WO2024119360A1 - 一种基于多源信息融合的电梯安全故障诊断方法 - Google Patents

一种基于多源信息融合的电梯安全故障诊断方法 Download PDF

Info

Publication number
WO2024119360A1
WO2024119360A1 PCT/CN2022/136820 CN2022136820W WO2024119360A1 WO 2024119360 A1 WO2024119360 A1 WO 2024119360A1 CN 2022136820 W CN2022136820 W CN 2022136820W WO 2024119360 A1 WO2024119360 A1 WO 2024119360A1
Authority
WO
WIPO (PCT)
Prior art keywords
elevator
car
abnormal
vibration
traction rope
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2022/136820
Other languages
English (en)
French (fr)
Inventor
张强
马忠君
田莹
顾颉颖
赵广浩
刘峻铭
张赫哲
谢飞
孙常亮
冯辉
王海舰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qingdao Special Equipment Inspection And Research Institute
Shandong University of Science and Technology
Original Assignee
Qingdao Special Equipment Inspection And Research Institute
Shandong University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Qingdao Special Equipment Inspection And Research Institute, Shandong University of Science and Technology filed Critical Qingdao Special Equipment Inspection And Research Institute
Priority to PCT/CN2022/136820 priority Critical patent/WO2024119360A1/zh
Priority to CN202280053511.2A priority patent/CN117916185A/zh
Publication of WO2024119360A1 publication Critical patent/WO2024119360A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
    • B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
    • B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
    • B66B5/02—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
    • B66B7/00—Other common features of elevators
    • B66B7/12—Checking, lubricating, or cleaning means for ropes, cables or guides

