CN118816994B - A perception monitoring system for smart factories - Google Patents
A perception monitoring system for smart factories Download PDFInfo
- Publication number
- CN118816994B CN118816994B CN202411266246.3A CN202411266246A CN118816994B CN 118816994 B CN118816994 B CN 118816994B CN 202411266246 A CN202411266246 A CN 202411266246A CN 118816994 B CN118816994 B CN 118816994B
- Authority
- CN
- China
- Prior art keywords
- photoelectric sensor
- data
- data vector
- neural network
- sampling time
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/048—Activation functions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention relates to the technical field of fault detection, and discloses a perception monitoring system of an intelligent factory, which comprises the following components: a first data collection module for collecting first data and encoding the first data into a first set of data vectors; the distance prediction module is used for inputting the first data vector set into the first neural network and outputting a distance reference value of a corresponding time point; the fault detection module is used for combining the first data vector and the distance reference value into a second data vector set according to time, inputting the second data vector set into the second neural network, outputting the probability of fault occurrence at each moment, and judging whether the fault occurs according to the probability; the fault elimination module is used for switching the standby photoelectric sensor to eliminate faults when faults exist; and the early warning module is used for triggering early warning when the standby photoelectric sensor fails. The invention fully considers the problem of abnormal fluctuation of the actual measurement distance caused by equipment or environmental factors, and the fault detection and early warning effects are more obvious.
Description
Technical Field
The invention relates to the technical field of fault detection, in particular to a perception monitoring system of an intelligent factory.
Background
In a certain logistics center intelligent factory, a production line comprises a conveyor belt, a sorting mechanical arm, a sensor system and bar code scanning equipment, and articles can be scanned by the bar code scanning equipment only in a limited position range, wherein a plurality of photoelectric sensors are distributed on two sides of the conveyor belt and used for judging the actual measured distance of the articles passing through the conveyor belt on the surfaces of the articles through light reflection. However, the actual measured distance of the object on the conveyor belt is affected by various factors, such as the temperature of the photoelectric sensor, the intensity of light emitted by the photoelectric sensor, the speed of the conveyor belt, the vibration frequency of the conveyor belt, the ambient temperature, and the humidity, so that the actual measured distance of the object cannot be compared with the fixed measured distance reference value.
Since the internal components of the photoelectric sensor are sensitive to temperature and humidity, if the ambient temperature and humidity change greatly, the sensitivity of the internal components of the photoelectric sensor is reduced or damaged, and the actual measured distance value of an object passing through the area generates abnormal fluctuation, but the fluctuation may not be enough to exceed the measured distance reference value, so that an alarm cannot be triggered.
If the speed of the conveyor belt and the vibration frequency change greatly, the light reflection measurement distance of the photoelectric sensor may cause abnormal fluctuation of the actual measurement distance value, but the fluctuation may not be enough to exceed the measurement distance reference value, so that an alarm cannot be triggered.
Disclosure of Invention
The invention provides a perception monitoring system of an intelligent factory, which solves the technical problems that in the related art, the actual measurement distance generates abnormal fluctuation for some other factors due to the fact that the actual measurement distance is compared with a measurement distance reference value based on the measurement distance value, but the fluctuation is insufficient to exceed the measurement distance reference value, so that the fault is difficult to find.
The invention provides a perception monitoring system of an intelligent factory, which comprises the following modules:
a first data collection module for collecting first data, the first data comprising:
Once at each sampling time point, the first data includes monitoring parameters of the photosensor, operating parameters of the conveyor belt, and environmental parameters.
The monitoring parameters of the photoelectric sensor comprise: the temperature, the emitted light intensity and the voltage of the photoelectric sensor, and the operation parameters of the conveyor belt comprise: the speed of the conveyor belt, the frequency of vibration of the conveyor belt, and environmental parameters include: light intensity, temperature, humidity; the sampling time point is the moment of data acquisition according to a preset time interval, and each time interval is one minute;
the first data collected at each sampling time point is encoded into a first data vector.
As a further refinement of the invention, the first data vector comprises dimensions representing values of all acquired parameters.
