Pattern Search Optimization Research Articles

Comparison of traditional and deep learning optimisation for the design of acoustic metamaterials

This paper presents a comparative analysis of traditional and Deep learning (DL) optimisation approaches for AMM design. A case study comparing a generalised pattern search (GPS) optimisation to a previously trained inverse design deep neural network (IDDNN) of an acoustic meta-absorber using various metrics is considered. For benchmarking, a baseline design is realised with arbitrarily chosen geometric parameters in a CAD software. This design is 3D printed and experimentally tested. The peak absorption and corresponding resonant frequency obtained from experimental absorption, along with other geometric parameters, are used as inputs to IDDNN and GPS. Results show that IDDNN provides superior inference speed, requiring 7.24e-3 [s] to generate predictions on a single input instance, compared to GPS which took about 30.5464 [s]. While GPS exhibits slightly closer alignment to experimental curve based on Kullback-Leibler divergence (2.79e-3) compared to IDDNN (3.31e-3), IDDNN still achieves a slightly higher accuracy of 92%, compared to 90% for GPS. It is emphasised that trade-offs may be necessary in practice to determine the most suitable design approach for a given application. Traditional optimisation offers ease of design flexibility, allowing for on-the-fly model improvement and modification. In contrast, DL may require re-training, leading to delays in process workflow

Quantum-Enhanced Generalized Pattern Search Optimization

While the development of quantum computers promises a myriad of advantages over their classical counterparts, care must be taken when designing algorithms that substitute a classical technique with a potentially advantageous quantum method. The probabilistic nature of many quantum algorithms may result in new behavior that could negatively impact the performance of the larger algorithm. The purpose of this work is to preserve the advantages of applying quantum search methods for generalized pattern search algorithms (GPSs) without violating the convergence criteria. It is well known that quantum search methods are able to reduce the expected number of oracle calls needed for finding the solution to a search problem from O(N) to O(N) However, the number of oracle calls needed to determine that no solution exists with certainty is exceedingly high and potentially infinite. In the case of GPS, this is a significant problem since overlooking a solution during an iteration will violate a needed assumption for convergence. Here, we overcome this problem by introducing the quantum improved point search (QIPS), a classical–quantum hybrid variant of the quantum search algorithm QSearch. QIPS retains the O(N) oracle query complexity of QSearch when a solution exists. However, it is able to determine when no solution exists, with certainty, using only O(N) oracle calls.

Automatic Generation Control of a Multi-Area Hybrid Renewable Energy System Using a Proposed Novel GA-Fuzzy Logic Self-Tuning PID Controller

Human activities overwhelm our environment with CO2 and other global warming issues. The current electricity landscape necessitates a superior, continuous power supply and addressing such environmental concerns. These issues can be resolved by incorporating renewable energy sources (RESs) into the utility grid. Thus, this paper presents an optimized hybrid fuzzy logic self-tuning PID controller to control the automatic generation control (AGC) of various renewable sources. This controller regulates the frequency deviations of the power system and governs the change in the tie-line load of a multi-area hybrid energy system composed of wind, biomass, and photovoltaic energy sources. MATLAB Simulink software was applied to design and test the system. The PID controller has been tuned using four algorithms, namely, genetic algorithm (GA), pattern search (PS), simulated annealing (SA), and particle swarm optimization (PSO), and we compared the results with the proposed novel optimized PID controller (GA-fuzzy logic self-tuning technique) to validate it. The results show the superiority of the proposed hybrid GA-fuzzy logic self-tuning algorithm over the other algorithms in bringing the power system back to its regular operation. The paper also proposes an operation strategy to lower the utilization of biomass energy in the presence of other renewable energy sources.

An approximating state-dependent control method based on modified pattern search optimization for nonlinear optimal control problem

In this paper, an approximating state-dependent control (ASC) method with modified pattern search (MPS) optimization for nonlinear optimal control problem is proposed. First, by converting the nonlinear optimal control problem into a number of interrelated time-varying linear quadratic regulator subproblems, the ASC method can solve each subproblem iteratively until the approximate solution is obtained. Second, in each iterative control process, the MPS is used to solve the controllability optimization problem. The optimal state-dependent weighting coefficients are obtained during the MPS optimization. Moreover, the MPS uses simplex gradient to design the search direction, which makes the optimization process efficient and fast. The convergence of MPS optimization is also proved in this paper. Finally, two simulation examples are given to illustrate the effectiveness of ASC method using the MPS optimization. The result shows that the ASC method can reduce the iterations of the approximate solution, and the MPS optimization can optimize the control performance of the ASC method.