Definitions

  • the present invention relates to the technical field of elevator fault diagnosis, and in particular to an elevator safety fault diagnosis method based on multi-source information fusion.
  • elevator fault diagnosis usually targets a single component of the elevator and cannot guarantee the normal operation of multiple components at the same time.
  • Single safety assurance mainly relies on regular maintenance and annual inspections.
  • problems such as failure to fully and timely discover elevator safety hazards and waste of manpower and material resources. Therefore, it is of great significance to conduct safety assessments on elevators based on their online monitoring data and to implement elevator status maintenance.
  • the technical problem to be solved by the present invention is to provide an elevator safety fault diagnosis method based on multi-source information fusion in view of the above-mentioned deficiencies in the prior art, so as to realize the diagnosis of elevator safety faults.
  • an elevator safety fault diagnosis method based on multi-source information fusion comprising the following steps:
  • Step 1 respectively collect vibration signals of the car and the reduction box and visual signals of the car traction rope during the operation of the elevator;
  • the data collection part includes the time domain signals of the elevator car and the reduction box vibration and the image visual signals of the traction rope, wherein the collected time domain signals include the peak value, mean value and pulse index of the time domain signals;
  • Step 2 preprocess the collected vibration signals and visual signals of the car and the reduction box, and extract the vibration signal characteristics of the car and the reduction box and the lateral displacement signal characteristics of the car traction rope respectively;
  • the signal preprocessing part includes the time domain signal preprocessing of the elevator car and reduction box vibration and the image visual signal preprocessing of the traction rope, specifically:
  • the time domain signals of the elevator car and the reduction box vibration are collected and subjected to noise reduction processing by the median filtering method, the processed time domain signals are converted into frequency domain signals by Fourier transform, and the frequency domain signals are subjected to feature extraction by the wavelet analysis method to obtain the vibration signal characteristics of the car and the reduction box;
  • the collected visual signals of the car traction rope image are scanned and denoised, and the current traction rope image information is calibrated using the Zhang Zhengyou calibration method.
  • the target tracking method in computer vision is used to track the light strip area illuminated on the wire rope in the camera lens, and then the lateral vibration displacement of the wire rope is obtained as the lateral displacement signal feature of the car traction rope;
  • the target tracking adopts the CSRT target tracking method, whose input signal is the image signal of the elevator traction rope, and the output result is the image signal after virtual marking;
  • the image signal detection model adopted by the CSRT target tracking method uniformly uses CNN for classification and prediction of the border position;
  • Step 3 Perform abnormal vibration signal detection based on the vibration signal characteristics of the car and the reduction box to obtain abnormal vibration signal detection results, and perform abnormal lateral displacement detection based on the lateral displacement signal characteristics of the car traction rope to obtain abnormal lateral detection results;
  • An abnormal vibration signal detection model is used to perform abnormal vibration signal detection processing on the vibration signals of the elevator car and the reducer, and an abnormal image state detection model is used to perform abnormal image signal detection processing on the image visual signal of the elevator traction rope;
  • the abnormal vibration signal detection model adopts the LightGBM model, whose input is the vibration signal characteristics of the car and the reduction box, and the output is the abnormal vibration signal detection result;
  • the abnormal image detection model adopts an R-CNN model, whose input is the image signal of the elevator car traction rope, and the output is the abnormal image signal detection result, and then the abnormal image signal is calibrated and tracked to obtain the abnormal lateral displacement signal;
  • Step 4 Fusing the abnormal vibration signal detection result and the abnormal lateral displacement signal detection result, and judging whether to generate an emergency stop fault alarm for the elevator based on the fusion result;
  • the abnormal vibration signal detection result and the abnormal lateral displacement signal detection result are fused based on fuzzy information to obtain a diagnosis result, as shown in the following formula:
  • A is the set of diagnosed elevator fault points, namely, three types of faults: severe vibration of elevator car and reducer, deformation of car traction rope exceeding the safety value
  • B is the set of elevator operation status, namely, three elevator operation status signals: vibration signal of car and reducer and lateral displacement signal of traction rope
  • X is the set of credibility values for judging the operation status of different elevator components
  • x 1 , x 2 , x 3 are the credibility values assigned to the operation status of different elevator components based on experience or fault database
  • Y obtained by fuzzy transformation is the possibility of each decision after fusion
  • the decision target to be determined should have the largest membership value
  • the possibility set of each fault inference decision is selected according to a certain weight to obtain the optimal fault inference result.
  • the elevator safety fault diagnosis method based on multi-source information fusion uses fuzzy information fusion theory to realize real-time centralized monitoring of the operating status of multiple components of the elevator, assigns different weights to the fault decisions of different components, and improves the speed and accuracy of elevator fault diagnosis.
  • FIG1 is a flow chart of an elevator safety fault diagnosis method based on multi-source information fusion provided by an embodiment of the present invention
  • FIG. 2 is a schematic diagram of the process of an elevator safety fault diagnosis method based on multi-source information fusion provided by an embodiment of the present invention
  • FIG3 is a flow chart of obtaining a fault diagnosis result based on fuzzy information fusion provided by an embodiment of the present invention.
  • a method for diagnosing elevator safety faults based on multi-source information fusion includes the following steps:
  • Step 1 respectively collect vibration signals of the car and the reduction box and visual signals of the car traction rope during the operation of the elevator;
  • the data acquisition part includes the time domain signals of the elevator car and the reduction box vibration and the image visual signals of the traction rope, wherein the acquired time domain signals include the peak value, mean value and pulse index of the time domain signals;
  • the arrangement of the elevator vibration signal monitoring points is as follows: the elevator vibration status monitoring points are arranged on the car and the reduction box respectively, and the sensor nodes in the car are mainly arranged at the center position of the walls of the car in three directions.
  • the vibration signal of the elevator is collected by installing independent acceleration sensors on the outside of the elevator mechanical parts.
  • the image status monitoring point of the elevator car traction rope is installed on the elevator shaft wall.
  • the detection method based on line structured light is used.
  • the imaging system includes a line light source and an industrial CCD camera.
  • the line light source and the camera have a relatively fixed spatial position.
  • the light plane of the line light source is perpendicular to the axis of the traction rope.
  • Step 2 preprocess the collected vibration signals and visual signals of the car and the reduction box, and extract the vibration signal characteristics of the car and the reduction box and the lateral displacement signal characteristics of the car traction rope respectively;
  • the signal preprocessing part includes the time domain signal preprocessing of the elevator car and reduction box vibration and the image visual signal preprocessing of the traction rope, specifically:
  • the time domain signals of the elevator car and the reduction box vibration are collected and subjected to noise reduction processing by the median filtering method, the processed time domain signals are converted into frequency domain signals by Fourier transform, and the frequency domain signals are subjected to feature extraction by the wavelet analysis method to obtain the vibration signal characteristics of the car and the reduction box;
  • the collected visual signals of the car traction rope image are scanned and denoised, and the current traction rope image information is calibrated using the Zhang Zhengyou calibration method.
  • the target tracking method in computer vision is used to track the light strip area illuminated on the wire rope in the camera lens, and then the lateral vibration displacement of the wire rope is obtained as the lateral displacement signal feature of the car traction rope;
  • the target tracking adopts the CSRT target tracking method, whose input signal is the image (video) signal of the elevator traction rope, and the output result is the image (video) signal after virtual marking;
  • the image signal detection model adopted by the CSRT target tracking method uniformly uses CNN for classification and prediction of the border position, wherein CNN is a convolutional neural network with a seven-layer structure, which is a type of feedforward neural network algorithm that includes convolution calculation and has a deep structure, and its first layer is the input layer; the second and third layers are both convolution layers, which are used for image feature extraction, that is, converting the image into a feature vector of fixed dimension; the fourth, fifth and sixth layers are all pooling layers, which are used for feature compression and extraction of main features; the seventh layer is a fully connected layer, which classifies the original image by extracting feature parameters; the eighth layer is the output layer;
  • Step 3 Perform abnormal vibration signal detection based on the vibration signal characteristics of the car and the reduction box to obtain abnormal vibration signal detection results, and perform abnormal lateral displacement detection based on the lateral displacement signal characteristics of the car traction rope to obtain abnormal lateral detection results;
  • An abnormal vibration signal detection model is used to perform abnormal vibration signal detection processing on the vibration signals of the elevator car and the reducer, and an abnormal image state detection model is used to perform abnormal image signal detection processing on the image visual signal of the elevator traction rope;
  • the abnormal vibration signal detection model adopts the LightGBM model, whose input is the vibration signal characteristics of the car and the reduction box, and the output is the abnormal vibration signal detection result;
  • the abnormal image detection model adopts an R-CNN model, whose input is the image signal of the elevator car traction rope, and the output is the abnormal image signal detection result, and then the abnormal image signal is calibrated and tracked to obtain the abnormal lateral displacement signal;
  • Step 4 Fuse the abnormal vibration signal detection result and the abnormal lateral displacement signal detection result, and determine whether to generate an emergency stop fault alarm for the elevator based on the fusion result;
  • the abnormal vibration signal detection results and the abnormal lateral displacement signal detection results are subjected to fault diagnosis based on fuzzy information fusion, as shown in FIG3 , specifically:
  • the elevator operation status data is collected and fuzzy processed, the relationship matrix between the operation status and the fault type is established, the weight vector of each sensor is obtained, and then multiplied with the relationship matrix to further obtain the diagnosis membership result matrix.
  • the possibility of each fault decision after fusion is obtained according to the fault judgment principle, as shown in the following formula:
  • A is the set of diagnosed elevator fault points, namely, three types of faults: severe vibration of elevator car and reducer, deformation of car traction rope exceeding the safety value
  • B is the set of elevator operation status, namely, three elevator operation status signals: vibration signal of car and reducer and lateral displacement signal of traction rope
  • X is the set of credibility values for judging the operation status of different elevator components
  • x 1 , x 2 , x 3 are the credibility values assigned to the operation status of different elevator components based on experience or fault database
  • Y obtained by fuzzy transformation is the possibility of each decision after fusion
  • the decision target to be determined should have the largest membership value
  • the possibility set of each fault inference decision is selected according to a certain weight to obtain the optimal fault inference result.

