PSO-based Algorithm Research Articles

Collaborative optimization of conical cutter sequence for efficient multi-axis machining of deep curved cavities

As an essential part of the composite fan blade, the leading edge protection cap (LPC) has a very narrow and deep cavity which, in order to avoid the interference, requires the conical ball-end cutter to be long and thin. It calls to improve the cutter’s stiffness so to increase the roughing efficiency of the whole cavity. Targeting this important problem, this paper proposes a novel method to collaboratively optimize a conical cutter sequence, the machining ranges, and the cutting parameters for achieving the highest material removal efficiency. The main difficulty is that all the relevant entities – i.e., the LPC cavity geometry, tool size, tool axis, machining ranges, minimum tool length, tool stiffness, and cutting performance – collaboratively affect the total cutting efficiency. To facilitate decoupling them, the 3D LPC cavity is characterized by its medial surface. To identify all the geometrically accessible conical tools of each medial point, an efficient algorithm is proposed, which is featured by searching characteristic points on the cavity surface. By considering both the geometric and physical accessibilities, the tool sequence domain is constructed. To obtain the appropriate cutting parameters, the best cutting performance of a weak conical cutter is modeled under the constraint of deflection due to the cutting force. Finally, the complicated collaborative optimization model aiming for the minimum cutting time is established and solved by a PSO-based algorithm. The experimental results confirm that the proposed model and algorithm are practicable for improving the roughing efficiency of LPC cavity.

Form-finding of frame-supported tensile membrane structures using stochastic optimisation

For tensile membrane structures (TMS), form-finding is the first step in design, in which the structure adopts a unique equilibrium shape based on the initial configuration of the structure and the applied initial prestress. However, the choice of a single robust numerical method for form-finding remains debatable, particularly due to the intensive computation involved. The present work proposes a novel form-finding method for TMS using a swarm intelligence algorithm. Particle swarm optimisation (PSO) is used here as the optimiser in the form-finding analysis, formulated as a (area or potential energy) minimisation problem. The computations involve stochastic PSO runs, until the structure converges to an equilibrium configuration. The proposed form-finding method is demonstrated for two study structures with different optimisation objectives.The proposed PSO-based form-finding technique is found to save significant computation cost, when compared to the most commonly used form-finding algorithm of dynamic relaxation. The PSO-based algorithm is also found to lessen arbitrary parameter selections and to be more robust to the selection of an initial TMS configuration.

Site layout planning for prefabricated components subject to dynamic and interactive constraints

The component site layout for the prefabricated construction subject to constraints such as site constraints, component constraints, construction schedule and component arrival schedule that may change over time and interact has seldom been studied. This study is aimed to solve the dynamic prefabricated component site layout problem (DPCSLP) subject to the above constraints. Based on analyses on the DPCSLP concerning the dynamic and interactive constraints, multi-phase layout, and dynamic allocation of yard locations, a mathematical model for the DPCSLP is proposed. Then the particle swarm optimization (PSO)-based solving algorithm regarding DPCSLP mapping and optimization framework incorporated with multi-phase layout and two-step optimization is developed. A case study is presented to demonstrate and justify the developed method. This study contributes to the domain knowledge for solving the DPCSLP regarding the mathematical model and the PSO-based algorithm as well as reducing project duration of the prefabricated construction by enhancing site coordination.

Evolutionary Multi-Objective Optimization for Web Service Location Allocation Problem

With the ever increasing number of functionally similar web services being available on the Internet, the market competition is becoming intense. Web service providers (WSPs) realize that good Quality of Service (QoS) is a key of business success and low network latency is a critical measurement of good QoS. Because network latency is related to location, a straightforward way to reduce network latency is to allocate services to proper locations. However, Web Service Location Allocation Problem (WSLAP) is a challenging task since there are multiple objectives potentially conflicting with each other and the solution search space has a combinatorial nature. In this paper, we consider minimizing the network latency and total cost simultaneously and model the WSLAP as a multi-objective optimization problem. We develop a new PSO-based algorithm to provide a set of trade-off solutions. The results show that the new algorithm can provide a more diverse range of solutions than the compared three well known multi-objective optimization algorithms. Moreover, the new algorithm performs better especially on large problems.

