CN109019210B - Lifting system tail rope health monitoring system and method based on convolutional neural network - Google Patents
Lifting system tail rope health monitoring system and method based on convolutional neural network Download PDFInfo
- Publication number
- CN109019210B CN109019210B CN201810694764.3A CN201810694764A CN109019210B CN 109019210 B CN109019210 B CN 109019210B CN 201810694764 A CN201810694764 A CN 201810694764A CN 109019210 B CN109019210 B CN 109019210B
- Authority
- CN
- China
- Prior art keywords
- rope
- tail rope
- tail
- neural network
- convolutional neural
- 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.)
- Active
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
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0018—Devices monitoring the operating condition of the elevator system
- B66B5/0031—Devices monitoring the operating condition of the elevator system for safety reasons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention relates to a convolutional neural network-based lifting system tail rope health monitoring system and method, and belongs to the field of mechanical system health monitoring. A roller of the monitoring system pulls a hoisting steel wire rope to move; the top ends of the two lifting containers are respectively connected with a steel wire rope; the bottom ends of the two lifting containers are connected through tail ropes respectively; the method is characterized in that: the system also comprises an image acquisition system, a mobile wireless sensor network and an upper computer; the image acquisition system is used for acquiring the tail rope state, the image acquisition system transmits the tail rope state to the upper computer through the mobile wireless sensor network, the upper computer carries out deep excavation on image data, and fault analysis and early warning are carried out on the tail rope state. The system and the method for monitoring the health of the tail rope of the lifting system based on the convolutional neural network can replace manual inspection, and realize the automation of the whole process of acquisition, learning, prediction and early warning of tail rope data information.
Description
Technical Field
The invention relates to a convolutional neural network-based lifting system tail rope health monitoring system and method, and belongs to the field of mechanical system health monitoring.
Background
The tail rope is mainly set for balancing the gravity of a hoisting steel wire rope and obtaining equal torque in a mine hoisting system, is an important part of the hoisting system, and the working condition of the tail rope directly influences the hoisting safety. The balance tail rope is positioned at the bottom of the lifting container and is positioned in the shaft without illumination for most of time. Common tail rope faults include uneven intervals, rope twisting, strand breaking, rope breaking and the like. Research and production experience shows that the failure reasons of the tail rope comprise: vibration in operation, rotation in starting and stopping, ore falling impact, shaft wind field, corrosion and the like; the accident hazard is as follows: the stability of the lifting system is influenced, broken ropes break equipment in the well, life safety of personnel is threatened, production stop is caused, and the like.
However, at present, the management problem of the lifting tail rope is always in the situation of 'paying attention to but having no effective means', the traditional balance tail rope overhauling work depends on the flashlight to check by a worker, the difficulty is high, the efficiency is low, and the personnel safety problem exists. The faults of the balance tail rope occur frequently, and serious threats are brought to the safe operation of a lifting system. For example, the lifting system of the main shaft of the coal mine of the children pavilion adopts two tail ropes, the tail ropes are broken and damaged for 5 times in 1989 and 1998, and 10 tail ropes are replaced due to accidents, so that serious manpower, material resources and property loss is caused; the main well of the north Minghe Hexie iron ore adopts three tail ropes, and the balance hammer side tail rope breakage falling accident happens in 14 days 2 month 2011, so that a shaft is damaged, and a large amount of economic loss is caused. Therefore, the fault of the balance tail rope is identified and early warning is given so as to conveniently adopt reasonable measures to eliminate the fault, which has great significance for the safe and efficient production of the lifting system.
Machine vision is to acquire object information with various imaging devices and to analyze the acquired information by an arithmetic processor. The machine vision detection system takes an arithmetic processor as a core and comprises modules of image information acquisition and digitization, image processing and decision making, execution processing control and the like, wherein an image processing and recognition algorithm is the core technology of the machine vision detection system. The machine vision has the characteristics of high processing speed, capability of performing nondestructive detection on an object, intuitional and comprehensive acquisition of characteristic information and the like, and therefore, the machine vision has important application in the fields of aerospace, machining, agricultural production, food detection, health monitoring and the like.