Landscapes

  • Maintenance And Inspection Apparatuses For Elevators (AREA)
  • Lift-Guide Devices, And Elevator Ropes And Cables (AREA)

Abstract

本发明提供一种基于多源信息融合的电梯安全故障诊断方法，涉及电梯故障诊断技术领域。该方法首先采集电梯运行过程中轿厢与减速箱的振动信号以及轿厢牵引绳的视觉信号，并对采集信号进行数据预处理后提取轿厢与减速箱的振动信号特征和轿厢牵引绳的横向位移信号特征；然后依据轿厢与减速箱的振动信号特征进行异常振动信号检测，获得异常的振动信号检测结果，基于轿厢牵引绳的横向位移信号特征进行异常横向位移检测并获得异常横向检测结果；最后对异常振动信号检测结果和异常横向位移信号检测结果进行融合，并基于融合结果判断是否针对电梯产生急停故障报警。该方法利用模糊信息融合理论，实现了电梯多部件运行状态的实时集中监测。

Description

一种基于多源信息融合的电梯安全故障诊断方法 技术领域
本发明涉及电梯故障诊断技术领域，尤其涉及一种基于多源信息融合的电梯安全故障诊断方法。
背景技术
随着城市化进程的发展，为了满足人们日常生活的垂直运输的需求，电梯的应用越来越广泛，但其使用量的增加也造成了很多的安全事故和安全隐患。电梯事故不仅造成经济损失，且严重威胁乘客的人身安全，因此需要更加全面有效的检测电梯的安全状况。
目前电梯的故障诊断通常针对电梯单一部件进行检测，不能同时保证多个部件正常工作，单一安全保障主要依靠定期维护和年检的方式，但是存在不能够全面及时发现电梯安全隐患、浪费人力物力等问题，因此根据电梯的在线监测数据，对电梯进行安全性评估，实现电梯的状态检修具有重要意义。
发明内容
本发明要解决的技术问题是针对上述现有技术的不足，提供一种基于多源信息融合的电梯安全故障诊断方法，实现对电梯安全故障的诊断。
为解决上述技术问题，本发明所采取的技术方案是：一种基于多源信息融合的电梯安全故障诊断方法，包括以下步骤：
步骤1、分别采集电梯运行过程中轿厢与减速箱的振动信号以及轿厢牵引绳的视觉信号；
数据采集部分包括电梯轿厢和减速箱振动的时域信号和牵引绳的图像视觉信号，其中，采集的时域信号包括时域信号的峰值、均值和脉冲指标；
步骤2、对采集的轿厢与减速箱的振动信号和视觉信号进行数据预处理，并分别提取轿厢与减速箱的振动信号特征和轿厢牵引绳的横向位移信号特征；
信号预处理部分包括电梯轿厢和减速箱振动的时域信号预处理和牵引绳的图像视觉信号预处理两部分，具体为：
将采集到电梯轿厢和减速箱振动的时域信号利用中值滤波法进行时域信号的降噪处理，将处理过的时域信号进行傅里叶变换转化为频域信号，再利用小波分析法对所述频域信号进行特征提取，得到轿厢与减速箱的振动信号特征；
将采集到的轿厢牵引绳图像视觉信号进行扫描和去噪处理，并利用张正友标定法将当前牵引绳图像信息进行标定，采用计算机视觉中的目标追踪方法，追踪相机镜头内照射在钢丝 绳上的光条区域，然后获取得到钢丝绳的横向振动位移量作为轿厢牵引绳的横向位移信号特征；
所述目标追踪采用CSRT目标追踪法，其输入信号为电梯牵引绳的图像信号，输出结果为经过虚拟标记后的图像信号；CSRT目标追踪法所采用的图像信号检测模型统一使用CNN来做分类和预测边框位置；
步骤3、依据轿厢与减速箱的振动信号特征进行异常振动信号检测，获得异常的振动信号检测结果，基于轿厢牵引绳的横向位移信号特征进行异常横向位移检测并获得异常横向检测结果；
采用异常振动信号检测模型对所述电梯轿厢和减速器的振动信号进行异常振动信号检测处理，采用异常图像状态检测模型对所述电梯牵引绳的图像视觉信号进行异常图像信号检测处理；
所述异常振动信号检测模型采用LightGBM模型，其输入为轿厢与减速箱的振动信号特征，输出为异常振动信号检测结果；
所述异常图像检测模型采用R-CNN模型，其输入为电梯轿厢牵引绳的图像信号，输出为异常图像信号检测结果，进而对异常图像信号进行标定与追踪，获得异常横向位移信号；
步骤4、对异常振动信号检测结果和异常横向位移信号检测结果进行融合，并基于融合结果判断是否针对电梯产生急停故障报警；
对所述异常振动信号检测结果和异常横向位移信号检测结果进行基于模糊信息融合，得到诊断结果，如下公式所示：
Y＝X×R AB
其中，
X＝[x 1 x 2 x 3]
Figure PCTCN2022136820-appb-000001
其中，A为被诊断电梯故障点集合，即电梯轿厢和减速器剧烈振动、轿厢牵引绳发生超过安全值的形变3种故障类型；B为电梯运行状态集合，即轿厢与减速箱的振动信号和牵引绳横向位移信号三种电梯运行状态信号；A和B的关系矩阵R AB中的元素μ ij为由电梯不同部件运行状态i推断决策故障类型为j的可能性，i＝1、2、3，j＝1、2、3，X为各个电梯不同部件运行状态判断的可信度值集合，x 1、x 2、x 3分别为根据经验或者故障数据库赋予电梯不同部件 运行状态的可信度值，经模糊变换得到的Y为融合后各决策的可能性；
对故障进行推断决策时，采用基于规则的方法，基本原则如下：
(1)判定的决策目标应具有最大的隶属度值；
(2)判定的决策目标的隶属度值要大于设定阈值；
(3)判定的决策目标和其他目标的隶属度值之差要大于某个门限；
最后，对各个故障推断决策的可能性集合按照一定的权重进行选择，得出最优故障推断结果。
采用上述技术方案所产生的有益效果在于：本发明提供的一种基于多源信息融合的电梯安全故障诊断方法，针对电梯轿厢、减速箱和牵引绳的运行状况，利用模糊信息融合理论，实现了电梯多部件运行状态的实时集中监测，对不同部件的故障决策赋予不同的权重，提高了电梯故障诊断的速度和精准度。
