CN108204892B - Roller set equipment fault detection method based on flexible array type pressure sensor - Google Patents
Roller set equipment fault detection method based on flexible array type pressure sensor Download PDFInfo
- Publication number
- CN108204892B CN108204892B CN201810070059.6A CN201810070059A CN108204892B CN 108204892 B CN108204892 B CN 108204892B CN 201810070059 A CN201810070059 A CN 201810070059A CN 108204892 B CN108204892 B CN 108204892B
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- roller
- fault diagnosis
- roller set
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L5/00—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes
-
- 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
- G06N3/047—Probabilistic or stochastic networks
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to a roller set equipment fault detection method based on a flexible array type pressure sensor, belonging to the technical field of fault diagnosis, and the method comprises the following steps: s1: collecting pressure distribution data between roller sets by using an array type pressure sensor; s2: the pressure distribution data is used as the fault data of the roller set, and the collected fault data is input into a probabilistic neural network to be trained to construct a fault diagnosis model; s3: and inputting the real-time collected fault data of the roller set into a fault diagnosis model for fault diagnosis and judgment to obtain the fault state of the roller set. The invention predicts the fault by utilizing the probabilistic neural network, and avoids the defects of local optimization, long training time and the like of the BP neural network. The method and the device have the advantage that the fault diagnosis of the roller set is well improved in real time and accuracy.
Description
Technical Field
The invention belongs to the technical field of fault diagnosis, and relates to a method for detecting faults of roller set equipment based on a flexible array type pressure sensor.
Background
The roller set is a common mechanical device in industrial production, such as a film pressing machine, a grinding machine and the like. The detection of mechanical failure can be realized by analyzing signals such as vibration, sound and the like of mechanical equipment to judge whether the mechanical equipment fails, but the vibration and sound signals are unstable and are easily interfered by external factors.
Disclosure of Invention
In view of this, the present invention provides a method for detecting a failure of a roller set device based on a flexible array type pressure sensor, in which the pressure sensor is used to obtain pressure data between roller sets, and the pressure data is analyzed to accurately determine a failure of the roller set.
In order to achieve the purpose, the invention provides the following technical scheme:
a roller set equipment fault detection method based on a flexible array type pressure sensor comprises the following steps:
s1: collecting pressure distribution data between roller sets by using an array type pressure sensor;
s2: the pressure distribution data is used as the fault data of the roller set, and the collected fault data is input into a probabilistic neural network to be trained to construct a fault diagnosis model;
s3: and inputting the real-time collected fault data of the roller set into a fault diagnosis model for fault diagnosis and judgment to obtain the fault state of the roller set.
Further, step S1 specifically includes:
s101: connecting the array type pressure sensor with a data acquisition single chip microcomputer, and connecting the single chip microcomputer to a computer;
s102: turning on a power supply and setting parameters;
s103: the computer sends an acquisition instruction to the single chip microcomputer;
s104: the singlechip acquires data, performs filtering and denoising, and transmits the data to the computer according to a communication protocol;
s105: and the computer checks and stores the data and continuously sends the acquisition instruction.
Further, the single chip microcomputer is connected with the computer through a serial port or a USB.
Further, step S2 includes the steps of:
s201: adding an identifier to the collected pressure distribution data, and marking the fault category to which the fault data belongs;
s202: dividing the pressure distribution data into a training data set and a testing data set, converting the training data set and the testing data set into vectors, and performing normalization processing;
s203: selecting a smoothing factor, and establishing a probabilistic neural network fault diagnosis model;
s204: inputting the training data set into a probabilistic neural network fault diagnosis model for training;
s205: testing the trained fault diagnosis model by using a test data set;
s206: and (5) repeatedly executing the steps S202-S205 through the pressure distribution data acquired for multiple times, and adjusting the smoothing factor of the probabilistic neural network until the accuracy of the fault diagnosis model meets the requirement.
Further, the failure types in step S201 are that the roller is tilted left, the roller is normal, the roller is tilted right, and a foreign object is on the roller.
The invention has the beneficial effects that: in data acquisition, compared with a method for diagnosing mechanical faults by using vibration and sound signals, the method greatly reduces the influence of external factors on data accuracy; in the aspect of algorithm processing, the invention utilizes the probabilistic neural network to predict the fault, thereby avoiding the defects of local optimization, long training time and the like of the BP neural network. The method and the device have the advantage that the fault diagnosis of the roller set is well improved in real time and accuracy.
