Fault monitoring method and system for pick-up mechanism of toothed plastic film residue recycling machine

Abstract

The invention discloses a fault monitoring method and system for a pick-up mechanism of a toothed plastic film residue recycling machine, and relates to the technical field of agricultural machinery fault monitoring. Hall sensors are used for collecting the running rotating speeds of a film lifting shaft, a tensioning shaft and a film stripping shaft respectively, and an acceleration vibration sensor is used for collecting the shaft end vibration value of a tensioning positioning shaft; acquiring a time trigger value of the film picking spike teeth by utilizing a correlation type infrared photoelectric sensor; based on the operation rotating speed, the shaft end vibration value and the time trigger value, a K nearest neighbor algorithm is adopted for classification; predicting the falling condition of the film picking spike teeth or the breakage condition of the film collecting belt by using a linear regression algorithm; an alarm is triggered when a fault is detected. Predicting the fault of the film collecting belt by using a linear regression algorithm; according to the difference value of the left and right vibration values, longitudinal tearing and abrasion prediction of the belt is achieved. According to the invention, picking mechanism faults caused by breakage of a picking film collecting belt and falling of film picking belt teeth of the toothed residual film recycling machine can be found, and the requirement for monitoring the faults of the toothed residual film recycling machine in real time is met.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery fault monitoring, and more particularly to a method and system for monitoring the fault of a picking mechanism of a toothed residual film recycling machine. Background Art

[0002] The mulching cultivation technology is beneficial to the quality and yield increase, and has been widely used in agricultural production. With the continuous growth of the use of mulch, the natural degradation of mulch in the soil is gradually inconsistent with the current emphasis on ecological protection and sustainable development. In order to solve the problem of residual mulch increasing the harm to agricultural production, mechanized recycling has become the main way to solve the problem of residual mulch recycling.

[0003] With the development of technology, the toothed film recovery machine has gradually become the main mechanical method of film recovery with its efficient picking ability. However, during the working hours of the toothed film recovery machine, the picking mechanism often fails due to the noise generated by the tractor's mechanical transmission and the breakage of the film collection belt and the falling off of the film picking belt teeth. In the absence of an effective fault monitoring method for the toothed film recovery machine, it is difficult for the driver to effectively distinguish between normal working conditions in a short period of time. This will cause the overall quality of film recovery operations to decline, aggravate the degree of damage to the toothed film recovery machine, further increase its maintenance cost, and form a chain reaction of toothed film recovery machine failures.

[0004] Therefore, how to provide a fault monitoring method and system for the picking mechanism of a toothed residual film recovery machine to detect the picking mechanism fault caused by the broken film picking belt and the falling off of the film picking belt teeth of the toothed residual film recovery machine, and help the driver to detect the fault in time is a problem that technical personnel in this field urgently need to solve. Summary of the invention

[0005] In view of this, the present invention provides a method and system for monitoring the fault of the picking mechanism of a toothed residual film recovery machine, which is applied to the current toothed residual film recovery machine models. It can detect the fault of the picking mechanism caused by the breakage of the film picking belt and the falling off of the film picking belt teeth of the toothed residual film recovery machine, thereby achieving the requirement of real-time monitoring of the fault of the toothed residual film recovery machine and helping the driver to detect the fault in time.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a method for monitoring the fault of a picking mechanism of a toothed residual film recycling machine, comprising:

[0007] Collect the operating data of the picking mechanism of the toothed residual film recycling machine;

[0008] The Hall sensor is used to collect the running speed of the film lifting shaft, tensioning shaft and film stripping shaft respectively, the acceleration vibration sensor is used to collect the shaft end vibration value of the tensioning and positioning shaft, and the infrared photoelectric sensor is used to obtain the time trigger value of the film lifting nail teeth;

[0009] Based on the running speed, shaft end vibration value and time trigger value, the K nearest neighbor algorithm is used for classification;

[0010] Use linear regression algorithm to predict the shedding of film picking nails or the breakage of film collecting belts;

[0011] The falling off of the film picking nail teeth or the breaking of the film collecting belt is judged according to the time trigger value; when the falling off of the film picking nail teeth or the breaking of the film collecting belt is detected, it is judged as a fault;

[0012] An alarm is triggered when a fault is detected.

[0013] Preferably, the time trigger value of the film picking nail teeth is obtained by using a beam-type infrared photoelectric sensor, including:

[0014] There are a plurality of film-lifting nail teeth, and there is a certain distance between each film-lifting nail tooth during operation;

[0015] When not triggered, the through-beam infrared photoelectric sensor is in a high level state;

[0016] When triggered, the film-picking nails block the infrared photoelectric sensor and put it in a low-level state;

[0017] According to the captured falling edge generated from high level to low level, the preset interrupt mode is triggered to obtain the current time trigger value.