Comparative Analysis of Intelligence Optimization Algorithms in the Thermo-Economic Performance of an Energy Recovery System Based on Organic Rankine Cycle

Abstract This paper compares the performance of a group of intelligent algorithms such as the genetic algorithm (GA), particle swarm optimization (PSO), and repulsive particle swarm optimization (RPSO) based on the optimization of thermo-economic indicators such as the payback period (PBP), the levelized energy cost (LEC), the specific investment cost (SIC), and also in the optimization of the thermodynamic process (net power output) of an energy recovery system in a 2 MW natural gas internal combustion engine based on an organic Rankine cycle. Four parameters were considered to analyze and compare the performance of these algorithms: integral of squared error (ISE), integral of absolute error (IAE), integral of time-weighted absolute error (ITAE), and the integral of time-weighted squared error (ITSE). Analyses of variances (ANOVA) were proposed for each of the parameters studied. The PSO and RPSO algorithms presented the best performance in terms of the mean and the standard deviation of the ISE, IAE, ITAE, and ITSE parameters. Significant differences were not found between the three algorithms in terms of the parameters considered. However, significant differences did exist when comparing groups (pairs) of algorithms considering a significance level of 5%. The ANOVA analysis showed that ITAE was the most affected parameter by population size, while the IAE and ITSE parameters were the less affected. In the optimization, the PSO algorithm obtained the best performance in terms of convergence with values of 0.1110 USD/kWh (LCOE), 4.6971 years (PBP), 1114 USD/kWh (SIC), and 173.64 kW (Wnet). PSO-based algorithms obtained better performance in computational terms compared with the genetic algorithms.

Enhanced Differential Crossover and Quantum Particle Swarm Optimization for IoT Applications

An optimized design with real-time and multiple realistic constraints in complex engineering systems is a crucial challenge for designers. In the non-uniform Internet of Things (IoT) node deployments, the approximation accuracy is directly affected by the parameters like node density and coverage. We propose a novel enhanced differential crossover quantum particle swarm optimization algorithm for solving nonlinear numerical problems. The algorithm is based on hybrid optimization using quantum PSO. Differential evolution operator is used to circumvent group moves in small ranges and falling into the local optima and improves global searchability. The cross operator is employed to promote information interchange among individuals in a group, and exceptional genes can be continued moderately, accompanying the evolutionary process's continuance and adding proactive and reactive features. The proposed algorithm's performance is verified as well as compared with the other algorithms through 30 classic benchmark functions in IEEE CEC2017, with a basic PSO algorithm and improved versions. The results show the smaller values of fitness function and computational efficiency for the benchmark functions of IEEE CEC2019. The proposed algorithm outperforms the existing optimization algorithms and different PSO versions, and has a high precision and faster convergence speed. The average location error is substantially reduced for the smart parking IoT application.

A Continuous Space Location Model and a Particle Swarm Optimization-Based Heuristic Algorithm for Maximizing the Allocation of Ocean-Moored Buoys

Ocean-moored buoys play an important role in global ocean environment monitoring. Motivated by building a sustainable ocean buoy observational network, a spatial optimization approach is proposed to site buoy stations to maximize spatial monitoring efficiency (SME). To achieve this goal, a non-linear, continuous maximum coverage location model named CMCP-Ocean was established, associated with a measurement method of the SME. Meanwhile, a heuristic framework based on the particle swarm optimization (PSO) algorithm was built to solve the CMCP-Ocean model, and optimization strategies including the multi-core parallel computing strategy, the particle velocity updating strategy based on spatial matching, and two potential station selection strategies related to the centroid-based random radiation method (CRRM) and random grid division method (RGDM) were established to improve computing performance. The effectiveness and efficiency of the PSO-based algorithm and the CMCP-Ocean model were verified by a series of experiments; the proposed computing schema named PSO-for-CMCP-Ocean has also proven to be practical and efficient. Finally, the PSO-for-CMCP-Ocean was applied to the buoy station selection of water mass monitoring in the Laizhou Bay of China, and a multi-scale sustainable site planning solution is reported.

Easy Particle Swarm Optimization for Nonlinear Constrained Optimization Problems

Particle swarm optimization (PSO) is a popular stochastic approach for solving practical optimal problems from industries due to its effective performance and few hyperparameters. Nonlinear constrained optimization (NCO) problems frequently cause multiple optimal regions and can cause many infeasible regions in the search space. The state-of-the-art approaches for handling the infeasible regions generated by problems’ constraints either block particles’ paths or penalize NCO problems’ objective values based on the standard updating velocity formula. The standard updating velocity formula introduces difficulties for particles in searching the undiscovered optimal solutions separated by infeasible regions and being mutually restrained on directions by social and cognitive factors. Afterward, the particles cause premature convergence and difficulty searching the undiscovered optimal regions to improve their solutions. Observing the biological ant colony and inspired by lazy ant behavior, this study proposes an easy particle that simulates the lazy ant to diversify the moving direction. Finally, this study integrates the proposed easy particles with referenced PSO-based approaches for solving NCO problems. The experiment results show that the proposed easy particles can effectively reinforce exploration abilities and improve the performances of all referenced PSO-based algorithms to reduce the status of premature convergence in solving NCO problems.

