A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Identifying daily water consumption patterns based on K-means Clustering, Agglomerative Hierarchical Clustering, and Spectral Clustering algorithms
Understanding daily water consumption patterns is crucial for efficient management and distribution of water resources, as well as for promoting energy conservation and achieving carbon peaking and neutrality targets. It compares performance of three clustering algorithms, K-means Clustering (KC), Agglomerative Hierarchical Clustering (AHC), and Spectral Clustering (SC), using Silhouette Coefficient Index (SCI) and Calinski–Harabasz Index (CHI) as evaluation metrics. We conducted a case study using original hourly flow series of a water distribution division. It aims to identify typical daily water consumption patterns and explore factors that influence them. Findings are as follows: (1) among the three algorithms, KC demonstrates the best, with SCI of 0.6315, 0.5922, and 0.6272, and CHI of 305.9207, 274.1120, and 302.4738 for KC, AHC, and SC, respectively. (2) KC successfully identifies three distinct typical daily water consumption patterns. (3) Results indicate a significant impact of seasons on daily water consumption patterns. (4) Conversely, weekdays and holidays have minimal effect on daily water consumption patterns. It highlights the importance of comprehending daily water consumption patterns and underscores the effectiveness of KC in identifying such patterns. Furthermore, it emphasizes the significant influence of seasons while revealing limited impact of weekdays and holidays on daily water consumption patterns. HIGHLIGHTS K-means Clustering performs the best among the three clustering algorithms.; Three typical daily water consumption patterns were identified in the case study.; Season was found to be a significant influencing factor for the three patterns.;
Identifying daily water consumption patterns based on K-means Clustering, Agglomerative Hierarchical Clustering, and Spectral Clustering algorithms
Understanding daily water consumption patterns is crucial for efficient management and distribution of water resources, as well as for promoting energy conservation and achieving carbon peaking and neutrality targets. It compares performance of three clustering algorithms, K-means Clustering (KC), Agglomerative Hierarchical Clustering (AHC), and Spectral Clustering (SC), using Silhouette Coefficient Index (SCI) and Calinski–Harabasz Index (CHI) as evaluation metrics. We conducted a case study using original hourly flow series of a water distribution division. It aims to identify typical daily water consumption patterns and explore factors that influence them. Findings are as follows: (1) among the three algorithms, KC demonstrates the best, with SCI of 0.6315, 0.5922, and 0.6272, and CHI of 305.9207, 274.1120, and 302.4738 for KC, AHC, and SC, respectively. (2) KC successfully identifies three distinct typical daily water consumption patterns. (3) Results indicate a significant impact of seasons on daily water consumption patterns. (4) Conversely, weekdays and holidays have minimal effect on daily water consumption patterns. It highlights the importance of comprehending daily water consumption patterns and underscores the effectiveness of KC in identifying such patterns. Furthermore, it emphasizes the significant influence of seasons while revealing limited impact of weekdays and holidays on daily water consumption patterns. HIGHLIGHTS K-means Clustering performs the best among the three clustering algorithms.; Three typical daily water consumption patterns were identified in the case study.; Season was found to be a significant influencing factor for the three patterns.;
Identifying daily water consumption patterns based on K-means Clustering, Agglomerative Hierarchical Clustering, and Spectral Clustering algorithms
Hongyuan Guo (author) / Xingpo Liu (author) / Qichen Zhang (author)
2024
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
An agglomerative hierarchical clustering tool for renewable energy sources
| BASE | 2011
Characterization of Vertical Wind Speed Profiles Based on Ward's Agglomerative Clustering Algorithm
| DOAJ | 2023
| DOAJ | 2022