[0018] Preferably, the K nearest neighbor algorithm is used for classification, including:

[0019] Use Euclidean distance to calculate the distance from the trigger value of the film-picking nail to all training data points;

[0020] The cross-validation method is used to select the optimal K value, and the optimal K nearest neighbor points are selected in the K nearest neighbor algorithm;

[0021] Classify according to the corresponding tags;

[0022] Determine whether a fault occurs based on the identified tags.

[0023] Preferably, historical storage data is obtained to train the trigger value of the film picking nail teeth, and a working data set D is obtained according to the linear node relationship between the film stripping shaft speed and the film collecting belt trigger time;

[0024] The working data set D includes the rotation speed, the trigger time and the vibration values ​​on both sides, which is expressed as:

[0025] D={(x i ,y i )|i=1,2,…,n};

[0026] Among them, xi Represents historical storage data, y i Indicates the corresponding structure label;

[0027] x i Defined as a tuple:

[0028] x i =(v i ,t i ,d i );

[0029] Where: v i Indicates the speed of the stripping shaft, t i Indicates the trigger time of the film collection belt, d i Indicates the vibration value of the shaft end;

[0030] The dataset D is represented as:

[0031] D={((v i ,t i ,d i ),y i |i=1,2,...,n};

[0032] Among them, y i There are only two categories: normal and abnormal; 0 and 1 are used to represent these two categories, y i =0 means normal, y i =1 indicates abnormality;

[0033] The Euclidean distance is used to calculate the distance between sample points;

[0034] The number of neighboring points K is selected as a hyperparameter based on the calculated distance;

[0035] According to the updated K nearest neighbor point distribution, the new data is classified to determine whether the film picking nail teeth are faulty or the film collecting belt is broken.

[0036] Preferably, a fault monitoring system for a picking mechanism of a toothed residual film recycling machine comprises: a Hall sensor, an acceleration vibration sensor, a beam-type infrared photoelectric sensor, a data acquisition module, a control module and an alarm module;

[0037] The Hall sensor, the acceleration vibration sensor and the opposing infrared photoelectric sensor are respectively installed on the picking mechanism of the toothed residual film recycling machine;

[0038] The Hall sensor is used to detect the running speed of the film-forming shaft, the tensioning shaft, and the film-removing shaft;

[0039] The acceleration vibration sensor is used to detect the vibration value of the shaft end of the tensioning and positioning shaft;

[0040] The infrared photoelectric sensor is used to capture the time trigger value of the film picking nail teeth;

[0041] The data acquisition module is respectively connected to the Hall sensor, the acceleration vibration sensor and the opposing infrared photoelectric sensor;

[0042] The control module is connected to the data acquisition module and the alarm module respectively.

[0043] Preferably, the acceleration vibration sensor adopts MPU6050 acceleration vibration sensor;

[0044] The acceleration vibration sensor is fixedly connected to the left and right sides of the tensioning and positioning shaft;

[0045] The acceleration vibration values ​​on both sides are collected using the IIC protocol, and the vibration values ​​in the three directions of X, Y, and Z are accumulated using the absolute cumulative value method, and sent to the control module in real time via the USART serial port protocol.

[0046] Preferably, a plurality of film picking nail teeth are fixed on the outer surface of the film collecting belt, and the film picking nail teeth are located in the middle of the opposing infrared photoelectric sensor;

[0047] When the film-lifting nails block the infrared photoelectric sensor, the level changes;

[0048] The data acquisition module captures the level change data, obtains the time trigger value between the intervals of two film-picking nail teeth, and sends it to the control module according to the USART serial port protocol.

[0049] Preferably, the opposing infrared photoelectric sensor comprises a transmitting end and a receiving end, wherein the transmitting end is mounted on one side of the film collecting belt, and the receiving end is mounted on the other side of the film collecting belt.

[0050] Preferably, a data display module and a data storage module;

[0051] The data display module and the data storage module are respectively connected to the control module;

[0052] The data display module is used to display the detection results in real time;

[0053] The data storage module is used to store the collected data.

[0054] Preferably, it also includes: an adjustable DC power supply and a start switch; the start switch is used to control opening and closing; the adjustable DC power supply is used to provide electrical energy.