Visual Robot Relocalization Based on Multi-Task CNN and Image-Similarity Strategy.

The traditional CNN for 6D robot relocalization which outputs pose estimations does not interpret whether the model is making sensible predictions or just guessing at random. We found that convnet representations trained on classification problems generalize well to other tasks. Thus, we propose a multi-task CNN for robot relocalization, which can simultaneously perform pose regression and scene recognition. Scene recognition determines whether the input image belongs to the current scene in which the robot is located, not only reducing the error of relocalization but also making us understand with what confidence we can trust the prediction. Meanwhile, we found that when there is a large visual difference between testing images and training images, the pose precision becomes low. Based on this, we present the dual-level image-similarity strategy (DLISS), which consists of two levels: initial level and iteration-level. The initial level performs feature vector clustering in the training set and feature vector acquisition in testing images. The iteration level, namely, the PSO-based image-block selection algorithm, can select the testing images which are the most similar to training images based on the initial level, enabling us to gain higher pose accuracy in testing set. Our method considers both the accuracy and the robustness of relocalization, and it can operate indoors and outdoors in real time, taking at most 27 ms per frame to compute. Finally, we used the Microsoft 7Scenes dataset and the Cambridge Landmarks dataset to evaluate our method. It can obtain approximately 0.33 m and 7.51 accuracy on 7Scenes dataset, and get approximately 1.44 m and 4.83 accuracy on the Cambridge Landmarks dataset. Compared with PoseNet, our CNN reduced the average positional error by 25% and the average angular error by 27.79% on 7Scenes dataset, and reduced the average positional error by 40% and the average angular error by 28.55% on the Cambridge Landmarks dataset. We show that our multi-task CNN can localize from high-level features and is robust to images which are not in the current scene. Furthermore, we show that our multi-task CNN gets higher accuracy of relocalization by using testing images obtained by DLISS.

Hybrid algorithms for generating optimal designs for discriminating multiple nonlinear models under various error distributional assumptions.

Finding a model-based optimal design that can optimally discriminate among a class of plausible models is a difficult task because the design criterion is non-differentiable and requires 2 or more layers of nested optimization. We propose hybrid algorithms based on particle swarm optimization (PSO) to solve such optimization problems, including cases when the optimal design is singular, the mean response of some models are not fully specified and problems that involve 4 layers of nested optimization. Using several classical examples, we show that the proposed PSO-based algorithms are not models or criteria specific, and with a few repeated runs, can produce either an optimal design or a highly efficient design. They are also generally faster than the current algorithms, which are generally slow and work for only specific models or discriminating criteria. As an application, we apply our techniques to find optimal discriminating designs for a dose-response study in toxicology with 5 possible models and compare their performances with traditional and a recently proposed algorithm. In the supplementary material, we provide a R package to generate different types of discriminating designs and evaluate efficiencies of competing designs so that the user can implement an informed design.

A digital watermarking method based on NSCT transform and hybrid evolutionary algorithms with neural networks

This study aims to determine the watermark resistance to different attacks as well as the PSNR level, both of which are essential requirements of watermarking. In our research, we came up with an intelligent design based on NSCT-SVD that fulfills these requirements to a great extent and we managed to use different-sized images for watermark instead of using logos on the host images. Yet we were able to improve PSNR levels and resistance to various attacks. In this paper an NSCT-SVD-based smart watermark model is proposed. We first compare the PSO and PSO-GA algorithms for greater stability using larger SFs obtained by the PSO-GA-AI algorithm. The resulting host image is then decomposed by NSCT transform to obtain images below the low frequency range. Stationary Wavelet Transform (SWT) is performed once on these coefficients and the low frequency coefficients are fed to SVD. Afterwards, SWT transform is performed on the watermark image and the transform is once again taken from the HL coefficients and the LL frequencies are given to the SVD conversion. The rest of image process is insertion. This insertion process dramatically increases the visual transparency and PSNR value. The experiment shows that such a model is able to resist the repeated image attacks with better visibility and power. These results are compared before and after using SWT. We have used a PSO-based algorithm for better results on the False Positive rate in the embedding phase.

Hybrid particle swarm optimization with particle elimination for the high school timetabling problem

In this paper, a PSO-based algorithm that hybridized Particle Swarm Optimization (PSO) and Hill Climbing (HC) is applied to high school timetabling problem. This hybrid has two features, a novel solution transformation and particle elimination. The proposed methodologies are tested on the XHSTT-2014 dataset (which is relatively new for the school timetabling problem) plus other additional instances. The experimental results show that the proposed algorithm is effective in solving small and medium instances compared to standalone HC and better than the conventional PSO for most instances. In a comparison to the state of the art methods, it achieved the lowest mean of soft constraint violations for 7 instances and the lowest mean of hard constraint violations for 1 instance.