A predictive modelling strategy for warpage and shrinkage defects in plastic injection molding using fuzzy logic and pattern search optimization

A predictive modelling strategy for warpage and shrinkage defects in plastic injection molding using fuzzy logic and pattern search optimization

Deep Learning-Based Metasurface Design for Smart Cooling of Spacecraft

A reconfigurable metasurface constitutes an important block of future adaptive and smart nanophotonic applications, such as adaptive cooling in spacecraft. In this paper, we introduce a new modeling approach for the fast design of tunable and reconfigurable metasurface structures using a convolutional deep learning network. The metasurface structure is modeled as a multilayer image tensor to model material properties as image maps. We avoid the dimensionality mismatch problem using the operating wavelength as an input to the network. As a case study, we model the response of a reconfigurable absorber that employs the phase transition of vanadium dioxide in the mid-infrared spectrum. The feed-forward model is used as a surrogate model and is subsequently employed within a pattern search optimization process to design a passive adaptive cooling surface leveraging the phase transition of vanadium dioxide. The results indicate that our model delivers an accurate prediction of the metasurface response using a relatively small training dataset. The proposed patterned vanadium dioxide metasurface achieved a 28% saving in coating thickness compared to the literature while maintaining reasonable emissivity contrast at 0.43. Moreover, our design approach was able to overcome the non-uniqueness problem by generating multiple patterns that satisfy the design objectives. The proposed adaptive metasurface can potentially serve as a core block for passive spacecraft cooling applications. We also believe that our design approach can be extended to cover a wider range of applications.

Smart carrot chasing guidance law for path following of unmanned surface vehicles

Carrot chasing guidance law is one of the most widely used path following algorithms due to its simplicity and ease of implementation; however, it has a fixed parameter which leads to large cross-tracking errors during different navigational conditions. This study proposes an innovative approach to carrot chasing algorithm to minimize cross-tracking errors. Pattern search optimization technique is integrated with carrot chasing guidance law to determine unique virtual target points obtained by flexible parameters instead of a fixed parameter. Proposed smart carrot chasing guidance law (SCCGL) provides stable and accurate path following even for different navigational conditions of unmanned surface vehicle (USV). To the best of our knowledge, we are the first to apply pattern search optimization technique to carrot chasing guidance law while USV is performing multi-tasks of predefined paths. This novelty significantly reduces both cross tracking errors and computational costs. Firstly, SCCGL is tested and compared with traditional carrot chasing algorithm in the numerical simulator for several navigational conditions such as different lists of waypoints, different initial locations, and different maximum turning rates of USV. SCCGL automatically determines optimal parameters to make stable and accurate navigation. SCCGL significantly reduces cross tracking errors compared to classical carrot chasing algorithm. This is the first contribution of this paper. Secondly, genetic algorithm optimization method has been implemented to carrot chasing guidance law instead of pattern search optimization technique. Genetic algorithm causes the total simulation time to be quite long. The proposed SCCGL (pattern search integrated carrot chasing guidance law) gives optimum results 20 times faster than the genetic algorithm. This is the second and main contribution of developed SCCGL method. It is observed that SCCGL provides best navigation with minimum cross-tracking errors and minimum computational cost compared to the classical carrot chasing algorithm and other optimization technique.

Broadband RCS reduction by means of disordered coding metasurfaces

Pattern Search (PS) optimization algorithm has been used for the design of disordered coding metasurfaces with the aim of reducing both monostatic and bistatic Radar Cross Section (RCS) in the 26–40 GHz frequency range. On this basis, high and low phase-state unit-cells have been defined with special attention to the minimization of coupling between the elementary patterns, thus allowing a quasi-constant 180° phase difference over a broad frequency range. Then, finite size metasurfaces have been designed and fabricated using the printed circuit board technological process. Finally, the experimental characterization of various disordered samples leads to −10 dB monostatic RCS reduction over a relative frequency band broader than 35% with a good agreement with both simulated and analytic theoretical predictions. This illustrates the relevance of disordered coded patterns for RCS reduction as well as the PS optimization approach as a design method of coding metasurfaces.

Robust Speed Control of Uncertain Two-Mass System

The main purpose of this work is to present a robust speed control structure for a two-mass system. The tested system consists of a PI controller with two additional feedback. The coefficients of the control system are selected using a pattern-search optimization method in order to obtain robustness to changes in the system parameters. The control system requires information about non-measurable state variables. For this purpose, it is proposed to use a multilayer observer. In order to show the advantages of the MLO system, this article also presents comparative studies with a classical observer. A number of simulation and experimental tests are carried out. The obtained results confirmed a much higher quality of control in the system cooperating with a multilayer observer compared to the system with a classical observer.

