CN110696049A - Predictive maintenance method for palletizing robot - Google Patents
Predictive maintenance method for palletizing robot Download PDFInfo
- Publication number
- CN110696049A CN110696049A CN201911064556.6A CN201911064556A CN110696049A CN 110696049 A CN110696049 A CN 110696049A CN 201911064556 A CN201911064556 A CN 201911064556A CN 110696049 A CN110696049 A CN 110696049A
- Authority
- CN
- China
- Prior art keywords
- palletizing robot
- image data
- servo motor
- loss function
- predictive maintenance
- 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.)
- Granted
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 20
- 239000013598 vector Substances 0.000 claims abstract description 37
- 238000001914 filtration Methods 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 238000011478 gradient descent method Methods 0.000 claims abstract description 7
- 230000000877 morphologic effect Effects 0.000 claims abstract description 7
- 238000006243 chemical reaction Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 42
- 238000012545 processing Methods 0.000 claims description 3
- 230000002035 prolonged effect Effects 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 description 7
- 238000012544 monitoring process Methods 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 239000002904 solvent Substances 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
- B25J19/0066—Means or methods for maintaining or repairing manipulators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Manipulator (AREA)
- Numerical Control (AREA)
Abstract
The invention discloses a predictive maintenance method for a palletizing robot. The robot palletizer image signal acquisition method comprises the steps of acquiring an image signal of the robot palletizer through a camera, carrying out format conversion on the image signal, storing image data, and uploading the image data to a cloud server; carrying out filtering pretreatment on the image data, carrying out target detection on the image data subjected to filtering pretreatment and carrying out morphological treatment on the image data; collecting coordinates of objects grabbed by the palletizing robot, and determining the comprehensive movement speed deviation of a servo motor of the palletizing robot; determining fault parameter vectors of the palletizing robot according to the motion speed deviation of a servo motor of the palletizing robot; establishing a coefficient vector and a loss function, and iterating the loss function by adopting a gradient descent method to minimize the loss function; performing predictive maintenance according to the minimized loss function.
Description
Technical Field
The invention relates to the technical field of predictive maintenance of a palletizing robot, in particular to a predictive maintenance method of the palletizing robot.
Background
The palletizing robot is more and more widely applied in the fields of packaging, logistics and automation. Once the palletizing robot breaks down, the electromechanical system of the palletizing robot is not easy to locate, thereby not only bringing certain difficulty to the maintenance of workers, but also greatly influencing the production schedule and the production efficiency.
Currently, most palletizing robot application occasions have no predictive maintenance. Some application occasions depend on experienced workers to perform predictive maintenance, the subjectivity is strong, the maintenance effect is greatly influenced by the experience of maintenance personnel, and the labor cost is high.
And constructing an operation state monitoring model according to historical data of operation and faults of the palletizing robot. And monitoring the real-time running state of the palletizing robot according to the running state monitoring model. And once the equipment running state does not meet the set normal running condition, performing predictive maintenance, starting maintenance early warning or entering a low-workload state, thereby avoiding the influence on the production schedule and the production efficiency caused by equipment damage.
Disclosure of Invention
The invention provides a predictive maintenance method for a palletizing robot in order to give out early warning of the palletizing robot so as to carry out predictive maintenance, and the invention provides the following technical scheme:
a method of predictive maintenance of a palletizing robot, comprising the steps of:
step 1: collecting image signals of the palletizing robot through a camera, performing format conversion on the image signals, storing the image data, and uploading the image data to a cloud server; carrying out filtering pretreatment on the image data, carrying out target detection on the image data subjected to filtering pretreatment and carrying out morphological treatment on the image data;
step 2: collecting coordinates of objects grabbed by the palletizing robot, and determining the comprehensive movement speed deviation of a servo motor of the palletizing robot;
and step 3: determining fault parameter vectors of the palletizing robot according to the motion speed deviation of a servo motor of the palletizing robot;
and 4, step 4: establishing a coefficient vector and a loss function, and iterating the loss function by adopting a gradient descent method to minimize the loss function;
and 5: performing predictive maintenance according to the minimized loss function.
