A method and system for detecting water pump operation status based on voiceprint analysis

Abstract

The present application provides a method and system for detecting the operating status of a water pump based on voiceprint analysis, which belongs to the field of water pump operating status monitoring and is used to solve the problem of poor efficiency and accuracy of water pump operating status monitoring schemes in related technologies. In this method and device, the frequency points that best reflect the status of the water pump can be efficiently and accurately selected, and the operating status of the water pump can be determined based on the signal-to-noise ratio data of these frequency points through intelligent and accurate analysis, which is beneficial to improving the efficiency and accuracy of the water pump operating status monitoring scheme.

Description

Technical Field

[0001] The present application relates to the field of water pump operating status monitoring, and in particular to a water pump operating status detection method and system based on voiceprint analysis. Background Art

[0002] In order to ensure the reliability of water pump operation, detect potential faults in time, and avoid production interruptions and economic losses caused by sudden water pump shutdown, it is necessary to monitor the operating status of the water pump. Water pump operating status monitoring based on voiceprint is a non-contact water pump operating status monitoring solution. Its main principles include: analyzing the voiceprint characteristics during the operation of the water pump to determine the operating status of the water pump. The efficiency and accuracy of the water pump operating status monitoring solution based on voiceprint analysis mainly rely on the voiceprint analysis algorithm. Summary of the invention

[0003] The present application provides a method and system for detecting the operating status of a water pump based on voiceprint analysis, which is beneficial to improving the efficiency and accuracy of a water pump operating status monitoring solution.

[0004] In a first aspect, the present application provides a method for detecting the operating status of a water pump based on voiceprint analysis. The method comprises:

[0005] Acquire the sound monitoring data of the target water pump under the operating state;

[0006] Substituting the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information;

[0007] Determine the operating status score of the target water pump according to the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information;

[0008] Determining the operating status score of the target water pump according to the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information includes:

[0009] Determine a first state score according to the sound amplitude data and preset amplitude data;

[0010] Determine a second state score according to the signal-to-noise ratio data of each selected frequency point information;

[0011] The operating state score of the target water pump is determined according to the first state score and the second state score.

[0012] By adopting the above technical solution, it is possible to efficiently and accurately select the frequency points that best reflect the status of the water pump, and to intelligently and accurately analyze the signal-to-noise ratio data of these frequency points to determine the operating status of the water pump, thereby helping to improve the efficiency and accuracy of the water pump operating status monitoring solution.

[0013] Furthermore, the step of substituting the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information includes:

[0014] Performing time integration on the sound monitoring data to obtain the sound amplitude data;

[0015] Performing a fast Fourier transform on the sound monitoring data to obtain sound spectrum data, wherein the sound spectrum data includes multiple frequency point information;

[0016] The signal-to-noise ratio data of each frequency point information is analyzed, and a specified number of frequency point information with the highest signal-to-noise ratio is determined as the selected frequency point information.

[0017] Further, determining the first state score according to the sound amplitude data and the preset amplitude data includes:

[0018] Assume the sound amplitude data is A, and the preset amplitude data is , the first state score is ,but

[0019] ;

[0020] In the formula, a is a preset constant and .

[0021] Further, determining the second state score according to the signal-to-noise ratio data of each selected frequency point information includes:

[0022] Assume that there are n selected frequency information, and the i-th frequency information is , the signal-to-noise ratio data of the i-th frequency point information is , the preset signal-to-noise ratio is ;

[0023] Compare the signal-to-noise ratio data of the selected frequency point information with the preset signal-to-noise ratio, determine the selected frequency point information whose signal-to-noise ratio data is higher than the preset signal-to-noise ratio, suppose there are m selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio, and the jth selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio is , The signal-to-noise ratio data is , the second state score is ,but

[0024] ;

[0025] Where c is a preset constant and , for The actual signal-to-noise ratio rank is, for The preset signal-to-noise ratio order is For the corresponding The preset weight factor.

[0026] Further, determining the operating state score of the target water pump according to the first state score and the second state score includes:

[0027] Assume the running status score is W, then .