A combined scheme of parallel-reaction kinetic model and multi-layer artificial neural network model on pyrolysis of Reed Canary

A comprehensive understanding of pyrolysis kinetics is crucial for the design of biomass pyrolysis. In this study, a combined scheme was proposed for biomass pyrolysis. A kinetic model, serving as a knowledge-based model, was established to capture the primary trend of biomass pyrolysis. An ANN model, as an experience-based model, was developed to provide “details” on the process. The kinetic model incorporated a least-square optimization model to calculate optimal kinetic parameters for pyrolysis reaction steps. The ANN model utilized metaheuristic algorithms such as Particle Swarm Optimization, Pattern Search, Genetic Algorithms, and Surrogate Optimization to optimize the hyperparameters of the ANN model. By combining the kinetic model with the ANN model, comprehensive predictions of biomass pyrolysis were obtained, and the combined Kinetic-ANN model was validated with experimental data at 4 heating rates. Finally, the combined model was compared with the solo-kinetic and solo-ANN models, confirming the advanced performance of the combined approach.

Effect of foundation embedment ratio in suppressing seismic-induced vibrations using optimum tuned mass damper

Tuned mass damper (TMD), which is used in structures to mitigate the dynamic responses caused by environmental loads, is tuned according to the dominant frequency of the structures. The dominant frequency of structure built on soil with low stiffness is significantly affected by the soil-structure interaction (SSI). In previous studies, only the surface foundation case is considered to investigate the design parameters of TMD considering SSI effects and the embedment depth of the foundation is neglected in these studies. Thus, this paper investigates the optimization of TMD for mitigation of responses of a high-rise building considering various depths of embedment and soil properties. The optimum parameters of TMD are investigated using five different optimization algorithms (i.e., simulated annealing, sequential quadratic programming, genetic algorithm, pattern search, and particle swarm optimization). The objective function minimizes the peak acceleration amplitude at the top of the structure. The effectiveness of optimum TMD is verified to use 20 near-fault pulse-like ground motions. The comparisons show that the embedment ratio of the foundation and the soil properties significantly affect the optimization of TMD.

A Comparison of Faulty Antenna Detection Methodologies in Planar Array

Broadcasting, radar, sonar and space telecommunication systems use phased arrays to produce directed signals to be transmitted at the desired angle. This system requires a large number of antenna elements. The presence of faulty element(s) in an array causes asymmetry, which results in a deformed radiation pattern with higher sidelobe levels. Higher sidelobe levels indicate waste of energy by transmitting and receiving signals in unwanted directions. Hence, it is important to develop a method that detects faulty elements and corrects the radiation pattern. To correct the failed radiation pattern, failed elements in an array must be identified first. There have been various studies conducted on linear array failed radiation pattern correction and the finding of faulty elements, but investigation on the planar array is limited. Further, the optimization suggested for linear arrays does not necessarily work for the planar array. In this study, planar array faulty antenna detection was developed with pattern search (PS), simulated annealing (SA), and particle swarm optimization (PSO) methods by reducing the Signal to Noise Ratio (SNR) as the objective function. The analysis was varied for 8 × 8 and 6 × 6 planar arrays with different types of failures. The results were compared to find the best method to identify the faulty element’s location in a planar array. The pattern search method produced outstanding results in finding the faulty element’s locations by providing 100% accuracy for all types of failure, while other methods failed to do the same.

Adaptive fuzzy approach for load frequency control using hybrid moth flame pattern search optimization with real time validation

Adaptive fuzzy approach for load frequency control using hybrid moth flame pattern search optimization with real time validation

Global sensitivity analysis and optimal design of heat recovery ventilation for zero emission buildings

Energy-efficient building services are necessary to realise zero-emission buildings while maintaining adequate indoor environmental quality. As the share of ventilation heating needs grow in well-insulated and airtight buildings, heat recovery in mechanical ventilation systems is increasingly common. Ventilation heat recovery is one of the most efficient and viable means to reduce ventilation heat losses and save energy. Highly efficient heat exchangers are being developed or applied to maximise the energy-saving potential of heat recovery ventilation. Nevertheless, the effects of practical operating conditions and the constraints of heat recovery – such as variations in ventilation rates, frost protection, and the prevention of an overheated air supply over a long-term period, which may significantly influence realistic recovery rates – have been less considered in efforts to maximise the energy savings. It is unclear which design parameters for heat recovery devices have the greatest effect on the annual energy savings from ventilation.This study proposes annual efficiency and annual net energy saving models for heat recovery ventilation that consider ventilation rate variations, the longitudinal heat conduction effect and operating controls. We use a global sensitivity analysis to quantify the contributions of various design input parameters to the variation in annual recovery efficiency and annual net energy savings. We identify the most influential parameters and their significant interaction effects for the annual energy performance of heat recovery ventilation. More attention should be paid to these most influential parameters during the design process. Furthermore, the optimal designs for rotary heat exchangers (as identified by a pattern-search optimisation algorithm) can improve annual net energy savings in demand-controlled ventilation by 33–48%, depending on the building areas. In combination with the reference year analysis presented in this study, heat recovery and demand-controlled ventilation can help to meet the need for highly efficient ventilation systems and zero-emission buildings.