The wireless sensor network is a distributed sensor network and consists of a plurality of sensor nodes for sensing and checking the outside world. The sensors in the WSN communicate in a wireless mode, so that the network setting is flexible, the position of equipment can be changed at any time, and wired or wireless connection can be carried out with the Internet. The mobile wireless sensor network has a changeable network topology structure due to the flexibility of the nodes, so that the environmental adaptability of the network is improved, and the mobile wireless sensor network is more and more concerned by academia and industry in recent years.
With the arrival of the big data era and the rapid development of deep learning, in recent years, a convolutional neural network becomes a core algorithm of machine vision, a complex artificial feature extraction process is not needed, feature information can be directly and automatically mined from an original image, feature self-adaptive identification and classification are completed, and the method has the characteristics of high precision, good instantaneity and the like.
Therefore, the invention combines the current health monitoring situation of the tail rope and advanced machine vision, wireless sensor network and convolutional neural network technologies to carry out real-time health monitoring and fault diagnosis on the tail rope of the lifting system. The machine vision, the wireless sensor network and the deep learning have wide application prospects in the field of mechanical equipment health monitoring, and have great significance in the safety production of mineral products when being applied in the field of mining mechanical equipment health monitoring.
Disclosure of Invention
The invention provides a system and a method for monitoring the health of a tail rope of a lifting system based on a convolutional neural network.
The invention adopts the following technical scheme:
the invention relates to a tail rope health monitoring system of a lifting system based on a convolutional neural network, wherein the lifting system comprises a roller, a lifting steel wire rope, a lifting container and a tail rope; the roller pulls the hoisting steel wire rope to move; the top ends of the two lifting containers are respectively connected with a steel wire rope; the bottom ends of the two lifting containers are connected through tail ropes respectively; the system also comprises an image acquisition system, a mobile wireless sensor network and an upper computer; the image acquisition system is used for acquiring the tail rope state, the image acquisition system transmits the tail rope state to the upper computer through the mobile wireless sensor network, the upper computer carries out deep excavation on image data, and fault analysis and early warning are carried out on the tail rope state.
The invention relates to a lifting system tail rope health monitoring system based on a convolutional neural network, which comprises an image acquisition system, a control system and a control system, wherein the image acquisition system comprises a background plate, a light source, a CCD (charge coupled device) camera, an acquisition card, a controller and a memory; a background plate is arranged between tail ropes connected between the two lifting containers and is arranged on the hanging surface of the tail ropes in parallel; the optical axis of the CCD camera is vertical to the background plate, and a light source is arranged on the CCD camera; the CCD camera is connected with an acquisition card, and the acquisition card is connected with the controller; the light source and the storage are respectively connected with the controller.
The invention relates to a lifting system tail rope health monitoring system based on a convolutional neural network, wherein a mobile wireless sensor network comprises a rope guide and a mobile sensor node; and one end of the mobile sensor node along the extending direction of the rope guide is connected with the image acquisition system, and the other end of the mobile sensor node is connected with an upper computer.
According to the tail rope health monitoring system of the lifting system based on the convolutional neural network, a convolutional neural network program and an early warning program are contained in the upper computer; and detecting and judging the state of the tail rope through a convolutional neural network program, feeding the result back to an early warning program, and controlling the start and stop of the lifting system by the early warning program.
The invention relates to a method for improving a system tail rope health monitoring system based on a convolutional neural network, which comprises the following steps of:
1) acquiring the tail rope state through an image acquisition system, and classifying the collected images;
2) carrying out convolutional neural network training, and carrying out forward propagation and backward propagation on the image data; determining errors and establishing a convolutional neural network model;
3) and adopting the convolution neural network model in the step 2) to predict and classify the input tail rope state image in real time, feeding the predicted and classified tail rope state image back to an early warning program, and judging and early warning the classification result by the early warning program.
The method for improving the system tail rope health monitoring system based on the convolutional neural network comprises the step 1) that an image acquisition system generates a training data set, a test data set, a training label set and a test label set aiming at acquired image data.