[0055] It can be seen from the above technical scheme that compared with the prior art, the present invention discloses a fault monitoring method and system for the picking-up mechanism of a toothed residual film recycling machine, including: collecting the operating data of the picking-up mechanism of the toothed residual film recycling machine; using Hall sensors to respectively collect the operating speeds of the film lifting shaft, the tensioning shaft, and the film stripping shaft, using an acceleration vibration sensor to collect the shaft end vibration value of the tensioning positioning shaft, and using a opposed infrared photoelectric sensor to obtain the time trigger value of the film picking nail teeth; based on the operating speed, the shaft end vibration value and the time trigger value, using the K nearest neighbor algorithm for classification; using a linear regression algorithm to predict the shedding of the film picking nail teeth or the breakage of the film collecting belt; judging the shedding of the film picking nail teeth or the breakage of the film collecting belt according to the time trigger value; when it is detected that the film picking nail teeth fall off or the film collecting belt is broken, it is judged as a fault; and triggering an alarm when a fault is detected. The present invention is applied to a toothed residual film recovery machine, and detects the failure of the picking mechanism caused by the breakage of the film picking belt and the falling off of the film picking belt teeth of the toothed residual film recovery machine, thereby achieving the requirement of real-time monitoring of the failure of the toothed residual film recovery machine and helping the driver to discover the failure in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0057] Figure 1 A schematic diagram of the program system structure principle provided by an embodiment of the present invention.

[0058] Figure 2 A schematic diagram of the sensor connection and data acquisition process provided by an embodiment of the present invention.

[0059] Figure 3 A schematic diagram of the picking mechanism structure and Hall sensor installation of a toothed residual film recovery machine provided in an embodiment of the present invention.

[0060] Figure 4 A schematic diagram of the installation of the left vibration sensor of the tensioning and positioning shaft provided in an embodiment of the present invention.

[0061] Figure 5 A schematic diagram of the installation of the right vibration sensor of the tensioning and positioning shaft provided in an embodiment of the present invention.

[0062] Figure 6 This is a schematic diagram of the installation of the through-beam infrared photoelectric sensor provided in an embodiment of the present invention.

[0063] Figure 7 A schematic diagram of the composition of a through-beam infrared photoelectric sensor provided in an embodiment of the present invention.

[0064] Among them, 101, film lifting shaft; 102, tensioning shaft; 103, film stripping shaft; 104, tensioning and positioning shaft; 105, film picking nail teeth; 106, film collecting belt; 107, first Hall sensor; 108, second Hall sensor; 109, third Hall sensor; 201, first MPU6050 acceleration vibration sensor; 301, second MPU6050 acceleration vibration sensor; 401, mounting bracket of opposed infrared photoelectric sensor; 402, opposed infrared photoelectric sensor; 501, screw; 502, pressing plate of opposed infrared photoelectric sensor; 503, transmitting end of opposed infrared photoelectric sensor; 504, fixing bolt; 505, fixing bracket of opposed infrared photoelectric sensor; 506, receiving end of opposed infrared photoelectric sensor. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] The embodiment of the present invention discloses a method and system for monitoring the fault of the picking mechanism of a toothed residual film recovery machine. The embodiment of the present invention is applied to the toothed residual film recovery machine to detect the fault of the picking mechanism caused by the breakage of the film picking belt and the falling off of the film picking belt teeth of the toothed residual film recovery machine, thereby achieving the requirement of real-time monitoring of the fault of the toothed residual film recovery machine and helping the driver to find the fault in time.

[0067] In a specific embodiment of the present invention, a method for monitoring faults of a picking mechanism of a toothed residual film recycling machine is provided. Figure 1 As shown, including:

[0068] Collect the operating data of the picking mechanism of the toothed residual film recycling machine;

[0069] The Hall sensor is used to collect the running speed of the film lifting shaft, tensioning shaft and film stripping shaft respectively, the acceleration vibration sensor is used to collect the shaft end vibration value of the tensioning and positioning shaft, and the infrared photoelectric sensor is used to obtain the time trigger value of the film lifting nail teeth;

[0070] Based on the running speed, shaft end vibration value and time trigger value, the K nearest neighbor algorithm is used for classification;

[0071] Use linear regression algorithm to predict the shedding of film picking nails or the breakage of film collecting belts;

[0072] The falling off of the film picking nail teeth or the breaking of the film collecting belt is judged according to the time trigger value; when the falling off of the film picking nail teeth or the breaking of the film collecting belt is detected, it is judged as a fault;

[0073] An alarm is triggered when a fault is detected.

[0074] Among them, the vibration of the tensioning positioning shaft affects the tensioning force of the entire film collecting belt. If the vibration value is too large, the contact area between the tensioning positioning shaft and the film collecting belt will be smaller, the tensioning force of the film collecting belt will be relatively smaller, and the volatility will increase. At the same time, the stability of the film collecting belt will be reduced, and it is easy to exceed the monitoring range of the reflected infrared photoelectric sensor.