And the distance prediction module is used for inputting the first data vector set into a first neural network, and the first neural network outputs an upper limit reference value and a lower limit reference value of the photoelectric sensor measuring distance corresponding to the v-59 th to the v sampling time points.
It should be noted that, the first data vector set is formed by combining first data vectors from v-59 to v sampling time points, v represents the current sampling time point, if the interval between the v sampling time points and the sampling time point from the initial acquisition is less than 59, performing interpolation operation on the missing first data vector, where the interpolated first data vector is a zero vector;
As a further optimization scheme of the invention, the first neural network is a long-term memory network, and the long-term memory network comprises a forgetting gate, an input gate, a cell state and an output gate, and the output of the forgetting gate The calculation formula of (2) is as follows:
；
Wherein, Is a weight matrix of the forgetting gate,Is an offset item of the forgetting door,Is the hidden state of the previous time stepAnd concatenation of the t first data vector of the first set of data vectors,Is a Sigmoid activation function;
a first data vector equal to the p-th sampling time point;
；
output of input gate And candidate cell statusThe calculation formula of (2) is as follows:
；
；
Wherein, Is a weight matrix of the input gates,Is a bias term of the input gate,Is a hyperbolic tangent activation function,Is a weight matrix that computes candidate cell states,Is a bias term for candidate cell states;
Cell status Is based on the previous cell stateAnd the candidate cell state is weighted and synthesized, and the calculation formula is as follows:
；
Wherein, Representing element-by-element multiplication operations;
Outputting the activation value of the gate Hidden state with current time stepThe calculation formula of (2) is as follows:
；
；
Wherein, Is a matrix of weights for the output gates,Is a bias term for the output gate;
hidden state for each time step Processing, respectively calculating upper limit reference values by two independent full connection layersAnd a lower limit reference valueThe calculation formula is as follows:
；
；
Wherein, AndIs a weight matrix of two fully connected layers,AndIs the bias vector.
As a further optimization of the invention, a main penalty term is usedAnd penalty termAs a function of lossThe calculation formula is as follows:
；
；
；
Wherein the method comprises the steps of 、The upper and lower reference values of the measured distance predicted for time step k,、For the upper and lower reference values of the measured distance of time step k real,Is a weight coefficient with a default value of 1.
The fault detection module is used for splicing the first data vector with an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance of a corresponding sampling time point output by the first neural network to generate a second data vector; all the second data vectors are ordered according to the corresponding sampling time points to generate a second data vector set, the second data vector set is input into a second neural network, and the second neural network outputs a probability value that a photoelectric sensor has faults;
as a further optimization scheme of the invention, the method for splicing the upper limit reference value and the lower limit reference value of the photoelectric sensor measuring distance of the corresponding sampling time point output by the first data vector and the first neural network is as follows: the upper limit reference value and the lower limit reference value of the photoelectric sensor measurement distance are spliced to the first data vector.
The calculation formula of the second neural network is as follows:
；
；
；
；
；
；
Wherein, AndRepresenting the 1 st and the L-th second data vector of the second set of data vectors respectively,、、、Representing the weight parameters that can be trained,、、、Representing the bias parameters that may be trained,The dot product is represented by a graph of the dot product,Representing the first intermediate feature of the L-th,Representing a second intermediate feature of the L-th,Representing a third intermediate feature of the L-th,AndRepresenting the L-1 th and L-th output features respectively,A value indicating whether the photosensor has a fault, tanh indicating a tanh function,Representing a Sigmoid function.
As a further preferred embodiment of the present invention,A probability value of greater than 0.5 indicates that the photosensor has a fault, otherwise indicates that the photosensor has no fault.
The fault elimination module is used for controlling the actuating mechanism to switch the photoelectric sensor with the fault into a standby photoelectric sensor;
As a further optimization scheme of the invention, the actuating mechanism is a turntable carrying the photoelectric sensor and a driving motor for driving the turntable, and the fault elimination module switches the standby photoelectric sensor by controlling the driving motor to rotate the turntable.
And the early warning module is used for triggering early warning if the standby photoelectric sensor fails after being switched into the standby photoelectric sensor.