Comparison of parallel optimization algorithms on computationally expensive groundwater remediation designs

Contaminated groundwater resources threaten human health and destroy ecosystems in many areas worldwide. Groundwater remediation is crucial for environmental recovery; however, it can be cost prohibitive. Planning a cost-effective remediation design can take a long time, as it may involve the evaluation of many management decisions, and the corresponding simulation models are computationally demanding. Parallel optimization can facilitate much faster management decisions for cost-effective groundwater remediation design using complex pollutant transport models. However, the efficiency of different parallel optimization algorithms varies depending on both the search strategy and parallelism. In this paper, we show the performance of a parallel surrogate-based optimization algorithm called parallel stochastic radial basis function (p-SRBF), which has not been previously used on contaminant remediation problems. The two case problems involve two superfund sites (i.e., the Umatilla Aquifer and Blaine Aquifer), and one objective evaluation takes 5 and 30 min for the two problems, respectively. Exceptional speedup (superlinear) is achieved with 4 to 16 cores, and excellent speedup is achieved using up to 64 cores, obtaining a good solution at 80 % efficiency. We compare our p-SRBF with three different parallel derivative-free optimization algorithms, including genetic algorithm, mesh adaptive direct search, and asynchronous parallel pattern search optimization, in terms of objective quality, computational reduction and robust behavior across multiple trials. p-SRBF outperforms the other algorithms, as it finds the best solution in both the Umatilla and Blaine cases and reduces the computational budget by at least 50 % in both problems. Additionally, statistical comparisons show that the p-SRBF results are better than those of the alternative algorithms at the 5 % significant level. This study enriches theoretical real-world groundwater remediation methods. The results demonstrate that p-SRBF is promising for environmental management problems that involve computationally expensive models.

Parametrisation Scheme for Multidisciplinary Design Analysis and Optimisation of a Floating Offshore Wind Turbine Substructure – OC3 5MW Case Study

The development of novel energy technologies is considered imperative in the provision of solutions to meet an increasing global demand for clean energy. Floating Offshore Wind Turbine (FOWT) is one of the emerging technologies to exploit the vast wind resources available in deeper waters. To lower the levelized cost of energy (LCOE) or optimise the performance response associated with a FOWT system, a detailed understanding of the different disciplines (Aero-Hydro-Servo-Elastic) within the system and the relationship between the FOWT system and the dynamics of the marine environment is required. This requires an efficient Multidisciplinary Design, Analysis and Optimisation (MDAO) framework for FOWT systems to reduce the capital cost and increase dynamic performance. A key component of any MDAO framework is the shape parameterisation scheme, as it enables the modelling of a large array of platform designs with different geometric shapes using limited number of parameters. This work focuses on the B-Spline parameterisation modelling technique of OC3 spar-buoy and the use pattern search optimization algorithm to select the optimal design variants. The parametrisation technique is implemented in an analysis framework, where a B-spline library from Sesam GeniE is used to model each design representation, and a potential flow frequency domain analysis solver (HydroD/Wadam) is used for the hydrodynamic analysis. Validation of the selected designs within the design space is conducted with a benchmark NREL5MW spar-buoy hydrodynamic response results in literature with the hydrodynamic response of the frequency domain modelling approach using Sesam GeniE and HydroD/Wadam. This analysis process shows a high accuracy in response results between the OC3 spar-buoy in literature and the OC3 spar-buoy model design using B-Spline parametrization technique. Key performance metrics like the cost of materials and root mean square (RMS) of the nacelle acceleration also show improvement with the design variants compared to estimation from OC3 design in literature.