The invention discloses a method for improving a system tail rope health monitoring system based on a convolutional neural network, which comprises the following steps of: setting training parameters of the network, and initializing the weight and the bias of the network; the input characteristic diagram is processed by a convolution layer, a sampling layer and a full connection layer and then transmitted to an output layer, wherein the output of each layer is the input of the next layer;
the data back propagation mode is as follows: and reversely propagating the error between the actual output and the expected output layer by layer through a BP algorithm, distributing the error to each layer, and adjusting the weight and the bias of the network until the convergence condition is met.
The method for improving the system tail rope health monitoring system based on the convolutional neural network comprises the steps of establishing a tail rope characteristic database aiming at collected image data; and expands the image data.
Advantageous effects
The system and the method for monitoring the health of the tail rope of the lifting system based on the convolutional neural network can replace manual inspection, and realize the automation of the whole process of acquisition, learning, prediction and early warning of tail rope data information.
The method for monitoring and early warning the health of the tail rope of the hoisting system adopts the CNN algorithm, can effectively avoid the complex steps of artificial feature extraction and description and the like, gets rid of the dependence on a large number of signal processing technologies and diagnosis experiences, completes the self-adaptive extraction of fault features and the intelligent diagnosis of health conditions, has the fault recognition rate of over 99 percent and the recognition precision far higher than the K nearest neighbor (thek-NearestNeighbor, KNN) and BP Neural networks (BPNN), which have high accuracy and real-time;
the method can find faults such as uneven tail rope spacing, strand breakage, rope twisting, rope breakage and the like in real time, solves the problems that the existing tail rope faults are difficult to detect, long in detection period and incapable of being found in time, and is a feasible scheme for safety monitoring of the tail rope of the mine hoisting system. Meanwhile, the method and conclusion can be popularized to health monitoring and fault diagnosis of other mechanical systems.
Drawings
FIG. 1 is a schematic diagram of a monitoring system according to the present invention;
FIG. 2 is a schematic flow chart of a monitoring method according to the present invention;
FIG. 3 is a diagram of a convolutional neural network structure of the present invention;
FIG. 4 is a schematic representation of a representative feature of the tail rope of the present invention;
FIG. 5 is a tail rope feature seed diagram of the present invention;
FIG. 6 is a flow chart of a data set expansion method of the present invention;
FIG. 7 is an exemplary diagram of 9 states of a data set according to the present invention;
FIG. 8 is a schematic diagram of an iteration of the convolutional neural network diagnostic process of the present invention;
FIG. 9 is a diagram illustrating a comparison of classification results of different algorithms according to the present invention.
Detailed Description
In order to make the purpose and technical solution of the embodiments of the present invention clearer, the technical solution of the embodiments of the present invention will be clearly and completely described below with reference to the drawings of the embodiments of the present invention. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention without any inventive step, are within the scope of protection of the invention.
As shown in fig. 1, the convolutional neural network-based lifting system tail rope health monitoring system of the present invention includes an image acquisition system, a mobile wireless sensor network, and an upper computer. The image acquisition system comprises a light source, a background plate, a CCD camera, an acquisition card, a storage and the like, and realizes the real-time acquisition of the image data of the balance tail rope; the mobile wireless transmission network comprises a rope cage guide and a mobile sensor node, and can transmit the acquired image data to an upper computer; the upper computer is composed of one or more high-performance deep learning workstations, comprises an upper computer interface, a convolutional neural network program and an early warning program, can finish deep excavation of image big data characteristics, analyzes and warns according to grades aiming at tail rope faults, immediately sends out alarm information once rope twisting and rope breaking faults occur in the tail rope are found, and informs parking maintenance to avoid fault expansion.
The image acquisition system comprises a background plate, a light source, a CCD camera, an acquisition card, a controller and a memory; the background plate is arranged on a plane vertical to the suspension plane of the single tail rope (the plane where the single tail rope is arranged), is positioned between the left vertical section and the right vertical section of the single tail rope, and does not interfere with the normal operation of the container and the tail rope; the optical axis of the CCD camera is vertical to the background plate, and the irradiation mode of the light source is forward irradiation, namely the light source and the camera are positioned at the same side; the CCD camera is connected with the acquisition card and then connected with the controller; the light source and the memory are also respectively connected with the controller.