附图说明
图1为本发明实施例提供的一种基于多源信息融合的电梯安全故障诊断方法的流程图；
图2为本发明实施例提供的基于多源信息融合的电梯安全故障诊断方法的过程示意图
图3为本发明实施例提供的基于模糊信息融合得到故障诊断结果的流程图。
具体实施方式
下面结合附图和实施例，对本发明的具体实施方式作进一步详细描述。以下实施例用于说明本发明，但不用来限制本发明的范围。
本实施例中，一种基于多源信息融合的电梯安全故障诊断方法，如图1、2所示，包括以下步骤：
步骤1、分别采集电梯运行过程中轿厢与减速箱的振动信号以及轿厢牵引绳的视觉信号；
数据采集部分包括电梯轿厢和减速箱振动的时域信号和牵引绳的图像视觉信号，其中，采集的时域信号包括时域信号的峰值、均值和脉冲指标；
本实施例中，对电梯振动信号监测点的布置为：将电梯振动状态监测点分别布置在轿厢和减速箱两个部件上，轿厢内传感器节点主要布置在轿厢的三个方向面壁的中心位置，通过采用在电梯机械部件外部安装独立加速度传感器的方式来采集电梯的振动信号。
将电梯轿厢牵引绳的图像状态监测点安装在电梯井壁上，利用基于线结构光的检测方法，其成像系统包括线光源和工业CCD摄像机，线光源与摄像机具有相对固定的空间位置，线光源光平面与牵引绳的轴线垂直，当将线状结构的视觉传感器激光平面投影到要测量的工件表面上时，由于工件表面尺寸的变化，激光平面会产生变形光条，摄像机获取变形光条的图像。 该方法采用独立传感器采集方式对电梯现有的机械结构和控制电路不会产生任何影响，也不会影响电梯正常运转。
步骤2、对采集的轿厢与减速箱的振动信号和视觉信号进行数据预处理，并分别提取轿厢与减速箱的振动信号特征和轿厢牵引绳的横向位移信号特征；
信号预处理部分包括电梯轿厢和减速箱振动的时域信号预处理和牵引绳的图像视觉信号预处理两部分，具体为：
将采集到电梯轿厢和减速箱振动的时域信号利用中值滤波法进行时域信号的降噪处理，将处理过的时域信号进行傅里叶变换转化为频域信号，再利用小波分析法对所述频域信号进行特征提取，得到轿厢与减速箱的振动信号特征；
将采集到的轿厢牵引绳图像视觉信号进行扫描和去噪处理，并利用张正友标定法将当前牵引绳图像信息进行标定，采用计算机视觉中的目标追踪方法，追踪相机镜头内照射在钢丝绳上的光条区域，然后获取得到钢丝绳的横向振动位移量作为轿厢牵引绳的横向位移信号特征；
所述目标追踪采用CSRT目标追踪法，其输入信号为电梯牵引绳的图像(视频)信号，输出结果为经过虚拟标记后的图像(视频)信号；CSRT目标追踪法所采用的图像信号检测模型统一使用CNN来做分类和预测边框位置，其中，CNN为七层结构的卷积神经网，是一类包含卷积计算且具有深度结构的前馈神经网络算法，其第一层为输入层；第二、三层均为卷积层，用于图像特征抽取，即将图片转换为固定维度的特征向量；第四、五、六层均为池化层，用于进行特征压缩，提取主要特征；第七层为全连接层，通过提取的特征参数对原始图像进行分类；第八层为输出层；
步骤3、依据轿厢与减速箱的振动信号特征进行异常振动信号检测，获得异常的振动信号检测结果，基于轿厢牵引绳的横向位移信号特征进行异常横向位移检测并获得异常横向检测结果；
采用异常振动信号检测模型对所述电梯轿厢和减速器的振动信号进行异常振动信号检测处理，采用异常图像状态检测模型对所述电梯牵引绳的图像视觉信号进行异常图像信号检测处理；
所述异常振动信号检测模型采用LightGBM模型，其输入为轿厢与减速箱的振动信号特征，输出为异常振动信号检测结果；
所述异常图像检测模型采用R-CNN模型，其输入为电梯轿厢牵引绳的图像信号，输出为异常图像信号检测结果，进而对异常图像信号进行标定与追踪，获得异常横向位移信号；
步骤4、对异常振动信号检测结果和异常横向位移信号检测结果进行融合，并基于融合 结果判断是否针对电梯产生急停故障报警；
本实施例中，对异常振动信号检测结果和异常横向位移信号检测结果进行基于模糊信息融合的故障诊断，如图3所示，具体为：
首先采集电梯运行状态数据并进行模糊化处理，建立运行状态与故障类型的关系矩阵，获取各个传感器的权重向量，再和关系矩阵做乘法运算，进一步得到诊断隶属度结果矩阵，最后依据故障判定原则得到融合后的各故障决策的可能性，如下公式所示：
Y＝X×R AB
其中，
X＝[x 1 x 2 x 3]
Figure PCTCN2022136820-appb-000002
其中，A为被诊断电梯故障点集合，即电梯轿厢和减速器剧烈振动、轿厢牵引绳发生超过安全值的形变3种故障类型；B为电梯运行状态集合，即轿厢与减速箱的振动信号和牵引绳横向位移信号三种电梯运行状态信号；A和B的关系矩阵R AB中的元素μ ij为由电梯不同部件运行状态i推断决策故障类型为j的可能性，i＝1、2、3，j＝1、2、3，X为各个电梯不同部件运行状态判断的可信度值集合，x 1、x 2、x 3分别为根据经验或者故障数据库赋予电梯不同部件运行状态的可信度值，经模糊变换得到的Y为融合后各决策的可能性；
对故障进行推断决策时，采用基于规则的方法，基本原则如下：
(1)判定的决策目标应具有最大的隶属度值；
(2)判定的决策目标的隶属度值要大于设定阈值；
(3)判定的决策目标和其他目标的隶属度值之差要大于某个门限；
最后，对各个故障推断决策的可能性集合按照一定的权重进行选择，得出最优故障推断结果。
最后应说明的是：以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明权利要求所限定的范围。