As a further optimization scheme of the invention, the early warning comprises light and sound warning and informs operation and maintenance personnel to perform on-site inspection and processing.
As a further optimized scheme of the present invention, a sensing and monitoring method of an intelligent factory is provided, which performs the following steps based on the sensing and monitoring system of an intelligent factory:
Step 301, collecting monitoring parameters of a photoelectric sensor, operation parameters of a conveyor belt and environmental parameters at each sampling time point as first data, and encoding and combining the first data collected at each sampling time point into a first data vector set;
Step 302, inputting a first data vector set into a first neural network, wherein the first neural network outputs an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance corresponding to a v-59 th to a v-th sampling time point;
Step 303, splicing the first data vector with an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance of a corresponding sampling time point output by the first neural network to generate a second data vector; all the second data vectors are ordered according to the corresponding sampling time points to generate a second data vector set, the second data vector set is input into a second neural network, and the second neural network outputs a probability value that a photoelectric sensor has faults;
Step 304, judging whether the photoelectric sensor has faults according to the probability value output by the second neural network, and controlling the executing mechanism to switch the photoelectric sensor with faults into a standby photoelectric sensor if the faults exist;
Step 305, if the standby photoelectric sensor fails after switching to the standby photoelectric sensor, triggering early warning.
The invention has the beneficial effects that:
According to the invention, a measurement distance reference value at the current moment is output according to the monitoring parameters of the photoelectric sensor, the running parameters of the conveyor belt and the environmental parameters, the measurement distance reference value is dynamically adjusted, and compared with an actual measurement distance value to judge whether the actual measurement distance is abnormal, and if the actual measurement distance is abnormal, the fault possibly exists.
According to the invention, the measured distance reference value at the current moment is obtained through calculation, the problem of abnormal fluctuation of the actual measured distance caused by equipment or environmental factors is fully considered, and the fault detection and early warning effects are more obvious.
Drawings
FIG. 1 is a schematic block diagram of a perception monitoring system of an intelligent plant of the present invention;
FIG. 2 is a flow chart of a method of perceptual monitoring of an intelligent plant of the present invention.
Detailed Description
The subject matter described herein will now be discussed with reference to example embodiments. It is to be understood that these embodiments are merely discussed so that those skilled in the art may better understand and implement the subject matter described herein and that changes may be made in the function and arrangement of the elements discussed without departing from the scope of the disclosure herein. Various examples may omit, replace, or add various procedures or components as desired. In addition, features described with respect to some examples may be combined in other examples as well.
At least one embodiment of the present invention discloses a perception monitoring system of an intelligent factory, as shown in fig. 1, comprising:
a first data collection module 100 for collecting first data, the first data comprising:
Once at each sampling time point, the first data includes monitoring parameters of the photosensor, operating parameters of the conveyor belt, and environmental parameters.
The monitoring parameters of the photoelectric sensor comprise: the temperature, the emitted light intensity and the voltage of the photoelectric sensor, and the operation parameters of the conveyor belt comprise: the speed of the conveyor belt, the frequency of vibration of the conveyor belt, and environmental parameters include: light intensity, temperature, humidity; the sampling time point is the moment of data acquisition according to a preset time interval, and each time interval is one minute;
the first data collected at each sampling time point is encoded into a first data vector.
In one embodiment of the invention, the first data vector comprises dimensions representing values of all acquired parameters.
The distance prediction module 200 is configured to input the first data vector set into a first neural network, where the first neural network outputs an upper limit reference value and a lower limit reference value of the measured distance of the photoelectric sensor corresponding to the v-59 th to the v-th sampling time point.