A two-stage cooperative scatter search algorithm with multi-population hierarchical learning mechanism

Scatter search (SS) is a population-based metaheuristic algorithm, which has been proved high efficiency and effective optimizer for complex continuous real value problems. A two-stage cooperative SS guided with the multi-population hierarchical learning mechanism (TCSSMH) to overcome the slow convergence speed of the original SS is proposed. Three strategies are applied to the original SS. Firstly, TCSSMH adopts an adaptive two-way selection search strategy based on the elite reference set (RefSet), which is elite-oriented and ensures the quality of the population. Secondly, the multi-group hierarchical learning mechanism is embedded in the updating process of the RefSet, and the population of the candidates is divided into three levels including excellent candidates, medium candidates, and inferior candidates according to the fitness value of the function. These three subpopulations cooperate to balance the exploration and exploitation ability of the algorithm in the process of evolution. Finally, each subpopulation adopts an interactive learning strategy to increase the diversity of the population and avoid premature convergence of solutions. The optimum of each subpopulation with high accuracy is obtained by the pattern search (PS) optimization. The stronger search ability and higher search efficiency of these additional proposed strategies are verified by extensive experiments. The TCSSMH algorithm is tested on the CEC2017 benchmark test suite and practical engineering problems. The experimental results show that the TCSSMH algorithm is superior to other state-of-the-art algorithms in global search ability and convergence on the benchmark problems.

Dielectric spectroscopy of nanofluids in deionized water: Method of removing electrode polarization effect

HypothesisThe electrode polarization observed in the measured complex dielectric frequency responses of the nanosuspensions could be isolated by using time-domain analysis. ExperimentsComplex impedance and dielectric frequency responses of various ferroelectric and magnetic nanofluids were ultrasonically dispersed in deionized water and measured within 20 Hz to 6 GHz range for the first time. A general Kramer-Krönig continuous relaxation model was fitted to the dielectric spectra in the least-squares sense using the Tikhonov’s regularization. The independent ζ-potential measurements of the BTO-C2 and Fe3O4 nanosuspensions were performed. FindingsThree distinct peaks were observed in the time-domain responses. Zero padding of the two slowest peaks, followed by the inversion of the corrected time domain responses, resulted in the flattening of the lower frequency portion of the dielectric constant. The effective medium theory model coupled with the pattern search optimization routine, were used to match these flattened responses. The surface charge on the nanoparticles was estimated from the measured ζ-potential using the Gouy-Chapman-Stern-Grahame model. The resulting values were comparable with the ones obtained from the effective medium theory fitting procedure, which proves effectiveness of the proposed approach at removing the electrode polarization effects. The proposed method has the advantage over the existing ones, because it is carried out in time domain and requires no prior knowledge about the underlying relaxation time amplitude distributions or specific equivalent circuits.

Optimization of Fuel Economy for a Multimode Plug-in Hybrid Electric Vehicle using Atkinson Thermodynamic Cycle Engine

Recently, Plug-in Hybrid electric vehicles become a sustainable solution to strike a balance between performance and fuel economy. For a multimode PHEV, the car switches among three operation modes; namely electric mode, series mode, and parallel mode to maximize fuel economy based on the driving conditions. Atkinson thermodynamic cycle has a higher expansion stroke compared to Otto cycle; which leads to more work, less emissions and higher thermal efficiency. In this work, the optimization of fuel economy for a multimode PHEV reference vehicle that resembles Honda Accord PHEV using Atkinson engine is conducted. The optimization is based on a combined driving cycle that includes both a city cycle and a highway cycle. Mapping technique is used to represent performance and fuel consumption of Atkinson engine. The mapping is calibrated to match Honda Accord PHEV performance data. Global generalized pattern search optimization method is utilized. The optimization is performed in two steps. In the first step, the driving mode-switching strategy is optimized to increase overall equivalent Miles-per-Gallon () for the combined driving cycle. In the second step, powertrain components are re-sized to further improve equivalent fuel economy. Optimization of driving mode-switching increased from 48 to 64.5 (30% increase) and a further 10% increase to 70.5 is achieved by powertrain components sizing optimization. The developed optimization method proved to be a viable method to improve fuel economy of hybrid vehicles.

Exploiting Problem Structure in Derivative Free Optimization

A structured version of derivative-free random pattern search optimization algorithms is introduced, which is able to exploit coordinate partially separable structure (typically associated with sparsity) often present in unconstrained and bound-constrained optimization problems. This technique improves performance by orders of magnitude and makes it possible to solve large problems that otherwise are totally intractable by other derivative-free methods. A library of interpolation-based modelling tools is also described, which can be associated with the structured or unstructured versions of the initial pattern search algorithm. The use of the library further enhances performance, especially when associated with structure. The significant gains in performance associated with these two techniques are illustrated using a new freely-available release of the Brute Force Optimizer (BFO) package firstly introduced in [Porcelli and Toint 2017 ], which incorporates them. An interesting conclusion of the numerical results presented is that providing global structural information on a problem can result in significantly less evaluations of the objective function than attempting to building local Taylor-like models.