Growth optimizer is a novel metaheuristic algorithm that has powerful numerical optimization capabilities. However, its parameters and search operators become crucial factors that significantly impact its optimization...
详细信息
Growth optimizer is a novel metaheuristic algorithm that has powerful numerical optimization capabilities. However, its parameters and search operators become crucial factors that significantly impact its optimization capability for engineering problems and benchmarks. Therefore, this paper proposes a quadruple parameter adaptation growth optimizer (QAGO) integrated with distribution, confrontation, and balance features. In QAGO, the quadruple parameter adaptation mechanism aims to reduce the algorithmic sensitivity for parameter setting and enhance the algorithmic adaptability. By employing parameter sampling that adheres to specific probability distributions, the parameter adaptation mechanism achieves dynamic tuning of the algorithm hyperparameters. Moreover, one-dimensional mapping and fitness difference methods are designed in the triple parameter self-adaptation mechanism based on the contradictory relationship to adjust the operator's parameters. After that, "spear" and "shield" are balanced based on the Jensen-Shannon divergence in information theory. Furthermore, the topological structure of the operators is redesigned, and by combining the parameter adaptation mechanism, operator refinement is achieved. Refined operators can effectively utilize different evolutionary information to improve the quality of the solution. The experiment evaluates the performance of QAGO on distinct optimization problems on the CEC 2017 and CEC 2022 test suites. To demonstrate the capability of QAGO in solving real-world applications, it was applied to tackle two specific problems: multilevel threshold image segmentation and wireless sensor network node deployment. The results demonstrated that QAGO delivers highly promising optimization results compared to seventy-one high-performance competing algorithms, including the five IEEE CEC competition winners. The source code of the QAGO algorithm is publicly available at https://***/tsingke/QAGO
To address the issues of the Dwarf Mongoose optimization algorithm (DMOA) being prone to local optima and exhibiting low convergence accuracy during its operation, this paper introduces the Chaotic Dwarf Mongoose Opti...
详细信息
ISBN:
(数字)9789819755783
ISBN:
(纸本)9789819755776;9789819755783
To address the issues of the Dwarf Mongoose optimization algorithm (DMOA) being prone to local optima and exhibiting low convergence accuracy during its operation, this paper introduces the Chaotic Dwarf Mongoose optimization algorithm (CDMOA). The CDMO algorithm employs a chaos mapping strategy to ensure a uniform distribution of the initial population across the solution space, thereby enhancing population diversity. Additionally, it utilizes an inverse learning strategy to bolster the global search capabilities of the algorithm. Comparative experiments conducted using benchmark test functions demonstrate that CDMOA outperforms the original DMO algorithm in terms of optimization performance, convergence accuracy, and algorithm stability. Finally, the application of CDMOA to drone flight path planning is presented. The simulation results indicate that the optimized flight paths generated by the improved algorithm are superior and more stable.
A new design of robust variable horizon model predictive control (VH-MPC) with move blocking is proposed for helicopter shipboard landings on moving decks in rough seas. The design introduces an efficient strategy to ...
详细信息
ISBN:
(数字)9781624107115
ISBN:
(纸本)9781624107115
A new design of robust variable horizon model predictive control (VH-MPC) with move blocking is proposed for helicopter shipboard landings on moving decks in rough seas. The design introduces an efficient strategy to implement the VH-MPC by evaluating multiple controllers at each iteration, corresponding to different horizon lengths, and selecting the one with the lowest cost. This approach avoids the computational burden of solving a batch of mixed integer quadratic programming problem for executing VH-MPC. Additionally, move blocking is introduced to reduce the number of decision variables in the MPC optimization problem. The new VH-MPC can be executed in parallel computation, which means the controller latency is determined by the slowest single optimization problem's solution time rather than the sum of all solution times. To evaluate the design, a nonlinear helicopter-ship dynamics interface is utilized, incorporating significant helicopter and ship dynamics, as well as ship airwake interactions.
