A method, system, device, and medium for removing ambient noise from a vibration wave filter

Abstract

The application discloses a kind of vibration wave filtering method, system, equipment and medium for eliminating environmental noise, the application includes the intensity peak data corresponding to vibration wave real-time waveform data in target vibration waveform data, preset waveform time data and vibration wave frequency are input into preset fault judgment model, and output fault value;Determine the fault result of fault value based on preset fault threshold value;Using the intensity peak data corresponding to second vibration waveform data in target vibration waveform data and vibration wave real-time waveform data respectively, calculate the true value of fault result;Judge whether true value is greater than or equal to preset true threshold value, determine the fault signal of fault result according to the result of judging.The technical problem that the present technology is less accurate is solved.The application improves the authenticity of judgment, and the sampling vibration wave filtering needs to have adaptive specific application requirements for the complex and complex interference wave of the installation distribution environment of medium voltage distribution switch.

Description

Technical Field

[0001] The present invention relates to the field of vibration wave filtering technology, and in particular to a vibration wave filtering method, system, equipment and medium for eliminating environmental noise. Background Art

[0002] With the development of distribution networks, the mechanical performance and quality of distribution switches have become increasingly prominent. Many electrical faults often originate from mechanical failures. Major mechanical failures of distribution switches include contact wear, ring jamming, loose static contacts, contact jamming, and bracket breakage. Due to the large number of switch manufacturers and the continuous evolution of switching methods and disconnecting technologies, diagnostic methods solely relying on electrical characteristics are increasingly unable to meet current requirements. Therefore, the use of mechanical vibration waves to monitor equipment has become an important monitoring method.

[0003] However, currently, vibration wave monitoring under the impact of opening and closing can effectively detect some mechanical abnormalities. However, when using the above method, since the distribution switches are installed outdoors, there are many interference factors in daily vibration wave monitoring, which to a certain extent restricts the reliability of monitoring switch mechanical problems through vibration wave characteristics.

[0004] Therefore, existing technologies usually use the vibration data of the fiber optic gyroscope for filtering processing, but the vibration noise suppression method of the above method is defective. It cannot use the filtering method for noisy outdoor environments such as bird sounds, car sounds and human voices, and cannot provide a relatively pure vibration wave of the mechanical state of the switch, resulting in low recognition accuracy. Summary of the Invention

[0005] The present invention provides a vibration wave filtering method, system, device and medium for eliminating environmental noise, which solves the technical problem that the existing technology cannot use filtering methods for noisy outdoor environments such as bird sounds, car sounds and human voices, and cannot provide relatively pure vibration waves of the mechanical state of the switch, resulting in low recognition accuracy.

[0006] A first aspect of the present invention provides a vibration wave filtering method for eliminating environmental noise, comprising:

[0007] In response to a vibration wave filtering instruction request, obtaining target vibration waveform data of a medium voltage distribution switch corresponding to the vibration wave filtering instruction request;

[0008] Inputting the target vibration waveform data into a preset target detection model, and extracting intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data;

[0009] Inputting the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model and outputting a fault value;

[0010] Determining a fault result of the fault value based on a preset fault threshold;

[0011] Calculating a true value of the fault result by using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration waveform data in the target vibration waveform data;

[0012] It is determined whether the true value is greater than or equal to a preset true threshold value, and a fault signal of the fault result is determined according to the determination result.

[0013] Optionally, the step of obtaining target vibration waveform data of the medium voltage distribution switch corresponding to the vibration wave filtering instruction request in response to the vibration wave filtering instruction request includes:

[0014] In response to a vibration wave filtering instruction request, determining a medium voltage distribution switch corresponding to the vibration wave filtering instruction request;

[0015] Vibration wave data acquisition equipment is used to respectively collect first vibration wave data and vibration wave real-time waveform data of the medium-voltage distribution switch in a noise-free environment and in an installation environment;

[0016] Waveform acquisition equipment is used to collect waveform data of bird sounds, vehicle sounds, and human voices within a preset area of ​​a medium-voltage distribution switch;

[0017] generating second vibration waveform data using the bird sound waveform data, the vehicle sound waveform data, and the human voice sound waveform data;

[0018] Target vibration waveform data is generated using the first vibration waveform data, the vibration wave real-time waveform data, and the second vibration waveform data.

[0019] Optionally, the step of inputting the target vibration waveform data into a preset target detection model and extracting intensity peak data, preset waveform time data and vibration wave frequency corresponding to the target vibration waveform data includes:

[0020] Divide the peak values ​​of each waveform data at different time points in the target vibration waveform data according to a preset proportion coefficient to generate a training set and a test set;

[0021] Inputting the waveform data corresponding to the training set into a preset initial detection model for training to generate an updated detection model;

[0022] Inputting the waveform data corresponding to the test set into the updated detection model for testing to generate a target detection model;

[0023] Inputting the real-time vibration wave waveform data into the target detection model, and extracting intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time vibration wave waveform data;

[0024] The second vibration waveform data is input into the target detection model, and intensity peak data corresponding to the second vibration waveform data is extracted.

[0025] Optionally, the step of inputting intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model and outputting a fault value includes:

[0026] The intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data are input into a preset fault judgment model; wherein the calculation formula of the preset fault judgment model is:

[0027]

[0028] Where F is the fault value, a1 is the time ratio, a2 is the peak ratio, a3 is the frequency ratio, and t max -t min is the set frequency wave time safety range, t m X is the value closest to t1 within the safe range of vibration wave time. max -X min is the set safety range value of the vibration wave intensity, X m The closest to X within the safe range of vibration wave intensity fz The value of K max -K min is the set vibration frequency safety range, K m The closest to K within the vibration frequency safety range fz The value of , 1 = a1 + a2 + a3;

[0029] The fault value of the real-time waveform data of the vibration wave is calculated using the preset fault judgment model.

[0030] Optionally, the step of determining the fault result of the fault value based on a preset fault threshold includes:

[0031] Determining whether the fault value is greater than or equal to a preset fault threshold;

[0032] If so, it is determined that a fault has occurred and a fault occurrence result is generated;

[0033] If not, no fault has occurred, and a no fault result is generated.

[0034] Optionally, the step of calculating the true value of the fault result by using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration waveform data in the target vibration waveform data, comprises:

[0035] Acquiring second vibration waveform data and vibration wave real-time waveform data from the target vibration waveform data;

[0036] Calculating the total number of bird sound waveform data, vehicle sound waveform data, and human voice sound waveform data in the second vibration waveform data, respectively, to generate a total number of bird sounds, a total number of vehicle sounds, and a total number of human voices;

[0037] Inputting a preset true value model using intensity peak data corresponding to the bird sound waveform data, the vehicle sound waveform data, the human voice waveform data, and the vibration wave real-time waveform data, as well as the total number of bird sounds, the total number of vehicle sounds, and the total number of human voices;

[0038] The true value of the fault result is calculated using the preset true value model.

[0039] Optionally, the step of judging whether the true value is greater than or equal to a preset true threshold value and determining a fault signal of the fault result according to the judgment result includes:

[0040] Determining whether the true value is greater than or equal to a preset true threshold;

[0041] If so, it is determined that the fault result is a false fault signal and no alarm is required;

[0042] If not, the fault result is determined to be a fault signal, and an alarm message is generated.

[0043] A second aspect of the present invention provides a vibration wave filtering system for eliminating environmental noise, comprising:

[0044] a target vibration waveform data module, configured to obtain target vibration waveform data of the medium voltage distribution switch corresponding to the vibration wave filtering instruction request in response to the vibration wave filtering instruction request;

[0045] a vibration wave frequency module, configured to input the target vibration waveform data into a preset target detection model and extract intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data;

[0046] A fault value module is used to input the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model and output a fault value;

[0047] A fault result module, configured to determine a fault result of the fault value based on a preset fault threshold;

[0048] a true value module, configured to calculate a true value of the fault result by using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration waveform data in the target vibration waveform data;

[0049] The fault signal module is used to determine whether the true value is greater than or equal to a preset true threshold value, and determine the fault signal of the fault result according to the determination result.

[0050] The third aspect of the present invention provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory. When the computer program is executed by the processor, the processor executes the steps of the vibration wave filtering method for eliminating environmental noise as described in any one of the above items.

[0051] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the vibration wave filtering method for eliminating environmental noise as described in any one of the above items.

[0052] It can be seen from the above technical solutions that the present invention has the following advantages:

[0053] The present invention obtains target vibration waveform data of a medium-voltage distribution switch corresponding to a vibration wave filtering instruction request in response to a vibration wave filtering instruction request; inputs the target vibration waveform data into a preset target detection model to extract intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data; inputs the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time vibration wave waveform data in the target vibration waveform data into a preset fault judgment model to output a fault value; determines the fault result of the fault value based on a preset fault threshold; calculates the true value of the fault result using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration wave waveform data in the target vibration waveform data; determines whether the true value is greater than or equal to the preset true threshold, and determines the fault signal of the fault result based on the judgment result. This solves the technical problem that the existing technology cannot use filtering methods for outdoor noisy environments such as bird sounds, car sounds, and human voices, cannot provide relatively pure vibration waves of the mechanical state of the switch, and thus has low recognition accuracy.

[0054] The present invention substitutes the collected real-time waveform data into a constructed medium-voltage distribution switch fault filtering model, monitors the vibration wave data in real time, identifies whether a fault occurs, substitutes the judgment result and the real-time waveform data into the true value recognition strategy, identifies the authenticity of the judgment result, outputs the fault type and the authenticity of the judgment, compares the true value with the true threshold, and determines whether to issue an alarm, thereby improving the authenticity of the judgment. In view of the diverse and complex installation and distribution environment of medium-voltage distribution switches and the large number of interference waves, the filtering of sampled vibration waves needs to have adaptive specific application requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0056] Figure 1 A flowchart of a vibration wave filtering method for eliminating environmental noise provided in Example 1 of the present invention;

[0057] Figure 2 A flowchart of a vibration wave filtering method for eliminating environmental noise provided in the second embodiment of the present invention;

[0058] Figure 3 This is a structural block diagram of a vibration wave filtering system for eliminating environmental noise provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0059] Embodiments of the present invention provide a vibration wave filtering method, system, device, and medium for eliminating environmental noise, which are used to solve the technical problem that existing technologies cannot use filtering methods for noisy outdoor environments such as bird sounds, car sounds, and human voices, and cannot provide relatively pure vibration waves of the mechanical state of switches, resulting in low recognition accuracy.

[0060] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0061] See also Figure 1 , Figure 1 This is a flowchart of the steps of a vibration wave filtering method for eliminating environmental noise provided in Example 1 of the present invention.

[0062] The present invention provides a vibration wave filtering method for eliminating environmental noise, comprising the following steps:

[0063] Step 101: In response to a vibration wave filtering instruction request, obtain target vibration waveform data of a medium voltage distribution switch corresponding to the vibration wave filtering instruction request.

[0064] It should be noted that the vibration wave filtering instruction request refers to performing vibration wave filtering processing on the vibration wave of the medium voltage distribution switch itself and the noise near the medium voltage distribution switch, so as to determine whether there is a fault in the medium voltage distribution switch.

[0065] The target vibration waveform data refers to the vibration wave data of the medium-voltage distribution switch in a noise-free environment, and the real-time vibration wave waveform data of the medium-voltage distribution switch in the installation environment at the same time. It also includes noise data such as bird sound wave waveform data, vehicle sound wave waveform data, and human voice sound wave waveform data near the medium-voltage distribution switch.

[0066] In specific implementation, when responding to a vibration wave filtering instruction request, the vibration wave data of the medium-voltage distribution switch in the noise environment corresponding to the vibration wave filtering instruction request is obtained, and the real-time vibration wave waveform data of the medium-voltage distribution switch in the installation environment at the same time is obtained, and also includes noise data such as bird sound wave waveform data, vehicle sound wave waveform data and human voice sound wave waveform data near the medium-voltage distribution switch.

[0067] Step 102: Input the target vibration waveform data into a preset target detection model, and extract the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data.

[0068] It should be noted that the target detection model refers to the regression detection model obtained by training and testing the initial regression model. The regression detection model network is one of the linear regression, decision tree, support vector machine, or random forest models.

[0069] The intensity peak data refers to the size of the intensity peak point extracted from the vibration waveform data.

[0070] The preset waveform time refers to the time during which the waveform that accounts for 80% of the peak intensity exists.

[0071] In specific implementation, the target vibration waveform data are input into the target detection model respectively, and the target detection model is used to extract the intensity peak point size of the target vibration waveform data, the time when the waveform accounting for 80% of the intensity peak exists, and the vibration wave frequency.

[0072] Step 103: Input the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model, and output a fault value.

[0073] It should be noted that the preset fault judgment model refers to a model generated by a fault judgment calculation formula.

[0074] In specific implementation, the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time waveform data of the target vibration waveform data are input into the preset fault judgment model and substituted into the fault judgment calculation formula, which is:

[0075]

[0076] Where F is the fault value, a1 is the time ratio, a2 is the peak ratio, a3 is the frequency ratio, and t max -t min is the set frequency wave time safety range, t m X is the value closest to t1 within the safe range of vibration wave time. max -X min is the set safety range value of the vibration wave intensity, X m The closest to X within the safe range of vibration wave intensity fz The value of K max -K min is the set vibration frequency safety range, K m The closest to K within the vibration frequency safety range fz The value of , 1 = a1 + a2 + a3.

[0077] The above calculation formula is used to calculate the fault value of the medium voltage distribution switch corresponding to the current real-time waveform data of the vibration wave.

[0078] Step 104: Determine a fault result of the fault value based on a preset fault threshold.

[0079] It should be noted that the preset fault threshold is set according to actual conditions and is not limited here.

[0080] In a specific implementation, the fault value is compared with a preset fault threshold. If the fault value is greater than or equal to the preset fault threshold, it indicates that a fault has occurred; otherwise, it indicates that no fault has occurred.

[0081] Step 105 : Calculate the true value of the fault result using the intensity peak data corresponding to the second vibration waveform data in the target vibration waveform data and the vibration wave real-time waveform data.

[0082] It should be noted that the second vibration waveform data refers to bird sound waveform data, vehicle sound waveform data, and human voice sound waveform data.

[0083] In specific implementation, the intensity peak data corresponding to the bird sound wave waveform data, vehicle sound wave waveform data, human voice wave waveform data and vibration wave real-time waveform data are input into the true value calculation formula to calculate the true value of the fault result.

[0084] Step 106: Determine whether the true value is greater than or equal to a preset true threshold, and determine a fault signal of the fault result according to the determination result.

[0085] It should be noted that the preset real threshold is set according to the actual situation and is not limited here.

[0086] In specific implementation, the true value is compared with the preset true threshold. If the true value is greater than or equal to the preset true threshold, it means that the fault result is a false fault signal. Otherwise, it means that the fault result is a fault signal and an alarm information needs to be generated to issue an alarm.

[0087] The present invention obtains target vibration waveform data of a medium-voltage distribution switch corresponding to a vibration wave filtering instruction request in response to a vibration wave filtering instruction request; inputs the target vibration waveform data into a preset target detection model to extract intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data; inputs the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time vibration wave waveform data in the target vibration waveform data into a preset fault judgment model to output a fault value; determines the fault result of the fault value based on a preset fault threshold; calculates the true value of the fault result using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration wave waveform data in the target vibration waveform data; determines whether the true value is greater than or equal to the preset true threshold, and determines the fault signal of the fault result based on the judgment result. This solves the technical problem that the existing technology cannot use filtering methods for outdoor noisy environments such as bird sounds, car sounds, and human voices, cannot provide relatively pure vibration waves of the mechanical state of the switch, and thus has low recognition accuracy.

[0088] The present invention substitutes the collected real-time waveform data into a constructed medium-voltage distribution switch fault filtering model, monitors the vibration wave data in real time, identifies whether a fault occurs, substitutes the judgment result and the real-time waveform data into the true value recognition strategy, identifies the authenticity of the judgment result, outputs the fault type and the authenticity of the judgment, compares the true value with the true threshold, and determines whether to issue an alarm, thereby improving the authenticity of the judgment. In view of the diverse and complex installation and distribution environment of medium-voltage distribution switches and the large number of interference waves, the filtering of sampled vibration waves needs to have adaptive specific application requirements.

[0089] See also Figure 2, Figure 2 A step flow chart of a vibration wave filtering method for eliminating environmental noise is provided for the second embodiment of the present application.

[0090] The vibration wave filtering method for eliminating environmental noise provided by the present application comprises:

[0091] Step 201, in response to a vibration wave filtering instruction request, determine the medium voltage distribution switch corresponding to the vibration wave filtering instruction request.

[0092] In the embodiment of the present application, the specific implementation process of step 201 is similar to step 101, which will not be repeated here.

[0093] Step 202, use a vibration wave data acquisition device to respectively collect first vibration wave data and vibration wave real-time waveform data of the medium voltage distribution switch in a noise-free environment and in an installed environment.

[0094] It should be noted that the first vibration wave data refers to the vibration wave data of the medium voltage distribution switch in a noise-free environment.

[0095] The vibration wave real-time waveform data refers to the vibration wave data of the medium voltage distribution switch collected in real time in the installed environment.

[0096] In specific implementation, the vibration wave data acquisition device is used to collect the vibration wave data of the medium voltage distribution switch in a noise-free environment, and the vibration wave data of the medium voltage distribution switch in the installed environment at the same time.

[0097] Step 203, use a waveform acquisition device to collect bird sound wave waveform data, vehicle sound wave waveform data and human sound wave waveform data in a preset area of the medium voltage distribution switch.

[0098] It should be noted that the medium voltage distribution switch preset area refers to the vicinity / periphery of the medium voltage distribution switch, and the specific range is planned according to the actual situation, which is not limited here.

[0099] In specific implementation, the waveform acquisition device is used to respectively collect bird sound wave waveform data, vehicle sound wave waveform data and human sound wave waveform data in the vicinity of the medium voltage distribution switch.

[0100] Step 204, use the bird sound wave waveform data, the vehicle sound wave waveform data and the human sound wave waveform data to generate second vibration waveform data.

[0101] In specific implementation, the bird sound wave waveform data, the vehicle sound wave waveform data and the human sound wave waveform data are all classified as second vibration waveform data.

[0102] Step 205, use the first vibration waveform data, the vibration wave real-time waveform data and the second vibration waveform data to generate target vibration waveform data.

[0103] In the specific implementation, the vibration wave data of the medium-voltage distribution switch in a noise-free environment, the real-time vibration wave waveform data, the bird sound wave waveform data, the vehicle sound wave waveform data and the human voice wave waveform data are combined to obtain the target vibration waveform data.

[0104] Step 206: Input the target vibration waveform data into a preset target detection model, and extract the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data.

[0105] Optionally, step 206 includes the following steps S11-S15:

[0106] S11, dividing the peak values ​​of each waveform data at different time points in the target vibration waveform data according to a preset proportion coefficient to generate a training set and a test set;

[0107] S12, inputting the waveform data corresponding to the training set into a preset initial detection model for training to generate an updated detection model;

[0108] S13, inputting the waveform data corresponding to the test set into the updated detection model for testing to generate a target detection model;

[0109] S14, inputting the real-time vibration wave waveform data into the target detection model, and extracting the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time vibration wave waveform data;

[0110] S15. Input the second vibration waveform data into the target detection model, and extract the intensity peak data corresponding to the second vibration waveform data.

[0111] It should be noted that the preset proportion coefficients are divided into a 70% proportion coefficient and a 30% proportion coefficient. The peak values ​​of the waveform data at different time points of the target vibration waveform data, including the vibration wave data of a medium-voltage distribution switch in a noise-free environment, the real-time vibration wave waveform data of a medium-voltage distribution switch in an installation environment, the bird sound waveform data, the vehicle sound waveform data, and the human voice waveform data, are used as the training set with a 70% proportion coefficient, and the 30% proportion coefficient is used as the test set.

[0112] The preset initial detection model is a regression model network.

[0113] The updated detection model refers to the regression model network after training and optimizing the initial detection model.

[0114] In specific implementation, the training set with the proportion coefficient of 70% is input into the initial detection model for training to obtain an updated detection model, the test set with the proportion coefficient of 30% is used to test the updated detection model, and the optimal detection model meeting the preset waveform test accuracy is output as the target detection model.

[0115] In specific implementation, the waveform data of the vibration wave real-time waveform data, the bird sound wave waveform data, the vehicle sound wave waveform data and the human sound wave waveform data collected in the installation environment of the medium-voltage distribution switch are input into the target detection model, the target detection model analyzes the input waveform data, and the target detection model extracts the intensity peak point size X fz , the time t1 of the waveform existing in 80% of the intensity peak value and the vibration wave frequency K fz .

[0116] The target detection model extracts the intensity peak point size of the bird sound wave waveform data, the vehicle sound wave waveform data and the human sound wave waveform data respectively.

[0117] Step 207, input the intensity peak value data, the preset waveform time data and the vibration wave frequency corresponding to the vibration wave real-time waveform data in the target vibration waveform data into the preset fault judgment model, and output a fault value.

[0118] Optionally, step 207 includes the following steps S21-S22:

[0119] S21, input the intensity peak value data, the preset waveform time data and the vibration wave frequency corresponding to the vibration wave real-time waveform data in the target vibration waveform data into the preset fault judgment model; wherein, the calculation formula of the preset fault judgment model is:

[0120]

[0121] In the formula, F is the fault value, a1 is the time proportion coefficient, a2 is the peak value proportion coefficient, a3 is the frequency proportion coefficient, t max -t min is the set frequency wave time safety range, t m is the value closest to t1 in the vibration wave time safety range, X max -X min is the set vibration wave intensity safety range value, X m is the value closest to X fz in the vibration wave intensity safety range, K max -K min is the set vibration frequency safety range, K m is the value closest to K fz in the vibration frequency safety range, and 1=a1+a2+a3.

[0122] S22. Calculate the fault value of the real-time waveform data of the vibration wave using a preset fault judgment model.

[0123] In the specific implementation, the target detection model is used to extract the intensity peak value X in the real-time waveform data of the vibration wave. fz , the time t1 of the waveform that accounts for 80% of the intensity peak and the vibration wave frequency K fz .

[0124] The intensity peak value, the time when the waveform that accounts for 80% of the intensity peak value exists, and the vibration wave frequency in the real-time vibration wave waveform data are input into the preset fault judgment model. The calculation formula of the preset fault judgment model is:

[0125]

[0126] Where F is the fault value, a1 is the time ratio, a2 is the peak ratio, a3 is the frequency ratio, and t max -t min is the set frequency wave time safety range, t m X is the value closest to t1 within the safe range of vibration wave time. max -X min is the set safety range value of the vibration wave intensity, X m The closest to X within the safe range of vibration wave intensity fz The value of K max -K min is the set vibration frequency safety range, K m The closest to K within the vibration frequency safety range fz The value of , 1 = a1 + a2 + a3;

[0127] The fault value of the medium-voltage distribution switch corresponding to the real-time waveform data of the vibration wave is calculated through the calculation formula of the preset fault judgment model.

[0128] Step 208: Determine a fault result of the fault value based on the preset fault threshold.

[0129] Optionally, step 208 includes the following steps S31-S33:

[0130] S31, determining whether the fault value is greater than or equal to a preset fault threshold;

[0131] S32: If yes, determine that a fault has occurred and generate a fault result;

[0132] S33. If not, then no fault has occurred, and a no fault result is generated.

[0133] During specific implementation, the fault value is compared with the preset fault threshold. If the true value is greater than or equal to the preset true threshold, it means that the fault judgment is accurate and a fault result is generated and sent. If the true value is less than the preset fault threshold, it means that the fault judgment is inaccurate and a no fault result is generated or a re-judgment is required. Therefore, it is necessary to further judge the true value of the fault result.

[0134] Step 209 : Calculate the true value of the fault result using the intensity peak data corresponding to the second vibration waveform data in the target vibration waveform data and the vibration wave real-time waveform data.

[0135] Optionally, step 209 includes the following steps S41-S44:

[0136] S41, acquiring second vibration waveform data and vibration wave real-time waveform data in the target vibration waveform data;

[0137] S42, calculating the total number of bird sound waveform data, vehicle sound waveform data, and human voice waveform data in the second vibration waveform data, respectively, to generate a total number of bird sounds, a total number of vehicle sounds, and a total number of human voices;

[0138] S43, using the intensity peak data corresponding to the bird sound waveform data, vehicle sound waveform data, human voice waveform data, and vibration wave real-time waveform data, as well as the total number of bird sounds, vehicle sounds, and human voices, to input a preset true value model;

[0139] S44. Calculate the true value of the fault result using a preset true value model.

[0140] It should be noted that the intensity peak points of the bird sound waveform data extracted by the target detection model are (m1, m2, m3, ...m n1 ), the intensity peak points in the vehicle sound wave waveform data are (k1, k2, k3, ...k n2 ), the intensity peak points of the human voice waveform data are (P1, P2, P3, ... P n3 ) and the intensity peak value X in the real-time waveform data of the vibration wave fz , calculate the total number of bird sound waveform data, car sound waveform data and human sound waveform data respectively, where n1 is the total number of bird sounds, m i is the peak intensity of the i-th bird sound waveform, i∈(1,n1), n2 is the total number of car sounds, k j is the peak intensity of the jth car sound waveform, j∈(1,n2), n3 is the total number of human voices, p z is the peak intensity of the z-th individual voice waveform, z∈(1,n3).

[0141] In the specific implementation, the bird sound waveform data, vehicle sound waveform data, human sound waveform data and vibration wave real-time waveform data in the target vibration waveform data are obtained, and the intensity peak points in the bird sound waveform data are respectively (m1, m2, m3, ...m n1 ), the intensity peak points in the vehicle sound wave waveform data are (k1, k2, k3, ...k n2 ), the intensity peak points of the human voice waveform data are (P1, P2, P3, ... P n3 ) and the intensity peak value X in the real-time waveform data of the vibration wave fz Substitute the total number of bird sounds, the total number of car sounds, and the total number of human voices into the calculation formula of the preset true value model, and the calculation formula is:

[0142]

[0143] Where T is the true value, m i is the peak intensity of the i-th bird sound waveform, k j is the peak intensity of j vehicle sound waveforms, p z is the peak intensity of the z-person voice waveform, X fz is the intensity peak value in the real-time waveform data of the vibration wave, n1 is the total number of bird sounds, n2 is the total number of car sounds, and n3 is the total number of human voices.

[0144] Step 210: Determine whether the true value is greater than or equal to a preset true threshold, and determine a fault signal of the fault result according to the determination result.

[0145] Optionally, step 210 includes the following steps S51-S53:

[0146] S51, determining whether the true value is greater than or equal to a preset true threshold;

[0147] S52: If yes, the fault result is determined to be a false fault signal and no alarm is required;

[0148] S53: If not, determine that the fault result is a fault signal and generate an alarm message.

[0149] It should be noted that the alarm information refers to the information of the fault signal of the medium voltage distribution switch.

[0150] In a specific implementation, the true value calculated in step 209 is compared with a preset true threshold. If the true value is greater than or equal to the preset true threshold, it is a false fault signal and no alarm processing is required. If the true value is less than the preset true threshold, it is a fault signal and an alarm processing is required. The generated alarm information is sent to the management personnel or maintenance personnel. This improves the authenticity of the judgment and allows for timely reporting and processing.

[0151] The present invention obtains target vibration waveform data of a medium-voltage distribution switch corresponding to a vibration wave filtering instruction request in response to a vibration wave filtering instruction request; inputs the target vibration waveform data into a preset target detection model to extract intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data; inputs the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time vibration wave waveform data in the target vibration waveform data into a preset fault judgment model to output a fault value; determines the fault result of the fault value based on a preset fault threshold; calculates the true value of the fault result using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration wave waveform data in the target vibration waveform data; determines whether the true value is greater than or equal to the preset true threshold, and determines the fault signal of the fault result based on the judgment result. This solves the technical problem that the existing technology cannot use filtering methods for outdoor noisy environments such as bird sounds, car sounds, and human voices, cannot provide relatively pure vibration waves of the mechanical state of the switch, and thus has low recognition accuracy.

[0152] The present invention substitutes the collected real-time waveform data into a constructed medium-voltage distribution switch fault filtering model, monitors the vibration wave data in real time, identifies whether a fault occurs, substitutes the judgment result and the real-time waveform data into the true value recognition strategy, identifies the authenticity of the judgment result, outputs the fault type and the authenticity of the judgment, compares the true value with the true threshold, and determines whether to issue an alarm, thereby improving the authenticity of the judgment. In view of the diverse and complex installation and distribution environment of medium-voltage distribution switches and the large number of interference waves, the filtering of sampled vibration waves needs to have adaptive specific application requirements.

[0153] See also Figure 3 , Figure 3 This is a structural block diagram of a vibration wave filtering system for eliminating environmental noise provided in Example 3 of the present invention.

[0154] The present invention provides a vibration wave filtering system for eliminating environmental noise, comprising:

[0155] The target vibration waveform data module 301 is used to obtain target vibration waveform data of the medium voltage distribution switch corresponding to the vibration wave filtering instruction request in response to the vibration wave filtering instruction request;

[0156] The vibration wave frequency module 302 is used to input the target vibration waveform data into a preset target detection model and extract the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the target vibration waveform data;

[0157] A fault value module 303 is used to input the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model and output a fault value;

[0158] The fault result module 304 is configured to determine a fault result of the fault value based on a preset fault threshold value.

[0159] The true value module 305 is configured to calculate a true value of the fault result by using intensity peak value data corresponding to the second vibration waveform data and the vibration real-time waveform data in the target vibration waveform data.

[0160] The fault signal module 306 is configured to determine whether the true value is greater than or equal to a preset true threshold value, and determine a fault signal of the fault result according to a determination result.

[0161] Optionally, the target vibration waveform data module 301 comprises:

[0162] The medium-voltage power distribution switch submodule is configured to determine a medium-voltage power distribution switch corresponding to the vibration wave filtering instruction request in response to the vibration wave filtering instruction request.

[0163] The vibration real-time waveform data submodule is configured to collect first vibration wave data and vibration real-time waveform data of the medium-voltage power distribution switch in a noise-free environment and in an installation environment, respectively, by using a vibration wave data collection device.

[0164] The human voice sound wave waveform data submodule is configured to collect bird sound wave waveform data, vehicle sound wave waveform data and human voice sound wave waveform data in a preset area of the medium-voltage power distribution switch by using a waveform collection device.

[0165] The second vibration waveform data submodule is configured to generate second vibration waveform data by using the bird sound wave waveform data, the vehicle sound wave waveform data and the human voice sound wave waveform data.

[0166] The target vibration waveform data submodule is configured to generate target vibration waveform data by using the first vibration waveform data, the vibration real-time waveform data and the second vibration waveform data.

[0167] Optionally, the vibration wave frequency module 302 comprises:

[0168] The test set submodule is configured to divide peak values of each waveform data in the target vibration waveform data at different time points according to a preset proportion coefficient to generate a training set and a test set.

[0169] The update detection model submodule is configured to input waveform data corresponding to the training set into a preset initial detection model to train the initial detection model and generate an updated detection model.

[0170] The target detection model submodule is configured to input waveform data corresponding to the test set into the updated detection model to test the updated detection model and generate a target detection model.

[0171] The vibration wave frequency submodule is used to input the real-time vibration wave waveform data into the target detection model and extract the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time vibration wave waveform data;

[0172] The intensity peak data submodule is used to input the second vibration waveform data into the target detection model and extract the intensity peak data corresponding to the second vibration waveform data.

[0173] Optionally, the fault value module 303 includes:

[0174] The fault judgment model submodule is used to input the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into the preset fault judgment model; wherein, the calculation formula of the preset fault judgment model is:

[0175]

[0176] Where F is the fault value, a1 is the time ratio, a2 is the peak ratio, a3 is the frequency ratio, and t max -t min is the set frequency wave time safety range, t m X is the value closest to t1 within the safe range of vibration wave time. max -X min is the set safety range value of the vibration wave intensity, X m The closest to X within the safe range of vibration wave intensity fz The value of K max -K min is the set vibration frequency safety range, K m The closest to K within the vibration frequency safety range fz The value of , 1 = a1 + a2 + a3;

[0177] The fault value submodule is used to calculate the fault value of the real-time waveform data of the vibration wave through a preset fault judgment model.

[0178] Optionally, the fault result module 304 includes:

[0179] A first judgment submodule is used to judge whether the fault value is greater than or equal to a preset fault threshold;

[0180] A fault occurrence submodule is used to determine if a fault occurs and generate a fault occurrence result;

[0181] The no fault submodule is used to generate a no fault result if no fault has occurred.

[0182] Optionally, the real value module 305 includes:

[0183] An acquisition submodule, configured to acquire the second vibration waveform data and the vibration wave real-time waveform data in the target vibration waveform data;

[0184] a calculation submodule, configured to calculate the total number of bird sound waveform data, vehicle sound waveform data, and human voice sound waveform data in the second vibration waveform data, and generate a total number of bird sounds, a total number of vehicle sounds, and a total number of human voices, respectively;

[0185] A preset true value model submodule is used to input the preset true value model using the intensity peak data corresponding to the bird sound waveform data, the vehicle sound waveform data, the human voice waveform data and the vibration wave real-time waveform data, as well as the total number of bird sounds, the total number of vehicle sounds and the total number of human voices;

[0186] The true value submodule is used to calculate the true value of the fault result through a preset true value model.

[0187] Optionally, the fault signal module 306 includes:

[0188] The second judgment submodule is used to judge whether the true value is greater than or equal to a preset true threshold;

[0189] A false fault signal submodule is used to determine that the fault result is a false fault signal and no alarm is required if the fault result is true;

[0190] The fault signal submodule is used to determine that the fault result is a fault signal and generate an alarm message if no.

[0191] An electronic device provided in a fourth embodiment of the present invention includes a memory and a processor. A computer program is stored in the memory. When the computer program is executed by the processor, the processor executes the steps of the vibration wave filtering method for eliminating environmental noise as described in any of the above embodiments.

[0192] A fifth embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed, the vibration wave filtering method for eliminating environmental noise as described in any of the above embodiments is implemented.

[0193] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0194] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0195] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0196] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0197] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0198] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vibration wave filtering method for eliminating environmental noise, characterized in that: include: In response to a vibration wave filtering instruction request, obtaining target vibration waveform data of a medium voltage distribution switch corresponding to the vibration wave filtering instruction request; Inputting the target vibration waveform data into a preset target detection model, and extracting intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data; Inputting the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model and outputting a fault value; Determining a fault result of the fault value based on a preset fault threshold; Calculating a true value of the fault result by using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration waveform data in the target vibration waveform data; Determine whether the true value is greater than or equal to a preset true threshold, and determine a fault signal of the fault result according to the determination result; The step of obtaining target vibration waveform data of the medium voltage distribution switch corresponding to the vibration wave filtering instruction request in response to the vibration wave filtering instruction request comprises: In response to a vibration wave filtering instruction request, determining a medium voltage distribution switch corresponding to the vibration wave filtering instruction request; Using vibration wave data acquisition equipment to respectively collect first vibration waveform data and vibration wave real-time waveform data of the medium voltage distribution switch in a noise-free environment and in an installation environment; A waveform acquisition device is used to collect waveform data of bird sounds, vehicle sounds, and human voices within a preset area of ​​a medium-voltage distribution switch; generating second vibration waveform data using the bird sound waveform data, the vehicle sound waveform data, and the human voice sound waveform data; Target vibration waveform data is generated using the first vibration waveform data, the vibration wave real-time waveform data, and the second vibration waveform data.

2. The vibration wave filtering method for eliminating environmental noise according to claim 1, characterized in that: The step of inputting the target vibration waveform data into a preset target detection model and extracting intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data includes: Divide the peak values ​​of each waveform data at different time points in the target vibration waveform data according to a preset proportion coefficient to generate a training set and a test set; Inputting the waveform data corresponding to the training set into a preset initial detection model for training to generate an updated detection model; Inputting the waveform data corresponding to the test set into the updated detection model for testing to generate a target detection model; Inputting the real-time vibration wave waveform data into the target detection model, and extracting intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time vibration wave waveform data; The second vibration waveform data is input into the target detection model, and intensity peak data corresponding to the second vibration waveform data is extracted.

3. The vibration wave filtering method for eliminating environmental noise according to claim 1, characterized in that: The step of inputting the intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model and outputting a fault value includes: The intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data are input into a preset fault judgment model; wherein the calculation formula of the preset fault judgment model is: ； Where, is the fault value, is the time proportion coefficient, is the peak ratio, is the frequency ratio, For the set frequency wave time safety range, The closest to the safety range of vibration wave time The value of is the set safety range value of the vibration wave intensity, The closest to the safe range of vibration wave intensity The value of is the set vibration frequency safety range, The closest to the vibration frequency safety range The value of ; The fault value of the real-time waveform data of the vibration wave is calculated using the preset fault judgment model.

4. The vibration wave filtering method for eliminating environmental noise according to claim 1, characterized in that: The step of determining the fault result of the fault value based on the preset fault threshold comprises: Determining whether the fault value is greater than or equal to a preset fault threshold; If so, it is determined that a fault has occurred and a fault occurrence result is generated; If not, no fault has occurred, and a no fault result is generated.

5. The vibration wave filtering method for eliminating environmental noise according to claim 1, characterized in that: The step of calculating the true value of the fault result by using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration waveform data in the target vibration waveform data, comprises: Acquiring second vibration waveform data and vibration wave real-time waveform data from the target vibration waveform data; Calculating the total number of bird sound waveform data, vehicle sound waveform data, and human voice sound waveform data in the second vibration waveform data, respectively, to generate a total number of bird sounds, a total number of vehicle sounds, and a total number of human voices; Inputting a preset true value model using intensity peak data corresponding to the bird sound waveform data, the vehicle sound waveform data, the human voice waveform data, and the vibration wave real-time waveform data, as well as the total number of bird sounds, the total number of vehicle sounds, and the total number of human voices; The true value of the fault result is calculated using the preset true value model.

6. The vibration wave filtering method for eliminating environmental noise according to claim 1, characterized in that: The step of judging whether the true value is greater than or equal to a preset true threshold value and determining the fault signal of the fault result according to the judgment result includes: Determining whether the true value is greater than or equal to a preset true threshold; If so, it is determined that the fault result is a false fault signal and no alarm is required; If not, the fault result is determined to be a fault signal, and an alarm message is generated.

7. A vibration wave filtering system for eliminating environmental noise, characterized in that: include: a target vibration waveform data module, configured to obtain target vibration waveform data of the medium voltage distribution switch corresponding to the vibration wave filtering instruction request in response to the vibration wave filtering instruction request; a vibration wave frequency module, configured to input the target vibration waveform data into a preset target detection model and extract intensity peak data, preset waveform time data, and vibration wave frequency corresponding to the target vibration waveform data; A fault value module is used to input the intensity peak data, preset waveform time data and vibration wave frequency corresponding to the real-time waveform data of the vibration wave in the target vibration waveform data into a preset fault judgment model and output a fault value; A fault result module, configured to determine a fault result of the fault value based on a preset fault threshold; a true value module, configured to calculate a true value of the fault result by using the intensity peak data corresponding to the second vibration waveform data and the real-time vibration waveform data in the target vibration waveform data; A fault signal module is used to determine whether the true value is greater than or equal to a preset true threshold value, and determine a fault signal of the fault result according to the determination result; The target vibration waveform data module includes: A medium voltage distribution switch submodule, configured to respond to a vibration wave filtering instruction request and determine a medium voltage distribution switch corresponding to the vibration wave filtering instruction request; A vibration wave real-time waveform data submodule is used to collect first vibration waveform data and vibration wave real-time waveform data of a medium-voltage distribution switch in a noise-free environment and in an installation environment respectively using a vibration wave data acquisition device; The human voice sound wave waveform data submodule is used to collect the sound wave waveform data of bird sounds, vehicle sounds and human voices in the preset area of ​​the medium voltage distribution switch using a waveform acquisition device; A second vibration waveform data submodule is configured to generate second vibration waveform data using the bird sound waveform data, the vehicle sound waveform data, and the human voice sound waveform data; The target vibration waveform data submodule is used to generate target vibration waveform data using the first vibration waveform data, the vibration wave real-time waveform data and the second vibration waveform data.

8. An electronic device, characterized in that: It includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the vibration wave filtering method for eliminating environmental noise as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the vibration wave filtering method for eliminating environmental noise as described in any one of claims 1 to 6 is implemented.

Images