Rough sets and social ski-driver optimization for drug toxicity analysis

Background and objectives: Toxicity testing is an important step for developing new drugs, and animals are widely used in this step by exposing them to the toxicants. Zebrafishes are widely used for measuring and detecting the toxicity. However, measuring and testing toxicity manually is not feasible due to the large number of embryos. This work presents an automated model to investigate the toxicity of two toxicants (3, 4-Dichloroaniline (34DCA) and p-Tert-Butylphenol (PTBP)).Methods: The proposed model consists of two steps. In the first step, a set of features is extracted from microscopic images of zebrafish embryos using the Segmentation-Based Fractal Texture Analysis (SFTA) technique. Secondly, a novel rough set-based model using Social ski-driver (SSD) is used to find a global minimal subset of features that preserves important information of the original features. In the third step, the AdaBoost classifier is used to classify an unknown sample to alive or coagulant after exposing the embryo to a toxic compound.Results: For detecting the toxicity, the proposed model is compared with (i) three deterministic rough set reduction algorithms and (ii) the PSO-based algorithm. The classification performance rate of our model was ranged from 97.1% to 99.5% and it outperformed the other algorithms.Conclusions: The results of our experiments proved that the proposed drug toxicity model is efficient for rough set-based feature selection and it obtains a high classification performance.

A multiobjective evolutionary optimization method based critical rainfall thresholds for debris flows initiation

At present, most researches on the critical rainfall threshold of debris flow initiation use a linear model obtained through regression. With relatively weak fault tolerance, this method not only ignores nonlinear effects but also is susceptible to singular noise samples, which makes it difficult to characterize the true quantization relationship of the rainfall threshold. Besides, the early warning threshold determined by statistical parameters is susceptible to negative samples (samples where no debris flow has occurred), which leads to uncertainty in the reliability of the early warning results by the regression curve. To overcome the above limitations, this study develops a data-driven multiobjective evolutionary optimization method that combines an artificial neural network (ANN) and a multiobjective evolutionary optimization implemented by particle swarm optimization (PSO). Firstly, the Pareto optimality method is used to represent the nonlinear and conflicting critical thresholds for the rainfall intensity I and the rainfall duration D. An ANN is used to construct a dual-target (dual-task) predictive surrogate model, and then a PSO-based multiobjective evolutionary optimization algorithm is applied to train the ANN and stochastically search the trained ANN for obtaining the Pareto front of the I-D surrogate prediction model, which is intended to overcome the limitations of the existing linear regression-based threshold methods. Finally, a double early warning curve model that can effectively control the false alarm rate and negative alarm rate of hazard warnings are proposed based on the decision space and target space maps. This study provides theoretical guidance for the early warning and forecasting of debris flows and has strong applicability.

A PSO-based algorithm for mining association rules using a guided exploration strategy

Association rule mining is one of the most important and active research areas in data mining. In the literature, several association rule miners have been proposed; among them, those based on particle swarm optimization (PSO) have reported the best results. However, these algorithms tend to prematurely fall into local solutions, avoiding a wide exploration that could produce even better results. In this paper, an algorithm based on PSO, called PSO-GES, for mining association rules using a Guided Exploration Strategy is introduced. Our experiments, over real-world transactional databases, show that our proposed algorithm mines better quality association rules than the most recent PSO-based algorithms for mining association rules of the state of the art.

Dynamic Resource Allocation in Non-orthogonal Multiple Access Using Weighted Maximin Fairness Strategy for a UAV Network

Non-orthogonal multiple access (NOMA) with power domain multiplexing and successive interference cancellation (SIC) is one of the promising technologies for future wireless communication. The performance of NOMA is highly dependent on resource allocation such as power allocation and channel assignment. In this paper, we investigate the power allocation (PA) scheme, to optimize the weighted maximin fairness (MMF) for 2-user and 3-user clusters. We utilize the particle swarm optimization (PSO) based algorithm for power allocation due to its promising behavior. Application area of NOMA is becoming broader, then, we considered a cellular network, assisted by an unmanned aerial vehicle (UAV) as the base station (BS) which is integrated with the NOMA system. The PA for weighted MMF problem in NOMA is non-convex, it is difficult to find out the optimal solution directly. Simulation results show the performance of PSO-based algorithm in different adaptive weights and its convergence characteristics. We have also shown that the rate and fairness tradeoff using weighted maximin fairness. Numerical results compare the performance of NOMA and orthogonal multiple access (OMA) and prove the significance of the proposed algorithm.

