In this paper, we propose a three-layered neural network controller (NC) optimized using an improved bat algorithm (BA) for a rotary crane system. In our previous study, the simulation results showed that an NC optimi...
详细信息
In this paper, we propose a three-layered neural network controller (NC) optimized using an improved bat algorithm (BA) for a rotary crane system. In our previous study, the simulation results showed that an NC optimized using the original BA exhibits good control and evolutionary performance. However, the simulation execution time was long. Therefore, to address this problem, we propose an improved BA that reduces the execution time. We show that the NC optimized by the improved BA exhibits the same control performance as that optimized via conventional methods. It is also shown that the time for evolutionary calculations can be reduced.
Deep reinforcement learning (DRL) has achieved notable success in continuous control tasks. However, it faces challenges that limit its applicability to a wider array of tasks, including sparse rewards and limited exp...
详细信息
Deep reinforcement learning (DRL) has achieved notable success in continuous control tasks. However, it faces challenges that limit its applicability to a wider array of tasks, including sparse rewards and limited exploration. Recently, the integration of evolutionaryalgorithms (EAs) with deep reinforcement learning has emerged as a significant area of research. evolutionary reinforcement learning (ERL) methods can help address specific challenges inherent in conventional reinforcement learning algorithms. However, the introduction of evolutionary computation algorithms increases the number of hyperparameters, and sensitivity to these hyperparameters continues to pose a significant challenge. This paper proposes an evolutionary reinforcement learning method incorporating evolutionary mutation rates. This method integrates a self-adaptive mutation rate mechanism into the ERL framework, which maintains two populations: one consisting of individuals (agents) and the other one comprising mutation rates. This represents our original contribution to this research. The actor population is categorized into several groups, each assigned a specific mutation rate. After mutation, the mutation rate of the population evolves based on the performance of the mutations within the actor population. This approach addresses the challenge of selecting mutation rates in ERL. Experimental results demonstrate superior performance compared to the standard ERL framework across six continuous control tasks.
Recently, computer vision tasks such as classification and object detection have been dominated by deep neural network (DNN) approaches. As DNN methodologies have matured, researchers have found that some of the most ...
详细信息
ISBN:
(数字)9781728186719
ISBN:
(纸本)9781728186719
Recently, computer vision tasks such as classification and object detection have been dominated by deep neural network (DNN) approaches. As DNN methodologies have matured, researchers have found that some of the most common DNN techniques result in models that are highly dependent upon the textures and colors of the imagery, rather than the shape, leading to suboptimal network performance. This problem can be especially problematic in the remote sensing domain, where the discrimination of objects for classification or detection may rely heavily on their shape. To combat this lack of shape bias in DNNs, a network was developed to integrate the Differential Morphological Profile (DMP), an image processing technique for shape extraction, with standard convolutional DNNs for performing computer vision tasks on High Resolution Remote Sensing Imagery (HR-RSI). Previously, this network, known as DMPNet, has been applied to both classification and object detection in HR-RSI with high levels of success. However, the hyper-parametric nature of DMPNet structure required researchers to carefully select the parameters of shape extraction, a choice that could greatly help or hinder DMPNet performance. In this study, we utilize a evolutionary computation algorithm (ECA) to learn the parameters of shape extraction from the data presented to the DMPNet for object detection. Our results show that our DMP-enabled detection models perform better object detection in HR-RSI using an ECA to learn shape extraction parameters than manually selected parameters on the same dataset.
As a kind of new emerging optimization technology, distributed evolutionarycomputation (DEC) algorithms have fast developed in recent years. The DEC algorithms, which make use of multiple computers or resources to en...
详细信息
ISBN:
(纸本)9781665412544
As a kind of new emerging optimization technology, distributed evolutionarycomputation (DEC) algorithms have fast developed in recent years. The DEC algorithms, which make use of multiple computers or resources to enhance the optimization capabilities of algorithms, have received widespread attention. Among the DEC algorithms, a cloud-based distributed differential evolution (Cloudde) algorithm has shown excellent performance. The Cloudde has a double-layered heterogeneous distribution structure, which can run different differential evolution (DE) variants with various parameters andJor operators in different populations. Moreover, the Cloudde can adaptively migrate individuals among the populations to make best use of the computational resources among multiple populations. However, since the proposal of the Cloudde, there are still some questions remained to he discussed. The first is how to choose the basic DE algorithms to form various DE variants (i.e., the various populations). The second is how to evaluate the performance of different populations of individuals hence we can rank the populations. The third is how to design an efficient migration strategy to make full use of computing resources among multiple populations. This paper makes investigation on these issues and studies the performance of Cloudde variants with various configurations for these three aspects. The experimental results in this paper are useful for researchers who want to conduct further research on Cloulde and other related DEC algorithms. Moreover, based on the investigation results, an improved Cloudde (1-Cloudde) is proposed and the experimental results show the superiority of 1-Cloudde when compared with Cloudde.
