Improved Binary Artificial Fish Swarm Algorithm and Fast Constraint Processing for Large Scale Unit Commitment

Zhu, Yongli; Gao, Hui

doi:10.1109/access.2020.3015585

Cited by 20 publications

(13 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generally, max pooling is exploited as pooling in CNN and it is shown in eq. (10). , correspondingly.…”

Section: Cnn Modelmentioning

confidence: 99%

“…The proposed Improved Binary Artificial Fish Swarm Algorithm (IBASFS) optimization model follows top-down optimization which enthuses clustering, searching, chasing, as well as arbitrary behavior of fish in nature to attain the global optimum [10]. It chooses searching behavior by evaluating the consistency of food and the congestion factor within visual distance.…”

Section: Proposed Optimization Modelmentioning

confidence: 99%

See 1 more Smart Citation

Improved Binary Artificial Fish Swarm Algorithm for Diagnosis of Thyroid Disease

2022

View full text Add to dashboard Cite

The thyroid gland generates the thyroid hormones, which help the appropriate body's metabolism regulation. Generally, the abnormalities of thyroid function are categorized into two classifications such as small thyroid hormone production called hypothyroidism or large thyroid hormone production named hyperthyroidism. The thyroid diagnosis using suitable thyroid data interpretation is considered a very important classification problem. Till now, few contributions are performed in the automatic diagnosis of thyroid disease. The main aim of this research is to work on a novel diagnosis of thyroid approach, which follows a new working model. Here, it dealt with the integration of both the thyroid data and image. Moreover, two important working procedures are carried out such as classification as well as feature extraction. In the initial stage, two feature sorts are extracted. Image features such as gradient as well as neighborhood-based features are extracted and the principle component analysis (PCA) process is included for the feature extraction from thyroid data. Then, the process of classification is carried out in two segments. Here, the Convolutional Neural Network (CNN) is exploited to obtain the classified outcomes bypassing the image itself. At the same time, in order to obtain the data features and image as the input, the Neural Network (NN) is exploited for the classification procedure. At last, both the classified outcomes such as NN as well as CNN are integrated to raise the precise diagnosis rate. Furthermore, the main contribution of this paper is to raise the rate of accuracy, therefore this work attempts to activate the optimization model. In CNN, the convolution layer is optimally chosen and when classifying in the NN model the subjected feature must be optimal. Therefore, the necessary features are chosen optimally. For this purpose, a novel enhanced approach called Improved Binary Artificial Fish Swarm Algorithm (IBASFS) is the enhancement model of Fish Swarm Algorithm is presented in this paper. At last, the adopted model performances are evaluated with the existing models and it reveals that the enhancement of the adopted model to detect the thyroid.

show abstract

“…Generally, max pooling is exploited as pooling in CNN and it is shown in eq. (10). , correspondingly.…”

Section: Cnn Modelmentioning

confidence: 99%

Section: Proposed Optimization Modelmentioning

confidence: 99%

Improved Binary Artificial Fish Swarm Algorithm for Diagnosis of Thyroid Disease

2022

View full text Add to dashboard Cite

show abstract

“…Xie et al [19] combined the improved particle swarm algorithm with the artificial fish swarm algorithm, using the local convergence of the particle swarm optimization algorithm and the global convergence of the artificial fish swarm algorithm to improve the convergence speed and accuracy of the hybrid algorithm, so that the path The planning is optimal, but the model of the algorithm to solve the problem is too simple, and the algorithm does not have sufficient validity during verification. Zhu et al [20] proposed an improved Binary Artificial Fish Swarm Algorithm (IBAFSA), which used a dual-threshold selection strategy to enhance the effectiveness of population evolution, and proposed heuristic greed among the best individuals in each generation during the optimization iteration process. The search algorithm improves the computational convergence.…”

Section: Artificial Fish Swarm Algorithmmentioning

confidence: 99%

Research Status and Development Trend of UAV Path Planning Algorithms

Zeng¹,

Liu²

2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

This paper introduces the basic concepts of UAV track planning, the relationship between track planning and algorithms, and focuses on the analysis and induction of several commonly used algorithms at home and abroad in recent years. The advantages and disadvantages of current path planning algorithms in different application scenarios are summarized. Finally, the future development trend of UAV path planning algorithm is prospected, and it is pointed out that real-time online path planning algorithm research, multi-UAV swarm planning algorithm, perfect path planning problem model and algorithm hybrid optimization are the next research trends.

show abstract

“…But the issue of DR and cyber security is not considered. Also, the proposed method by Zhu and Gao (2020) is a centralized fusion, so it is expensive and highly vulnerable. A summary of the literature review is presented in Table 1.…”

Section: Introductionmentioning

confidence: 99%

Adaptive distributed scheduling of the resources in smart grids with considering demand response

Farkhad

Foroud

2023

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

Unit commitment (UC), a non-linear and non-convex problem, is one of the important problems in power system operation. Solving the UC problem centrally has problems such as security concerns. On the contrary, solving it in a fully distributed manner reduces performance speed. This paper presents a method for solving the UC problem that addresses both security concerns and speeds up performance. This paper intends to solve the UC problem by using a hybrid structure and with the help of accelerator and cyber security parameters of the power system. This problem is solved for the first time in a power system despite its non-linear nature and non-convexity. Also, the demand response (DR) issue is considered as a contributing factor in resource scheduling management. This is important because DR is an inherently distributed issue. The presented method uses a developed decentralized algorithm called the accelerated hybrid alternating direction method of multipliers (AHADMM). AHADMM is a combination of the ADMM framework and hybrid network architecture that its parameters are tuned. Then for decentralized optimization, the generalized Benders decomposition (GBD) algorithm and the AHADMM algorithm are utilized to speed up the computation and provide a framework to exploit the distributed nature of the problem. The proposed algorithm consists of two main problems, primal and master problems. The primal problem, which is solved in a distributed way, is responsible for providing a feasible solution such that an optimality cut can be generated. In contrast, the master problem is responsible for accumulating the optimality cuts and approximating the feasible region of the original non-linear problem. Regarding the improvement of the solution method, we have used the accelerator parameters and the limitation parameter of fusion centers in such a way as to guarantee the speed of calculations and the cyber security of the power system.

show abstract

Improved Binary Artificial Fish Swarm Algorithm and Fast Constraint Processing for Large Scale Unit Commitment

Cited by 20 publications

References 41 publications

Improved Binary Artificial Fish Swarm Algorithm for Diagnosis of Thyroid Disease

Improved Binary Artificial Fish Swarm Algorithm for Diagnosis of Thyroid Disease

Research Status and Development Trend of UAV Path Planning Algorithms

Adaptive distributed scheduling of the resources in smart grids with considering demand response

Contact Info

Product

Resources

About