As shown in fig. 2, the schematic flow chart of the method for improving the system tail rope health monitoring system based on the convolutional neural network of the present invention includes three parts, namely, preprocessing of image data, training of the convolutional neural network, and fault diagnosis and early warning. Firstly, tail rope image data are collected, and a training data set, a test data set, a training label set and a test label set are generated after preprocessing. Secondly, training the convolutional neural network, including the forward propagation of data and the backward propagation of errors, wherein the forward propagation process of data is as follows: setting training parameters of a network, initializing weight and bias of the network, inputting data in a batch sample input mode, processing an input characteristic diagram by a convolution layer, a sampling layer and a full connection layer, and transmitting the input characteristic diagram to an output layer, wherein the output of each layer is the input of the next layer; the back propagation process of the error is as follows: and reversely propagating errors between the actual output and the expected output layer by layer through a BP algorithm, distributing the errors to each layer, adjusting the weight and the bias of the network until a convergence condition is met, so as to realize supervised training of the network, and further obtain a convolutional neural network model meeting the requirements on precision and instantaneity. And finally, embedding the model into a tail rope health monitoring system of a lifting system, when a new tail rope image is input, performing real-time prediction classification on the convolutional neural network model, and performing early warning by an early warning program according to a classification result.
As shown in fig. 3, which is a structure diagram of the convolutional neural network of the present invention, includes an input layer, a hidden layer, a full link layer and an output layer,
the detailed structure is Input [ 28X 28] -64C [ 3X 3] -64P [ 2X 2] -128C [ 3X 3] -128P [ 2X 2] -FC [200-64-32] -Output [9 ].
The input characteristic map is a 28 × 28 gray scale map; the hidden layer is composed of two convolutional layers and two sampling layers alternately, the number of convolution kernels of the first convolutional layer and the second convolutional layer is 64 and 128 respectively, the size of the convolution kernels is 3 multiplied by 3, and the step length is 1; matching with 'same' filling operation before convolution, the convolution result at the boundary can be retained, so that the output shape is the same as the input shape; selecting mean value sampling for the sampling mode of the sampling layer, namely, calculating the mean value in a 2 x 2 area of the characteristic diagram; the full-connection layer is set to be 3 layers, and each layer is respectively 200 neurons, 64 neurons and 32 neurons; selecting a Rectisected Linear Unit (ReLU) function as an activation function of the convolutional layer and the full connection layer; selecting a softmax classifier by an output layer; to prevent overfitting, a full connectivity layer F1 was performedL 2And (6) regularizing.
As shown in fig. 4, the schematic diagram of typical characteristics of the tail rope of the present invention includes five typical characteristic state picture data of normal (a), uneven spacing (b), stranded rope (c), defect (d) and break (e). Wherein normal is defined as having a pitch greater than 3/4 normal; uneven spacing is defined as the minimum spacing between two ropes is less than 1/2 spacing; the twisted rope is defined as two ropes which are entangled together in various forms, and the two ropes are also defined as the twisted rope when just contacting; the defects mainly simulate the broken strand defect and are divided into a left broken strand (d1), a right broken strand (d2) and a double broken strand (d 3); the breaks are divided into a left cord break (e1), a right cord break (e2) and a double cord break (e 3). The normal interval of the tail rope is set asDThe length of the field of view of the picture taken by the cameraLIs wide and wideWThen, then
Fig. 5 shows a characteristic seed picture of the tail rope according to the present invention.
The simulation picture is adopted to replace a field picture, and the simulation picture comprises 9 characteristic pictures, namely a normal picture (a), an uneven interval picture (b), a stranded rope picture (c), a left broken rope picture (d1), a right broken rope picture (d2), a double broken rope picture (d3), a left broken rope picture (e1), a right broken rope picture (e2) and a double broken rope picture (e 3).