[0075] Specifically, the time trigger value of the film picking nail teeth is obtained by using the infrared photoelectric sensor, including:

[0076] There are a plurality of film-lifting nail teeth, and there is a certain distance between each film-lifting nail tooth during operation;

[0077] When not triggered, the infrared photoelectric sensor is in the normally open mode and maintains a high level state;

[0078] When triggered, the film-picking nails block the infrared photoelectric sensor and put it in a low-level state;

[0079] According to the captured falling edge generated from high level to low level, the preset interrupt mode is triggered to obtain the current time trigger value.

[0080] Specifically, the K nearest neighbor algorithm is used for classification, including:

[0081] Use Euclidean distance to calculate the distance from the trigger value of the film-picking nail to all training data points;

[0082] The cross-validation method is used to select the optimal K value, and the optimal K nearest neighbor points are selected in the K nearest neighbor algorithm;

[0083] Classify according to the corresponding tags;

[0084] Determine whether a fault occurs based on the identified tags.

[0085] Specifically, the falling off of the film picking nail teeth or the breaking of the film collecting belt is determined according to the time trigger value; when the falling off of the film picking nail teeth or the breaking of the film collecting belt is detected, it is determined to be a fault.

[0086] Among them, the trigger value is labeled, such as the trigger value of the first belt and the trigger value of the second belt. If one belt has no trigger value for a period of time, it can be determined that the belt has a fault, such as breakage or slippage.

[0087] Specifically, historical storage data is obtained to train the trigger value of the film picking nail teeth, and the working data set D is obtained according to the linear node relationship between the film stripping shaft speed and the film collection belt trigger time;

[0088] The working data set D includes the rotation speed, the trigger time and the vibration values ​​on both sides, which is expressed as:

[0089] D={(x i ,y i )|i=1,2,…,n};

[0090] Among them, x i Represents historical storage data, y i Indicates the corresponding structure label;

[0091] x i Defined as a tuple:

[0092] x i =(v i ,t i ,d i );

[0093] Where: v i Indicates the speed of the stripping shaft, t i Indicates the trigger time of the film collection belt, d i Indicates the vibration value of the shaft end;

[0094] The dataset D is represented as:

[0095] D={((v i ,t i ,d i ),y i |i=1,2,...,n};

[0096] Among them, y i There are only two categories: normal and abnormal; 0 and 1 are used to represent these two categories, y i =0 means normal, y i =1 indicates abnormality;

[0097] The Euclidean distance is used to calculate the distance between sample points;

[0098] In the K-nearest neighbor algorithm, the commonly used distance metric is usually Euclidean distance, and the distance between sample points reflects the similarity between the sample points.

[0099] Select the number of neighboring points K as a hyperparameter;

[0100] According to the updated K nearest neighbor point distribution, the new data is classified to determine whether the film picking nail teeth are faulty or the film collecting belt is broken.

[0101] Among them, the choice of K value affects the generalization ability and accuracy of the model. If the K value is too small, the model is divided into multiple categories, and each point is fully considered, the accuracy is high, but the decision boundary is fragmented, and the generalization ability is weak; if the K value is too large, even the points far away are judged as neighboring points, the accuracy is low, but the decision boundary is smooth, and the generalization ability is strong.

[0102] Specifically, the sample data is divided into K parts, one of which is used as the test set, and the other K-1 parts are used as the training set, so that an error E i , and then find the average of these K accuracy rates, the formula is as follows:

[0103]

[0104] The higher the average accuracy of E, the stronger the generalization ability. The generalization error obtained by subtracting this average from 1 can be used to select the optimal K value.

[0105] In a specific embodiment of the present invention, x i Designed as a vector containing the stripping shaft speed and the film collection belt trigger time, it represents the characteristics of each sample. Obtain historical storage data to train the film picking nail tooth trigger value, and obtain the working data set D based on the linear node relationship between the stripping shaft speed and the film collection belt trigger time;

[0106] The working data set D includes the rotation speed, the trigger time and the vibration values ​​on both sides, which is expressed as:

[0107] D={(x i ,y i )|i=1,2,…,n};

[0108] Among them, x i Represents historical storage data, y i Indicates the corresponding structure label;

[0109] x i Defined as a tuple:

[0110] x i =(v i ,t i ,d i );

[0111] Where: v i Indicates the speed of the stripping shaft, t i Indicates the trigger time of the film collection belt, d i Indicates the vibration value of the shaft end;

[0112] The dataset D is represented as:

[0113] D={((v i,t i ,d i ),y i |i=1,2,...,n};

[0114] Among them, y i There are only two categories: normal and abnormal; 0 and 1 are used to represent these two categories, y i =0 means normal, y i =1 indicates abnormality;

[0115] Faults are detected by calculating the Euclidean distance. For two samples x i and x j , the calculation formula of Euclidean distance d is:

[0116]

[0117] By comparing the distance between the new sample and the known samples, it is evaluated whether the sample belongs to the normal or abnormal category.