Drawings
In order to make the object, technical scheme and beneficial effect of the invention more clear, the invention provides the following drawings for explanation:
FIG. 1 is a block diagram of data interaction for a data acquisition system according to the present invention;
FIG. 2 is a schematic diagram of four failure types of a roller set;
FIG. 3 is a pressure profile of four failure types of a roller set;
FIG. 4 is a partial result of a roller set fault detection;
wherein the reference numerals are:
the device comprises a roller, a roller left-leaning device, a roller normal device, a roller right-leaning device, a.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
The invention provides a method for diagnosing and detecting faults of roller sets by utilizing flexible array type pressure sensors, which comprises the steps of collecting pressure distribution data among the roller sets by utilizing the flexible array type pressure sensors, taking the pressure distribution data as fault data of the roller sets, inputting the collected fault data into a probability neural network for training to construct a fault diagnosis model, testing the fault diagnosis model constructed by utilizing the probability neural network, and finally inputting the fault data of the roller sets collected in real time into the fault diagnosis model for fault diagnosis and judgment to obtain the fault state of the roller sets, thereby realizing the fault diagnosis of the roller sets.
The flexible array type pressure sensor can be customized according to the needs of actual conditions.
Training and building the fault diagnosis model comprises classifying fault data and training the probabilistic neural network fault diagnosis model.
When a fault diagnosis model is trained and constructed, firstly, fault data of a roller set are collected, then identifiers are added to the data to mark fault categories to which the fault data belong, and a data set is divided into a training data set and a testing data set according to a certain proportion.
And training a probabilistic neural network fault diagnosis model, respectively converting the training data set and the test data set into vectors, and performing normalization processing. And inputting the training data set into a probabilistic neural network fault diagnosis model, training the model, and testing the trained fault diagnosis model by using the test data set. And a proper smoothing factor of the probabilistic neural network is selected through multiple tests, so that the accuracy of the fault diagnosis model is higher.
And applying the obtained fault diagnosis model to a real-time diagnosis system to realize fault diagnosis of the roller set.
In practical application, the array pressure sensors are customized according to the size of an actual roller set, the array pressure sensors sense pressure distribution between the roller set, the array pressure sensors are connected with a data acquisition single chip microcomputer, the single chip microcomputer is connected with a computer through a serial port or a USB (universal serial bus), the computer sends instructions to control the single chip microcomputer to acquire pressure distribution data, the single chip microcomputer acquires the pressure distribution data between the roller set, the acquired data are filtered and denoised, then the data are transmitted to the computer according to a protocol format, the computer preprocesses the data, the acquired data are input into a trained fault diagnosis algorithm model, and decision-making judgment is carried out on the fault type of the roller set. A data interaction block diagram is shown in fig. 1.
Step S1 specifically includes:
s101: connecting the array type pressure sensor with a data acquisition single chip microcomputer, and connecting the single chip microcomputer to a computer, as shown in figure 1;
s102: turning on a power supply and setting parameters;
s103: the computer sends an acquisition instruction to the single chip microcomputer;
s104: the singlechip acquires data, performs filtering and denoising, and transmits the data to the computer according to a communication protocol;
s105: and the computer checks and stores the data and continuously sends the acquisition instruction.
The failure types are that the roller inclines left 1, the roller normally 2, the roller inclines right 3, and the roller has foreign matters 4, as shown in fig. 2 and fig. 3, wherein 5 is a legend.
The fault decision specifically comprises the following steps:
1. reading a data set, adding an identifier to each piece of data in the data set, and marking the data fault category;
2. dividing a data set into a training data set and a testing data set according to a certain proportion;
3. converting the data set into vectors and carrying out normalization processing;
4. selecting a proper smoothing factor and establishing a probabilistic neural network;
5. inputting a training data set into a neural network, and training a probabilistic neural network fault diagnosis model;
6. testing the fault diagnosis model;
7. carrying out a plurality of test experiments, and adjusting a smoothing factor to ensure that the accuracy is highest as possible;
8. collecting fault data, and inputting the fault data into a fault diagnosis model to perform fault prediction and judgment;
9. and obtaining a prediction result.