[0028] In the second aspect, the present application provides a water pump operation status detection system based on voiceprint analysis. The device comprises:

[0029] An acquisition module, used to acquire sound monitoring data of a target water pump under operating conditions;

[0030] An extraction module, used for substituting the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information; and

[0031] An analysis module, used to determine the operating status score of the target water pump according to the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information;

[0032] The analysis module is further configured to:

[0033] Determine a first state score according to the sound amplitude data and preset amplitude data;

[0034] Determine a second state score according to the signal-to-noise ratio data of each selected frequency point information;

[0035] The operating state score of the target water pump is determined according to the first state score and the second state score.

[0036] Furthermore, the extraction module is further configured to:

[0037] Performing time integration on the sound monitoring data to obtain the sound amplitude data;

[0038] Performing a fast Fourier transform on the sound monitoring data to obtain sound spectrum data, wherein the sound spectrum data includes multiple frequency point information;

[0039] The signal-to-noise ratio data of each frequency point information is analyzed, and a specified number of frequency point information with the highest signal-to-noise ratio is determined as the selected frequency point information.

[0040] Furthermore, the analysis module is further configured to:

[0041] Assume the sound amplitude data is A, and the preset amplitude data is , the first state score is ,but

[0042] ;

[0043] In the formula, a is a preset constant and .

[0044] Furthermore, the analysis module is further configured to:

[0045] Assume that there are n selected frequency information, and the i-th frequency information is , the signal-to-noise ratio data of the i-th frequency point information is , the preset signal-to-noise ratio is ;

[0046] Compare the signal-to-noise ratio data of the selected frequency point information with the preset signal-to-noise ratio, determine the selected frequency point information whose signal-to-noise ratio data is higher than the preset signal-to-noise ratio, suppose there are m selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio, and the jth selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio is , The signal-to-noise ratio data is , the second state score is ,but

[0047] ;

[0048] Where c is a preset constant and , for The actual signal-to-noise ratio rank is, for The preset signal-to-noise ratio order is For the corresponding The preset weight factor.

[0049] Furthermore, the analysis module is further configured to:

[0050] Assume the running status score is W, then .

[0051] In summary, this application at least has the following beneficial effects:

[0052] A method and system for detecting the operating status of a water pump based on voiceprint analysis are provided, which can intelligently select frequency points representing the operating status of the water pump and analyze the signal-to-noise ratio data of these frequency points to determine the operating status data of the water pump.

[0053] It should be understood that the contents described in the Summary of the Invention are not intended to limit the key or important features of the embodiments of the present application, nor are they intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0055] Figure 1 A flow chart of a method for detecting the operating status of a water pump based on voiceprint analysis in an embodiment of the present application is shown;

[0056] Figure 2 A block diagram of a water pump operation status detection system based on voiceprint analysis in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0058] In addition, the term "and / or" in this article is only a description of the association relationship between the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0059] The present application provides a method and system for detecting the operating status of a water pump based on voiceprint analysis, which can intelligently select frequency points representing the operating status of the water pump, and analyze the signal-to-noise ratio data of these frequency points to determine the operating status data of the water pump.

[0060] In a first aspect, the present application provides a method for detecting the operating status of a water pump based on voiceprint analysis.

[0061] Specifically, an audio acquisition device can be used to collect sound monitoring data of the target water pump under operating conditions, and a state analysis device can be used to analyze the operating conditions of the target water pump based on the sound monitoring data. The audio acquisition device is configured near the target water pump and can collect the score of the target water pump operating conditions. The state analysis device has an analysis and calculation function and can analyze the operating conditions of the target water pump based on the sound monitoring data. When the target water pump is in different operating conditions, such as when operating at different powers, the sound monitoring data of the target water pump itself reflects different operating conditions. In the embodiment of the present application, the operating condition monitoring scheme of the target water pump under one operating condition is disclosed, and those skilled in the art can understand and extend this scheme to any operating condition.