Claims (7)

  1. 一种基于多源信息融合的电梯安全故障诊断方法，其特征在于：包括以下步骤：
    步骤1、分别采集电梯运行过程中轿厢与减速箱的振动信号以及轿厢牵引绳的视觉信号；
    步骤2、对采集的轿厢与减速箱的振动信号和视觉信号进行数据预处理，并分别提取轿厢与减速箱的振动信号特征和轿厢牵引绳的横向位移信号特征；
    步骤3、依据轿厢与减速箱的振动信号特征进行异常振动信号检测，获得异常的振动信号检测结果，基于轿厢牵引绳的横向位移信号特征进行异常横向位移检测并获得异常横向检测结果；
    步骤4、对异常振动信号检测结果和异常横向位移信号检测结果进行融合，并基于融合结果判断是否针对电梯产生急停故障报警。
  2. 根据权利要求1所述一种基于多源信息融合的电梯安全故障诊断方法，其特征在于：所述步骤1采集电梯轿厢和减速箱振动的时域信号和牵引绳的图像视觉信号，其中，采集的时域信号包括时域信号的峰值、均值和脉冲指标。
  3. 根据权利要求2所述一种基于多源信息融合的电梯安全故障诊断方法，其特征在于：所述步骤2的数据预处理包括电梯轿厢和减速箱振动的时域信号预处理和牵引绳的图像视觉信号预处理两部分，具体为：
    将采集到电梯轿厢和减速箱振动的时域信号利用中值滤波法进行时域信号的降噪处理，将处理过的时域信号进行傅里叶变换转化为频域信号，再利用小波分析法对所述频域信号进行特征提取，得到轿厢与减速箱的振动信号特征；
    将采集到的轿厢牵引绳图像视觉信号进行扫描和去噪处理，并利用张正友标定法将当前牵引绳图像信息进行标定，采用计算机视觉中的目标追踪方法，追踪相机镜头内照射在钢丝绳上的光条区域，然后获取得到钢丝绳的横向振动位移量作为轿厢牵引绳的横向位移信号特征。
  4. 根据权利要求3所述一种基于多源信息融合的电梯安全故障诊断方法，其特征在于：所述目标追踪采用CSRT目标追踪法，其输入信号为电梯牵引绳的图像信号，输出结果为经过虚拟标记后的图像信号；CSRT目标追踪法所采用的图像信号检测模型统一使用CNN来做分类和预测边框位置。
  5. 根据权利要求4所述一种基于多源信息融合的电梯安全故障诊断方法，其特征在于：所述步骤3采用异常振动信号检测模型对所述电梯轿厢和减速器的振动信号进行异常振动信号检测处理，采用异常图像状态检测模型对所述电梯牵引绳的图像视觉信号进行异常图像信号检测处理；
    所述异常振动信号检测模型采用LightGBM模型，其输入为轿厢与减速箱的振动信号特征，输出为异常振动信号检测结果；
    所述异常图像检测模型采用R-CNN模型，其输入为电梯轿厢牵引绳的图像信号，输出为异常图像信号检测结果，进而对异常图像信号进行标定与追踪，获得异常横向位移信号。
  6. 根据权利要求5所述一种基于多源信息融合的电梯安全故障诊断方法，其特征在于：所述步骤4基于模糊信息融合得出诊断结果，如下公式所示：
    Y＝X×R AB
    其中，
    X＝[x 1 x 2 x 3]
    Figure PCTCN2022136820-appb-100001
    其中，A为被诊断电梯故障点集合，即电梯轿厢和减速器剧烈振动、轿厢牵引绳发生超过安全值的形变3种故障类型；B为电梯运行状态集合，即轿厢与减速箱的振动信号和牵引绳横向位移信号三种电梯运行状态信号；A和B的关系矩阵R AB中的元素μ ij为由电梯不同部件运行状态i推断决策故障类型为j的可能性，i＝1、2、3，j＝1、2、3，X为各个电梯不同部件运行状态判断的可信度值集合，x 1、x 2、x 3分别为根据经验或者故障数据库赋予电梯不同部件运行状态的可信度值，经模糊变换得到的Y为融合后各决策的可能性；
    最后，对各个故障推断决策的可能性集合按照一定的权重进行选择，得出最优故障推断结果。
  7. 根据权利要求6所述一种基于多源信息融合的电梯安全故障诊断方法，其特征在于：由电梯不同部件运行状态i推断决策故障类型为j的可能性时，采用基于规则的方法，基本原则如下：
    (1)判定的决策目标应具有最大的隶属度值；
    (2)判定的决策目标的隶属度值要大于设定阈值；
    (3)判定的决策目标和其他目标的隶属度值之差要大于某个门限。
PCT/CN2022/136820 2022-12-06 2022-12-06 一种基于多源信息融合的电梯安全故障诊断方法 Ceased WO2024119360A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2022/136820 WO2024119360A1 (zh) 2022-12-06 2022-12-06 一种基于多源信息融合的电梯安全故障诊断方法
CN202280053511.2A CN117916185A (zh) 2022-12-06 2022-12-06 一种基于多源信息融合的电梯安全故障诊断方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/136820 WO2024119360A1 (zh) 2022-12-06 2022-12-06 一种基于多源信息融合的电梯安全故障诊断方法

Publications (1)

