Scholarlink

<p>With the rapid pace of urbanization and the rise of smart cities in vertical growth, elevators are becoming more important and more critical in their safety and reliability, making continuous operation and passenger safety imperative. Traditional elevator maintenance approaches are largely reactive, often relying on periodic inspections or failures, which can lead to unpredictable failures, longer downtime, and even safety risks because there are no real-time fault detection and predictive mechanisms in place. This paper proposes a new model, AIDE-Lift (Adaptive Intelligent Diagnostic and Early-warning System for Lifts) that combines an IoT-based multi-sensor data acquisition and a hybrid deep learning model with Temporal Convolutional Networks (TCN) and Attention-based Bidirectional Long Short-Term Memory (BiLSTM). The system performs real-time preprocessing on edge computing and predictive analytics and anomaly classification on a cloud platform. The model updates itself to the operational patterns to detect fault signs in advance and issue timely warnings. Experimental analysis shows that AIDE-Lift reaches a high fault detection rate of 97.3%, minimizes false alarms, and detects potential failures much earlier than traditional approaches. The proposed system enhances elevator safety, reduces downtime, and enables predictive maintenance, moving towards smarter and more reliable building management systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Design of elevator fault monitoring and early warning system based on Internet of Things and deep learning

  • 期刊论文

  • 发表日期：

    2026年8月1日

查看全文
  • Chen Niu

摘要

With the rapid pace of urbanization and the rise of smart cities in vertical growth, elevators are becoming more important and more critical in their safety and reliability, making continuous operation and passenger safety imperative. Traditional elevator maintenance approaches are largely reactive, often relying on periodic inspections or failures, which can lead to unpredictable failures, longer downtime, and even safety risks because there are no real-time fault detection and predictive mechanisms in place. This paper proposes a new model, AIDE-Lift (Adaptive Intelligent Diagnostic and Early-warning System for Lifts) that combines an IoT-based multi-sensor data acquisition and a hybrid deep learning model with Temporal Convolutional Networks (TCN) and Attention-based Bidirectional Long Short-Term Memory (BiLSTM). The system performs real-time preprocessing on edge computing and predictive analytics and anomaly classification on a cloud platform. The model updates itself to the operational patterns to detect fault signs in advance and issue timely warnings. Experimental analysis shows that AIDE-Lift reaches a high fault detection rate of 97.3%, minimizes false alarms, and detects potential failures much earlier than traditional approaches. The proposed system enhances elevator safety, reduces downtime, and enables predictive maintenance, moving towards smarter and more reliable building management systems.