Preferably, the step 1 specifically comprises:
step 1.1: collecting image signals of the palletizing robot through a camera, converting the image signals into RGB image data in a format, storing the RGB image data into a memory, and uploading the image data to a cloud server;
step 1.2: carrying out Gaussian filtering pretreatment on the RGB image data stored in the memory to inhibit noise in the RGB image data;
step 1.3: and carrying out target detection on the image data subjected to filtering pretreatment and carrying out morphological processing.
Preferably, the step 2 specifically comprises:
step 2.1: acquiring coordinates of objects grabbed by the palletizing robot, comparing the coordinates of the objects grabbed by the palletizing robot with image data, and determining the grabbing success rate of the palletizing robot;
step 2.2: determining the movement speed deviation of a servo motor of the palletizing robot, and expressing the comprehensive movement speed deviation of the servo motor of the palletizing robot by the following formula:
wherein x is(2)Deviation of the movement speed of a servomotor of a palletizing robot, DiIs the motion speed deviation of the ith servo motor, n is the number of servo motors, ViThe set movement speed for the ith servo motor.
Preferably, the step 3 specifically comprises: determining a fault parameter vector of the palletizing robot according to the movement speed deviation of a servo motor of the palletizing robot, and expressing the fault parameter vector of the palletizing robot through the following formula:
x＝(x(1)，x(2)，x(3)，1) (2)
wherein x is fault parameter vector of the palletizing robot, and x(1)To get success rate, x(3)The service time of the palletizing robot is prolonged.
Preferably, the step 4 specifically includes:
step 4.1: establishing a coefficient vector, representing the coefficient vector by:
w＝(w(1)，w(2)，w(3)，b) (3)
wherein w is a coefficient vector;
establishing a hypothesis function from the coefficient vector, the hypothesis function being represented by:
wherein h isw(x) For a hypothetical function, T is transposed;
step 4.2: from the hypothesis function, a loss function is found, which is expressed by:
wherein J (w) is a loss function, m is the number of samples in the dataset, and y is a class label;
step 4.3: iterating the loss function by adopting a gradient descent method, minimizing the loss function, solving the optimal value of w, and expressing the optimal value of w after the loss function is minimized by the following formula:
where α is the step size.
Preferably, two states of "faulty" and "non-faulty" are represented by the category label y, and when y is 1, it represents "faulty", and when y is 0, it represents "non-faulty".
Preferably, the step size α is 0.15.
Preferably, the step 5 specifically comprises:
step 5.1: performing predictive maintenance according to the minimized loss function, determining whether to perform early warning on the palletizing robot or not according to the fault parameter vector, and when wx is used0＞γwxcThen, a warning is given, wherein x0Is a fault parameter vector, x, of the palletizing robot at the current momentcThe fault parameter vector of the palletizing robot when leaving the factory is gamma which is a threshold coefficient;
step 5.2: when the palletizing robot enters an early warning state, a servo motor of the palletizing robot is adjusted to enter a low working load state, and the speed of the servo motor entering the low working load state is represented by the following formula:
wherein, VLowV is the speed of the servo motor in a low working load state, and is the movement speed set by the servo motor;
step 5.3: adjusting the maximum working time of the palletizing robot in an early warning state, and expressing the maximum working time through the following formula:
wherein, T0Is the maximum working time T of the palletizing robot in an early warning stateWThe maximum working time of the day is specified when the palletizing robot leaves a factory.
Preferably, the threshold coefficient ranges from 15% to 20%.
The invention has the following beneficial effects:
according to the invention, an operation state monitoring model is constructed according to the historical data of operation and faults of the palletizing robot. And monitoring the real-time running state of the palletizing robot according to the running state monitoring model. And once the equipment running state does not meet the set normal running condition, performing predictive maintenance, starting maintenance early warning or entering a low-workload state, thereby avoiding the influence on the production schedule and the production efficiency caused by equipment damage. The defects of strong subjectivity and high labor cost of manual maintenance are overcome.
Drawings
Fig. 1 is a flow chart of a palletizing robot predictive maintenance.
Detailed Description
The present invention will be described in detail with reference to specific examples.