[0118] In view of the lack of relative fault monitoring methods in the current picking structure of the toothed residual film recycling machine, in the past the driver only relied on driving experience to stop the machine for inspection. However, due to the interference of mechanical noise in the working state, it is not easy to effectively distinguish the normal state.

[0119] In this regard, Figure 1 As shown, when the data acquisition module starts working, the timer function is used to configure the simultaneous interrupt mode. The data acquisition module uses the Hall speed sensor to collect the speed of the film-forming shaft 101, the tensioning shaft 102, and the film-removing shaft 103. In view of the fact that the working speed of the residual film recovery machine is about 3000 revolutions, the speed measurement method is used for speed collection. The timing mode of 1ms timing time interval is adopted, the maximum timing counting cycle is 32 bits, and the counting value duration is about 2 32 -1=4,294,967,295ms, the film picking nail teeth 105 are fixed on the outer surface of the film collecting belt 106, and the trigger time of the film picking nail teeth 105 is obtained by the infrared photoelectric sensor 402. And the vibration of the tension wheel is monitored in real time according to the first MPU6050 acceleration vibration sensor 201 (left) and the second MPU6050 acceleration vibration sensor 301 (right), and the control module is transmitted to determine the level characteristics and influences such as the film collecting belt breaking and the film picking nail teeth falling off in the film picking belt teeth.

[0120] In order to solve the problem that at the same time, multiple infrared photoelectric sensors 402 monitor whether the film pick-up nail teeth 105 are in a normal state, the K nearest neighbor algorithm is used to classify the film pick-up nail teeth 105 to identify whether a fault occurs, and an alarm is issued after a fault is determined.

[0121] At this point, the control module has completed its work and returns to the preliminary classification and processing data. When the data acquisition module starts working, the timer function is used to configure the simultaneous interrupt mode, and the Hall sensors installed on one side of the film-forming shaft 101, the tensioning shaft 102, and the film-removing shaft 103 are used to collect the rotation speed, and the collected rotation speed is sent using the USART serial port protocol. For the MPU6050 acceleration vibration sensor, it is fixedly connected to the left and right sides of the tensioning and positioning shaft 104, and the acceleration vibration values ​​on both sides are obtained using the IIC protocol. The absolute cumulative value method is used to accumulate the vibration values ​​in the three directions of X, Y, and Z, and send them in real time using the USART serial port protocol.

[0122] The data acquisition module collects the time interval trigger value of the film picking nail teeth 105, uses the 1ms timing mode and interrupt mode, sets the maximum timing counting cycle to 32 bits, and the counting value duration is about 2 32 -1=4,294,967,295ms. The multiple film picking nail teeth 105 fixed on the outer surface of the film collecting belt 106 are respectively connected in the middle of each opposing infrared photoelectric sensor 402. During operation, the opposing infrared photoelectric sensor 402 will generate a low level when the film picking nail teeth 105 block the opposing infrared photoelectric sensor 402 at a certain interval between the two film picking nail teeth 105. The data acquisition module will be triggered by the low level to obtain the time trigger value between the two film picking nail teeth 105 and send it to the control module according to the USART serial port protocol. Then the control module determines that the belt in the film belt is broken and the film picking nail teeth fall off according to the rotation speed, vibration value, and the time interval trigger value of the film picking nail teeth.

[0123] The embodiment of the present invention has a reasonable layout, a compact structure, and saves resources. By comprehensively utilizing multiple sensors and advanced algorithms, not only the accuracy of fault detection is improved, but also an alarm can be issued in time when a fault occurs, thereby reducing equipment damage and downtime and improving production efficiency.

[0124] In another specific embodiment of the present invention, a fault monitoring system for a picking mechanism of a toothed residual film recycling machine is provided. Figure 2 As shown, it includes: a Hall sensor, an acceleration vibration sensor, a beam-type infrared photoelectric sensor, a data acquisition module, a control module and an alarm module;

[0125] The Hall sensor, the acceleration vibration sensor and the opposing infrared photoelectric sensor are respectively installed on the picking mechanism of the toothed residual film recycling machine;

[0126] The Hall sensor is used to detect the running speed of the film-forming shaft, the tensioning shaft, and the film-removing shaft;

[0127] The acceleration vibration sensor is used to detect the vibration value of the shaft end of the tensioning and positioning shaft;

[0128] The infrared photoelectric sensor is used to capture the time trigger value of the film picking nail teeth;

[0129] The data acquisition module is respectively connected to the Hall sensor, the acceleration vibration sensor and the opposing infrared photoelectric sensor;

[0130] The control module is connected to the data acquisition module and the alarm module respectively.

