Fast grouping fusion method of dual carbon monitoring data based on DBSCAN clustering algorithm

•Introducing advanced data fusion methods for dual-carbon monitoring.•Improving clustering accuracy and reducing misclassification rates.•Characterizing carbon monitoring data to improve control performance. Due to the serious redundant interference of dual-carbon monitoring data resources, threshol...

Full description

Saved in:
Bibliographic Details
Published in:Results in engineering Vol. 26; p. 105057
Main Authors: Ma, Rui, Sha, Jiangbo, Zhang, Shuang, Zhu, Dongge, Kang, Wenni, Liu, Jia
Format: Journal Article
Language:English
Published: Elsevier B.V 01.06.2025
Elsevier
Subjects:
Abnormal data
DBSCAN clustering algorithm
Dual carbon monitoring data
Packet fusion
Support function
Weighting coefficient
Support function
Abnormal data
Packet fusion
Dual carbon monitoring data
DBSCAN clustering algorithm
Weighting coefficient
ISSN:2590-1230, 2590-1230
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Introducing advanced data fusion methods for dual-carbon monitoring.•Improving clustering accuracy and reducing misclassification rates.•Characterizing carbon monitoring data to improve control performance. Due to the serious redundant interference of dual-carbon monitoring data resources, threshold setting in the fusion process is inaccurate. To realize the sharing and utilization of monitoring data resources, a fast group fusion method of dual-carbon monitoring data based on DBSCAN (Ensity - Based Spatial Clustering of Applications with Noise) clustering algorithm is proposed. After collecting dual-carbon monitoring data through the wireless sensor network, DBSCAN clustering algorithm calculates the neighborhood distance threshold of each data object. Sensor nodes compare the dual-carbon monitoring data and the abnormal data, determine the threshold value, assess whether abnormal data exists and is valid through integrated support, rejecting if invalid. Least squares method realizes fusion of normal data monitored by sensor nodes in the group. Cluster head node determines weight of the normal dual-carbon data monitored by each sensor according to different confidence levels. Variance estimation learning algorithm achieves weighted fusion of data from sensors within the group. The experimental results show that the clustering algorithm has a good clustering effect on the dual-carbon monitoring data and can realize the rapid group fusion. In the complex environment with frequent invalid abnormal data, the mean square error of the fusion results is small.
ISSN:2590-1230
2590-1230
DOI:10.1016/j.rineng.2025.105057