approach.

Enhancement of a viscous-inviscid-interaction airfoil analysis code using the parallel direct search algorithm

The coupling mechanism for an existing viscous-inviscid-interaction (VII) code, developed for the analysis of two-dimensional, turbulent, attached flow around airfoils, is enhanced using the parallel direct search (PDS) optimization algorithm. It is demonstrated that this parallel processing implemented optimization scheme leads to faster convergence of the VII code, and therefore, requires less computational time when the number of optimization (or design) variables is low, and a moderate number of processors are available. As the number of design variables increases, more processors are required to maintain this advantage. Results are presented for the NACA-0012 and the RAE-2822 airfoils. The quality of the results obtained is satisfactory and confirms that the enhanced VII code can be an acceptable alternative to reduced Navier-Stokes solvers as an airfoil analysis tool.

Designing for optimal energy absorption, III: numerical minimization of the spectral abscissa

We address the optimal design problem associated with introducing passive damping devices into a structure in such a way thatall vibration frequencies are moved as far as possible into the left halfplane. As this natural objective function lacks a bounded derivative we are forced to adopt a derivative-free direct search method. We report on our application of parallel direct search to the optimal design of both shear buildings and strings.

A comparison of a direct search method and a genetic algorithm for conformational searching

We present results from the application of two conformational searching methods: genetic algorithms (GA) and direct search methods for finding low energy conformations of organic molecules. GAs are in a class of biologically motivated optimization methods that evolve a population of individuals in which individuals who are more “fit” have a higher probability of surviving into subsequent generations. The parallel direct search method (PDS) is a type of pattern search method that uses an adaptive grid to search for minima. Both methods found energies equal to or lower than the energy of the relaxed crystal structure in all cases, at a relatively small cost in CPU time. We suggest that either method would be a good candidate to find 3-D conformations in a large scale screening application. © 1996 by John Wiley & Sons, Inc.

Numerical analysis of electrical defibrillation: The parallel approach

Numerical modeling offers a viable tool for studying electrical defibrilation, allowing the behavior of field quantities to be observed easily as the different system parameters are varied. One numerical technique, namely the finite-element method, has been found particularly effective for modeling complex thoracic anatomies. However, an accurate finite-element model of the thorax often requires a large number of elements and nodes, leading to a large set of equations that cannot be solved effectively with the computational power of conventional computers. This is especially true if many finite-element solutions need to be achieved within a reasonable time period (eg, electrode configuration optimization). In this study, the use of massively parallel computers to provide the memory and reduction in solution time for solving these large finite-element problems is discussed. Both the uniform and unstructured grid approaches are considered. Algorithms that allow efficient mapping of uniform and unstructured grids to data-parallel and message-passing parallel computers are discussed. An automatic iterative procedure for electrode configuration optimization is presented. The procedure is based on the minimization of an objective function using the parallel direct search technique. Computational performance results are presented together with simulation results.