With the wide use of electromagnetic information equipment, a larg number of wireless radiation systems coexisting in the same region produce intentional or unintentional interference on electronic receivers. For the ...
详细信息
With the wide use of electromagnetic information equipment, a larg number of wireless radiation systems coexisting in the same region produce intentional or unintentional interference on electronic receivers. For the purpose of intentional electromagnetic interference, it is necessary to realisethe efficient suppression of other receivers at little cost. When multipletransmitting sources are used to interfere with multiple receivers, the parameters of multiple transmitting sources are required to be comprehensivelyoptimised and set so as to achieve a desired high-efficiency ***, we propose a novel method to optimise the setting of parameters of a multi-source, multi-object and multi-domain (M-SOD) interference system based on intelligent optimisation approaches. Furthermore, this study also builds anintelligent optimisation model, which contains multiple transmitters andreceivers which involved many parameters include position, direction of space domain, frequency, bandwidth, and power. Then the model is abstracted to theproblem of single-objective optimisation with constraints and optimised through a traditional GA and an improved FWA method. The extensive experiments andcomparisons show that the proposed algorithm is an effective approach forsetting the parameters of an M-SOD electromagnetic interference system andsuperior to the conventional method.
A data-mining approach for the optimization of a HVAC (heating, ventilation, and air conditioning) system is presented. A predictive model of the HVAC system is derived by data-mining algorithms, using a dataset colle...
详细信息
A data-mining approach for the optimization of a HVAC (heating, ventilation, and air conditioning) system is presented. A predictive model of the HVAC system is derived by data-mining algorithms, using a dataset collected from an experiment conducted at a research facility. To minimize the energy while maintaining the corresponding IAQ (indoor air quality) within a user-defined range, a multi-objective optimization model is developed. The solutions of this model are set points of the control system derived with an evolutionary computation algorithm. The controllable input variables supply air temperature and supply air duct static pressure set points are generated to reduce the energy use. The results produced by the evolutionary computation algorithm show that the control strategy saves energy by optimizing operations of an HVAC system. (C) 2011 Elsevier Ltd. All rights reserved.
An evolutionarycomputation approach for optimization of power factor and power output of wind turbines is discussed. Data-mining algorithms capture the relationships among the power output, power factor, and controll...
详细信息
An evolutionarycomputation approach for optimization of power factor and power output of wind turbines is discussed. Data-mining algorithms capture the relationships among the power output, power factor, and controllable and non-controllable variables of a 1.5 MW wind turbine. An evolutionary strategy algorithm solves the data-derived optimization model and determines optimal control settings. computational experience has demonstrated opportunities to improve the power factor and the power output by optimizing set points of blade pitch angle and generator torque. It is shown that the pitch angle and the generator torque can be controlled to maximize the energy capture from the wind and enhance the quality of the power produced by the wind turbine with a DFIG generator. These improvements are in the presence of reactive power remedies used in modern wind turbines. The concepts proposed in this paper are illustrated with the data collected at an industrial wind farm. (C) 2009 Elsevier Ltd. All rights reserved.
The paper presents an intelligent wind turbine control system based on models integrating the following three approaches: data mining, model predictive control, and evolutionarycomputation. To enhance the control str...
详细信息
The paper presents an intelligent wind turbine control system based on models integrating the following three approaches: data mining, model predictive control, and evolutionarycomputation. To enhance the control strategy of the intelligent system, a multi-objective model is proposed. The model involves five different objectives with different weights controlling the wind turbine performance. These weights are adjusted in response to the variable wind conditions and operational requirements. Three control factors, wind speed, turbulence intensity, and electricity demand are considered in eight computational scenarios. The performance of each scenario is illustrated with numerical results. Published by Elsevier Ltd.
暂无评论