[0131] Among them, the data acquisition module adopts a 1ms timing time interval timing mode and is connected to multiple infrared photoelectric sensors, such as Figure 6 and Figure 7 As shown, the main body of the infrared photoelectric sensor 402 is a split type infrared photoelectric sensor, including an infrared photoelectric sensor transmitter 503 and an infrared photoelectric sensor transmitter receiver 506, which are respectively installed on both sides of each film collection belt 106 on the infrared photoelectric sensor fixing frame 505. The maximum timing counting cycle of the data acquisition module is 32 bits, and the counting value duration is about 2 32 -1=4,294,967,295ms, and the longest time for a single continuous operation is 49.71 days.

[0132] Specifically, the opposed infrared photoelectric sensor mounting frame 401 is fixed on the inner side of the film collecting belt (the inner side of the loose side of the tensioning positioning shaft close to the film stripping shaft); a plurality of opposed infrared photoelectric sensor fixing frames 505 are fixed on the opposed infrared photoelectric sensor mounting frame 401 by fixing bolts 504; the number of the opposed infrared photoelectric sensor fixing frames 505 is the same as the number of film collecting belts; the opposed infrared photoelectric sensor fixing frame 505 is a U-shaped structure; the bottom of the U-shaped structure is fixed on the opposed infrared photoelectric sensor mounting frame 401. Film picking nail teeth pass through the U-shaped structure; the opposed infrared photoelectric sensor transmitting end 503 and the opposed infrared photoelectric sensor transmitting end receiving end 506 are respectively fixed by the opposed infrared photoelectric sensor pressing plate 502, and the opposed infrared photoelectric sensor pressing plate 502 is fixed on the opposed infrared photoelectric sensor fixing frame 505 by screws 501.

[0133] Specifically, the Hall sensor includes: a first Hall sensor 107, a second Hall sensor 108 and a third Hall sensor 109;

[0134] The first Hall sensor 107 is used to monitor the running speed of the film stripping shaft;

[0135] The second Hall sensor 108 is used to monitor the running speed of the tensioning shaft;

[0136] The third Hall sensor 109 is used to monitor the running speed of the film-forming shaft.

[0137] Specifically, the acceleration vibration sensor adopts MPU6050 acceleration vibration sensor;

[0138] The acceleration vibration sensor is fixedly connected to the left and right sides of the tensioning and positioning shaft;

[0139] The acceleration vibration values ​​on both sides are collected using the IIC protocol, and the vibration values ​​in the three directions of X, Y, and Z are accumulated using the absolute cumulative value method, and sent to the control module in real time via the USART serial port protocol.

[0140] Specifically, Figure 4 and Figure 5 As shown, the MPU6050 acceleration vibration sensor includes a first MPU6050 acceleration vibration sensor 201 (left) and a second MPU6050 acceleration vibration sensor 301 (right); the first MPU6050 acceleration vibration sensor 201 is connected to the left side of the tensioning positioning shaft, and the second MPU6050 acceleration vibration sensor 301 is connected to the right side of the tensioning positioning shaft.

[0141] The larger the vibration value, the smaller the contact area between the film collecting belt and the film lifting shaft, tensioning shaft and film stripping shaft is. The larger the vibration value, the easier it is to exceed the trigger range of the film picking nail teeth monitored by the opposing infrared photoelectric sensor.

[0142] In a specific embodiment of the present invention, by detecting the vibration values ​​on the left and right sides of the film collecting belt, the tension of the film collecting belt can be checked, and the tension can be used to determine whether it is within the front and rear range of the infrared photoelectric sensor.

[0143] Based on this, the embodiment of the present invention uses the linear regression algorithm in machine learning to build a layer of prediction of film collection belt failure on the basis of the above. Using the symmetry of the tensioning and positioning axis, a film collection belt failure prediction model is established, and the longitudinal tearing and wear of the belt are predicted based on the difference in the left and right vibration values.

[0144] Specifically, the vibration values ​​on the left and right sides of each film collecting belt are collected and named d_left and d_right respectively.

[0145] Record the tension of the film collection belt.

[0146] The actual fault conditions of the film collecting belt (such as tearing and wear) are marked to form a binary target variable.