It should be noted that, the first data vector set is formed by combining first data vectors from v-59 to v sampling time points, v represents the current sampling time point, if the interval between the v sampling time points and the sampling time point from the initial acquisition is less than 59, performing interpolation operation on the missing first data vector, where the interpolated first data vector is a zero vector;
in one embodiment of the invention, the first neural network is a long-term memory network comprising a forgetting gate, an input gate, a cell state and an output gate, the output of the forgetting gate The calculation formula of (2) is as follows:
；
Wherein, Is a weight matrix of the forgetting gate,Is an offset item of the forgetting door,Is the hidden state of the previous time stepAnd concatenation of the t first data vector of the first set of data vectors,Is a Sigmoid activation function;
a first data vector equal to the p-th sampling time point;
；
output of input gate And candidate cell statusThe calculation formula of (2) is as follows:
；
；
Wherein, Is a weight matrix of the input gates,Is a bias term of the input gate,Is a hyperbolic tangent activation function,Is a weight matrix that computes candidate cell states,Is a bias term for candidate cell states;
Cell status Is based on the previous cell stateAnd the candidate cell state is weighted and synthesized, and the calculation formula is as follows:
；
Wherein, Representing element-by-element multiplication operations;
Outputting the activation value of the gate Hidden state with current time stepThe calculation formula of (2) is as follows:
；
；
Wherein, Is a matrix of weights for the output gates,Is a bias term for the output gate;
hidden state for each time step Processing, respectively calculating upper limit reference values by two independent full connection layersAnd a lower limit reference valueThe calculation formula is as follows:
；
；
Wherein, AndIs a weight matrix of two fully connected layers,AndIs the bias vector.
Using master penalty entriesAnd penalty termAs a function of lossThe calculation formula is as follows:
；
；
；
Wherein the method comprises the steps of 、The upper and lower reference values of the measured distance predicted for time step k,、For the upper and lower reference values of the measured distance of time step k real,Is a weight coefficient with a default value of 1.
The fault detection module 300 is configured to splice the first data vector with an upper limit reference value and a lower limit reference value of a measurement distance of the photoelectric sensor at a corresponding sampling time point output by the first neural network to generate a second data vector; all the second data vectors are ordered according to the corresponding sampling time points to generate a second data vector set, the second data vector set is input into a second neural network, and the second neural network outputs a probability value that a photoelectric sensor has faults;
In one embodiment of the present invention, the method for splicing the first data vector with the upper limit reference value and the lower limit reference value of the photoelectric sensor measurement distance of the corresponding sampling time point output by the first neural network is: the upper limit reference value and the lower limit reference value of the photoelectric sensor measurement distance are spliced to the first data vector.
The calculation formula of the second neural network is as follows:
；
；
；
；
；
；
Wherein, AndRepresenting the 1 st and the L-th second data vector of the second set of data vectors respectively,、、、Representing the weight parameters that can be trained,、、、Representing the bias parameters that may be trained,The dot product is represented by a graph of the dot product,Representing the first intermediate feature of the L-th,Representing a second intermediate feature of the L-th,Representing a third intermediate feature of the L-th,AndRepresenting the L-1 th and L-th output features respectively,A value indicating whether the photosensor has a fault, tanh indicating a tanh function,Representing a Sigmoid function.
In one embodiment of the present invention,A probability value of greater than 0.5 indicates that the photosensor has a fault, otherwise indicates that the photosensor has no fault.
A failure elimination module 400 for controlling the actuator to switch the failed photoelectric sensor to a spare photoelectric sensor;
In one embodiment of the invention, the actuator is a turntable carrying a photosensor and a drive motor for driving the turntable, and the fault elimination module 400 switches the spare photosensor by controlling the drive motor to rotate the turntable.
And the early warning module 500 is used for triggering early warning if the standby photoelectric sensor fails after being switched into the standby photoelectric sensor.
In one embodiment of the invention, the pre-warning includes light and audible warnings and notifies the service personnel to perform on-site inspections and processes.
