Vibration testing device and testing method for production of electromechanical special equipment

Abstract

The present invention relates to the field of vibration detection technology for traction equipment, and in particular to a vibration inspection device and detection method for the production of electromechanical special equipment, including setting i sampling points on the passenger compartment, guide rails and traction mechanism of the traction equipment, installing vibration sensors on the i sampling points, and collecting vibration data; establishing a vibration data set based on the vibration data collected at each sampling point on the passenger compartment, recorded as a first vibration data set; obtaining a frequency spectrum of the vibration data of each sampling point in the first vibration data set; inputting the obtained frequency spectrum of each sampling point into a trained vibration type recognition model to obtain the type of vibration data. The present invention comprehensively analyzes vibration characteristics through multi-point collection and data collected multiple times at each sampling point, and uses machine learning methods for classification, and then judges the vibration condition of the equipment, thereby improving the representativeness and reliability of the data, and can effectively help to timely discover and solve traction equipment operation problems.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration detection of traction equipment, and in particular to a vibration detection device and a detection method for production of electromechanical special equipment. Background Art

[0002] At present, for special equipment, such as traction machines and passenger ropeways, the monitoring equipment inside them can only be used for video monitoring and complete some analysis and detection services, and then send the detection results to the client, or initiate an alarm to the client, and the management personnel will handle the event according to the alarm situation. That is, the current monitoring equipment detection service for special equipment is relatively simple, and can only alarm based on video conditions. It is impossible to perform all-round vibration detection based on special equipment and supporting facilities that affect the safety of special equipment to ensure the safety of special equipment operation;

[0003] The patent application with publication number CN118565858A discloses a method, device, storage medium and engineering machinery for detecting the driving stability of engineering machinery. The method includes: when the current posture of the engineering machinery reaches the preset posture, collecting the time parameters and operation parameters of the engineering machinery in each detection stage; determining the driving evaluation parameters and vibration evaluation parameters of the engineering machinery in each detection stage according to the time parameters and the operation parameters; determining the driving detection results and vibration detection results of the engineering machinery in each detection stage based on the driving evaluation parameters and the vibration evaluation parameters; determining whether the driving stability of the engineering machinery is qualified according to the driving detection results and vibration detection results of all detection stages, so that the driving stability test of the engineering machinery is more accurate;

[0004] As in the above application, the prior art is based on vibration detection of special equipment, which generally relies solely on installing vibration sensors directly on the special equipment to perform vibration detection, analyzing the vibration detection results based on timing or state posture, and cooperating with an alarm to sound an alarm. However, it is impossible to accurately analyze the operating status of the special equipment by simply collecting the vibration of the special equipment itself for alarm. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a vibration inspection device and a detection method for the production of electromechanical special equipment.

[0006] The present invention adopts the following technical scheme, which is used for a vibration detection method for the production of electromechanical special equipment and is applied to traction equipment, including:

[0007] Set i sampling points on the passenger compartment, guide rails and traction mechanism of the traction equipment, install vibration sensors on i (i>3) sampling points to collect vibration data;

[0008] The vibration data collected from each sampling point on the passenger compartment are used to establish a vibration data set, which is recorded as a first vibration data set;

[0009] Obtain a frequency spectrum of vibration data of each sampling point in the first vibration data set;

[0010] Inputting the obtained spectrum of each sampling point into the trained vibration type recognition model to obtain the type of vibration data, wherein the type of vibration data includes passenger compartment operation vibration and human vibration;

[0011] Establishing a passenger compartment operation vibration data set from the vibration data of the sampling points belonging to the passenger compartment operation vibration in the first vibration data set, and generating a passenger compartment operation stability coefficient based on the evaluation of the passenger compartment operation vibration data set;

[0012] The vibration data collected from each sampling point on the guide rail is used to establish a guide rail vibration data set, which is recorded as the guide rail vibration data set, and a guide rail stability coefficient is generated based on the guide rail vibration data set evaluation;

[0013] The vibration data collected from each sampling point on the traction mechanism are used to establish a traction mechanism vibration data set, which is recorded as the traction mechanism vibration data set, and the traction mechanism operation stability coefficient is generated based on the traction mechanism vibration data set evaluation;

[0014] The operating stability coefficient of the traction equipment is generated based on the analysis of the operating stability coefficient of the passenger car, the guide rail stability coefficient and the traction mechanism operating stability coefficient. The operating stability coefficient of the traction equipment is compared and analyzed with the preset operating stability coefficient gradient threshold to generate an alarm instruction.

[0015] As a further description of the above technical solution: the method for evaluating and generating the passenger compartment operation stability coefficient based on the passenger compartment operation vibration data set includes:

[0016] The passenger compartment operation vibration data set is expressed as: ;in, represents the vibration data of the zth sampling at the Fth sampling point;

[0017] Obtain the root mean square value of the vibration data of each sampling point in the passenger compartment operation vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Jp and Jf respectively;

[0018] The calculation formula of passenger compartment operation stability coefficient is: ;

[0019] In the formula, is the passenger compartment operation stability coefficient, and is the weight coefficient, and Both are greater than zero.

[0020] As a further description of the above technical solution: the method for evaluating and generating the guide rail stability coefficient based on the guide rail vibration data set includes:

[0021] The guide rail vibration data set is: .in, Represents the vibration data of the pth sampling at the Tth sampling point;

[0022] Obtain the root mean square value of the vibration data of each sampling point in the guide rail vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Dp and Df respectively;

[0023] The calculation formula of guide rail stability coefficient is: ;

[0024] In the formula, is the guide rail stability factor, and is the weight coefficient, and Both are greater than zero.

[0025] As a further description of the above technical solution: the method for evaluating and generating the traction mechanism operation stability coefficient based on the traction mechanism vibration data set includes:

[0026] The traction mechanism vibration data set is expressed as: ,in, Represents the vibration data of the nth sampling at the mth sampling point;

[0027] Obtain the root mean square value of the vibration data of each sampling point in the traction mechanism vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Yp and Yf ​​respectively;

[0028] The calculation formula of the traction mechanism operation stability coefficient is: ;

[0029] In the formula, is the operating stability coefficient of the traction mechanism, and is the weight coefficient, and Both are greater than zero.

[0030] As a further description of the above technical solution: the alarm instruction includes an alarm reminder instruction and an alarm shutdown instruction;

[0031] The method of comparing and analyzing the running stability coefficient of the traction equipment with a preset running stability coefficient gradient threshold and generating an alarm instruction includes:

[0032] The obtained passenger car operation stability coefficient, guide rail stability coefficient and traction mechanism operation stability coefficient are weightedly summed to obtain the traction equipment operation stability coefficient, which is recorded as DTyxws;

[0033] Preset operation stability coefficient gradient thresholds DT1 and DT2, and DT1<DT2;

[0034] When DTyxws<DT1, no alarm instruction is generated;

[0035] When DT1≤DTyxws<DT2, an alarm reminder instruction is generated;

[0036] When DTyxws≥DT2, an alarm shutdown command is generated.

[0037] As a further description of the above technical solution: it also includes:

[0038] The vibration data belonging to artificial vibration in the first vibration data set is used to establish an artificial vibration data set;

[0039] Obtaining an artificial vibration coefficient of vibration data of each sampling point in an artificial vibration data set;

[0040] The obtained artificial vibration coefficient is compared and analyzed with the preset artificial vibration coefficient threshold, and a voice warning command is generated.

[0041] As a further description of the above technical solution: the parameters affecting the artificial vibration coefficient include the RMS value, peak value and kurtosis of the vibration data of each sampling point in the artificial vibration data set;

[0042] The calculation formula of the artificial vibration coefficient is:

[0043]

[0044] In the formula, is the artificial vibration coefficient, is the RMS value, is the peak value, is the kurtosis, , and is the weight coefficient, , and Both are greater than zero.

[0045] As a further description of the above technical solution: the training method of the vibration type recognition model includes:

[0046] The spectrum of the vibration data of the sampling points in the passenger compartment of the traction equipment is collected, and the vibration type of the spectrum is labeled, that is, the label is the passenger compartment operation vibration and artificial vibration, the spectrum and the label corresponding to the spectrum constitute a set of training data, X groups of training data constitute a sample set, X is an integer greater than 1, and the sample set is divided into a training set and a test set; the spectrum in the training set is used as the input of the vibration type recognition model, and the label in the training set is used as the output of the vibration type recognition model, and the vibration type recognition model is trained to obtain an initial vibration type recognition model, and the minimization of the sum of prediction errors is the training goal, and the initial vibration type recognition model is evaluated using the test set, and the initial vibration type recognition model when the sum of prediction errors reaches convergence is used as the constructed vibration type recognition model;

[0047] The vibration type recognition model is a deep neural network model.

[0048] As a further description of the above technical solution: the voice warning command includes a voice reminder command and a voice alarm command;

[0049] The method of comparing and analyzing the acquired artificial vibration coefficient with a preset artificial vibration coefficient threshold to generate a voice warning command includes:

[0050] Preset artificial vibration coefficient thresholds RW1 and RW2, where RW1<RW2;

[0051] when When <RW1, no voice warning command is generated;

[0052] When RW1≤ <RW2, generate voice reminder command;

[0053] when ≥RW 2， When the voice alarm command is generated.

[0054] Vibration testing device for the production of electromechanical special equipment, including vibration sensor and processor;

[0055] The vibration sensor is used to be installed on the passenger compartment, guide rails and traction mechanism of the traction equipment to collect vibration data;

[0056] The processor is used to execute the vibration detection method for the production of electromechanical special equipment.

[0057] Beneficial effects:

[0058] The vibration inspection device and detection method for the production of electromechanical special equipment are used to perform vibration detection on traction equipment. When in use, the device collects vibration data by setting sampling points on the passenger compartment, guide rails and traction mechanism of the traction equipment, and distinguishes the collected passenger compartment vibration data into passenger compartment operation vibration and human vibration by training a vibration type recognition model. Then, the stability of the traction equipment operation is judged by comprehensively analyzing the passenger compartment operation vibration data, guide rail vibration data and traction mechanism vibration data. Compared with the prior art that only collects vibration data by installing a vibration sensor on the passenger compartment to analyze the operation stability of the traction equipment, the data collected by the device is more comprehensive. Vibration characteristics are comprehensively analyzed through multi-point collection and data collected multiple times at each sampling point, and machine learning methods are used for classification. Then, the vibration condition of the equipment is judged, thereby improving the representativeness and reliability of the data, which can effectively help to timely discover and solve the operation problems of the traction equipment.

[0059] Furthermore, through the trained vibration type recognition model, the artificial vibration data is classified, that is, the vibration data belonging to artificial vibration in the first vibration data set is used to establish an artificial vibration data set, the artificial vibration coefficient of the vibration data of each sampling point in the artificial vibration data set is obtained, the obtained artificial vibration coefficient is compared and analyzed with the preset artificial vibration coefficient threshold, and a voice warning command is generated, that is, intelligent judgment of the influence of artificial vibration is realized, and reasonable voice warning commands are generated based on the influence of artificial vibration, so as to avoid alarming and panic just because of slight vibration caused by human entry and exit of traction equipment, and overcome the influence of artificial vibration on the judgment of stable operation of traction equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments:

[0061] Figure 1 A flow chart of a vibration detection method for producing electromechanical special equipment provided in Example 1 of the present invention;

[0062] Figure 2 This is a flow chart of a vibration detection method for the production of electromechanical special equipment provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0063] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention is further described below with reference to specific diagrams. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0064] Example 1

[0065] See also Figure 1The embodiment of the present invention provides a technical solution: a vibration detection method for the production of electromechanical special equipment, applied to traction equipment, comprising:

[0066] Set i sampling points on the passenger compartment, guide rails and traction mechanism of the traction equipment, install vibration sensors at i sampling points, i>3, and collect vibration data;

[0067] The vibration data collected from each sampling point on the passenger compartment are used to establish a vibration data set, which is recorded as a first vibration data set;

[0068] Obtain a frequency spectrum of vibration data of each sampling point in the first vibration data set;

[0069] Specifically, the method for obtaining the spectrum graph is to use the fast Fourier transform (FFT) to convert the time domain signal to the frequency domain, and then use the Matplotlib library to draw the frequency domain data into a spectrum graph;

[0070] It should be noted that the vibrations generated by the operation of the traction equipment usually have specific frequency characteristics. These frequencies are related to the mechanical movement of the traction equipment, the rotation of the traction mechanism, etc. The vibrations generated by humans (such as the footsteps and jumping of passengers) are usually low in frequency and irregular, and can be intuitively displayed on the spectrum graph.

[0071] Inputting the obtained spectrum of each sampling point into the trained vibration type recognition model to obtain the type of vibration data, wherein the type of vibration data includes passenger compartment operation vibration and human vibration;

[0072] The training method of the vibration type recognition model includes:

[0073] The spectrum of the vibration data of the sampling points in the passenger compartment of the traction equipment is collected, and the vibration type of the spectrum is labeled, that is, the label is the passenger compartment operation vibration and artificial vibration, the spectrum and the label corresponding to the spectrum constitute a set of training data, X groups of training data constitute a sample set, X is an integer greater than 1, and the sample set is divided into a training set and a test set; the spectrum in the training set is used as the input of the vibration type recognition model, and the label in the training set is used as the output of the vibration type recognition model, and the vibration type recognition model is trained to obtain an initial vibration type recognition model, and the minimization of the sum of prediction errors is the training goal, and the initial vibration type recognition model is evaluated using the test set, and the initial vibration type recognition model when the sum of prediction errors reaches convergence is used as the constructed vibration type recognition model;

[0074] The vibration type recognition model is a deep neural network model;

[0075] It should be noted that in the machine learning model, the calculation formula for prediction error is: ,in, is the number of the characteristic data (i.e., the spectrum graph), is the prediction error, For the The predicted state value corresponding to the group feature data, the state value is the vibration type, For the The actual status value corresponding to the group feature data.

[0076] Establishing a passenger compartment operation vibration data set from the vibration data of the sampling points belonging to the passenger compartment operation vibration in the first vibration data set, and generating a passenger compartment operation stability coefficient based on the evaluation of the passenger compartment operation vibration data set;

[0077] The method for evaluating and generating the passenger compartment operation stability coefficient based on the passenger compartment operation vibration data set includes:

[0078] The passenger compartment operation vibration data set is expressed as: ;in, represents the vibration data of the zth sampling at the Fth sampling point;

[0079] Obtain the root mean square value of the vibration data of each sampling point in the passenger compartment operation vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Jp and Jf respectively;

[0080] The calculation formula of passenger compartment operation stability coefficient is: ;

[0081] In the formula, is the passenger compartment operation stability coefficient, and is the weight coefficient, and Both are greater than zero.

[0082] It should be noted that the size of the weight coefficient is a specific value obtained by quantifying each data to facilitate subsequent comparison. The size of the weight coefficient depends on the number of comprehensive parameters and the preliminary setting of the corresponding weight coefficient for each set of comprehensive parameters by technical personnel in this field.

[0083] Specifically, the root mean square value represents the average energy of the vibration data and is an indicator to measure the vibration intensity. Therefore, the larger the Jp, the greater the vibration intensity; the larger the Jf, the greater the vibration volatility, and vice versa. In summary, the larger the passenger compartment operation stability coefficient, the worse the passenger compartment operation stability.

[0084] The formula for calculating the RMS value is: , where R represents the root mean square value, Indicates the sampling point The data collected, Indicates the total number of samples at the sampling point.

[0085] For example, the formula for calculating the RMS value of the first sampling point in the passenger compartment is:

[0086] .

[0087] The vibration data collected from each sampling point on the guide rail is used to establish a guide rail vibration data set, which is recorded as the guide rail vibration data set, and a guide rail stability coefficient is generated based on the guide rail vibration data set evaluation;

[0088] Methods for evaluating and generating guide rail stability factors based on guide rail vibration data sets include:

[0089] The guide rail vibration data set is: .in, Represents the vibration data of the pth sampling at the Tth sampling point;

[0090] Obtain the root mean square value of the vibration data of each sampling point in the guide rail vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Dp and Df respectively;

[0091] The calculation formula of guide rail stability coefficient is: ;

[0092] In the formula, is the guide rail stability factor, and is the weight coefficient, and Both are greater than zero.

[0093] It should be noted that the size of the weight coefficient is a specific value obtained by quantifying each data to facilitate subsequent comparison. The size of the weight coefficient depends on the number of comprehensive parameters and the preliminary setting of the corresponding weight coefficient for each set of comprehensive parameters by technical personnel in this field.

[0094] Specifically, the root mean square value represents the average energy of the vibration data and is an indicator for measuring the vibration intensity. Therefore, the larger the Dp, the greater the vibration intensity, and the larger the Df, the greater the vibration volatility, and vice versa. In summary, the larger the guide rail stability coefficient, the worse the stability of the guide rail.

[0095] The vibration data collected from each sampling point on the traction mechanism are used to establish a traction mechanism vibration data set, which is recorded as the traction mechanism vibration data set, and the traction mechanism operation stability coefficient is generated based on the traction mechanism vibration data set evaluation;

[0096] The method for evaluating and generating the running stability coefficient of the traction mechanism based on the traction mechanism vibration data set includes:

[0097] The traction mechanism vibration data set is expressed as: ,in, Represents the vibration data of the nth sampling at the mth sampling point;

[0098] Obtain the root mean square value of the vibration data of each sampling point in the traction mechanism vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Yp and Yf ​​respectively;

[0099] The calculation formula of the traction mechanism operation stability coefficient is: ;

[0100] In the formula, is the operating stability coefficient of the traction mechanism, and is the weight coefficient, and Both are greater than zero.

[0101] It should be noted that the size of the weight coefficient is a specific value obtained by quantifying each data to facilitate subsequent comparison. The size of the weight coefficient depends on the number of comprehensive parameters and the preliminary setting of the corresponding weight coefficient for each set of comprehensive parameters by technical personnel in this field.

[0102] Specifically, the root mean square value represents the average energy of the vibration data and is an indicator for measuring the vibration intensity. Therefore, the larger the Yp, the greater the vibration intensity; the larger the Yf, the greater the vibration volatility, and vice versa. In summary, the larger the operating stability coefficient of the traction mechanism, the worse the stability of the traction mechanism.

[0103] Generate the traction equipment operation stability coefficient based on the passenger compartment operation stability coefficient, guide rail stability coefficient and traction mechanism operation stability coefficient, compare and analyze the traction equipment operation stability coefficient with the preset operation stability coefficient gradient threshold, and generate an alarm instruction;

[0104] The alarm instructions include alarm reminder instructions and alarm shutdown instructions;

[0105] The method of comparing and analyzing the running stability coefficient of the traction equipment with a preset running stability coefficient gradient threshold and generating an alarm instruction includes:

[0106] The obtained passenger car operation stability coefficient, guide rail stability coefficient and traction mechanism operation stability coefficient are weightedly summed to obtain the traction equipment operation stability coefficient, which is recorded as DTyxws;

[0107] Preset operation stability coefficient gradient thresholds DT1 and DT2, and DT1<DT2;

[0108] When DTyxws<DT1, no alarm instruction is generated;

[0109] When DT1≤DTyxws<DT2, an alarm reminder instruction is generated;

[0110] When DTyxws≥DT2, an alarm shutdown command is generated.

[0111] In the present embodiment, the vibration detection method for the production of electromechanical special equipment is used to perform vibration detection on traction equipment. When in use, it collects vibration data by setting sampling points on the passenger car, guide rails and traction mechanism of the traction equipment, and distinguishes the collected passenger car vibration data into passenger car operation vibration and human vibration by training a vibration type recognition model. Then, the passenger car operation vibration data, guide rail vibration data and traction mechanism vibration data are comprehensively analyzed to judge the stability of the traction equipment operation. Compared with the prior art that only collects vibration data by installing a vibration sensor on the passenger car to analyze the operation stability of the traction equipment, the data collected is more comprehensive, realizes multi-point collection and multiple data collection at each sampling point, comprehensively analyzes the vibration characteristics, and uses machine learning methods for classification, and then judges the vibration condition of the equipment, thereby improving the representativeness and reliability of the data, which can effectively help to timely discover and solve traction equipment operation problems.

[0112] Example 2

[0113] See also Figure 1 and Figure 2 Based on the above embodiment, this embodiment adds:

[0114] Establishing an artificial vibration data set from the vibration data belonging to artificial vibration in the first vibration data set;

[0115] Obtaining an artificial vibration coefficient of vibration data of each sampling point in an artificial vibration data set;

[0116] The obtained artificial vibration coefficient is compared and analyzed with the preset artificial vibration coefficient threshold, and a voice warning command is generated.

[0117] The parameters affecting the artificial vibration coefficient include the RMS value, peak value and kurtosis of the vibration data of each sampling point in the artificial vibration data set;

[0118] The calculation formula of the artificial vibration coefficient is:

[0119]

[0120] In the formula, is the artificial vibration coefficient, is the RMS value, is the peak value, is the kurtosis, , and is the weight coefficient, , and Both are greater than zero.

[0121] It should be noted that the size of the weight coefficient is a specific value obtained by quantifying each data to facilitate subsequent comparison. The size of the weight coefficient depends on the number of comprehensive parameters and the preliminary setting of the corresponding weight coefficient for each set of comprehensive parameters by technical personnel in this field.

[0122] It should be noted that the methods for obtaining the RMS value, peak value and kurtosis of vibration data are all existing technologies and will not be elaborated on again. The calculation method of the RMS value has been explained above, and the peak value is the maximum absolute value in the vibration data.

[0123] Specifically, the larger the RMS value, the higher the overall energy of the vibration signal and the greater the vibration intensity; the smaller the RMS value, the lower the overall energy of the vibration signal and the smaller the vibration intensity, which may indicate that the equipment is running smoothly without significant vibration problems;

[0124] The larger the peak value, the more significant the instantaneous high amplitude is in the vibration signal, which may be caused by shock or impact. On the contrary, the smaller the peak value, the smaller the instantaneous amplitude is in the vibration signal, which may be caused by shock or impact.

[0125] The greater the kurtosis, the more sharp peaks there are in the vibration signal. The higher the kurtosis of the signal, the higher the kurtosis value is usually a sign of abnormal vibration.

[0126] The voice warning command includes a voice reminder command and a voice alarm command.

[0127] The method of comparing and analyzing the acquired artificial vibration coefficient with a preset artificial vibration coefficient threshold to generate a voice warning command includes:

[0128] Preset artificial vibration coefficient thresholds RW1 and RW2, where RW1<RW2;

[0129] when When <RW1, no voice warning command is generated;

[0130] When RW1≤ <RW2, generate voice reminder command;

[0131] when ≥RW 2，When the voice alarm command is generated.

[0132] In this embodiment, the artificial vibration data is classified through the trained vibration type recognition model, that is, the vibration data belonging to artificial vibration in the first vibration data set is used to establish an artificial vibration data set, and the artificial vibration coefficient of the vibration data of each sampling point in the artificial vibration data set is obtained. The obtained artificial vibration coefficient is compared and analyzed with the preset artificial vibration coefficient threshold, and a voice warning command is generated. In this way, intelligent judgment of the impact of artificial vibration is achieved, and reasonable voice warning commands are generated based on the impact of artificial vibration, so as to avoid alarming and causing panic just because of slight vibration caused by human entry and exit of traction equipment.

[0133] Example 3

[0134] Vibration testing device for the production of electromechanical special equipment, including vibration sensor and processor;

[0135] The vibration sensor is used to be installed on the passenger compartment, guide rails and traction mechanism of the traction equipment to collect vibration data;

[0136] The processor is used to execute the vibration detection method for the production of electromechanical special equipment.

[0137] The basic principles, main features and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and the description in the specification are only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A vibration detection method for the production of electromechanical special equipment, applied to traction equipment, characterized in that: include: Set i sampling points on the passenger compartment, guide rails and traction mechanism of the traction equipment, install vibration sensors at i sampling points, i>3, and collect vibration data; The vibration data collected from each sampling point on the passenger compartment are used to establish a vibration data set, which is recorded as a first vibration data set; Obtain a frequency spectrum of vibration data of each sampling point in the first vibration data set; Inputting the obtained spectrum of each sampling point into the trained vibration type recognition model to obtain the type of vibration data, wherein the type of vibration data includes passenger compartment operation vibration and human vibration; Establishing a passenger compartment operation vibration data set from the vibration data of the sampling points belonging to the passenger compartment operation vibration in the first vibration data set, and generating a passenger compartment operation stability coefficient based on the evaluation of the passenger compartment operation vibration data set; The vibration data collected from each sampling point on the guide rail is used to establish a guide rail vibration data set, which is recorded as the guide rail vibration data set, and a guide rail stability coefficient is generated based on the guide rail vibration data set evaluation; The vibration data collected from each sampling point on the traction mechanism are used to establish a traction mechanism vibration data set, which is recorded as the traction mechanism vibration data set, and the traction mechanism operation stability coefficient is generated based on the traction mechanism vibration data set evaluation; The operating stability coefficient of the traction equipment is generated based on the analysis of the operating stability coefficient of the passenger car, the guide rail stability coefficient and the traction mechanism operating stability coefficient. The operating stability coefficient of the traction equipment is compared and analyzed with the preset operating stability coefficient gradient threshold to generate an alarm instruction.

2. The vibration detection method for the production of electromechanical special equipment according to claim 1, characterized in that: The method for evaluating and generating the passenger compartment operation stability coefficient based on the passenger compartment operation vibration data set includes: The passenger compartment operation vibration data set is expressed as: ;in, represents the vibration data of the zth sampling at the Fth sampling point; Obtain the root mean square value of the vibration data of each sampling point in the passenger compartment operation vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Jp and Jf respectively; The calculation formula of passenger compartment operation stability coefficient is: ; In the formula, is the passenger compartment operation stability coefficient, and is the weight coefficient, and Both are greater than zero.

3. The vibration detection method for the production of electromechanical special equipment according to claim 2, characterized in that: Methods for evaluating and generating guide rail stability factors based on guide rail vibration data sets include: The guide rail vibration data set is: ,in, Represents the vibration data of the pth sampling at the Tth sampling point; Obtain the root mean square value of the vibration data of each sampling point in the guide rail vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Dp and Df respectively; The calculation formula of guide rail stability coefficient is: ; In the formula, is the guide rail stability factor, and is the weight coefficient, and Both are greater than zero.

4. The vibration detection method for the production of electromechanical special equipment according to claim 3 is characterized in that: The method for evaluating and generating the running stability coefficient of the traction mechanism based on the traction mechanism vibration data set includes: The traction mechanism vibration data set is expressed as: ,in, Represents the vibration data of the nth sampling at the mth sampling point; Obtain the root mean square value of the vibration data of each sampling point in the traction mechanism vibration data set, perform statistical analysis on the root mean square values ​​of the vibration data of all sampling points, and obtain their mean and variance, which are recorded as Yp and Yf ​​respectively; The calculation formula of the traction mechanism operation stability coefficient is: ; In the formula, is the operating stability coefficient of the traction mechanism, and is the weight coefficient, and Both are greater than zero.

5. The vibration detection method for the production of electromechanical special equipment according to claim 4, characterized in that: The alarm instructions include alarm reminder instructions and alarm shutdown instructions; The method of comparing and analyzing the running stability coefficient of the traction equipment with a preset running stability coefficient gradient threshold and generating an alarm instruction includes: The obtained passenger car operation stability coefficient, guide rail stability coefficient and traction mechanism operation stability coefficient are weightedly summed to obtain the traction equipment operation stability coefficient, which is recorded as DTyxws; Preset operation stability coefficient gradient thresholds DT1 and DT2, and DT1<DT2; When DTyxws<DT1, no alarm instruction is generated; When DT1≤DTyxws<DT2, an alarm reminder instruction is generated; When DTyxws≥DT2, an alarm shutdown command is generated.

6. The vibration detection method for electromechanical special equipment production according to claim 1, characterized in that: Also includes: Establishing an artificial vibration data set from the vibration data belonging to artificial vibration in the first vibration data set; Obtaining an artificial vibration coefficient of vibration data of each sampling point in an artificial vibration data set; The obtained artificial vibration coefficient is compared and analyzed with the preset artificial vibration coefficient threshold, and a voice warning command is generated.

7. The vibration detection method for the production of electromechanical special equipment according to claim 6, characterized in that: The parameters affecting the artificial vibration coefficient include the RMS value, peak value and kurtosis of the vibration data of each sampling point in the artificial vibration data set; The calculation formula of the artificial vibration coefficient is: ； In the formula, is the artificial vibration coefficient, is the RMS value, is the peak value, is the kurtosis, , and is the weight coefficient, , and Both are greater than zero.

8. The vibration detection method for the production of electromechanical special equipment according to claim 1, characterized in that: The training method of the vibration type recognition model includes: The spectrum graph of the vibration data of the sampling points in the passenger compartment of the traction equipment is collected, and the vibration type of the spectrum graph is labeled, that is, the label is the passenger compartment operation vibration and artificial vibration, the spectrum graph and the label corresponding to the spectrum graph constitute a set of training data, X groups of training data constitute a sample set, X is an integer greater than 1, and the sample set is divided into a training set and a test set; the spectrum graph in the training set is used as the input of the vibration type recognition model, and the label in the training set is used as the output of the vibration type recognition model, and the vibration type recognition model is trained to obtain an initial vibration type recognition model, and the minimization of the sum of prediction errors is the training goal, and the initial vibration type recognition model is evaluated using the test set, and the initial vibration type recognition model when the sum of prediction errors reaches convergence is used as the constructed vibration type recognition model; The vibration type recognition model is a deep neural network model.

9. The vibration detection method for the production of electromechanical special equipment according to claim 7, characterized in that: The voice warning command includes a voice reminder command and a voice alarm command; The method of comparing and analyzing the acquired artificial vibration coefficient with a preset artificial vibration coefficient threshold to generate a voice warning command includes: Preset artificial vibration coefficient thresholds RW1 and RW2, where RW1<RW2; when When <RW1, no voice warning command is generated; When RW1≤ <RW2, generate voice reminder command; when ≥RW 2， When the voice alarm command is generated.

10. A vibration test device for the production of electromechanical special equipment, characterized in that: including a vibration sensor and a processor; The vibration sensor is used to be installed on the passenger compartment, guide rails and traction mechanism of the traction equipment to collect vibration data; The processor is used to execute the vibration detection method for the production of electromechanical special equipment as described in any one of claims 1 to 9.

Images