Performance of improved Gaussian Extreme Learning Machine for water pipeline leak recognition

• Published in: IEEE sensors journal Vol. 24; no. 6; p. 1
• Main Authors: Liu, Mingyang, Guo, Guancheng, Xu, Yuexia, Yang, Yang, Liu, Ning
• Format: Journal Article
• Language: English
• Published: New York IEEE 15.03.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Algorithms; Artificial neural networks; Classification; Deep learning; ELM; Extreme learning machines; GELM; GMM; Leak detection; Leakage; Leakage recognition; Machine learning; Machine learning algorithms; Pipelines; Probabilistic models; Recognition; Sensors; Vibrations; Water conservation; Water distribution; Water distribution networks; Water engineering; Water pipelines
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2024.3360185

At present, many machine or deep learning algorithms have been applied to water distribution networks (WDNs) leakage recognition. The higher the algorithm classification accuracy, the higher the leakage accidents recognition accuracy for WDNs. This is of great significance for the green water-saving systems construction. As a popular machine learning algorithm, the Extreme Learning Machine (ELM) is also applied in WDNs leakage recognition due to its fast learning speed and high accuracy. However, the traditional ELM obtains the classification model output weights based on the mapping calculation of the randomly initialized input weights. Due to the random initialization action, the traditional ELM input weights do not consider the statistical characteristics of the leak samples data, which can easily lead to low accuracy or classification errors phenomena. In this paper, we will use the Gaussian Mixture Model (GMM) framework to evaluate the leak samples data, and the new input weights will be generated by GMM evaluation. The new input weights can prompt the ELM to avoid classification error and low accuracy drawbacks. We have named this improved ELM algorithm as Gaussian Extreme Learning Machine (GELM) algorithm. And we also compared GELM with traditional ELM and other machine learning algorithms, the classification accuracy of GELM was significantly higher.