The first embodiment is as follows:
according to fig. 1, the invention provides a palletizing robot predictive maintenance method, which comprises the following steps:
step 1: collecting image signals of the palletizing robot through a camera, performing format conversion on the image signals, storing the image data, and uploading the image data to a cloud server; carrying out filtering pretreatment on the image data, carrying out target detection on the image data subjected to filtering pretreatment and carrying out morphological treatment on the image data;
the step 1 specifically comprises the following steps:
step 1.1: collecting image signals of the palletizing robot through a camera, converting the image signals into RGB image data in a format, storing the RGB image data into a memory, and uploading the image data to a cloud server;
step 1.2: carrying out Gaussian filtering pretreatment on the RGB image data stored in the memory to inhibit noise in the RGB image data;
step 1.3: and carrying out target detection on the image data subjected to filtering pretreatment and carrying out morphological processing.
Step 2: collecting coordinates of objects grabbed by the palletizing robot, and determining the comprehensive movement speed deviation of a servo motor of the palletizing robot;
the step 2 specifically comprises the following steps:
step 2.1: acquiring coordinates of objects grabbed by the palletizing robot, comparing the coordinates of the objects grabbed by the palletizing robot with image data, and determining the grabbing success rate of the palletizing robot;
step 2.2: determining the movement speed deviation of a servo motor of the palletizing robot, and expressing the comprehensive movement speed deviation of the servo motor of the palletizing robot by the following formula:
wherein x is(2)Deviation of the movement speed of a servomotor of a palletizing robot, DiIs the motion speed deviation of the ith servo motor, n is the number of servo motors, ViIs the ithThe set movement speed of the servo motor.
And step 3: determining a fault parameter vector of the palletizing robot according to the movement speed deviation of a servo motor of the palletizing robot, and expressing the fault parameter vector of the palletizing robot through the following formula:
x＝(x(1)，x(2)，x(3)，1)
wherein x is fault parameter vector of the palletizing robot, and x(1)To get success rate, x(3)The service time of the palletizing robot is prolonged.
And 4, step 4: establishing a coefficient vector and a loss function, and iterating the loss function by adopting a gradient descent method to minimize the loss function;
the step 4 specifically comprises the following steps:
step 4.1: establishing a coefficient vector, representing the coefficient vector by:
w＝(w(1)，w(2)，w(3)，b) (3)
wherein w is a coefficient vector;
establishing a hypothesis function from the coefficient vector, the hypothesis function being represented by:
wherein h isw(x) For a hypothetical function, T is transposed;
step 4.2: from the hypothesis function, a loss function is found, which is expressed by:
wherein J (w) is a loss function, m is the number of samples in the dataset, and y is a class label;
step 4.3: iterating the loss function by adopting a gradient descent method, minimizing the loss function, solving the optimal value of w, and expressing the optimal value of w after the loss function is minimized by the following formula:
wherein α is the step size, and α is 0.15; the iterative calculation number R is 900.
This step size keeps a moderate iteration speed; it does not happen that the iteration is too fast and the optimal solution may be missed. It does not happen that the iteration speed is too slow to cause the iteration to fail to end.
And 5: performing predictive maintenance according to the minimized loss function;
the step 5 specifically comprises the following steps:
step 5.1: performing predictive maintenance according to the minimized loss function, determining whether to perform early warning on the palletizing robot or not according to the fault parameter vector, and when wx is used0＞γwxcThen, a warning is given, wherein x0Is a fault parameter vector, x, of the palletizing robot at the current momentcThe fault parameter vector of the palletizing robot when leaving the factory is gamma which is a threshold coefficient;
step 5.2: when the palletizing robot enters an early warning state, a servo motor of the palletizing robot is adjusted to enter a low working load state, and the speed of the servo motor entering the low working load state is represented by the following formula:
wherein, VLowV is the speed of the servo motor in a low working load state, and is the movement speed set by the servo motor;
step 5.3: adjusting the maximum working time of the palletizing robot in an early warning state, and expressing the maximum working time through the following formula:
wherein, T0Is the maximum working time T of the palletizing robot in an early warning stateWThe maximum working time of the day specified by the delivery of the palletizing robotAnd (3) removing the solvent.
After the early warning of the starting maintenance of the equipment, if the equipment is still necessary to be used because of the urgent production task, the equipment is recommended to be set and operated in a low working load state, the speed of a servo motor is reduced, and the daily maximum working time of the equipment is reduced, so that the service life of the equipment is prevented from being seriously influenced by the over-strain of the equipment.