[0147] Calculating the difference and percentage of the vibration values ​​on the left and right sides, and dividing the data set according to the difference in vibration values; the percentage is calculated based on the ratio of the difference to the tension;

[0148] constructing a first model and a second model;

[0149] If the difference percentage between the two sides is less than or equal to 10%, the corresponding fault condition is predicted by the first model.

[0150] If the difference percentage between the two sides is greater than 10% and less than or equal to 20%, the corresponding fault condition is predicted by the second model.

[0151] When a new vibration value is received, the corresponding model is selected for prediction based on its percentage to achieve the prediction of longitudinal tearing and wear of multiple coaxial belts.

[0152] Specifically, a plurality of film picking nail teeth are fixed on the outer surface of the film collecting belt, and the film picking nail teeth are located in the middle of the opposing infrared photoelectric sensor;

[0153] When the film-lifting nails block the infrared photoelectric sensor, the level changes;

[0154] The data acquisition module captures the level change data, obtains the time trigger value between the intervals of two film-picking nail teeth, and sends it to the control module according to the USART serial port protocol.

[0155] In the embodiment of the present invention, monitoring the film collection belt and the film picking nail teeth failure depends on the interrupt mode of the data acquisition module. During the operation, there is a certain distance between the film picking nail teeth. When not triggered, the infrared photoelectric sensor is in the normally open mode and maintains a high level state. When triggered, the film picking nail teeth block the infrared photoelectric sensor and put it in a low level state. The data acquisition module will capture the falling edge generated from the high level to the low level, triggering the interrupt mode set in the data acquisition module to obtain the current time trigger value. And pass it to the control module through the USART serial communication protocol.

[0156] Specifically, the opposing infrared photoelectric sensor includes a transmitting end and a receiving end. The transmitting end is installed on one side of the film collecting belt, and the receiving end is installed on the other side of the film collecting belt.

[0157] Specifically, a data display module and a data storage module;

[0158] The data display module and the data storage module are respectively connected to the control module;

[0159] The data display module is used to display the detection results in real time;

[0160] The data storage module is used to store the collected data.

[0161] Specifically, it also includes: an adjustable DC power supply and a start switch; the start switch is used to control opening and closing; the adjustable DC power supply is used to provide electrical energy.

[0162] Specifically, Figure 3 As shown, the picking mechanism of the toothed residual film recycling machine is composed of a film-lifting shaft 101, a tensioning shaft 102, a film-removing shaft 103, a tensioning and positioning shaft 104, a film-collecting belt 107, film-picking nail teeth 105, a film-feeding fan and other transmission and connecting parts. A Hall sensor is connected to one side of the film-lifting shaft, the tensioning shaft and the film-removing shaft to collect the operating speed; the MPU6050 acceleration vibration sensor is connected to both sides of the tensioning and positioning shaft to collect the vibration value; the opposing infrared photoelectric sensor 402 is connected to both sides of each film-collecting belt and film-picking nail teeth to obtain the time trigger value of the film-picking nail teeth. The information is transmitted to the control module through the USART serial port protocol, and the program and algorithm calculation are used to monitor in real time whether the picking mechanism of the toothed residual film recycling machine has any faults, and alarm is issued according to the fault. Figure 2 As shown, the data acquisition module is used to connect the Hall sensor and the infrared photoelectric sensor. The MPU6050 acceleration vibration sensor and the data acquisition module adopt the IIC communication method to obtain the data of each sensor. The control module is used to transmit information between the data display, data storage, and alarm modules and control the alarm module signal transmission; the information transmission between the control module and the data acquisition module adopts USART serial communication to transmit data.

[0163] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0164] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for monitoring faults of a picking mechanism of a toothed residual film recycling machine, characterized in that: include: Collect the operating data of the picking mechanism of the toothed residual film recycling machine; The Hall sensor is used to collect the running speed of the film lifting shaft, tensioning shaft and film stripping shaft respectively, the acceleration vibration sensor is used to collect the shaft end vibration value of the tensioning and positioning shaft, and the infrared photoelectric sensor is used to obtain the time trigger value of the film lifting nail teeth; Based on the running speed, shaft end vibration value and time trigger value, the K nearest neighbor algorithm is used for classification; Use linear regression algorithm to predict the shedding of film picking nails or the breakage of film collecting belts; The falling off of the film picking nail teeth or the breaking of the film collecting belt is judged according to the time trigger value; when the falling off of the film picking nail teeth or the breaking of the film collecting belt is detected, it is judged as a fault; An alarm is triggered when a fault is detected.

