Original Flower Pollination Algorithm Research Articles

A novel robust design optimization method combining improved flower pollination algorithm and dual Kriging

PurposeThis study aims to propose a more precise method for robust design optimization of mechanical structures with black-box problems, while also considering the efficiency of uncertainty analysis.Design/methodology/approachThe method first introduces a dual adaptive chaotic flower pollination algorithm (DACFPA) to overcome the shortcomings of the original flower pollination algorithm (FPA), such as its susceptibility to poor accuracy and convergence efficiency when dealing with complex optimization problems. Furthermore, a DACFPA-Kriging model is developed by optimizing the relevant parameter of Kriging model via DACFPA. Finally, the dual Kriging model is constructed to improve the efficiency of uncertainty analysis, and a robust design optimization method based on DACFPA-Dual-Kriging is proposed.FindingsThe DACFPA outperforms the FPA, particle swarm optimization and gray wolf optimization algorithms in terms of solution accuracy, convergence speed and capacity to avoid local optimal solutions. Additionally, the DACFPA-Kriging model exhibits superior prediction accuracy and robustness contrasted with the original Kriging and FPA-Kriging. The proposed method for robust design optimization based on DACFPA-Dual-Kriging is applied to the motor hanger of the electric multiple units as an engineering case study, and the results confirm a significant reduction in the fluctuation of the maximum equivalent stress.Originality/valueThis study represents the initial attempt to enhance the prediction accuracy of the Kriging model using the improved FPA and to combine the dual Kriging model for uncertainty analysis, providing an idea for the robust optimization design of mechanical structure with black-box problem.

Flower pollination algorithm with pollinator attraction

The Flower Pollination Algorithm (FPA) is a highly efficient optimization algorithm that is inspired by the evolution process of flowering plants. In the present study, a modified version of FPA is proposed accounting for an additional feature of flower pollination in nature that is the so-called pollinator attraction. Pollinator attraction represents the natural tendency of flower species to evolve in order to attract pollinators by using their colour, shape and scent as well as nutritious rewards. To reflect this evolution mechanism, the proposed FPA variant with Pollinator Attraction (FPAPA) provides fitter flowers of the population with higher probabilities of achieving pollen transfer via biotic pollination than other flowers. FPAPA is tested against a set of 28 benchmark mathematical functions, defined in IEEE-CEC’13 for real-parameter single-objective optimization problems, as well as structural optimization problems. Numerical experiments show that the modified FPA represents a statistically significant improvement upon the original FPA and that it can outperform other state-of-the-art optimization algorithms offering better and more robust optimal solutions. Additional research is suggested to combine FPAPA with other modified and hybridized versions of FPA to further increase its performance in challenging optimization problems.

Intelligent diagnosis of natural gas pipeline defects using improved flower pollination algorithm and artificial neural network

With the increasing service life of pipelines, natural gas pipelines can gradually age and produce various corrosion defects. Hence, in order to ensure the efficiency and safety of pipeline transportation in the peak period of natural gas consumption, the intelligent diagnosis technology of the pipelines is of vital importance. In this paper, iterative chaotic map with infinite collapses (ICMIC) and comprehensive opposition (CO) learning strategy are used to optimize flower pollination algorithm (FPA) to enhance search abilities of original FPA algorithm. Among them, the ICMIC enhances the diversity of population, and the local CO learning strategy enhances its exploitation ability. Fifteen classical benchmark functions are used to test the optimization performance of the improved flower pollination algorithm (IFPA). Considering the category, complexity and application of test functions, a more reasonable evaluation formula is proposed. The test shows that the performance of IFPA algorithm is obviously better than other classical intelligent algorithms. Based on the excellent performance of IFPA, the IFPA algorithm is used to optimize the initial weights and thresholds of back propagation (BP) neural network. Therefore, a comprehensive IFPA-BP network model is constructed for the intelligent diagnosis of natural gas pipeline defects. The results show that the proposed model can effectively overcome the problem that BP neural network is prone to fall into local optimal value, and it can accurately identify the pipelines defects. This will facilitate intelligent diagnosis of natural gas pipelines defects.

DYNAMIC PROBABILITY SELECTION FOR FLOWER POLLINATION ALGORITHM BASED ON METROPOLISHASTINGS CRITERIA

Flower Pollination Algorithm (FPA) is a relatively new meta-heuristic algorithm that adopts its metaphor from the proliferation role of flowers in plants. Having only one parameter control (i.e. the switch probability, pa) to choose from the global search (i.e. exploration) and local search (i.e. exploitation) is the main strength of FPA as compared to other meta-heuristic algorithms. However, FPA still suffers from variability of its performance as there is no one size that fits all values for pa, depending on the characteristics of the optimisation function. This paper proposed flower pollination algorithm metropolis-hastings (FPA-MH) based on the adoption of Metropolis-Hastings criteria adopted from the Simulated Annealing (SA) algorithm to enable dynamic selection of the pa probability. Adopting the problem of t-way test suite generation as the case study and with the comparative evaluation with the original FPA, FPA-MH gave promising results owing to its dynamic and adaptive selection of search operators based on the need of the current search.