As shown in fig. 2, in one embodiment of the present invention, there is provided a perception monitoring method of an intelligent plant, which performs the following steps based on the perception monitoring system of an intelligent plant as described above:
Step 301, collecting monitoring parameters of a photoelectric sensor, operation parameters of a conveyor belt and environmental parameters at each sampling time point as first data, and encoding and combining the first data collected at each sampling time point into a first data vector set;
Step 302, inputting a first data vector set into a first neural network, wherein the first neural network outputs an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance corresponding to a v-59 th to a v-th sampling time point;
Step 303, splicing the first data vector with an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance of a corresponding sampling time point output by the first neural network to generate a second data vector; all the second data vectors are ordered according to the corresponding sampling time points to generate a second data vector set, the second data vector set is input into a second neural network, and the second neural network outputs a probability value that a photoelectric sensor has faults;
Step 304, judging whether the photoelectric sensor has faults according to the probability value output by the second neural network, and controlling the executing mechanism to switch the photoelectric sensor with faults into a standby photoelectric sensor if the faults exist;
Step 305, if the standby photoelectric sensor fails after switching to the standby photoelectric sensor, triggering early warning.
While the embodiments of the present invention have been described above, the embodiments are not limited to the above-described embodiments, which are intended to be illustrative only and not limiting, and many equivalents thereof may be made by those of ordinary skill in the art in light of the present disclosure, which fall within the scope of the embodiments.
Claims (8)
1. A perception monitoring system of an intelligent factory, comprising the following modules:
A first data collection module for collecting first data and encoding the first data into a first set of data vectors;
The distance prediction module is used for inputting the first data vector set into a first neural network, and the first neural network outputs an upper limit reference value and a lower limit reference value of the photoelectric sensor measuring distance corresponding to the v-59 th to the v sampling time points;
The fault detection module is used for splicing the first data vector with an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance of a corresponding sampling time point output by the first neural network to generate a second data vector; all the second data vectors are ordered according to the corresponding sampling time points to generate a second data vector set, the second data vector set is input into a second neural network, and the second neural network outputs a probability value that a photoelectric sensor has faults;
the fault elimination module is used for controlling the actuating mechanism to switch the photoelectric sensor with the fault into a standby photoelectric sensor;
the early warning module is used for triggering early warning if the standby photoelectric sensor fails after being switched into the standby photoelectric sensor;
The first data in the first data collection module includes:
collecting monitoring parameters of a primary photoelectric sensor, operation parameters of a conveyor belt and environmental parameters at each sampling time point;
Wherein the monitoring parameters of the photoelectric sensor include: temperature, emitted light intensity, voltage of the photosensor;
The operating parameters of the conveyor belt include: the speed of the conveyor belt and the frequency of vibration of the conveyor belt;
the environmental parameters include: light intensity, temperature, humidity;
The sampling time point is the moment of data acquisition according to a preset time interval, and each time interval is one minute;
The first data vector set is formed by combining first data vectors from v-59 to v sampling time points, v represents the current sampling time point, if the interval between the v sampling time points and the sampling time point from the initial acquisition is smaller than 59, interpolation operation is carried out on the missing first data vector, and the interpolated first data vector is zero vector.
2. The intelligent plant perception monitoring system of claim 1, wherein the first neural network in the distance prediction module is calculated as follows:
the first neural network is a long-term memory network comprising a forgetting gate, an input gate, a cell state and an output gate, and the output of the forgetting gate The calculation formula of (2) is as follows:
；
Wherein, Is a weight matrix of the forgetting gate,Is an offset item of the forgetting door,Is the hidden state of the previous time stepAnd concatenation of the t first data vector of the first set of data vectors,Is a Sigmoid activation function;
a first data vector equal to the p-th sampling time point;
；
output of input gate And candidate cell statusThe calculation formula of (2) is as follows:
；
；
Wherein, Is a weight matrix of the input gates,Is a bias term of the input gate,Is a hyperbolic tangent activation function,Is a weight matrix that computes candidate cell states,Is a bias term for candidate cell states;
Cell status Is based on the previous cell stateAnd the candidate cell state is weighted and synthesized, and the calculation formula is as follows:
；
Wherein, Representing element-by-element multiplication operations;
Outputting the activation value of the gate Hidden state with current time stepThe calculation formula of (2) is as follows:
；
；
Wherein, Is a matrix of weights for the output gates,Is a bias term for the output gate;
hidden state for each time step Processing, respectively calculating upper limit reference values by two independent full connection layersAnd a lower limit reference valueThe calculation formula is as follows:
；
；
Wherein, AndIs a weight matrix of two fully connected layers,AndIs the bias vector.
