Multiobjective optimization of microgrid based on SOM clustering and Markov chain

Huang, Junshuo; Chen, Jian; Fang, Juhao; Hou, Shiliang

doi:10.1088/1755-1315/512/1/012130

Cited by 1 publication

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This algorithm still has the problem of not being able to balance the anti-noise ability and detail retention, and there is still room for improvement in the anti-noise ability. On the basis of this algorithm, aiming at the problem that the existing fuzzy clustering algorithm [22][23][24][25][26][27]is not perfect enough, this paper further improves the existing clustering algorithm on the basis of the existing clustering algorithm, in order to enhance the original algorithm's anti-noise robustness and segmentation accuracy. Based on the FLICMLNLI algorithm, this paper introduces the Gaussian kernel distance metric and uses iterative self-learning to assign the weight of the current pixel and its neighborhood information to solve the problems of the FLICMLNLI algorithm to further improve the robustness of the algorithm.…”

Section: Introductionmentioning

confidence: 99%

Kernel-Based Fuzzy Local Information Clustering Algorithm Self-integrating Non-Local Information

Song¹,

Wu²,

Tian³

et al. 2021

Preprint

View full text Add to dashboard Cite

The application of fuzzy clustering algorithms in image segmentation is a hot research topic nowadays. Existing fuzzy clustering algorithms have the following three problems: (1)The parameters of spatial information constraints can$'$t be selected adaptively; (2)The image corrupted by high noise can$'$t be segmented effectively; (3)It is difficult to achieve a balance between noise removal and detail preservation. In the fuzzy clustering based on the optimization model, the choice of distance metric is very important. Since the use of Euclidean distance will lead to sensitivity to outliers and noise, it is difficult to obtain satisfactory segmentation results, which will affect the clustering performance. This paper proposes an optimization algorithm based on the kernel-based fuzzy local information clustering integrating non-local information (KFLNLI). The algorithm adopts a self-integration method to introduce local and non-local information of images, which solves the common problems of current clustering algorithm. Firstly, the self-integration method solves the problem of selecting spatial constraint parameters. The algorithm uses continuous self-learning iteration to calculate the weight coefficients; Secondly, the distance metric uses Gaussian kernel function to induce the distance to further enhance the robustness against noise and the adaptivity of processing different images; Finally, both local and non-local information are introduced to achieve a segmentation effect that can eliminate most of the noise and retain the original details of the image. Experimental results show that the algorithm is superior to existing state-of-the-art fuzzy clustering-related algorithm in the presence of high noise.

show abstract