The partial prediction results are shown in fig. 4.
By adopting the technical scheme, compared with a method for diagnosing mechanical faults by using vibration and sound signals, the method greatly reduces the influence of external factors on the data accuracy in data acquisition; in the aspect of algorithm processing, the invention utilizes the probabilistic neural network to predict the fault, thereby avoiding the defects of local optimization, long training time and the like of the BP neural network. The method and the device have the advantage that the fault diagnosis of the roller set is well improved in real time and accuracy.
Finally, it is noted that the above-mentioned preferred embodiments illustrate rather than limit the invention, and that, although the invention has been described in detail with reference to the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims.
Claims (3)
1. A roller set equipment fault detection method based on a flexible array type pressure sensor is characterized by comprising the following steps: the method comprises the following steps:
s1: collecting pressure distribution data between roller sets by using an array type pressure sensor;
s2: the pressure distribution data is used as the fault data of the roller set, and the collected fault data is input into a probabilistic neural network to be trained to construct a fault diagnosis model;
s3: inputting the real-time collected fault data of the roller set into a fault diagnosis model for fault diagnosis and judgment to obtain the fault state of the roller set;
step S1 specifically includes:
s101: connecting the array type pressure sensor with a data acquisition single chip microcomputer, and connecting the single chip microcomputer to a computer;
s102: turning on a power supply and setting parameters;
s103: the computer sends an acquisition instruction to the single chip microcomputer;
s104: the method comprises the following steps that a single chip microcomputer collects pressure distribution data between roller sets, carries out filtering and denoising, and sends the pressure distribution data to a computer according to a communication protocol;
s105: the computer checks and stores the pressure distribution data and continues to send an acquisition instruction;
step S2 includes the following steps:
s201: adding an identifier to the collected pressure distribution data, and marking the fault category to which the fault data belongs;
s202: dividing the pressure distribution data into a training data set and a testing data set, converting the training data set and the testing data set into vectors, and performing normalization processing;
s203: selecting a smoothing factor, and establishing a probabilistic neural network fault diagnosis model;
s204: inputting the training data set into a probabilistic neural network fault diagnosis model for training;
s205: testing the trained fault diagnosis model by using a test data set;
s206: and (5) repeatedly executing the steps S202-S205 through the pressure distribution data acquired for multiple times, and adjusting the smoothing factor of the probabilistic neural network until the accuracy of the fault diagnosis model meets the requirement.
2. The method for detecting the faults of the roller set equipment based on the flexible array type pressure sensor as claimed in claim 1, wherein the method comprises the following steps: the single chip microcomputer is connected with the computer through a serial port or a USB.
3. The method for detecting the faults of the roller set equipment based on the flexible array type pressure sensor as claimed in claim 1, wherein the method comprises the following steps: the failure types in step S201 are that the roller is tilted left, the roller is normal, the roller is tilted right, and a foreign object is on the roller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810070059.6A CN108204892B (en) | 2018-01-24 | 2018-01-24 | Roller set equipment fault detection method based on flexible array type pressure sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810070059.6A CN108204892B (en) | 2018-01-24 | 2018-01-24 | Roller set equipment fault detection method based on flexible array type pressure sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108204892A CN108204892A (en) | 2018-06-26 |
| CN108204892B true CN108204892B (en) | 2020-02-07 |
Family
ID=62606387
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810070059.6A Active CN108204892B (en) | 2018-01-24 | 2018-01-24 | Roller set equipment fault detection method based on flexible array type pressure sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108204892B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108995225A (en) * | 2018-07-04 | 2018-12-14 | 合肥欧语自动化有限公司 | A kind of automation rolling device |
| CN113392936B (en) * | 2021-07-09 | 2022-09-02 | 四川英创力电子科技股份有限公司 | Oven fault diagnosis method based on machine learning |