3. The intelligent plant perception monitoring system of claim 2, wherein a primary loss term is usedAnd penalty termAs a function of lossThe calculation formula is as follows:
；
；
；
Wherein the method comprises the steps of 、The upper and lower reference values of the measured distance predicted for time step k,、For the upper and lower reference values of the measured distance of time step k real,Is a weight coefficient with a default value of 1.
4. The intelligent plant's perception monitoring system of claim 1, wherein the second neural network in the fault detection module is calculated as follows:
；
；
；
；
；
；
Wherein, AndRepresenting the 1 st and the L-th second data vector of the second set of data vectors respectively,、、、Representing the weight parameters that can be trained,、、、Representing the bias parameters that may be trained,The dot product is represented by a graph of the dot product,Representing the first intermediate feature of the L-th,Representing a second intermediate feature of the L-th,Representing a third intermediate feature of the L-th,AndRepresenting the L-1 th and L-th output features respectively,A value indicating whether the photosensor has a fault, tanh indicating a tanh function,Representing a Sigmoid function.
5. A perception monitoring system for an intelligent plant as claimed in claim 4, wherein,A probability value of greater than 0.5 indicates that the photosensor has a fault, otherwise indicates that the photosensor has no fault.
6. The intelligent plant perception monitoring system of claim 1, wherein the actuator comprises:
the fault elimination module switches the standby photoelectric sensor by controlling the driving motor to rotate the turntable.
7. The intelligent factory perception monitoring system of claim 1, wherein the early warning of the early warning module comprises:
light and sound warnings and notices to the service personnel for on-site inspection and processing.
8. A method of perception monitoring of an intelligent plant, characterized in that the following steps are performed based on a perception monitoring system of an intelligent plant as claimed in any of claims 1-7:
Step 301, collecting monitoring parameters of a photoelectric sensor, operation parameters of a conveyor belt and environmental parameters at each sampling time point as first data, and encoding and combining the first data collected at each sampling time point into a first data vector set;
Step 302, inputting a first data vector set into a first neural network, wherein the first neural network outputs an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance corresponding to a v-59 th to a v-th sampling time point;
Step 303, splicing the first data vector with an upper limit reference value and a lower limit reference value of a photoelectric sensor measuring distance of a corresponding sampling time point output by the first neural network to generate a second data vector; all the second data vectors are ordered according to the corresponding sampling time points to generate a second data vector set, the second data vector set is input into a second neural network, and the second neural network outputs a probability value that a photoelectric sensor has faults;
Step 304, judging whether the photoelectric sensor has faults according to the probability value output by the second neural network, and controlling the executing mechanism to switch the photoelectric sensor with faults into a standby photoelectric sensor if the faults exist;
Step 305, if the standby photoelectric sensor fails after switching to the standby photoelectric sensor, triggering early warning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411266246.3A CN118816994B (en) | 2024-09-11 | 2024-09-11 | A perception monitoring system for smart factories |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411266246.3A CN118816994B (en) | 2024-09-11 | 2024-09-11 | A perception monitoring system for smart factories |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118816994A CN118816994A (en) | 2024-10-22 |
| CN118816994B true CN118816994B (en) | 2024-11-26 |
Family
ID=93073082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411266246.3A Active CN118816994B (en) | 2024-09-11 | 2024-09-11 | A perception monitoring system for smart factories |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118816994B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119066893B (en) * | 2024-11-06 | 2025-01-17 | 合肥航太电物理技术有限公司 | Aviation arc fault damage simulation method |