Publication Number Publication Date
WO2024119360A1 true WO2024119360A1 (zh) 2024-06-13

Family

ID=90691153

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/136820 Ceased WO2024119360A1 (zh) 2022-12-06 2022-12-06 一种基于多源信息融合的电梯安全故障诊断方法

Country Status (2)

Country Link
CN (1) CN117916185A (zh)
WO (1) WO2024119360A1 (zh)

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118561117A (zh) * 2024-08-01 2024-08-30 浙江新再灵科技股份有限公司 基于多源数据融合的电梯安全监测方法、系统及电子设备
CN118797487A (zh) * 2024-09-11 2024-10-18 爱动超越人工智能科技（北京）有限责任公司 基于多源状态数据和神经网络模型进行叉车碰撞的智能检测系统
CN119026088A (zh) * 2024-10-28 2024-11-26 北京芯盾时代科技有限公司 一种基于人工智能的电力设备诊断方法及其系统
CN119046771A (zh) * 2024-10-31 2024-11-29 江苏省特种设备安全监督检验研究院 用于机电类特种设备生产的振动检验装置及检测方法
CN119211681A (zh) * 2024-11-26 2024-12-27 中国特种设备检测研究院 一种基于视觉检测的电梯安全管理方法及系统
CN119299332A (zh) * 2024-10-09 2025-01-10 中国科学院昆明植物研究所 一种基于机器视觉的安全监测系统及监测方法
CN119284683A (zh) * 2024-12-04 2025-01-10 中国计量大学 基于数字孪生的电梯故障诊断方法
CN119377650A (zh) * 2024-09-23 2025-01-28 安徽鼎立网络科技有限公司 一种基于物联网的智慧楼宇数据采集分析方法及系统
CN119762444A (zh) * 2024-12-10 2025-04-04 安字汽车紧固件(启东)有限公司 一种基于人工智能的汽车紧固件状态检测方法及系统
CN120217267A (zh) * 2025-05-27 2025-06-27 西安理工大学 一种基于多源信号的直拉硅单晶炉升降丝杠故障诊断方法
CN120238646A (zh) * 2025-03-26 2025-07-01 国网四川省电力公司内江供电公司 一种面向变电站场景的摄像头自检方法
CN120628625A (zh) * 2025-05-30 2025-09-12 上海铠铂云信息科技有限公司 一种用于半导体制造的高架穿梭车自我故障检测系统
CN120932680A (zh) * 2025-08-19 2025-11-11 广州励心物联科技有限公司 一种机房噪声分析判断方法及噪声分析判断装置
TWI906102B (zh) * 2024-12-25 2025-11-21 國立勤益科技大學 雲端電梯運行監測與健康狀態管理戰情系統
CN121433172A (zh) * 2025-12-29 2026-01-30 浙江常青树信息技术有限责任公司 基于多协议融合的工业设备控制方法及相关装置
CN121524893A (zh) * 2025-12-31 2026-02-13 煤炭工业规划设计研究院有限公司 基于多模态融合的洗选振动筛设备故障诊断方法及设备

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118999412B (zh) * 2024-10-23 2025-04-15 通用电梯股份有限公司 一种基于双目视觉的电梯导轨自动测量系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060243538A1 (en) * 2005-03-24 2006-11-02 Josef Husmann Elevator with vertical vibration compensation
JP2008074536A (ja) * 2006-09-20 2008-04-03 Mitsubishi Electric Corp エレベータのロープ横揺れ検出装置及びエレベータの管制運転装置
CN110642109A (zh) * 2019-04-26 2020-01-03 深圳市豪视智能科技有限公司 升降设备的振动检测方法、装置、服务器及存储介质
CN111929014A (zh) * 2020-08-10 2020-11-13 广州广日电梯工业有限公司 牵引绳垂直振动测量方法及装置
US20210206597A1 (en) * 2020-01-07 2021-07-08 Kone Corporation Method for operating an elevator
US20210371245A1 (en) * 2018-03-27 2021-12-02 Inventio Ag Method and device for monitoring properties of a supporting-means arrangement in an elevator system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060243538A1 (en) * 2005-03-24 2006-11-02 Josef Husmann Elevator with vertical vibration compensation
JP2008074536A (ja) * 2006-09-20 2008-04-03 Mitsubishi Electric Corp エレベータのロープ横揺れ検出装置及びエレベータの管制運転装置
US20210371245A1 (en) * 2018-03-27 2021-12-02 Inventio Ag Method and device for monitoring properties of a supporting-means arrangement in an elevator system
CN110642109A (zh) * 2019-04-26 2020-01-03 深圳市豪视智能科技有限公司 升降设备的振动检测方法、装置、服务器及存储介质
US20210206597A1 (en) * 2020-01-07 2021-07-08 Kone Corporation Method for operating an elevator
CN111929014A (zh) * 2020-08-10 2020-11-13 广州广日电梯工业有限公司 牵引绳垂直振动测量方法及装置