The above description is only a preferred embodiment of the predictive maintenance method for the palletizing robot, and the protection scope of the predictive maintenance method for the palletizing robot is not limited to the above embodiments, and all technical solutions belonging to the idea belong to the protection scope of the present invention. It should be noted that modifications and variations which do not depart from the gist of the invention will be those skilled in the art to which the invention pertains and which are intended to be within the scope of the invention.
Claims (9)
1. A predictive maintenance method for a palletizing robot is characterized by comprising the following steps: the method comprises the following steps:
step 1: collecting image signals of the palletizing robot through a camera, performing format conversion on the image signals, storing the image data, and uploading the image data to a cloud server; carrying out filtering pretreatment on the image data, carrying out target detection on the image data subjected to filtering pretreatment and carrying out morphological treatment on the image data;
step 2: collecting coordinates of objects grabbed by the palletizing robot, and determining the comprehensive movement speed deviation of a servo motor of the palletizing robot;
and step 3: determining fault parameter vectors of the palletizing robot according to the motion speed deviation of a servo motor of the palletizing robot;
and 4, step 4: establishing a coefficient vector and a loss function, and iterating the loss function by adopting a gradient descent method to minimize the loss function;
and 5: performing predictive maintenance according to the minimized loss function.
2. The predictive maintenance method of a palletizing robot as claimed in claim 1, wherein: the step 1 specifically comprises the following steps:
step 1.1: collecting image signals of the palletizing robot through a camera, converting the image signals into RGB image data in a format, storing the RGB image data into a memory, and uploading the image data to a cloud server;
step 1.2: carrying out Gaussian filtering pretreatment on the RGB image data stored in the memory to inhibit noise in the RGB image data;
step 1.3: and carrying out target detection on the image data subjected to filtering pretreatment and carrying out morphological processing.
3. The predictive maintenance method of a palletizing robot as claimed in claim 1, wherein: the step 2 specifically comprises the following steps:
step 2.1: acquiring coordinates of objects grabbed by the palletizing robot, comparing the coordinates of the objects grabbed by the palletizing robot with image data, and determining the grabbing success rate of the palletizing robot;
step 2.2: determining the movement speed deviation of a servo motor of the palletizing robot, and expressing the comprehensive movement speed deviation of the servo motor of the palletizing robot by the following formula:
wherein x is(2)Deviation of the movement speed of a servomotor of a palletizing robot, DiIs the motion speed deviation of the ith servo motor, n is the number of servo motors, ViThe set movement speed for the ith servo motor.
4. The predictive maintenance method of a palletizing robot as claimed in claim 1, wherein: the step 3 specifically comprises the following steps: determining a fault parameter vector of the palletizing robot according to the movement speed deviation of a servo motor of the palletizing robot, and expressing the fault parameter vector of the palletizing robot through the following formula:
x＝(x(1)，x(2)，x(3)，1) (2)
wherein the content of the first and second substances,x is fault parameter vector of the palletizing robot, x(1)To get success rate, x(3)The service time of the palletizing robot is prolonged.
5. The predictive maintenance method of a palletizing robot as claimed in claim 1, wherein: the step 4 specifically comprises the following steps:
step 4.1: establishing a coefficient vector, representing the coefficient vector by:
w＝(w(1)，w(2)，w(3)，b) (3)
wherein w is a coefficient vector;
establishing a hypothesis function from the coefficient vector, the hypothesis function being represented by:
wherein h isw(x) For a hypothetical function, T is transposed;
step 4.2: from the hypothesis function, a loss function is found, which is expressed by:
wherein J (w) is a loss function, m is the number of samples in the dataset, and y is a class label;
step 4.3: iterating the loss function by adopting a gradient descent method, minimizing the loss function, solving the optimal value of w, and expressing the optimal value of w after the loss function is minimized by the following formula:
where α is the step size.
6. The predictive maintenance method of a palletizing robot as claimed in claim 5, wherein: two states, "faulty" and "non-faulty" are indicated by the category label y, and when y is 1, it represents "faulty", and when y is 0, it represents "non-faulty".