When different faults are diagnosed, early warning is carried out according to different grades. Level 1: no early warning is carried out in a normal state; and 2, stage: the speed reduction operation is prompted due to uneven intervals; and 3, level: reminding overhaul when stock is broken; 4, level: the brake signal is sent out immediately when the rope is twisted or broken.
It should be noted that, in order to distinguish between the two characteristic states of normal and pitch irregularity, so that the division is made by the normal pitch larger than 3/4 and smaller than 1/2, when the pitch is between the 1/2 and 3/4 normal pitches, the recognition result may be normal (a) or pitch irregularity (b) for the state to be observed, but does not affect the failure diagnosis result (level 1 is a healthy state, level 2 is a state to be observed, level 3 is a mild failure state, and level 4 is a severe failure state).
FIG. 6 is a flow chart of the data set expansion method of the present invention. The data set establishing process comprises the following steps: firstly, establishing 9 typical characteristic pictures; then, 10 seed pictures of the same type are established for each type; then, the network model is multiplied by means of scaling, translation, rotation and the like so as to enhance the normalization capability of the network model. The data expansion method comprises the following steps:
Step1.rotating the seed picture from-5 degrees to 4 degrees with the increment of 1 degree;
Step2.will be provided withStep1The obtained picture has scale factor of 0.8,1.2]Scaling by 0.1 increments;
Step3.all pictures are uniformly scaled to 28 multiplied by 28 size by adopting a bilinear interpolation method;
Step4.graying all pictures to obtain a pixel matrix;
Step5.and adding a label to obtain a picture data set.
FIG. 7 is a diagram illustrating an example of 9 states of a data set according to the present invention. The method comprises a normal state (a), an uneven interval state (b), a stranded rope (c), a left rope broken strand state (d1), a right rope broken strand state (d2), a double rope broken strand state (d3), a left rope breakage state (e1), a right rope breakage state (e2) and a double rope breakage state (e3), wherein each type contains 500 samples, and the samples are sequentially coded according to 1-9.
Fig. 8 is a schematic diagram illustrating an iteration of the diagnostic process of the convolutional neural network of the present invention. The experimental environment was as follows: the hardware is CPU Intel (R) core (TM) i5-6200U 2.40GHz and the memory is 8.00 GB; the software is an operating system, Windows 64bit, and a development tool, Keras (Theano). By adopting a random gradient descent method (SGD), when the learning rate is 0.01 and the batch size is 10, the training precision is 99.11 percent, the testing precision is 99.33 percent, the training and testing precision is higher, the consumed time of each round is less (29s/epoch,8ms/step), and the requirements of system accuracy and real-time performance can be met.
FIG. 9 is a diagram illustrating a comparison of classification results of different algorithms according to the present invention. The k nearest neighbor algorithm (KNN) whenkWhen = 1, 5, 10, 20, 30, 40, 50 and 100, the diagnostic accuracy is 88.89%; the BPNN-1 has a structure of 200-64-32-9, adopts the SGD method same as the CNN, and has diagnosis precision of 73.56%; the BPNN-2 has a structure of 128-64-9, and the diagnosis precision is 90.11% by using an adam optimization method; the CNN has the diagnosis precision of99.33 percent. The accuracy of the CNN algorithm is far higher than that of the KNN and BPNN algorithms, and the real-time requirement is met.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. The monitoring method of the tail rope health monitoring system of the lifting system based on the convolutional neural network comprises the steps that the lifting system comprises a roller (1), a lifting steel wire rope (2), a lifting container (3), a tail rope (4), an image acquisition system, a mobile wireless sensor network and an upper computer; the roller (1) pulls the hoisting steel wire rope (2) to move; the top ends of the two lifting containers (3) are respectively connected with a lifting steel wire rope (2); the bottom ends of the two lifting containers (3) are connected through tail ropes (4) respectively; also includes; the image acquisition system is used for acquiring the state of the tail