CN111025994B

CN111025994B - Equipment monitoring system and method based on CNNs

Info

Publication number: CN111025994B
Application number: CN201911368212.4A
Authority: CN
Inventors: 吴文烨; 陈烨; 吴勇明; 郭宏记
Original assignee: Dianeng Technology Hangzhou Co ltd
Current assignee: Dianeng Technology Hangzhou Co ltd
Priority date: 2019-12-26
Filing date: 2019-12-26
Publication date: 2021-12-14
Anticipated expiration: 2039-12-26
Also published as: CN111025994A

Abstract

The invention provides a CNNs-based equipment monitoring system, which comprises: the sampling device is electrically connected to a device to be tested and used for collecting data of the device to be tested; the control device comprises a computing unit, the computing unit is electrically connected with the sampling device and is used for receiving the data of the equipment to be tested, processing and judging the data of the equipment to be tested through a convolutional neural network architecture and outputting a corresponding judgment result; and the server is electrically connected with the control device and used for receiving the judgment result output by the calculation unit and storing the judgment result. The monitoring system can realize intelligent identification, intelligent monitoring and intelligent control of the equipment to be tested. The invention also provides a device monitoring method based on the CNNs.

Description

Equipment monitoring system and method based on CNNs

Technical Field

The present invention relates to the field of device monitoring technologies, and in particular, to a device monitoring system and method based on Convolutional Neural Networks (CNNs).

Background

At present in the equipment production field, a lot of machine equipment, because the year of purchase is more for a long time, do not have relevant sensor or the ability of handling data mostly, consequently can't accomplish effective intelligent monitoring equipment's running state, comparatively inconvenient. In addition, if a dedicated monitoring device is used, the monitoring cost may be high, and the monitoring device is not practical.

Disclosure of Invention

In view of the above, it is desirable to provide a monitoring system and method for CNNs-based devices, which can achieve intelligent monitoring and have low cost.

An embodiment of the present invention provides an apparatus monitoring system based on Convolutional Neural Networks (CNNs), including:

the sampling device is electrically connected to a device to be tested and used for collecting data of the device to be tested;

the control device comprises a computing unit, the computing unit is electrically connected with the sampling device and is used for receiving the data of the equipment to be tested, processing and judging the data of the equipment to be tested through a convolutional neural network architecture and outputting a corresponding judgment result; and

and the server is electrically connected with the control device and used for receiving the judgment result output by the calculation unit and storing the judgment result.

As a preferred scheme, the control device further includes a control end, the control end is electrically connected to the computing unit, and the control end is configured to output different control instructions to the acquisition device according to the determination result.

As a preferred scheme, the system further includes a power supply unit, the power supply unit is configured to supply power to the acquisition device, and the server is further configured to configure the operation parameters of the acquisition device and the calculation unit.

As a preferable scheme, the collecting device is further configured to:

entering a registration request state after power-on starting, and sending a registration request message with a device ID to the control end at intervals of a first preset time;

entering a waiting configuration state after receiving a receiving request message sent by the control terminal;

entering a state of preparing acquisition after receiving the effective configuration message within a second preset time;

when the trigger condition for starting acquisition is met, entering an acquisition state to acquire the data of the equipment to be detected;

when the acquisition preparation state and the acquisition state are in the acquisition preparation state, a heartbeat response is requested to the control end at intervals of a third preset time; and

and when the effective configuration message is not received within the second preset time or the heartbeat response is not received for a continuous preset number of times, returning to the registration request state to continuously request for registration from the control terminal.

As a preferred scheme, the sampling device includes sampling unit and singlechip, the sampling unit includes sampling resistance and signal processing circuit, sampling resistance is used for gathering the equipment data that awaits measuring, and will gather the equipment data that awaits measuring convert voltage signal into, signal processing circuit with the sampling resistance electricity is connected, is used for with voltage signal handles the back and exports to the singlechip, the singlechip includes analog-to-digital converter at least, analog-to-digital converter is used for with voltage signal carries out behind the analog-to-digital conversion output to the computational element.

As a preferred scheme, the single chip microcomputer is further configured to convert the control instruction of the control end into a corresponding driving signal to drive the corresponding output device.

Preferably, the server is further electrically connected to the sampling device, and the server is further configured to:

receiving data of equipment to be tested of the sampling device, and storing the data of the equipment to be tested; and

and counting and analyzing the data of the equipment to be tested so as to monitor and manage the equipment to be tested.