7. The predictive maintenance method of a palletizing robot as claimed in claim 5, wherein: the step size alpha is 0.15.
8. The predictive maintenance method of a palletizing robot as claimed in claim 1, wherein: the step 5 specifically comprises the following steps:
step 5.1: performing predictive maintenance according to the minimized loss function, determining whether to perform early warning on the palletizing robot or not according to the fault parameter vector, and when wx is used0＞γwxcThen, a warning is given, wherein x0Is a fault parameter vector, x, of the palletizing robot at the current momentcThe fault parameter vector of the palletizing robot when leaving the factory is gamma which is a threshold coefficient;
step 5.2: when the palletizing robot enters an early warning state, a servo motor of the palletizing robot is adjusted to enter a low working load state, and the speed of the servo motor entering the low working load state is represented by the following formula:
wherein, VLowV is the speed of the servo motor in a low working load state, and is the movement speed set by the servo motor;
step 5.3: adjusting the maximum working time of the palletizing robot in an early warning state, and expressing the maximum working time through the following formula:
wherein, T0Is the maximum working time T of the palletizing robot in an early warning stateWThe maximum working time of the day is specified when the palletizing robot leaves a factory.
9. The predictive maintenance method of a palletizing robot as claimed in claim 8, wherein: the threshold coefficient range is 15% -20%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911064556.6A CN110696049B (en) | 2019-11-04 | 2019-11-04 | Predictive maintenance method for palletizing robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911064556.6A CN110696049B (en) | 2019-11-04 | 2019-11-04 | Predictive maintenance method for palletizing robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110696049A true CN110696049A (en) | 2020-01-17 |
| CN110696049B CN110696049B (en) | 2023-03-24 |
Family
ID=69203123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911064556.6A Active CN110696049B (en) | 2019-11-04 | 2019-11-04 | Predictive maintenance method for palletizing robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110696049B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021260595A1 (en) | 2020-06-23 | 2021-12-30 | G.D Societa' Per Azioni | Method for the predictive maintenance of an automatic machine for manufacturing or packing consumer articles |
-
2019
- 2019-11-04 CN CN201911064556.6A patent/CN110696049B/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021260595A1 (en) | 2020-06-23 | 2021-12-30 | G.D Societa' Per Azioni | Method for the predictive maintenance of an automatic machine for manufacturing or packing consumer articles |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110696049B (en) | 2023-03-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110208019B (en) | Dynamic threshold early warning method for monitoring state of mobile equipment | |
| CN110696049B (en) | Predictive maintenance method for palletizing robot | |
| CN113283344A (en) | Mining conveying belt deviation detection method based on semantic segmentation network | |
| CN111337789A (en) | Method and system for detecting fault electrical element in high-voltage transmission line | |
| CN112629905A (en) | Equipment anomaly detection method and system based on deep learning and computer medium | |
| CN116594328B (en) | Intelligent industrial robot management and control system based on 5G network | |
| CN110751270A (en) | Unmanned aerial vehicle wire fault detection method, system and equipment | |
| CN111666978B (en) | Intelligent fault early warning system for IT system operation and maintenance big data | |
| CN115796610A (en) | Comprehensive monitoring method and system for operation of branch pipe forming system and storage medium | |
| CN105416958A (en) | Intelligent warehouse-in and warehouse-out system | |
| CN111581425A (en) | Equipment sound classification method based on deep learning | |
| CN116663314A (en) | Digital twin system of wharf safety intelligent monitoring platform | |
| CN211278482U (en) | Predictive maintenance system for palletizing robot | |
| CN116079709A (en) | Closed-loop monitoring method, system and device for track type robot motion system | |
| CN111381567A (en) | Safety detection system and method for industrial control system | |
| CN110889395B (en) | Machine learning-based mechanical motion recognition method and system | |
| CN114446027A (en) | Equipment fault alarm method, system, equipment and medium based on Internet of things | |
| CN113806615A (en) | KPI (Key performance indicator) abnormity early warning method of intelligent IT operation and maintenance system | |
| CN112668912A (en) | Training method of artificial neural network, dynamic calculation segmentation scheduling method, storage medium and system | |
| CN111950508A (en) | Substation circuit breaker appearance damage identification method | |
| CN113898528B (en) | Abnormality detection method, model training method and related device for fan variable pitch bearing | |
| CN113494428A (en) | Fault detection method and device of wind generating set | |
| CN117365923B (en) | Intelligent control method and system for air compressor | |
| CN111273619B (en) | Remote alarm method for chemical equipment | |
| CN112651491B (en) | Industrial robot fault prediction method based on CNN convolution model |
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 |