rope, the image acquisition system transmits the state of the tail rope (4) to an upper computer through a mobile wireless sensor network, the upper computer performs deep mining on image data, and fault analysis and early warning are performed on the state of the tail rope; the image acquisition system comprises a background plate, a light source, a CCD camera, an acquisition card, a controller and a memory; a background plate is arranged between tail ropes (4) connected between the two lifting containers (3), and the background plate is arranged on the suspension surface of the tail ropes (4) in parallel; the optical axis of the CCD camera is vertical to the background plate, and a light source is arranged on the CCD camera; the CCD camera is connected with an acquisition card, and the acquisition card is connected with the controller; the light source and the storage are respectively connected with the controller; the method is characterized in that: the method comprises the following steps:
1) acquiring the tail rope state through an image acquisition system, and dividing the acquired image data set;
the tail rope is typically characterized by five typical characteristic state picture data of normal (a), uneven spacing (b), twisted rope (c), defect (D) and fracture (e), the normal spacing of the tail rope is D, the visual field length L and width W of a picture shot by a camera are defined, and the state expression is as follows:
in the formula, d1 is the distance between two ropes in a normal state, d2 is the distance between the two ropes in a non-uniform distance state, d3 is the distance between the two ropes in a rope biting state, h1 is the distance between a broken strand defect and the lower edge of a picture in a broken strand state, and h2 is the distance between a rope end and the lower edge of the picture in a broken strand state;
2) carrying out convolutional neural network training, and carrying out forward propagation and backward propagation on the image data; determining errors and establishing a convolutional neural network model;
3) and adopting the convolution neural network model in the step 2) to predict and classify the input tail rope state image in real time, feeding the predicted and classified tail rope state image back to an early warning program, and judging and early warning the classification result by the early warning program.
2. The monitoring method of the convolutional neural network-based lifting system tail rope health monitoring system as claimed in claim 1, wherein: the mobile wireless sensor network comprises a rope cage guide and mobile sensor nodes; and one end of the mobile sensor node along the extending direction of the rope guide is connected with the image acquisition system, and the other end of the mobile sensor node is connected with an upper computer.
3. The monitoring method of the convolutional neural network-based lifting system tail rope health monitoring system as claimed in claim 1, wherein: the upper computer internally contains a convolutional neural network program and an early warning program; and judging the state of the tail rope through a convolutional neural network program, feeding the result back to an early warning program, and controlling the starting, the stopping and the acceleration and the deceleration of the lifting system by the early warning program.
4. The monitoring method of the convolutional neural network-based lifting system tail rope health monitoring system as claimed in claim 1, wherein: in the step 1), the image acquisition system generates a training data set, a testing data set, a training label set and a testing label set aiming at the acquired image data.
5. The monitoring method of the convolutional neural network-based lifting system tail rope health monitoring system as claimed in claim 1, wherein: the data forward propagation mode is as follows: setting training parameters of the network, and initializing the weight and the bias of the network; the input characteristic diagram is processed by a convolution layer, a sampling layer and a full connection layer and then transmitted to an output layer, wherein the output of each layer is the input of the next layer;
the data back propagation mode is as follows: and reversely propagating the error between the actual output and the expected output layer by layer through a BP algorithm, distributing the error to each layer, and adjusting the weight and the bias of the network until the convergence condition is met.