Preferably, the convolutional neural network architecture is disposed in the computing unit or the server.

As a preferred scheme, the convolutional neural network architecture is used to import data of devices to be tested, classify the data of the devices to be tested, perform model training to form corresponding model files, and the computing unit outputs the determination result according to the data of the devices to be tested and the model files.

As a preferred scheme, the control device supports a plurality of acquisition devices to acquire data of the device to be tested, and the server is connected to the plurality of control devices through a network support.

The embodiment of the invention also provides a device monitoring method based on CNNs, which is applied to a device monitoring system based on a convolutional neural network, wherein the system comprises a sampling device, a control device and a server, and the method comprises the following steps:

collecting data of equipment to be tested through the sampling device;

receiving the data of the equipment to be tested through the control device, processing and judging the data of the equipment to be tested through a convolutional neural network architecture, and outputting a corresponding judgment result; and

and receiving the judgment result through the server, and storing the judgment result.

As a preferable aspect, the method further includes:

outputting different control instructions to the acquisition device through the control device according to the judgment result;

and converting the control instruction into a corresponding driving signal through the acquisition device so as to drive corresponding output equipment.

As a preferred solution, the method further performs, by the acquisition device:

entering a registration request state after power-on starting, and sending a registration request message with a device ID to the control device at intervals of a first preset time;

entering a waiting configuration state after receiving a receiving request message sent by the control device;

and when the valid configuration message is not received within the second preset time or no heartbeat response is received for a preset number of times, returning to the registration request state to continuously request for registration from the control device.

As a preferred solution, the method further performs, by the server:

Preferably, the convolutional neural network architecture is disposed within the control device or the server.

The equipment monitoring system and method based on the CNNs can collect, mark, train, identify modes and other operations on the acquired data, can effectively detect the running state of the equipment to be detected without changing the structure of the equipment to be detected, and realize the monitoring management of the equipment to be detected. That is to say, the CNNs-based device monitoring system and method can realize intelligent identification, intelligent monitoring and intelligent control on the device to be tested, i.e., can use less cost, so that the device to be tested becomes an intelligent device.

Drawings

Fig. 1 is a functional block diagram of a CNNs-based device monitoring system in a preferred embodiment of the present invention.

Fig. 2 is a schematic diagram of signal or data transmission of the CNNs-based device monitoring system shown in fig. 1.

Fig. 3 is a schematic diagram illustrating the transition of each state of the sampling device in the CNNs-based device monitoring system shown in fig. 1.

Fig. 4 is a functional block diagram of a CNNs-based device monitoring system according to another preferred embodiment of the present invention.

Fig. 5 is a flowchart of a CNNs-based device monitoring method according to a preferred embodiment of the present invention.

Fig. 6 is a flowchart illustrating sub-steps of step S1 shown in fig. 5.

Description of the main elements

Equipment monitoring system based on convolutional neural network	1
		Sampling device	100
Sampling unit	11
		Sampling resistor	111
Signal processing circuit	113
		Single chip microcomputer	13
Control device	200
		Computing unit	201
Control terminal	203
		Server	300
Output device	400
		Power supplyUnit cell	500

The following detailed description will further illustrate the invention in conjunction with the above-described figures.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.

Referring to fig. 1, fig. 1 is a functional block diagram of a Convolutional Neural Network (CNNs) based device monitoring system 1 according to a preferred embodiment of the present invention. The monitoring system 1 can be applied to automatic screw machine waiting equipment for monitoring the running state of the equipment to be monitored, so as to realize intelligent identification, intelligent monitoring and intelligent control. In this embodiment, the device under test is taken as an example of an electric screwdriver.

The equipment monitoring system 1 at least includes a sampling device 100, a control device 200 and a server 300. The sampling device 100 is electrically connected to the device under test for collecting data of the device under test. In this embodiment, the device under test is an electric screwdriver. The sampling device 100 collects a value of a current flowing through a load, i.e., the electric screwdriver. Of course, in other embodiments, the sampling device 100 is not limited to collecting current data, but may also collect voltage data, impedance data, spindle rotation ratio of the device under test, or other data.