Federated Learning (FL) algorithms commonly sample a random subset of clients to address the straggler issue and improve communication efficiency. While recent works have proposed various client sampling methods, they...
详细信息
ISBN:
(纸本)9781728190549
Federated Learning (FL) algorithms commonly sample a random subset of clients to address the straggler issue and improve communication efficiency. While recent works have proposed various client sampling methods, they have limitations in joint system and data heterogeneity design, which may not align with practical heterogeneous wireless networks. In this work, we advocate a new independent client sampling strategy to minimize the wall-clock training time of FL, while considering data heterogeneity and system heterogeneity in both communication and computation. We first derive a new convergence bound for non-convex loss functions with independent client sampling and then propose an adaptive bandwidth allocation scheme. Furthermore, we propose an efficient independent client sampling algorithm based on the upper bounds on the convergence rounds and the expected per-round training time, to minimize the wall-clock time of FL, while considering both the data and system heterogeneity. Experimental results under practical wireless network settings with real-world prototype demonstrate that the proposed independent sampling scheme substantially outperforms the current best sampling schemes under various training models and datasets.
In the process of construction and operation of the power grid, power load forecasting is a very important part, according to the forecast results, can be better grid scheduling and security monitoring and other impor...
详细信息
ISBN:
(纸本)9798350390780;9798350379228
In the process of construction and operation of the power grid, power load forecasting is a very important part, according to the forecast results, can be better grid scheduling and security monitoring and other important work, so it is very necessary to do the forecast as much as possible to improve the prediction accuracy, according to the selection of the data in the temperature, humidity, date type and load data analysis, the use of optimization of the hyperparameters of the LSTM neural network through the grey wolf optimization algorithm, and finally the use of the optimized model for prediction.
This article proposes a dot-shaped transmit beamforming algorithm by using interior point (IP) algorithm to design optimized frequency offset (FO). We compared the beam focusing performance of this scheme with linear ...
详细信息
ISBN:
(纸本)9798350353129;9798350353136
This article proposes a dot-shaped transmit beamforming algorithm by using interior point (IP) algorithm to design optimized frequency offset (FO). We compared the beam focusing performance of this scheme with linear FO, logarithmic (Log) FO, and genetic algorithm (GA) FO schemes, and provided the convergence process of the algorithm. The simulation results show that the algorithm converges quickly while ensuring good beam focusing effect. We extended it to two-dimensional planar array and further reduced the computational complexity of the algorithm by applying symmetric frequency offset on the array. The effectiveness of the algorithm was verified through numerical simulation
To address the increased energy demands and carbon emissions caused by global urbanization, it is imperative to seek high-performance urban design solutions. Urban form generation and optimization (UFGO) is a powerful...
详细信息
To address the increased energy demands and carbon emissions caused by global urbanization, it is imperative to seek high-performance urban design solutions. Urban form generation and optimization (UFGO) is a powerful way of supporting performance-driven urban design by strategically searching for a possible design space to approach optimal solutions. Relevant areas of urban form generative design, urban energy and environment simulation, and urban form optimization have been widely studied. However, UFGO, which integrates these parts into an effective workflow, is still an emerging and meaningful research field lacking a systematic review. We examined studies that utilized UFGO techniques for urban design at different scales and outlined the general workflow. An overview of the available methods and tools, as well as their basic principles for each step, namely, urban form generation, performance simulation, and optimization, is provided. The reader will be well versed in the key problems and technical paths of UFGO. According to the review, UFGO is technically feasible;nevertheless, existing limitations necessitate further exploration. Future studies should focus on developing userfriendly UFGO software packages for urban designers, systematic and flexible generative design methods, and efficient data-driven models for urban performance evaluation. In addition, an evaluation system for UFGO techniques is also required to facilitate comparative studies and the widespread application of UFGO techniques.
Verification for a complex engineering system is essential for ensuring confidence in the system's performance, but verification must often be carried out under rigid time and cost constraints. Verification planni...