A stochastic particle swarm based model for long term production planning of open pit mines considering the geological uncertainty

Long term production planning (LTPP) plays a critical role to achieve success of a mining operation. LTPP, as an optimization problem, aims to maximize the net present value (NPV) of a mine subject to a set of constraints. One of the main reasons for not achieving production targets is the uncertainty of the LTPP's inputs. Geological uncertainty as the main sources of uncertainty is considered in this research. In this regard, a set of equiprobable scenarios of orebody and derived two new block model called “risk block model” and “EType” were used as inputs. Then, a stochastic integer programming (SIP) model was developed to integrate the geological uncertainty. Finally, a PSO-based algorithm was developed to solve the SIP model. Four different strategies were developed, according to the population topology and how to use the risk block model. Population topology defines the subset of particles that effect on each particle. Implementation the proposed approach on a large scale mine demonstrate its performance to develop a unique schedule considering geological uncertainties with maximum NPV and minimum risk of not achieving production targets. Investigations show that Gbest based PSO is more susceptible to trap in local optima. Multiple risk based strategies are able to generate better solutions, however, single risk based strategies are good practices when companies are looking for flexible or specific risk based designs.

Particle swarm optimization algorithms with selective differential evolution for AUV path planning

Particle swarm optimization (PSO)-based algorithms are suitable for path planning of the Autonomous Underwater Vehicle (AUV) due to their high computational efficiency. However, such algorithms may produce sub-optimal paths or require higher computational load to produce an optimal path. This paper proposed a new approach that improves the ability of PSO-based algorithms to search for the optimal path while maintaining a low computational requirement. By hybridizing with differential evolution (DE), the proposed algorithms carry out the DE operator selectively to improve the search ability. The algorithms were applied in an offline AUV path planner to generate a near-optimal path that safely guides the AUV through an environment with a priori known obstacles and time-invariant non-uniform currents. The algorithm performances were benchmarked against other algorithms in an offline path planner because if the proposed algorithms can provide better computational efficiency to demonstrate the minimum capability of a path planner, then they will outperform the tested algorithms in a realistic scenario. Through Monte Carlo simulations and Kruskal-Wallis test, SDEAPSO (selective DE-hybridized PSO with adaptive factor) and SDEQPSO (selective DE-hybridized Quantum-behaved PSO) were found to be capable of generating feasible AUV path with higher efficiency than other algorithms tested, as indicated by their lower computational requirement and excellent path quality.

Assignment of duplicate storage locations in distribution centres to minimise walking distance in order picking

With the rapid development of e-commerce, the orders processed in B2C warehouses are characterised by heterogeneous and small volume. The traditional storage assignment strategies used in the picker-to-parts warehouses do not have advantage any more. In this case, the scattered storage strategy is a good alternative. In this paper, we study a new scattered storage strategy that allows the same product to be placed in multiple storage locations. The correlation between products which reflects how frequently any two products will be ordered together in the same order is considered. The problem is formulated as a 0-1 integer programming model to minimise the weighted sum of distances between the products, with weight being the elements of the correlation matrix. To solve large-scale problems, a GA and a basic PSO algorithm are developed. To improve solution quality, a new PSO algorithm based on the problem characteristic is designed and a hybrid algorithm combing it with GA is proposed. Experiments show that the solutions of these algorithms are close to the optimal solutions for the small-sale problems. For larger problems, the specially designed new PSO greatly improves solution quality as compared to the basic algorithms and the hybrid algorithm makes further improvement.

Pemodelan Prediksi Predikat Kelulusan Mahasiswa Menggunakan Fuzzy C-Means Berbasis Particle Swarm Optimization

Analysis and excavation of information on educational data is an inseparable part of organizing student academic systems in universities. The excavation of information is known as Educational Data mining (EDM), which is a discipline that focuses on the application of techniques and data mining devices specifically in the field of education. One of the EDM processes is related to the need for prediction especially to gain knowledge regarding the progress of student studies. Practically the activity of getting information / knowledge about the progress of student studies is difficult to do conventionally considering the size of the data volume is quite large. The approach to data mining clustering on student study progress data for prediction of predicate graduation tends to be not optimal. Fuzzy C-means (FCM) modeling optimized with Particle Swarm Optimization (PSO) can produce a more optimal prediction performance. In the case of prediction of student graduation prediction, the application of PSO-based FCM algorithm model produces more optimal results, with predictive accuracy of 86% while the FCM algorithm modeling is 79%.