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118561117A (zh) * 2024-08-01 2024-08-30 浙江新再灵科技股份有限公司 基于多源数据融合的电梯安全监测方法、系统及电子设备
CN118797487A (zh) * 2024-09-11 2024-10-18 爱动超越人工智能科技（北京）有限责任公司 基于多源状态数据和神经网络模型进行叉车碰撞的智能检测系统
CN119377650A (zh) * 2024-09-23 2025-01-28 安徽鼎立网络科技有限公司 一种基于物联网的智慧楼宇数据采集分析方法及系统
CN119299332A (zh) * 2024-10-09 2025-01-10 中国科学院昆明植物研究所 一种基于机器视觉的安全监测系统及监测方法
CN119026088A (zh) * 2024-10-28 2024-11-26 北京芯盾时代科技有限公司 一种基于人工智能的电力设备诊断方法及其系统
CN119046771A (zh) * 2024-10-31 2024-11-29 江苏省特种设备安全监督检验研究院 用于机电类特种设备生产的振动检验装置及检测方法
CN119211681A (zh) * 2024-11-26 2024-12-27 中国特种设备检测研究院 一种基于视觉检测的电梯安全管理方法及系统
CN119284683A (zh) * 2024-12-04 2025-01-10 中国计量大学 基于数字孪生的电梯故障诊断方法
CN119762444A (zh) * 2024-12-10 2025-04-04 安字汽车紧固件(启东)有限公司 一种基于人工智能的汽车紧固件状态检测方法及系统
TWI906102B (zh) * 2024-12-25 2025-11-21 國立勤益科技大學 雲端電梯運行監測與健康狀態管理戰情系統
CN120238646A (zh) * 2025-03-26 2025-07-01 国网四川省电力公司内江供电公司 一种面向变电站场景的摄像头自检方法
CN120217267A (zh) * 2025-05-27 2025-06-27 西安理工大学 一种基于多源信号的直拉硅单晶炉升降丝杠故障诊断方法
CN120628625A (zh) * 2025-05-30 2025-09-12 上海铠铂云信息科技有限公司 一种用于半导体制造的高架穿梭车自我故障检测系统
CN120932680A (zh) * 2025-08-19 2025-11-11 广州励心物联科技有限公司 一种机房噪声分析判断方法及噪声分析判断装置
CN121433172A (zh) * 2025-12-29 2026-01-30 浙江常青树信息技术有限责任公司 基于多协议融合的工业设备控制方法及相关装置
CN121524893A (zh) * 2025-12-31 2026-02-13 煤炭工业规划设计研究院有限公司 基于多模态融合的洗选振动筛设备故障诊断方法及设备

Also Published As

Publication number Publication date
CN117916185A (zh) 2024-04-19

Similar Documents

Publication Publication Date Title
WO2024119360A1 (zh) 一种基于多源信息融合的电梯安全故障诊断方法
Zhou et al. Automatic detection of industrial wire rope surface damage using deep learning-based visual perception technology
CN109859170B (zh) 一种基于lbp特征的钢丝绳表面损伤智能监测方法及系统
WO2019080367A1 (zh) 一种机械设备健康状态评估方法
CN116821840A (zh) 一种基于特征级融合多模态的滚动轴承故障检测方法
CN105069778B (zh) 基于目标特征显著图构建的工业产品表面缺陷检测方法
CN103247039A (zh) 一种基于复合视觉的高压线缆带电检测方法
CN120525826A (zh) 一种高铁零部件裂纹实时检测方法
CN115165362B (zh) 一种基于巡检及振动信号的轴承故障智能检测系统和方法
CN111767815A (zh) 一种隧道渗漏水识别方法
CN119018735A (zh) 一种基于物联网的电梯轿厢运行工况检测方法及系统
CN119142973A (zh) 基于多特征数据融合决策的扶梯安全监控系统和方法
Zhang et al. EdgeAD: Unsupervised learning model based on prior knowledge enhanced image anomaly detection of heavy railway freight cars
CN120509742B (zh) 一种航空发动机轴承故障风险预警方法
WO2024119359A1 (zh) 一种电梯曳引绳状态在线检测系统及其检测方法
CN117184812A (zh) 输送皮带的撕裂及异物识别检测方法
CN119887181A (zh) 基于图像识别的桥梁拉索病害检测方法及系统
CN119284746A (zh) 一种基于机器学习的起重机运维方法、电子设备及计算机可读存储介质
CN119418487A (zh) 绕包机器人智能故障诊断方法
CN117372926A (zh) 皮带机ai异物识别系统
CN117636228A (zh) 起重机的故障识别方法、装置、电子设备及存储介质
CN114572788A (zh) 一种电梯智能化远程故障监测方法、系统及可存储介质
CN107870084A (zh) 列车转向架的非接触式监测方法及系统
CN119379681B (zh) 一种接触网补偿装置卡滞检测及预测方法
CN120039733B (zh) 一种电梯故障检测与报警系统及方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22967522

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22967522

Country of ref document: EP

Kind code of ref document: A1