2. A method for monitoring faults of a picking mechanism of a toothed residual film recycling machine according to claim 1, characterized in that: The time trigger value of the film picking nail teeth is obtained by using the infrared photoelectric sensor, including: There are a plurality of film-lifting nail teeth, and there is a certain distance between each film-lifting nail tooth during operation; When not triggered, the through-beam infrared photoelectric sensor is in a high level state; When triggered, the film-picking nails block the infrared photoelectric sensor and put it in a low-level state; According to the captured falling edge generated from high level to low level, the preset interrupt mode is triggered to obtain the current time trigger value.

3. The method for monitoring faults of a picking mechanism of a toothed residual film recycling machine according to claim 1, characterized in that: The K nearest neighbor algorithm is used for classification, including: Use Euclidean distance to calculate the distance from the trigger value of the film-picking nail to all training data points; The cross-validation method is used to select the optimal K value; Classify according to the corresponding tags; Determine whether a fault occurs based on the identified tags.

4. A method for monitoring faults of a picking mechanism of a toothed residual film recycling machine according to claim 3, characterized in that: Obtain historical storage data to train the trigger value of the film picking nail teeth, and obtain the working data set D based on the linear node relationship between the film stripping shaft speed and the film collection belt trigger time; The working data set D includes the rotation speed, the trigger time and the vibration values ​​on both sides, which is expressed as: D={(x i ,y i )|i=1,2,…,n}; Among them, x i Represents historical storage data, y i Indicates the corresponding structure label; x i Defined as a tuple: x i =(v i ,t i ,d i ); Where: v i Indicates the speed of the stripping shaft, t i Indicates the trigger time of the film collection belt, d i Indicates the vibration value of the shaft end; The dataset D is represented as: d={((v i ,t i ,d i ),y i |i=1,2,...,n}; Among them, y i There are only two categories: normal and abnormal; 0 and 1 are used to represent these two categories, y i =0 means normal, y i =1 indicates abnormality; The Euclidean distance is used to calculate the distance between sample points; The number of neighboring points K is selected as a hyperparameter based on the calculated distance; According to the updated K nearest neighbor point distribution, the new data is classified to determine whether the film picking nail teeth are faulty or the film collecting belt is broken.

5. A fault monitoring system for the picking mechanism of a toothed residual film recycling machine, characterized in that: include: Hall sensor, acceleration vibration sensor, infrared photoelectric sensor, data acquisition module, control module and alarm module; The Hall sensor, the acceleration vibration sensor and the opposing infrared photoelectric sensor are respectively installed on the picking mechanism of the toothed residual film recycling machine; The Hall sensor is used to detect the running speed of the film-lifting shaft, the tensioning shaft, and the film-removing shaft; The acceleration vibration sensor is used to detect the vibration value of the shaft end of the tensioning and positioning shaft; The infrared photoelectric sensor is used to capture the time trigger value of the film picking nail teeth; The data acquisition module is respectively connected to the Hall sensor, the acceleration vibration sensor and the opposing infrared photoelectric sensor; The control module is connected to the data acquisition module and the alarm module respectively.

6. A fault monitoring system for the picking mechanism of a toothed residual film recycling machine according to claim 5, characterized in that: The acceleration vibration sensor adopts MPU6050 acceleration vibration sensor; The acceleration vibration sensor is fixedly connected to the left and right sides of the tensioning and positioning shaft; The acceleration vibration values ​​on both sides are collected using the IIC protocol, and the vibration values ​​in the three directions of X, Y, and Z are accumulated using the absolute cumulative value method, and sent to the control module in real time via the USART serial port protocol.

7. The fault monitoring system for the picking mechanism of a toothed residual film recycling machine according to claim 5 is characterized in that: A plurality of film picking nail teeth are fixed on the outer surface of the film collecting belt, and the film picking nail teeth are located in the middle of the opposing infrared photoelectric sensor; When the film-lifting nails block the infrared photoelectric sensor, the level changes; The data acquisition module captures the level change data, obtains the time trigger value between the intervals of two film-picking nail teeth, and sends it to the control module according to the USART serial port protocol.

8. A fault monitoring system for a picking mechanism of a toothed residual film recycling machine according to claim 7, characterized in that: The opposing infrared photoelectric sensor comprises a transmitting end and a receiving end. The transmitting end is installed on one side of the film collecting belt, and the receiving end is installed on the other side of the film collecting belt.

9. The fault monitoring system for the picking mechanism of a toothed residual film recycling machine according to claim 5 is characterized in that: Data display module and data storage module; The data display module and the data storage module are respectively connected to the control module; The data display module is used to display the detection results in real time; The data storage module is used to store the collected data.

10. The fault monitoring system for the picking mechanism of a toothed residual film recycling machine according to claim 5, characterized in that: Also includes: Adjustable DC power supply and start switch; the start switch is used to control opening and closing; The adjustable DC power supply is used to provide electrical energy.

Images