| US20230244946A1 (en) * | 2022-01-28 | 2023-08-03 | International Business Machines Corporation | Unsupervised anomaly detection of industrial dynamic systems with contrastive latent density learning |
| CN115329493B (en) * | 2022-08-17 | 2023-07-14 | 兰州理工大学 | Impeller machinery fault detection method based on digital twin model of centrifugal pump |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203719796U (en) * | 2014-02-28 | 2014-07-16 | 厦门乃尔电子有限公司 | Pressure sensor flexible package structure for multipoint measurement |
| CN105976021A (en) * | 2016-05-24 | 2016-09-28 | 北京工业大学 | Fault diagnosis method for roller assembly of belt conveyor |
| CN106546362A (en) * | 2016-10-27 | 2017-03-29 | 中国科学院重庆绿色智能技术研究院 | A kind of capacitance pressure transducer, based on Graphene |
| CN106680616A (en) * | 2016-10-27 | 2017-05-17 | 北京智芯微电子科技有限公司 | Fault detection terminal of power distribution network and system thereof |
| CN107505548A (en) * | 2017-08-29 | 2017-12-22 | 华北电力大学（保定） | A kind of type local-discharge ultrasonic localization method based on flexible array sensor |
-
2018
- 2018-01-24 CN CN201810070059.6A patent/CN108204892B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203719796U (en) * | 2014-02-28 | 2014-07-16 | 厦门乃尔电子有限公司 | Pressure sensor flexible package structure for multipoint measurement |
| CN105976021A (en) * | 2016-05-24 | 2016-09-28 | 北京工业大学 | Fault diagnosis method for roller assembly of belt conveyor |
| CN106546362A (en) * | 2016-10-27 | 2017-03-29 | 中国科学院重庆绿色智能技术研究院 | A kind of capacitance pressure transducer, based on Graphene |
| CN106680616A (en) * | 2016-10-27 | 2017-05-17 | 北京智芯微电子科技有限公司 | Fault detection terminal of power distribution network and system thereof |
| CN107505548A (en) * | 2017-08-29 | 2017-12-22 | 华北电力大学（保定） | A kind of type local-discharge ultrasonic localization method based on flexible array sensor |
Non-Patent Citations (1)
| Title |
|---|
| 基于概率神经网络的离心式制冷机故障诊断;梁晴晴等;《暖通空调》;20151115;第45卷(第11期);第103页左栏第4段-104页左栏第3段 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108204892A (en) | 2018-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108204892B (en) | Roller set equipment fault detection method based on flexible array type pressure sensor | |
| CN106897543B (en) | Beam structure damage identification method of modal compliance curvature matrix norm | |
| CN109858104B (en) | Rolling bearing health assessment and fault diagnosis method and monitoring system | |
| CN110929934A (en) | Equipment failure prediction method and device, computer equipment and storage medium | |
| CN111914883A (en) | Spindle bearing state evaluation method and device based on deep fusion network | |
| CN111947928A (en) | Multi-source information fusion bearing fault prediction system and method | |
| CN112734977B (en) | Equipment risk early warning system and algorithm based on Internet of things | |
| CN104880217A (en) | Fault sensor information reconstruction method based on measured value association degree | |
| CN114255784A (en) | Substation equipment fault diagnosis method based on voiceprint recognition and related device | |
| CN111678699A (en) | Early fault monitoring and diagnosing method and system for rolling bearing | |
| CN114429152A (en) | Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption | |
| CN110727669B (en) | Electric power system sensor data cleaning device and cleaning method | |
| CN110320802B (en) | Complex system signal time sequence identification method based on data visualization | |
| CN114662619B (en) | Bridge monitoring system based on multi-source data fusion | |
| CN103743477A (en) | Mechanical failure detecting and diagnosing method and apparatus | |
| CN115201609A (en) | Voltage fluctuation consistency fault diagnosis method | |
| CN106500831B (en) | Detection method and device of vibration sensor | |
| KR102417694B1 (en) | Anomaly detection system and method in smart manufacturing environment using multiple AI techniques | |
| CN114186586A (en) | Damage identification method and equipment based on two-dimensional convolutional neural network | |
| CN114186624A (en) | Method and device for establishing damage identification model based on two-dimensional convolutional neural network | |
| CN113627885A (en) | Power grid power supply abnormity monitoring system and monitoring method thereof | |
| CN109993211B (en) | Damage identification method based on structural acceleration AR model coefficient | |
| CN113869373A (en) | Equipment abnormality detection method and device, computer equipment and storage medium | |
| CN113095364A (en) | Method, medium, and apparatus for extracting high-speed rail seismic event using convolutional neural network | |
| CN109271948A (en) | A kind of multichannel remote oscillation signal monitor system and monitoring method based on clustering algorithm |
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 |