The control device 200 comprises at least a calculation unit 201. The computing unit 201 is electrically connected to the sampling device 100. The computing unit 201 is configured to receive the device data to be tested, process and determine the device data to be tested through a Convolutional Neural Networks (CNNs) architecture, and output a corresponding determination result.

It is understood that a convolutional Neural network is a kind of feed forward Neural network (fed forward Neural Networks) containing convolutional calculation and having a deep structure, and is one of the representative algorithms of deep learning (deep learning). Convolutional Neural Networks have a feature learning (representation learning) capability, and can perform Shift-Invariant classification (Shift-Invariant classification) on input information according to a hierarchical structure thereof, and are also called Shift-Invariant Artificial Neural Networks (SIANN).

In this embodiment, the convolutional neural network architecture may be used to import data of a device to be tested, classify the data of the device to be tested, and perform model training to form a corresponding model file. In this way, the computing unit 201 outputs the determination result according to the data of the device to be tested and the model file. For example, the computing unit 201 may determine whether a screw is driven normally or OK, an electric screwdriver idles, a screw is tilted, the screw is roughened, and the like according to the data of the device under test.

The server 300 is electrically connected to the control device 200. The server 300 is configured to receive the determination results output by the computing unit 201 in the control device 200, such as a screw driving OK, a screwdriver idling, a screw tilting, a screw roughening, and store the determination results.

It should be understood that, referring to fig. 2, in the present embodiment, the sampling device 100 includes a sampling unit 11 and a single chip 13. The sampling unit 11 includes a sampling resistor 111 and a signal processing circuit 113. The sampling resistor 111 is used for collecting the data of the equipment to be tested and converting the collected data of the equipment to be tested into a voltage signal. The signal processing circuit 113 is electrically connected to the sampling resistor 111. The signal processing circuit 113 is configured to process the voltage signal (for example, amplify the voltage signal) and output the processed voltage signal to the single chip 13.

It can be understood that, in the present embodiment, the single chip 13 at least includes an analog-to-digital converter (not shown). The analog-to-digital converter is configured to perform analog-to-digital conversion on the voltage signal and output the voltage signal to the computing unit 201. The calculating unit 201 processes and determines the voltage signal after the analog-to-digital conversion, and further outputs the determination result.

It is understood that, in the present embodiment, the control device 200 further includes a control terminal 203. The control end 203 is electrically connected with the computing unit 201 and the single chip microcomputer 13 in the sampling device 100. The control end 203 is configured to output different control instructions to the single chip 13 according to the determination result of the computing unit 201. Thus, when receiving the control instruction, the single chip 13 is further configured to convert the control instruction of the control terminal 203 into a corresponding driving signal to drive the corresponding output device 400. For example, in the present embodiment, the single chip 13 can generate different electrical signals to drive the output device 400 installed on the sampling apparatus 100, so as to generate an audible and visual signal to inform the operator.

It is understood that the output device 400 may be an LED indicator, a buzzer, a digital tube, or other electronic components, modules, or devices that can notify an operator.

It should be understood that fig. 3 is a schematic diagram of the state transition of the acquisition apparatus 100. When the acquisition device 100 acquires data, it enters a registration request state after power-on is started, and sends a registration request message with a device ID to the control device 200 every first preset time. After receiving the receiving request message sent by the control device 200, that is, when the request registration information sent by the acquisition device 100 is accepted by the control device 200, the acquisition device 100 enters a waiting configuration state. Then, the acquisition device 100 enters a ready-to-acquire state after receiving a valid configuration message within a second predetermined time. In the ready-to-collect state, when the trigger condition for starting collection is satisfied, the collection device 100 enters the collection state to collect the data of the device under test.

It is understood that, in the present embodiment, when the sampling device 100 is in the ready-to-collect state and the collecting state, it also requests a heartbeat response from the control device 200 every third preset time. When the acquisition device 100 does not receive the heartbeat response for a preset number of consecutive times, for example, four consecutive times, it will determine that the communication between the two has been interrupted, and then return to the registration request state to continue requesting registration from the control device 200.

Similarly, in this embodiment, when the sampling device 100 does not receive a valid configuration message within the second predetermined time, the sampling device also returns to the registration request state to continue requesting registration from the control device 200.

It can be understood that, in this embodiment, when the acquisition apparatus 100 is in the ready-to-acquire state, the sampling apparatus 100 will continuously read a voltage signal and determine whether the voltage signal satisfies the trigger condition for acquisition. For example, whether the value of the voltage signal is greater than a predetermined value is determined. When the condition is satisfied, that is, when it is determined that the value of the voltage signal is greater than a preset value, the acquisition device 100 enters the acquisition state, that is, the acquisition device 100 acquires data of the device to be tested, and transmits a sampling result to the control device 200.

It can be understood that, in this embodiment, the first preset time, the second preset time, and the third preset time may all be set according to requirements. For example, the first preset time and the second preset time may be set to 10 seconds, and the third preset time may be set to 30 seconds.

It should be understood that, referring to fig. 4, in other embodiments, the number of the sampling devices 100 and the number of the control devices 200 in the system 1 are not limited to one, and the specific number thereof may be adjusted according to the requirement. For example, the control device 200 may support a plurality of collecting devices 100 to collect data of the device under test, so as to reduce the cost. Also, the server 300 may support connection to a plurality of control devices 200 through a network. In this manner, the server 300 can receive data from the plurality of control apparatuses 200, and set or configure parameters of the plurality of control apparatuses 200 and the sampling apparatus 100. For example, the server 300 may configure the operation parameters of the sampling device 100 and the computing unit 201 through the control terminal 203, such as the type of the screwdriver, the torque, the number of screws, the threshold value of the current detection, and the like. Thus, when the sampling device 100 receives a valid configuration message or parameter from the server 300 within the second predetermined time, the sampling device may enter the ready-to-collect state.

It is understood that in other embodiments, the server 300 is further configured to receive the device under test data of the sampling apparatus 100 and store the device under test data. Thus, the server 300 may monitor and manage the device under test by counting and analyzing the data of the device under test. For example, the server 300 performs statistics and analysis on the received data of the device under test to implement:

(1) the screw that probably appears is bad to report an emergency and ask for help or increased vigilance, so quality control only need to detect the product of reporting an emergency and asking for help or increased vigilance can, effectively improved detection efficiency.

(2) The screw driven by the automatic screw machine is detected, and a warning is given to the driving failure. Therefore, whether the screwdriver runs normally can be identified according to the received current sequence value, and the state of the screwdriver equipment can be warned in advance, namely the screwdriver equipment is checked in advance.

(3) And (4) according to the information of the screwdriver, evaluating the screw driving skills of the staff, and the like.

It is understood that referring again to fig. 4, in other embodiments, the system 1 further comprises a power supply unit 500. The power supply unit 500 is electrically connected to the collection device 100 to supply power to the collection device 100.

It is understood that the convolutional neural network architecture may be provided within the computing unit 201 or the server 300. When the convolutional neural network architecture is provided in the computing unit 201, the system 1 can operate and perform offline determination even if the server 300 stops operating. Of course, this has a high requirement on the computing performance of the computing unit 201, and when the model is updated, the synchronous deployment at multiple control ends is required. In addition, when the convolutional neural network architecture is set in the server 300, the server 300 receives the data of the device to be tested, and performs classification and model training on the data of the device to be tested to form a corresponding model file.

It is understood that in the present embodiment, the sampling apparatus 100 is provided independently of the device under test. Of course, in other embodiments, the sampling device 100 may also be directly disposed in the device under test. In this way, the control device 200 can be directly electrically connected to the sampling device 100 on the device under test without an additional cable, which effectively simplifies the structure of the device monitoring system 1. Furthermore, when the sampling device 100 is directly disposed on the device under test, the single chip 13 in the sampling device 100 can directly control the output device 400 electrically connected thereto. Thus, the output device 400 may also be disposed on the device under test along with the sampling apparatus 100, so that a user may more intuitively monitor the operation state of the device under test through the output device 400.

Referring to fig. 5, another embodiment of the present invention further provides a device monitoring method based on CNNs. The equipment monitoring method can be applied to the equipment monitoring system 1 based on the CNNs and is used for monitoring the running state of the equipment to be tested, and intelligent identification, intelligent monitoring and intelligent control are achieved. In this embodiment, the device under test is taken as an example of an electric screwdriver. It will be understood that the order of the steps in the flow diagrams may be changed and certain steps may be omitted, depending on various requirements. It is understood that, in the present embodiment, the electronic component is taken as an example to specifically describe the method.

Step S1, acquiring data of the device under test by the sampling apparatus 100.

It can be understood that, in the present embodiment, the device under test is an electric screwdriver. The sampling device 100 collects a value of a current flowing through a load, i.e., the electric screwdriver.

Step S2, receiving the device data to be tested through the control device 200, processing and determining the device data to be tested through a convolutional neural network architecture, and outputting a corresponding determination result.

In this embodiment, the convolutional neural network architecture may be used to import data of a device to be tested, classify the data of the device to be tested, and perform model training to form a corresponding model file. In this way, the control device 200 outputs the determination result according to the data of the device to be tested and the model file. For example, the control device 200 may determine whether a screw is driven normally or OK, an electric screwdriver idles, a screw is tilted, the screw is roughened, and the like according to the data of the device under test.

Step S3, receiving the determination result through the server 300, and storing the determination result.

It is understood that in other embodiments, the method further comprises steps S4, S5.

Step S4, outputting different control commands to the collecting device 100 through the control device 200 according to the determination result.

Step S5, the acquisition device 100 converts the control command into a corresponding driving signal to drive a corresponding output device.

It is understood that the driving signal may be different electrical signals to drive an output device mounted on the collecting apparatus 100. For example, to generate an audible and visual signal to notify the operator.

It will be appreciated that the output device may be an LED indicator, buzzer, digital tube, or other electronic element, module, or device that may notify the operator.

It is understood that in other embodiments, the method further comprises steps S6, S7.

Step S6, receiving, by the server 300, the device data to be tested of the sampling apparatus 100, and storing the device data to be tested.

And step S7, counting and analyzing the data of the equipment to be tested so as to monitor and manage the equipment to be tested.

For example, the server 300 performs statistics and analysis on the received data of the device under test to implement:

It is understood that, referring to fig. 6, the step S1 specifically includes the following sub-steps:

in sub-step S11, the sampling device 100 is powered up.

In sub-step S12, a registration request state is entered, and a registration request message with a device ID is sent to the control device 200 at a first predetermined time interval.

In the sub-step S13, after receiving the reception request message sent by the control device 200, the control device enters a wait configuration state.

In sub-step S14, the acquisition ready state is entered after receiving a valid configuration message within a second predetermined time.

And a substep S15, entering a collection state to collect the data of the device to be tested when a trigger condition for starting collection is met.

It is understood that in the sub-steps S14 and S15, i.e., in the ready-to-acquire state and the acquisition state, a heartbeat response is requested from the control device 200 at every third predetermined time. When a valid configuration message is not received within the second predetermined time, the control device returns to the registration request state, i.e., returns to sub-step S12, so as to continue to request registration from the control device 200.

Similarly, in the present embodiment, in the sub-step S13, when the sampling device 100 does not receive a valid configuration message within the second predetermined time, the sampling device returns to the registration request state, i.e., returns to the sub-step S12, so as to continue to request registration from the control device 200.

Obviously, the CNNs-based equipment monitoring system and method performs pattern recognition on data collected by the sampling device 100 by setting the sampling device 100, the control device 200 and the server 300 and simultaneously using a convolutional neural network structure. Therefore, collected data can be collected, marked, trained, pattern recognized and the like, the running state of the equipment to be tested can be effectively detected without changing the structure of the equipment to be tested, and monitoring management of the equipment to be tested is realized. That is to say, the CNNs-based device monitoring system and method can realize intelligent identification, intelligent monitoring and intelligent control on the device to be tested, i.e., can use less cost, so that the device to be tested becomes an intelligent device.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is to be understood that the word "comprising" does not exclude other modules or steps, and the singular does not exclude the plural. Several modules or electronic devices recited in the electronic device claims may also be implemented by one and the same module or electronic device by means of software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A convolutional neural network-based device monitoring system, the system comprising: the device comprises a sampling device, a control device and a server;

the sampling device comprises a sampling unit and a single chip microcomputer, the sampling unit comprises a sampling resistor and a signal processing circuit, the sampling resistor is electrically connected to an electric screwdriver and used for acquiring data of the electric screwdriver and converting the acquired data of the electric screwdriver into a voltage signal, and the signal processing circuit is electrically connected with the sampling resistor and used for processing the voltage signal and outputting the processed voltage signal to the single chip microcomputer; the sampling device is further configured to:

when the trigger condition for starting acquisition is met, entering an acquisition state to acquire the data of the electric screwdriver;

requesting a heartbeat response to the control device at intervals of a third preset time when the acquisition preparation state and the acquisition state are in the acquisition preparation state; and

when the valid configuration message is not received within the second preset time or no heartbeat response is received for a preset number of times, returning to the registration request state to continue requesting registration from the control device;

the control device comprises a computing unit, the computing unit is electrically connected with the sampling device and used for receiving the voltage signals after analog-to-digital conversion, and the computing unit outputs a judgment result according to the voltage signals after analog-to-digital conversion and a model file, wherein the model file is formed by introducing the voltage signals through a convolutional neural network architecture so as to classify the voltage signals and performing model training, and the judgment result comprises the judgment of whether the driven screws are normal or not; and

the server is electrically connected with the control device and used for receiving the judgment result output by the calculation unit and storing the judgment result, and the server is also used for counting and analyzing the voltage signal so as to realize warning for driving abnormal screws and/or evaluating the screw driving skill of staff; the server is further configured to configure the sampling device and the operating parameters of the computing unit, so that the sampling device enters a state of being prepared for acquisition after receiving a valid configuration message within a second preset time.

2. The system of claim 1, wherein: the control device also comprises a control end, the control end is electrically connected with the computing unit, and the control end is used for outputting different control instructions to the sampling device according to the judgment result.

3. The system of claim 2, wherein: the system further comprises a power supply unit for supplying power to the sampling device.

4. A system according to claim 2 or 3, characterized in that: the sampling device further comprises a single chip microcomputer, the single chip microcomputer at least comprises an analog-to-digital converter, and the analog-to-digital converter is used for performing analog-to-digital conversion on the voltage signal output by the signal processing circuit and outputting the voltage signal to the computing unit.

5. The system of claim 4, wherein: the single chip microcomputer is further used for converting the control instruction of the control end into a corresponding driving signal so as to drive corresponding output equipment.

6. The system of claim 1, wherein: the server is also electrically connected to the sampling device, the server further configured to:

receiving data of the electric screwdriver of the sampling device, and storing the data of the electric screwdriver; and

and counting and analyzing the data of the electric screwdriver so as to monitor and manage the electric screwdriver.

7. The system of claim 1, wherein: the convolutional neural network architecture is disposed within the computing unit or server.

8. The system of claim 1, wherein: the control device supports a plurality of sampling devices to collect data of the electric screwdriver, and the server is connected to the plurality of control devices through network support.

9. The utility model provides an equipment monitoring method based on convolutional neural network, is applied to in an equipment monitoring system based on convolutional neural network, the system includes sampling device, controlling means and server, its characterized in that, the sampling device includes sampling unit and singlechip, the sampling unit includes sampling resistance and signal processing circuit, the controlling means includes the computational element, the computational element with the sampling device electricity is connected, the method includes:

the sampling resistor is electrically connected to an electric screwdriver for collecting data of the electric screwdriver and converting the collected data of the electric screwdriver into a voltage signal, and the signal processing circuit is electrically connected with the sampling resistor and used for processing the voltage signal and outputting the processed voltage signal to the single chip microcomputer;

the method is also performed by the sampling device:

when the valid configuration message is not received within the second preset time or no heartbeat response is received for a preset number of times, returning to the registration request state to continue requesting registration from the control device; receiving the voltage signal subjected to analog-to-digital conversion through the computing unit, and outputting a judgment result according to the voltage signal subjected to analog-to-digital conversion and a model file, wherein the model file is formed by introducing the voltage signal through a convolutional neural network architecture to classify the voltage signal and performing model training, and the judgment result comprises judging whether a driven screw is normal or not; and

the server is used for receiving the judgment result and storing the judgment result, and is also used for counting and analyzing the voltage signal so as to realize warning for driving abnormal screws and/or evaluating the screw driving skill of staff; the server is further configured to configure the sampling device and the operating parameters of the computing unit, so that the sampling device enters a state of being prepared for acquisition after receiving a valid configuration message within a second preset time.

10. The method of claim 9, wherein: the control device further comprises a control end, and the method further comprises the following steps:

the control end is used for outputting different control instructions to the sampling device according to the judgment result;

and converting the control instruction into a corresponding driving signal through the sampling device so as to drive corresponding output equipment.

11. The method of claim 9, wherein: the method further performs, by the server:

12. The method of claim 9, wherein: the convolutional neural network architecture is disposed within the control device or server.