Toward the modification of flower pollination algorithm in clustering-based image segmentation

Flower pollination algorithm (FPA) is a new bio-inspired optimization algorithm, which has shown an effective performance on solving many optimization problems. However, the effectiveness of FPA significantly depends on the balance achieved by the exploration and exploitation evolutionary stages. Since purely exploration procedure promotes non-accurate solutions, meanwhile, purely exploitation operation promotes sub-optimal solutions in the presence of multiple optima. In this study, three global search and two local search strategies have been designed to improve balance among evolutionary stages, increasing the efficiency and robustness of the original FPA methodology. Additionally, some parameter adaptation techniques are also incorporated in the proposed methodology. The modified FPA has been successfully applied for histopathological image segmentation problem. The experimental and computational effort results indicate its effectiveness over existing swarm intelligence algorithms and machine learning methods.

Exploiting flower constancy in flower pollination algorithm: improved biotic flower pollination algorithm and its experimental evaluation

Recent growth of metaheuristic search strategies has brought a huge progress in the domain of computational optimization. The breakthrough started since the well-known Particle Swarm Optimization algorithm had been introduced and examined. Optimization technique presented in this contribution mimics the process of flower pollination. It is build on the foundation of the first technique of this kind—known as Flower Pollination Algorithm (FPA). In this paper, its simplified and improved version, obtained after extensive performance testing, is presented. It is based on only one natural phenomena—called flower constancy—the natural mechanism allowing pollen carrying insects to remember the positions of the best pollen sources. Modified FPA, named as Biotic Flower Pollination Algorithm (BFPA) and relying solely on biotic pollinators, outperforms original FPA, which itself proved to be very effective approach. The paper first presents a short description of original FPA and the changes leading to Biotic Flower Pollination Algorithm. It also discusses performance of the modified algorithm on a full set of CEC17 benchmark functions. Furthermore, in that aspect, the comparison between BFPA and other optimization algorithms is also given. Finally, brief exemplary application of modified algorithm in the field of probabilistic modeling, related to physics and engineering, is also presented.

Calculation of maximum permitted capacity of photovoltaic based on flower pollination algorithm combined with genetic algorithm (GA-FPA) in distribution network

The interconnection of distributed photovoltaic (PV) power in the distribution network will bring plenty of adverse effects. It is critical of researching on maximum permitted capacity (MPC) of distribution network to maintain the safe operation of distribution network. A mathematical optimization model was constructed for the MPC of distributed PV with the objective of maximizing the difference between the sum of distributed PV introduction and the network loss, and the constraints of system power flow, node voltage, feeder current, and short-circuit current. At the same time, an improved Flower Pollination Algorithm (FPA) that is Flower Pollination Algorithm combined with Genetic Algorithm (GA-FPA) was proposed to analyse IEEE 33-bus system. Further, by comparing with other optimization algorithms in terms of convergence speed and convergence accuracy, like original FPA and Genetic Algorithm (GA), the correctness of the model and the effectiveness of the algorithm are verified.

Comparative Study of Adaptive Elitism and Mutation Operators in Flower Pollination Algorithm for Combinatorial Testing Problem

The performance of meta-heuristic search algorithms highly depends on their intensification and diversification abilities. Different algorithms adopt intensification and diversification strategies in order to obtain better results. Elitism and mutation are common operators that are used for increasing the diversity of the population. Flower Pollination Algorithm (FPA) is one of the recent meta-heuristic algorithms for global optimization. Although proven to be efficient, FPA is prone to get stuck into a local optimum due to the weakness of its population’s diversity especially for multimodal optimization problem. In this paper, first, we propose two strategies based on mutation-FPA (mFPA) and elitism-FPA (eFPA) for t-way test generation (t refer to interaction strength). Then, a comparison between mFPA and eFPA is studied to analysis the effect of introducing elitism and mutation operators on FPA’s performance. The results of the experiments show that both of eFPA and mFPA strategies appear to produce better results than original FPA strategy, however, eFPA performs much better than mFPA in term of tests size.

A Novel Flower Pollination Algorithm for Modeling the Boiler Thermal Efficiency

The flower pollination algorithm (FPA) is a nature-inspired optimization algorithm. To improve the solution quality and convergence speed of FPA, we proposed a novel flower pollination algorithm (NFPA) which is a hybrid algorithm based on original FPA and wind driven optimization algorithm. Simulation experiments demonstrate that NFPA has better search performance on classical numerical function optimizations compared with other the state-of-the-art optimization methods. In addition, the NFPA is adopted to optimize parameters of fast learning network to build thermal efficiency model of a 330 MW coal-fired boiler and a well-generalized model is obtained. Experimental results show that the tuned fast learning network model by NFPA has better prediction accuracy and generalization ability than other combination models.

Hybrid flower pollination algorithm strategies for t-way test suite generation.

The application of meta-heuristic algorithms for t-way testing has recently become prevalent. Consequently, many useful meta-heuristic algorithms have been developed on the basis of the implementation of t-way strategies (where t indicates the interaction strength). Mixed results have been reported in the literature to highlight the fact that no single strategy appears to be superior compared with other configurations. The hybridization of two or more algorithms can enhance the overall search capabilities, that is, by compensating the limitation of one algorithm with the strength of others. Thus, hybrid variants of the flower pollination algorithm (FPA) are proposed in the current work. Four hybrid variants of FPA are considered by combining FPA with other algorithmic components. The experimental results demonstrate that FPA hybrids overcome the problems of slow convergence in the original FPA and offers statistically superior performance compared with existing t-way strategies in terms of test suite size.

Parameter estimation for chaotic systems via a hybrid flower pollination algorithm

Parameter estimation is a fundamental research issue which has attracted great concern in the control and synchronization of chaotic systems. This problem can be mathematically described as a multi-dimensional continuous optimization problem through constructing an appropriate fitness function, and then solved via meta-heuristic algorithms. A hybrid flower pollination algorithm is proposed in this paper for solving this problem more efficiently. The proposed algorithm well combines the good global exploration ability of the original flower pollination algorithm and the powerful local exploitation ability of the Nelder–Mead simplex method together by integrating these two methods in a very simple way. The experimental results tested on three typical chaotic system parameter estimation problems with three unknown system parameters, including the Lorenz system, the Rossler system, and the Lorenz system under the noise condition, demonstrate that the algorithm can estimate the unknown parameters efficiently and accurately. The comparisons with the basic flower pollination algorithm and some other four reported methods including the quantum particle swarm optimization, the hybrid adaptive cuckoo search optimization algorithm, the oppositional backtracking search optimization algorithm, and the hybrid algorithm combining differential evolution with artificial bee colony suggest that the proposed algorithm is energy efficient and superior. The proposed algorithm can be used as a new and effective choice for parameter estimation of chaotic systems.

Flower Pollination Algorithm for Multimodal Optimization

This paper proposes a new algorithm called Multimodal Flower Pollination Algorithm (MFPA). Under MFPA, the original Flower Pollination Algorithm (FPA) is enhanced with multimodal capabilities in order to find all possible optima in an optimization problem. The performance of the proposed MFPA is compared to several multimodal approaches considering the evaluation in a set of well-known benchmark functions. Experimental data indicate that the proposed MFPA provides better results over other multimodal competitors in terms of accuracy and robustness.

Assembly sequence optimization using a flower pollination algorithm-based approach

One of the important decisions in assembly process planning is determination of assembly sequence. Choice of the optimum sequence is made difficult due to various reasons. There are various precedence constraints and optimization criteria. Moreover, a product may be possible to assemble in many alternative ways following different sequences, thus making assembly sequence optimization a multi-modal optimization problem with multiple optimum solutions. It is necessary to generate as many unique optimum solutions as possible in order to allow the process planner to take a decision. Moreover, with increase in part count, the number of feasible sequences rises staggeringly, thereby making assembly sequence optimization laborious and time consuming. Most conventional mathematical algorithms are known to perform poorly when used to obtain multiple optimum solutions. On the other hand, soft computing based evolutionary optimization algorithms are good candidates for multi-modal optimization. Another challenge is to develop an algorithm that can automatically maintain diversity in the optimum solutions found over the generations (i.e. optimum solutions having the same fitness but unique). Keeping the above in mind, in the present paper, an intelligent assembly sequence optimization methodology based on application of flower pollination algorithm (FPA) has been developed to automatically generate multiple unique optimal assembly sequences, subject to various precedence constraints, based on minimisation of number of orientation changes and tool changes. Since in the present paper, FPA has been applied for the first time to a discrete optimization problem like assembly sequence optimization, the main challenge before us in applying FPA was the continuous nature of the original FPA. Therefore, modifications have been made by us in the rules for local and global pollination of FPA to make it suited for solving the given discrete optimization problem. In order to evaluate the performance of FPA, the results have been compared with two other well-known soft computing techniques namely, Genetic Algorithm (GA) and Ant Colony Optimization (ACO) and also with a recently published soft computing based algorithm, Improved Harmony Search (IHS). It was found that the novelty of the proposed FPA lies in its capability to find multiple unique optimum solutions in one single simulation run and capability to automatically maintain diversity in the optimum solutions found over the generations. On the other hand, in case of GA, ACO and IHS, it is not possible to maintain the diversity in multiple optimum solutions as the complete population finally converges to a few unique optimum solutions. Therefore, it can be concluded that FPA performs better in solving the given multi-modal optimization problem of assembly sequence optimization.