详细信息
ISBN:
(数字)9781624107115
ISBN:
(纸本)9781624107115
Verification for a complex engineering system is essential for ensuring confidence in the system's performance, but verification must often be carried out under rigid time and cost constraints. Verification planning is the problem of deciding how to allocate resources in this process, and verification for some projects may use more time and money than is necessary to achieve a desired level of certainty. This paper proposes a systems engineering methodology for designing optimal verification plans to enable systems engineers to systematically plan verification activities. This methodology uses optimal Bayesian experimental design (OBED) to determine sets of activities that minimize effort and maximize certainty of system performance. The development of an OBED approach to verification planning is presented and its usefulness is demonstrated on a representative problem, planning verification activities for ensuring performance of a CCD.
Intensity-Modulated Radiation Therapy (IMRT) has gained prominence in cancer treatment due to its ability to deliver precise radiation doses to tumor regions while sparing adjacent healthy tissues. Optimizing the trea...
详细信息
ISBN:
(纸本)9783031671913;9783031671920
Intensity-Modulated Radiation Therapy (IMRT) has gained prominence in cancer treatment due to its ability to deliver precise radiation doses to tumor regions while sparing adjacent healthy tissues. Optimizing the treatment planning process in IMRT involves addressing complex, conflicting objectives and uncertainties inherent in medical decision-making. This paper proposes a novel Neutrosophic Fuzzy Hybrid Method (NFHM) integrated with metaheuristic algorithms to enhance the efficiency and robustness of IMRT treatment planning. The Neutrosophic Fuzzy Hybrid Method combines neutrosophic logic, fuzzy logic, and metaheuristic optimization techniques to model and handle uncertainties, imprecisions, and conflicting information present in the IMRT treatment planning domain. Neutrosophic logic provides a valuable framework for representing indeterminacy, while fuzzy logic aids in capturing vagueness in medical data. To optimize IMRT treatment plans, the proposed method incorporates metaheuristic algorithms such as Genetic algorithms (GA), Particle Swarm optimization (PSO), and Simulated Annealing (SA). These algorithms explore the solution space efficiently, seeking optimal trade-offs among conflicting treatment objectives, including target coverage, organ-at-risk sparing, and dose uniformity. The performance of the Neutrosophic Fuzzy Hybrid Method is evaluated using benchmark datasets and compared against traditional optimization approaches. The results demonstrate the method's superiority in generating high-quality IMRT plans that exhibit improved dose distribution and conformity while considering uncertainties and imprecisions. In conclusion, the integration of neutrosophic and fuzzy reasoning with metaheuristic optimization algorithms in the proposed NFHM presents a promising approach for enhancing the IMRT treatment planning process. The method's ability to handle uncertainties and find optimal solutions makes it a valuable tool for improving the efficacy and precision
In order to facilitate the high-quality advancement and digital innovation of the wine industry, a method based on principal component analysis (PCA) and improved quantum particle swarm optimization (IQPSO) to optimiz...
详细信息
In order to facilitate the high-quality advancement and digital innovation of the wine industry, a method based on principal component analysis (PCA) and improved quantum particle swarm optimization (IQPSO) to optimize support vector machine (PCA-IQPSO-SVM), was proposed to solve the wine classification problem. First, the feature extraction ability of PCA was used to reduce the input dimension of the model and improve the classification efficiency. At the same time, aiming at the problems that the quantum particle swarm optimization (QPSO) is easy to fall into local optimum and the convergence ability is decreased in the later stage of optimizing SVM, a variety of improvement strategies are used to improve QPSO to find the best parameters of SVM. The experimental results demonstrate that the model of PCA-IQPSO-SVM exhibits superior evaluation indices compared to other models. Moreover, the optimization efficiency of the PCA-IQPSO-SVM model is enhanced by 1.64% to reach an impressive 85.2%, showcasing its remarkable optimization effect. Simultaneously, this study provides a scientific approach for quality classification in the wine industry, thereby facilitating its high-quality development.
暂无评论