Elevator Door Installation Error Detection Using Sensor Fusion AI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing elevator installation monitoring systems rely on indirect sensor data, which are insufficient to enhance installation quality, leading to high costs and time consumption due to faulty installations, particularly with elevator doors, affecting customer satisfaction and necessitating frequent callbacks.
Innovation Solution
A computer-implemented method using a combination of Set Function and Fourier-Transform models trained with data from sensors directly installed on elevators to detect and classify installation errors, particularly in elevator doors, by extracting features from time series, static, and audio data to improve detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If indirect sensor data are used for monitoring, then the system complexity is reduced, but the installation quality detection precision deteriorates
Solution Approach 1:
The patent segments the monitoring system into multiple independent sensor units distributed at critical installation points (door operator, rails, motor, etc.). Each sensor captures local physical parameters directly related to installation quality, enabling precise detection without requiring a single complex centralized system. This segmentation allows the system to maintain low overall complexity while achieving high detection precision through distributed measurement points.
Solution Approach 2:
The patent introduces direct physical sensors as intermediaries between the installation structure and the monitoring system. Instead of using indirect data processing, sensors directly measure physical parameters (position, temperature, vibration, force) at critical points, serving as intermediaries that transfer installation quality information directly to the monitoring system. This eliminates the need for complex indirect inference while maintaining system simplicity.
2Measurement precision
If more sensors are installed to improve detection accuracy, then the installation quality monitoring improves, but the device complexity and cost increase
Solution Approach 1:
The patent applies local quality by placing sensors only at specific critical installation points where installation errors most frequently occur (door operator, rails, motor, etc.). Each sensor is strategically positioned to detect local physical parameters relevant to its specific location, rather than using a uniform dense sensor distribution. This approach achieves high monitoring accuracy while minimizing the total number of sensors and system complexity.
Solution Approach 2:
The patent monitors changes in physical parameters (position, temperature, vibration frequency, force) over time to detect installation errors. By tracking parameter variations rather than requiring absolute precision at all times, the system achieves effective error detection with fewer sensors. The machine learning model analyzes parameter change patterns to identify installation quality issues, reducing the need for excessive sensor deployment.
3Ease of manufacture
If traditional monitoring systems are used, then the implementation cost is low, but the number of callbacks due to poor installation quality remains high
Solution Approach 1:
The patent enables the installation monitoring system to automatically detect, classify, and report installation errors without requiring manual inspection. The machine learning model processes sensor data in real-time to identify installation quality issues, generating automatic alerts and classification reports. This self-service capability ensures reliable installation quality detection while keeping implementation costs low by eliminating the need for expensive manual verification processes.
Solution Approach 2:
The patent implements a feedback mechanism where sensor data continuously monitors installation parameters and feeds information to the machine learning model, which then provides feedback about installation quality status. This closed-loop feedback system automatically detects deviations from proper installation and alerts operators, ensuring high installation reliability without requiring expensive manual inspection protocols. The feedback enables real-time correction of installation errors before they cause callbacks.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances installation quality, reduces installation time and costs, and enables efficient maintenance scheduling by directly addressing installation errors, thereby improving customer satisfaction and reducing knock-on effects.
Implementation Method 1
the machine learning model being a combination of a Set Function model and a Fourier-Transform model
Data Source
AI summary
A computer-implemented method for training a machine learning model to detect installation errors in an elevator, in particular an elevator door. The machine learning model being a combination of a Set Function model and a Fourier-Transform model. The method including arranging a plurality of sensors at the elevator and each sensor being configured to detect a physical parameter. The method also includes detecting values of the physical parameters by the sensors so as to obtain a dataset comprising at least one time series and obtaining a first input layer by extracting features from the dataset. The method also includes obtaining a second input layer by extracting features from the dataset, feeding the Set Function model with the first input layer, and feeding the Fourier-Transform model with the second input layer.