| CN119622605B (en) * | 2025-02-12 | 2025-08-15 | 国网山西省电力公司超高压变电分公司 | Remote measurement data analysis method and device for transformer substation and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106338264A (en) * | 2016-08-19 | 2017-01-18 | 江苏大学 | Fault diagnosis method for switch reluctance BSG position sensor used for hybrid power vehicle |
| CN106586796A (en) * | 2016-11-15 | 2017-04-26 | 王蕊 | System and method for monitoring state of escalator |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3843890A (en) * | 1973-07-27 | 1974-10-22 | Du Pont | Optical-electrical web inspection system |
| GB2272327B (en) * | 1989-11-02 | 1994-09-28 | Combustion Eng | Display device |
| WO2017196821A1 (en) * | 2016-05-09 | 2017-11-16 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for the industrial internet of things |
| WO2019028269A2 (en) * | 2017-08-02 | 2019-02-07 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for detection in an industrial internet of things data collection environment with large data sets |
| KR102611399B1 (en) * | 2022-02-21 | 2023-12-08 | 다겸 주식회사 | Abnormal detection device using 1-demension autoencoder |
| CN116204825A (en) * | 2023-02-10 | 2023-06-02 | 湖北文理学院 | A data-driven fault detection method for production line equipment |
| CN116654390B (en) * | 2023-04-28 | 2025-06-20 | 广西中烟工业有限责任公司 | Cigarette Pack Appearance Inspection System |
| CN117353455B (en) * | 2023-10-17 | 2024-03-29 | 济南泉晓电气设备有限公司 | Supervision method of power transmission and transformation system based on artificial intelligence |
| CN118386024A (en) * | 2024-04-17 | 2024-07-26 | 万可工业科技(山东)有限公司 | Machine tool based on artificial intelligence and fault detection method thereof |
-
2024
- 2024-09-11 CN CN202411266246.3A patent/CN118816994B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106338264A (en) * | 2016-08-19 | 2017-01-18 | 江苏大学 | Fault diagnosis method for switch reluctance BSG position sensor used for hybrid power vehicle |
| CN106586796A (en) * | 2016-11-15 | 2017-04-26 | 王蕊 | System and method for monitoring state of escalator |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118816994A (en) | 2024-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118816994B (en) | A perception monitoring system for smart factories | |
| KR101967339B1 (en) | System and Method for Diagnosing Fault and Backup of ADAS Sensors based on Deep Learning | |
| Vachtsevanos et al. | Fault prognosis using dynamic wavelet neural networks | |
| CA2433941C (en) | Adaptive modeling of changed states in predictive condition monitoring | |
| KR102156858B1 (en) | Method for diagnosing and predicting robot arm's failure | |
| CN118921366B (en) | Circuit breaker control system based on cloud data | |
| CN119293664A (en) | A device operation evaluation method based on multi-source data fusion | |
| CN119721876B (en) | A metal casting process quality monitoring system and method for intelligent manufacturing | |
| CN119803822A (en) | A structural health monitoring system for long-span bridge beams based on deformation analysis | |
| CN120316629B (en) | Data anomaly detection method based on multi-scale feature extraction | |
| CN112052714A (en) | Data-driven machine learning for modeling aircraft sensors | |
| CN120524398B (en) | Crane vessel operability prediction method based on BP neural network | |
| JP2009259020A (en) | Anomaly detection apparatus | |
| CN120336973A (en) | Intelligent diagnosis method, system and medium for urban rail equipment health based on cloud platform | |
| CN119688009A (en) | Intelligent sensing method for state of pressure forming equipment by fusing multi-source sensor signals | |
| CN119312205A (en) | A method and system for identifying faults of intelligent fire fighting equipment | |
| CN119004012A (en) | Container loading and unloading detection control method and system based on artificial intelligence | |
| KR20230087661A (en) | Solar power generation forecasting method and integrated monitoring system based on it | |
| KR20220068799A (en) | System for detecting error of automation equipment and method thereof | |
| CN120449035A (en) | A method and system for intelligent monitoring and early warning of thermal power plant equipment defects based on a multimodal large model | |
| CN118816980B (en) | Plug-in terminal intelligent monitoring system | |
| CN120067868B (en) | Overhead crane fault prediction method, device, storage medium and electronic device | |
| CN118887044B (en) | RPA service processing method and system based on artificial intelligence | |
| KR102831727B1 (en) | Solar power generation equipment fault prediction system based on additive learning | |
| Li et al. | Research on Sensor Fault Isolation Method for Transmission Systems Based on Convolutional Autoencoder |
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 |