6. The monitoring method of the convolutional neural network-based lifting system tail rope health monitoring system as claimed in claim 1, wherein: establishing a tail rope characteristic database aiming at the collected image data; and expands the image data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810694764.3A CN109019210B (en) | 2018-06-29 | 2018-06-29 | Lifting system tail rope health monitoring system and method based on convolutional neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810694764.3A CN109019210B (en) | 2018-06-29 | 2018-06-29 | Lifting system tail rope health monitoring system and method based on convolutional neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109019210A CN109019210A (en) | 2018-12-18 |
| CN109019210B true CN109019210B (en) | 2021-03-23 |
Family
ID=65520853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810694764.3A Active CN109019210B (en) | 2018-06-29 | 2018-06-29 | Lifting system tail rope health monitoring system and method based on convolutional neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109019210B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111862041A (en) * | 2020-07-21 | 2020-10-30 | 华夏天信（北京）智能低碳技术研究院有限公司 | Bottom shaft tail rope fracture detection method based on template matching |
| CN112037197A (en) * | 2020-08-31 | 2020-12-04 | 中冶赛迪重庆信息技术有限公司 | Hot-rolled bar cold-shearing material accumulation detection method, system and medium |
| CN112087443B (en) * | 2020-09-04 | 2021-06-04 | 浙江大学 | Sensing data anomaly detection method under physical attack of industrial sensing network information |
| CN112330617A (en) * | 2020-10-28 | 2021-02-05 | 联鸾智能科技(上海)合伙企业(有限合伙) | Method and system for detecting tension state of elevator speed limiter rope |
| CN115393360B (en) * | 2022-10-28 | 2023-03-24 | 常州海图信息科技股份有限公司 | Tail rope monitoring control method and tail rope monitoring control device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203319437U (en) * | 2013-06-28 | 2013-12-04 | 洛阳威尔若普检测技术有限公司 | Elevator steel wire rope breakage monitoring device based on machine vision |
| CN204615967U (en) * | 2015-05-08 | 2015-09-02 | 中国矿业大学 | A kind of tail rope radio video monitoring device |
| CN105787439B (en) * | 2016-02-04 | 2019-04-05 | 广州新节奏智能科技股份有限公司 | A kind of depth image human synovial localization method based on convolutional neural networks |
| CN107826919B (en) * | 2017-10-20 | 2019-09-13 | 中国矿业大学 | A kind of lifting system critical component multimode health monitoring device and monitoring method |
| CN108010030A (en) * | 2018-01-24 | 2018-05-08 | 福州大学 | A kind of Aerial Images insulator real-time detection method based on deep learning |
-
2018
- 2018-06-29 CN CN201810694764.3A patent/CN109019210B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109019210A (en) | 2018-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109019210B (en) | Lifting system tail rope health monitoring system and method based on convolutional neural network | |
| CN111862083B (en) | Visual-electromagnetic detection-based steel wire rope state comprehensive monitoring system and method | |
| CN106870298B (en) | Blade root bolt fracture fault detection method based on machine learning | |
| CN108764601B (en) | Structural health monitoring abnormal data diagnosis method based on computer vision and deep learning technology | |
| CN113283344B (en) | Mining conveyor belt deviation detection method based on semantic segmentation network | |
| CN109490072B (en) | Detection system for civil engineering building and detection method thereof | |
| TW202200978A (en) | Systems and methods for artificial intelligence powered inspections and predictive analyses | |
| CN104477723A (en) | Disorder rope monitoring system for winding reel of mine hoist and disorder rope judging method | |
| CN109917213B (en) | Contact network detection fault early warning method based on dimensionality reduction fusion and factor analysis | |
| CN112989706B (en) | Tunnel lamp illumination attenuation prediction method | |
| CN109859170A (en) | A kind of steel wire rope surface damage intelligent monitoring method and system based on LBP feature | |
| CN115329446A (en) | Digital twinning modeling method for intelligent hoisting process of prefabricated parts of fabricated building | |
| CN110645899B (en) | Distributed optical fiber monitoring method and device for structural crack | |
| Zhou et al. | Visual sensing inspection for the surface damage of steel wire ropes with object detection method | |
| CN114218856A (en) | Method for judging abnormity of contact network based on autoregression and deep learning model | |
| CN104085789A (en) | Intelligent monitoring method of running state of crane | |
| CN115966074B (en) | Machine learning-based shield cutter disc mud cake early warning method and system | |
| CN113987905A (en) | Escalator braking force intelligent diagnosis system based on deep belief network | |
| CN117068692A (en) | Conveyer belt health diagnosis method, conveyer belt health diagnosis device, server and storage medium | |
| CN116086677B (en) | Multi-wire rope tension balance monitoring method and system and electronic equipment | |
| CN113392576B (en) | Method for evaluating and early warning tension state of main mooring rope of tethered balloon | |
| CN115082849A (en) | Template support safety intelligent monitoring method based on deep learning | |
| CN113567247A (en) | Bridge detection information management system | |
| CN117237357B (en) | Machine vision-based steel wire rope online monitoring system and method | |
| CN116612423B (en) | AI video identification method for coal mine transportation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |