Industrial safety counter-control equipment system and method

Abstract

The invention relates to the technical field of industrial safety counter-control, and discloses an industrial safety counter-control equipment system and method, which collect the vibration frequency and corresponding dust distribution map of a target device at fixed time intervals; pre-process the dust distribution map, and construct a dust vibration feature matrix according to the pre-processed dust distribution map; input the dust vibration feature matrix into a preset vibration monitoring model, and the vibration monitoring model outputs the vibration frequency of the target device; adjust the output power of the target device within a verification time, and determine the verification frequency of the target device through the vibration monitoring model according to the dust distribution map of the verification time; compare the vibration frequency with the verification frequency, and determine the state and processing plan of the target device; improve the reliability of the monitoring system and the detection accuracy of target device failures, thereby improving production safety and equipment service life.

Description

Technical Field

[0001] The present invention relates to the field of industrial safety counter-control, and more specifically, to an industrial safety counter-control equipment system and method. Background Art

[0002] Heavy machinery and equipment are widely used in industrial operations. Heavy machinery and equipment usually generate large vibrations during operation. Mechanical equipment vibration is closely related to the health status of the equipment. Abnormal vibration is usually an early signal of equipment wear, failure or unstable operation. Therefore, monitoring the vibration status of heavy machinery and equipment is an important means to ensure the normal operation of equipment and production safety.

[0003] However, vibration monitoring equipment is often placed in a harsh working environment with heavy machinery and equipment, which leads to frequent failures of the monitoring equipment itself, thus affecting the safety and stability of production. However, by monitoring the distribution of dust generated by the vibration of heavy machinery and equipment in the mine, effective feedback can be given on the vibration frequency of the machinery and equipment. Therefore, there is an urgent need for a system and method for obtaining the vibration frequency of heavy machinery and equipment by visually identifying dust. Summary of the invention

[0004] The present invention provides an industrial safety anti-control equipment system and method to solve the technical problems raised in the background technology.

[0005] In a first aspect, an industrial safety anti-control equipment system includes a data acquisition module, a data processing module, a vibration monitoring module, a verification module and an abnormality monitoring module:

[0006] A data acquisition module, used to collect the vibration frequency of the target device and the corresponding dust distribution map at fixed time intervals;

[0007] The data processing module is used to preprocess the dust distribution map and construct a dust vibration feature matrix according to the preprocessed dust distribution map, including:

[0008] Obtaining a non-working state image of the target device in the working area, and setting the non-working state image as a reference image;

[0009] The dust distribution map is used to register the reference image through the key points of the target device, and the size of the dust distribution map is kept the same as that of the reference image by scaling, cropping and padding the dust distribution map;

[0010] Gray-scale the reference image to obtain a gray-scale matrix, and set the gray-scale matrix as the reference matrix;

[0011] Gray-scale the collected dust distribution map to obtain a corresponding gray-scale matrix;

[0012] Calculate the difference between the grayscale matrix of the dust distribution map and the reference matrix to obtain a valid matrix;

[0013] The difference of the effective matrices obtained at adjacent time points is calculated and the difference is normalized to obtain the dust vibration feature matrix, the size of which is M×N;

[0014] The vibration monitoring module is used to input the dust vibration feature matrix into the preset vibration monitoring model; wherein the dust vibration feature matrix is ​​used as a training sample of the vibration monitoring model; and the vibration frequency corresponding to the dust vibration feature matrix is ​​used as a sample label of the vibration monitoring model, as follows:

[0015] The vibration monitoring model includes a first hidden layer and a second hidden layer;

[0016] The first hidden layer of the vibration monitoring model is used to extract feature maps from the dust vibration feature matrix;

[0017] The second hidden layer of the vibration monitoring model is used to input the feature map and output the hidden state to the classifier;

[0018] The classification space of the classifier represents the vibration frequency corresponding to the dust vibration feature matrix;

[0019] The vibration monitoring model is evaluated by the mean error variance loss function;

[0020] The output of the vibration monitoring model represents the vibration frequency of the target equipment;

[0021] The verification module is used to adjust the output power of the target device during the verification time and determine the verification frequency of the target device through the vibration monitoring model according to the dust distribution map of the verification time, including:

[0022] By stopping the operation of the target device during a preset test period, the time required for the dust to settle naturally is collected, and the time is set as the verification time u;

[0023] Through the vibration monitoring model, it is obtained that the target equipment has abnormal vibration frequency at the i-th moment;

[0024] Determine the verification power according to the output power of the target device at the i-th moment; the verification power is less than the output power of the target device at the i-th moment;

[0025] At the i-th moment, the target device adjusts its output power to the verification power; the target device runs at the verification power for a verification time of u;

[0026] The vibration frequency of the target device at the i+uth moment is obtained through the vibration monitoring model, and the vibration frequency is set as the verification frequency;

[0027] The abnormality monitoring module is used to compare the vibration frequency with the verification frequency and determine the state of the target device and a processing solution; wherein determining the state of the target device includes:

[0028] Compare the verification frequency with the vibration frequency of the target device judged by the vibration monitoring model at the i-th moment;

[0029] If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset equal value threshold range, it is determined that there is an interference entity in the scene where the target device is located;

[0030] If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset high threshold range, it is judged that the target device has a serious abnormality;

[0031] If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset low threshold range, it is judged ;in, and They represent the output power of the target device at the i-th moment and the vibration frequency of the target device determined by the vibration monitoring model. and , respectively represent the output power of the target device at the i+uth moment and the vibration frequency of the target device judged by the vibration monitoring model, represents the weight coefficient; if the equation is true, the target device has a slight abnormality; if the equation is not true, the target device has a serious abnormality;

[0032] in, The equation holds true within the preset equal value threshold range. It is obtained by collecting historical data and fitting it according to the linear least squares method.

[0033] Furthermore, the calculation formula of the first hidden layer of the vibration monitoring model is as follows:

[0034] ;

[0035] ;

[0036] ;

[0037] ;

[0038] in, represents the feature map of the first hidden layer output, represents the interaction feature of the nth row and mth column of the dust vibration feature matrix, represents the characteristic matrix of the nth row of the dust vibration characteristic matrix, represents the feature matrix of the mth column of the dust vibration feature matrix, represents the dust vibration feature matrix; n and m represent the row index and column index respectively, and N and M represent the number of rows and columns of the dust vibration feature matrix respectively; Represents the weight of the interaction feature between the nth row and the mth column of the dust vibration feature matrix in the feature map output by the first hidden layer, represents the weight between the nth row and the mth column of the dust vibration feature matrix, and Respectively represent the weights of the nth row and mth column of the dust vibration feature matrix; and They represent the offset of the nth row and mth column of the dust vibration feature matrix respectively; express Activation function;

[0039] The calculation formula of the second hidden layer of the vibration monitoring model is as follows:

[0040] ;

[0041] in, represents the hidden state of the second hidden layer output, represents the weight of the feature map output by the first hidden layer, represents the bias of the feature map output by the first hidden layer, a represents the index, A represents the number of data processing modules in the second hidden layer, express Activation function;

[0042] Each data processing module in the second hidden layer processes a row of vectors in the feature map output by the first hidden layer respectively.

[0043] Furthermore, the processing solution for the target device is determined, including:

[0044] Determine the processing plan for the target device according to the status of the target device;

[0045] If it is determined that there is an interfering entity in the scene where the target device is located, apply to the person in charge of the target device to change the security anti-control monitoring method of the target device;

[0046] If it is determined that the target device has a serious abnormality, the target device will be shut down and repaired immediately;

[0047] If it is determined that the target device has a slight abnormality, the target device is regularly inspected and repaired.

[0048] In a second aspect, an industrial safety counter-control device method is applied to the industrial safety counter-control device system, comprising:

[0049] Step S201, collecting the vibration frequency and corresponding dust distribution map of the target device at fixed time intervals;

[0050] Step S202, preprocessing the dust distribution map, and constructing a dust vibration feature matrix according to the preprocessed dust distribution map;

[0051] Step S203, inputting the dust vibration feature matrix into a preset vibration monitoring model, and the vibration monitoring model outputs the vibration frequency of the target device;

[0052] Step S204, adjusting the output power of the target device within the verification time, and determining the verification frequency of the target device through a vibration monitoring model according to the dust distribution map of the verification time;

[0053] Step S205, comparing the vibration frequency with the verification frequency, and determining the state of the target device and a processing solution.

[0054] According to a third aspect, a computing device includes:

[0055] one or more processors;

[0056] A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 3.

[0057] The beneficial effects of the present invention are as follows: by collecting the dust distribution diagram of the target device during operation, the vibration frequency of the device is analyzed, and the abnormal state of the target device is judged by the verification frequency obtained by adjusting the output power of the target device, thereby effectively improving the detection accuracy of equipment failures, and improving production safety and the service life of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a module diagram of an industrial safety counter-control equipment system of the present invention;

[0059] Figure 2 It is a flow chart of an industrial safety reverse control equipment method of the present invention.

[0060] In the figure, 101 is a data acquisition module; 102 is a data processing module; 103 is a vibration monitoring module; 104 is a verification module; and 105 is an abnormality monitoring module. DETAILED DESCRIPTION

[0061] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0062] like Figure 1-Figure 2 As shown, an industrial safety anti-control equipment system includes a data acquisition module 101, a data processing module 102, a vibration monitoring module 103, a verification module 104 and an abnormality monitoring module 105:

[0063] The data collection module 101 is used to collect the vibration frequency of the target device and the corresponding dust distribution map at fixed time intervals;

[0064] The data processing module 102 is used to pre-process the dust distribution map and construct a dust vibration feature matrix according to the pre-processed dust distribution map, including:

[0065] Obtaining a non-working state image of the target device in the working area, and setting the non-working state image as a reference image;

[0066] The dust distribution map is used to register the reference image through the key points of the target device, and the size of the dust distribution map is kept the same as that of the reference image by scaling, cropping and padding the dust distribution map;

[0067] Gray-scale the reference image to obtain a gray-scale matrix, and set the gray-scale matrix as the reference matrix;

[0068] Gray-scale the collected dust distribution map to obtain a corresponding gray-scale matrix;

[0069] Calculate the difference between the grayscale matrix of the dust distribution map and the reference matrix to obtain a valid matrix;

[0070] The difference of the effective matrices obtained at adjacent time points is calculated and the difference is normalized to obtain the dust vibration feature matrix, the size of which is M×N;

[0071] The vibration monitoring module 103 is used to input the dust vibration feature matrix into a preset vibration monitoring model; wherein the dust vibration feature matrix is ​​used as a training sample of the vibration monitoring model; and the vibration frequency corresponding to the dust vibration feature matrix is ​​used as a sample label of the vibration monitoring model, as follows:

[0072] The vibration monitoring model includes a first hidden layer and a second hidden layer;

[0073] The first hidden layer of the vibration monitoring model is used to extract feature maps from the dust vibration feature matrix;

[0074] The second hidden layer of the vibration monitoring model is used to input the feature map and output the hidden state to the classifier;

[0075] The classification space of the classifier represents the vibration frequency corresponding to the dust vibration feature matrix;

[0076] The vibration monitoring model is evaluated by the mean error variance loss function;

[0077] The output of the vibration monitoring model represents the vibration frequency of the target equipment;

[0078] The verification module 104 is used to adjust the output power of the target device within the verification time, and determine the verification frequency of the target device through the vibration monitoring model according to the dust distribution map of the verification time, including:

[0079] By stopping the operation of the target device during a preset test period, the time required for the dust to settle naturally is collected, and the time is set as the verification time u;

[0080] Through the vibration monitoring model, it is obtained that the target equipment has abnormal vibration frequency at the i-th moment;

[0081] Determine the verification power according to the output power of the target device at the i-th moment; the verification power is less than the output power of the target device at the i-th moment;

[0082] At the i-th moment, the target device adjusts its output power to the verification power; the target device runs at the verification power for a verification time of u;

[0083] The vibration frequency of the target device at the i+uth moment is obtained through the vibration monitoring model, and the vibration frequency is set as the verification frequency;

[0084] The abnormality monitoring module 105 is used to compare the vibration frequency with the verification frequency and determine the state of the target device and a processing solution; wherein determining the state of the target device includes:

[0085] Compare the verification frequency with the vibration frequency of the target device judged by the vibration monitoring model at the i-th moment;

[0086] If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset equal value threshold range, it is determined that there is an interference entity in the scene where the target device is located;

[0087] If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset high threshold range, it is judged that the target device has a serious abnormality;

[0088] If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset low threshold range, it is judged ;in, and They represent the output power of the target device at the i-th moment and the vibration frequency of the target device determined by the vibration monitoring model. and , respectively represent the output power of the target device at the i+uth moment and the vibration frequency of the target device judged by the vibration monitoring model, represents the weight coefficient; if the equation is true, the target device has a slight abnormality; if the equation is not true, the target device has a serious abnormality;

[0089] in, The equation holds true within the preset equal value threshold range. It is obtained by collecting historical data and fitting it according to the linear least squares method.

[0090] It should be noted that registering the dust distribution map with the reference image through the key points of the target device is a prior art and will not be described in detail here.

[0091] For example, the system will stop the operation of the target device within a preset test period in order to observe the natural settling time of dust on the target device in a stationary state. This time is called "verification duration u" and is used to determine the normal dust distribution state of the target device in the absence of vibration. For example, if the system collects the time required for the natural settling of dust as 10 minutes, the verification duration u is set to 10 minutes. During the normal operation of the target device, the system monitors the vibration frequency of the target device through the vibration monitoring model. If the system detects that the vibration frequency of the target device is abnormal at the i-th moment through the vibration monitoring model, it means that the target device may be faulty or unstable. After detecting the abnormal vibration frequency, the system will determine a lower verification power based on the output power of the target device at the i-th moment. The verification power must be less than the current output power of the target device to reduce the load of the target device so that the system can better evaluate the operation status of the target device in a lower power state. After the target device is operated within the verification duration and the output power of the target device is adjusted to the verification power, the system will let the target device run at the verification power for the entire verification duration u. At the i+uth moment, the system collects a new dust distribution map and obtains the vibration frequency of the target device at the i+uth moment through the vibration monitoring model. This vibration frequency is called the "verification frequency" and represents the vibration condition of the target device after adjusting the output power.

[0092] It should be noted that by collecting the power and corresponding rated vibration frequency of the target device at multiple gears, and verifying the corresponding rated vibration frequency under the power,

[0093] Bringing data into , by fitting the data with the linear least squares method, we get .

[0094] It should be noted that when the ratio of the verification frequency and the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset equivalence threshold range, it means that there is an external interference entity that has increased the dust vibration data of the scenario in which the target device is operating, causing the vibration monitoring model to be unable to identify the vibration frequency of the target device at the i-th moment and the actual vibration frequency at the i+u-th moment; at this time, the vibration monitoring model needs to apply to the person in charge of the target device to replace the safety counter-control monitoring method of the target device.

[0095] It should be noted that when the ratio of the verification frequency and the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset low threshold range, it means that the vibration frequency of the target device has not dropped significantly within the verification time u, which usually reflects that there are serious abnormalities inside the equipment. High vibration frequency may be caused by problems such as equipment component wear, mechanical imbalance, excessive load, etc. If it fails to improve significantly at a lower output power, it means that the equipment failure is serious and it is difficult for the system to restore to normal by simply adjusting the power. At this time, the vibration monitoring model will determine that the equipment has a serious abnormality and immediately issue a warning to the person in charge of the equipment, suggesting that the equipment be shut down and inspected immediately to prevent the failure from further deteriorating or causing a safety accident.

[0096] It should be noted that if the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset low threshold range, in the first case: if the ratio of the verification frequency to the vibration frequency meets the preset equation; it means that after the target device reduces the output power, the change in the vibration frequency meets the logical law; then it is determined that the target device has a slight abnormality, and it is recommended to continue to operate the device, but regular maintenance needs to be arranged to ensure the stable operation of the target device. In the second case: if the ratio of the verification frequency to the vibration frequency does not meet the preset equation; it means that after the target device reduces the power, the change in its vibration frequency does not meet the logical law. Then it is determined that the target device has a serious abnormality and needs to be shut down for maintenance.

[0097] In one embodiment of the present invention, the calculation formula of the first hidden layer of the vibration monitoring model is as follows:

[0098] ;

[0099] ;

[0100] ;

[0101] ;

[0102] in, represents the feature map of the first hidden layer output, represents the interaction feature of the nth row and mth column of the dust vibration feature matrix, represents the characteristic matrix of the nth row of the dust vibration characteristic matrix, represents the feature matrix of the mth column of the dust vibration feature matrix, represents the dust vibration feature matrix; n and m represent the row index and column index respectively, and N and M represent the number of rows and columns of the dust vibration feature matrix respectively; Represents the weight of the interaction feature between the nth row and the mth column of the dust vibration feature matrix in the feature map output by the first hidden layer, represents the weight between the nth row and the mth column of the dust vibration feature matrix, and Respectively represent the weights of the nth row and mth column of the dust vibration feature matrix; and They represent the offset of the nth row and mth column of the dust vibration feature matrix respectively; express Activation function;

[0103] The calculation formula of the second hidden layer of the vibration monitoring model is as follows:

[0104] ;

[0105] in, represents the hidden state of the second hidden layer output, represents the weight of the feature map output by the first hidden layer, represents the bias of the feature map output by the first hidden layer, a represents the index, A represents the number of data processing modules in the second hidden layer, express Activation function;

[0106] Each data processing module in the second hidden layer processes a row of vectors in the feature map output by the first hidden layer respectively.

[0107] It should be noted that if the size of the feature map generated by the first hidden layer is 2×2, the second hidden layer includes two data processing modules, each data processing module includes two data processing units; the output of the second data processing unit of each data processing module is added to obtain the final hidden state.

[0108] In one embodiment of the present invention, determining a processing scheme for a target device includes:

[0109] Determine the processing plan for the target device according to the status of the target device;

[0110] If it is determined that there is an interfering entity in the scene where the target device is located, apply to the person in charge of the target device to change the security anti-control monitoring method of the target device;

[0111] If it is determined that the target device has a serious abnormality, the target device will be shut down and repaired immediately;

[0112] If it is determined that the target device has a slight abnormality, the target device is regularly inspected and repaired.

[0113] An industrial safety counter-control device method, applied to an industrial safety counter-control device system, comprises:

[0114] Step S201, collecting the vibration frequency and corresponding dust distribution map of the target device at fixed time intervals;

[0115] Step S202, preprocessing the dust distribution map, and constructing a dust vibration feature matrix according to the preprocessed dust distribution map;

[0116] Step S203, inputting the dust vibration feature matrix into a preset vibration monitoring model, and the vibration monitoring model outputs the vibration frequency of the target device;

[0117] Step S204, adjusting the output power of the target device within the verification time, and determining the verification frequency of the target device through a vibration monitoring model according to the dust distribution map of the verification time;

[0118] Step S205, comparing the vibration frequency with the verification frequency, and determining the state of the target device and a processing solution.

[0119] A computing device comprising:

[0120] one or more processors;

[0121] A storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the industrial safety anti-control equipment method.

[0122] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. An industrial safety counter-control equipment system, characterized in that: It comprises a data acquisition module (101), a data processing module (102), a vibration monitoring module (103), a verification module (104) and an abnormality monitoring module (105): A data collection module (101) is used to collect the vibration frequency of the target device and the corresponding dust distribution map at fixed time intervals; The data processing module (102) is used to pre-process the dust distribution map and construct a dust vibration feature matrix according to the pre-processed dust distribution map, including: Obtaining a non-working state image of the target device in the working area, and setting the non-working state image as a reference image; The dust distribution map is used to register the reference image through the key points of the target device, and the size of the dust distribution map is kept the same as that of the reference image by scaling, cropping and padding the dust distribution map; Gray-scale the reference image to obtain a gray-scale matrix, and set the gray-scale matrix as the reference matrix; Gray-scale the collected dust distribution map to obtain a corresponding gray-scale matrix; Calculate the difference between the grayscale matrix of the dust distribution map and the reference matrix to obtain a valid matrix; The difference of the effective matrices obtained at adjacent time points is calculated and the difference is normalized to obtain the dust vibration feature matrix, the size of which is M×N; The vibration monitoring module (103) is used to input the dust vibration feature matrix into a preset vibration monitoring model; wherein the dust vibration feature matrix is ​​used as a training sample of the vibration monitoring model; and the vibration frequency corresponding to the dust vibration feature matrix is ​​used as a sample label of the vibration monitoring model, specifically as follows: The vibration monitoring model includes a first hidden layer and a second hidden layer; The first hidden layer of the vibration monitoring model is used to extract feature maps from the dust vibration feature matrix; The second hidden layer of the vibration monitoring model is used to input the feature map and output the hidden state to the classifier; The classification space of the classifier represents the vibration frequency corresponding to the dust vibration feature matrix; The vibration monitoring model is evaluated by the mean error variance loss function; The output of the vibration monitoring model represents the vibration frequency of the target equipment; The verification module (104) is used to adjust the output power of the target device within the verification time, and determine the verification frequency of the target device through a vibration monitoring model according to the dust distribution map of the verification time, including: By stopping the operation of the target device during a preset test period, the time required for the dust to settle naturally is collected, and the time is set as the verification time u; Through the vibration monitoring model, it is obtained that the target equipment has abnormal vibration frequency at the i-th moment; Determine the verification power according to the output power of the target device at the i-th moment; the verification power is less than the output power of the target device at the i-th moment; At the i-th moment, the target device adjusts its output power to the verification power; the target device runs at the verification power for a verification time of u; The vibration frequency of the target device at the i+uth moment is obtained through the vibration monitoring model, and the vibration frequency is set as the verification frequency; The abnormality monitoring module (105) is used to compare the vibration frequency with the verification frequency and determine the state of the target device and a processing solution; wherein determining the state of the target device includes: Compare the verification frequency with the vibration frequency of the target device judged by the vibration monitoring model at the i-th moment; If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset equal value threshold range, it is determined that there is an interference entity in the scene where the target device is located; If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset high threshold range, it is judged that the target device has a serious abnormality; If the ratio of the verification frequency to the vibration frequency judged by the vibration monitoring model for the target device at the i-th moment is within the preset low threshold range, it is judged ;in, and They represent the output power of the target device at the i-th moment and the vibration frequency of the target device determined by the vibration monitoring model. and , respectively represent the output power of the target device at the i+uth moment and the vibration frequency of the target device judged by the vibration monitoring model, represents the weight coefficient; if the equation is true, the target device has a slight abnormality; if the equation is not true, the target device has a serious abnormality; in, The equation holds true within the preset equal value threshold range. It is obtained by collecting historical data and fitting it according to the linear least squares method.

2. An industrial safety counter-control equipment system according to claim 1, characterized in that: The calculation formula of the first hidden layer of the vibration monitoring model is as follows: ； ； ； ； in, represents the feature map of the first hidden layer output, represents the interaction feature of the nth row and mth column of the dust vibration feature matrix, represents the characteristic matrix of the nth row of the dust vibration characteristic matrix, represents the feature matrix of the mth column of the dust vibration feature matrix, represents the dust vibration feature matrix; n and m represent the row index and column index respectively, and N and M represent the number of rows and columns of the dust vibration feature matrix respectively; Represents the weight of the interaction feature between the nth row and the mth column of the dust vibration feature matrix in the feature map output by the first hidden layer, represents the weight between the nth row and the mth column of the dust vibration feature matrix, and Respectively represent the weights of the nth row and mth column of the dust vibration feature matrix; and They represent the offset of the nth row and mth column of the dust vibration feature matrix respectively; express Activation function; The calculation formula of the second hidden layer of the vibration monitoring model is as follows: ； in, represents the hidden state of the second hidden layer output, represents the weight of the feature map output by the first hidden layer, represents the bias of the feature map output by the first hidden layer, a represents the index of the number of data processing modules of the second hidden layer, A represents the number of data processing modules of the second hidden layer, express Activation function; Each data processing module in the second hidden layer processes a row of vectors in the feature map output by the first hidden layer respectively.

3. An industrial safety counter-control equipment system according to claim 1, characterized in that: Determine the treatment plan for the target device, including: Determine the processing plan for the target device according to the status of the target device; If it is determined that there is an interfering entity in the scene where the target device is located, apply to the person in charge of the target device to change the security anti-control monitoring method of the target device; If it is determined that the target device has a serious abnormality, the target device will be shut down and repaired immediately; If it is determined that the target device has a slight abnormality, the target device is regularly inspected and repaired.

4. An industrial safety counter-control device method, applied to an industrial safety counter-control device system according to any one of claims 1 to 3, characterized in that: include: Step S201, collecting the vibration frequency and corresponding dust distribution map of the target device at fixed time intervals; Step S202, preprocessing the dust distribution map, and constructing a dust vibration feature matrix according to the preprocessed dust distribution map; Step S203, inputting the dust vibration feature matrix into a preset vibration monitoring model, and the vibration monitoring model outputs the vibration frequency of the target device; Step S204, adjusting the output power of the target device within the verification time, and determining the verification frequency of the target device through a vibration monitoring model according to the dust distribution map of the verification time; Step S205, comparing the vibration frequency with the verification frequency, and determining the state of the target device and a processing solution.

5. A computing device, characterized in that: include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to claim 4.

Images