[0062] The method can be specifically performed by a state analysis device, and the method specifically comprises the following steps:

[0063] S110: Acquire sound monitoring data of the target water pump under operating status.

[0064] Since the state analysis device is communicatively connected to the acquisition device, it can obtain the sound monitoring data of the target water pump operating state collected by the audio acquisition device. The audio acquisition device collects the sound monitoring data in real time, and the state analysis device also obtains the sound monitoring data in real time.

[0065] S120: Substituting the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information.

[0066] The method of this step specifically includes: performing time integration on the sound monitoring data to obtain the sound amplitude data; performing fast Fourier transform on the sound monitoring data to obtain sound spectrum data, wherein the sound spectrum data includes multiple frequency point information; analyzing the signal-to-noise ratio data of each frequency point information, and determining a specified number of frequency point information with the highest signal-to-noise ratio as the selected frequency point information.

[0067] The methods of performing time integration on the sound monitoring data to obtain the sound amplitude and performing fast Fourier transform on the sound monitoring data to obtain the sound frequency data are both well-known technologies and do not belong to the improvement of the present application, so they are not introduced here.

[0068] S130: Determine the operating status score of the target water pump according to the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information.

[0069] The method of this step specifically includes: determining a first state score according to the sound amplitude data and the preset amplitude data; determining a second state score according to the signal-to-noise ratio data of each selected frequency point information; and determining the operating state score of the target water pump according to the first state score and the second state score.

[0070] Wherein, determining the first state score according to the sound amplitude data and the preset amplitude data includes: assuming that the sound amplitude data is A, and the preset amplitude data is , the first state score is ,but

[0071] ;

[0072] In the formula, a is a preset constant and In the embodiment of the present application, a=0.2.

[0073] Determining the second state score according to the signal-to-noise ratio data of each selected frequency point information includes:

[0074] Assume that there are n selected frequency information, and the i-th frequency information is , the signal-to-noise ratio data of the i-th frequency point information is , the preset signal-to-noise ratio is ;

[0075] Compare the signal-to-noise ratio data of the selected frequency point information with the preset signal-to-noise ratio, determine the selected frequency point information whose signal-to-noise ratio data is higher than the preset signal-to-noise ratio, suppose there are m selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio, and the jth selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio is , The signal-to-noise ratio data is , the second state score is ,but

[0076] ;

[0077] Where c is a preset constant and , for The actual signal-to-noise ratio rank is, for The preset signal-to-noise ratio order is For the corresponding The preset weight coefficient: In the embodiment of the present application, c=0.8. is a constant greater than zero. The preset signal-to-noise ratio rank can be trained based on the sound data of the training set or the sound data of the target water pump under the ideal operating state. If the actual signal-to-noise ratio rank deviates from the preset signal-to-noise ratio rank, it means that the operating state of the target water pump is unbalanced, and the larger the deviation value, the more serious the imbalance of the operating state of the target water pump.

[0078] Determining the operating status score of the target water pump according to the first status score and the second status score includes:

[0079] Assume the running status score is W, then .

[0080] In the above formula, -100, 100, and 10 are all specific implementations of preset constants. and When taking a specific value, the value range of W is distinguished, that is, -100, 100, and 10 can also be configured as other constant values ​​if allowed.

[0081] exist and hour, The specific value range is In this range, the smaller the value of W is, the better the operating state of the target water pump is. Other values ​​are -90, 110, They are all special codes. Fixed values ​​represent specific states. For example, -90 represents that the sound amplitude data is abnormally low and the frequency signal-to-noise ratio is unbalanced. The range value also represents a specific state, but indicates other attributes, such as , indicating that the sound amplitude data is abnormally high and the frequency signal-to-noise ratio data is not unbalanced. The higher the data is, the more the frequency signal-to-noise ratio data tends to be unbalanced, and the worse the operating condition of the target water pump is.

[0082] In summary, the method can present the operating status score of the target water pump, and can determine different operating statuses of the target water pump according to different operating status scores, that is, it can intelligently determine the operating status of the target water pump.

[0083] It should be understood that this method is essentially a machine learning model. The parameters in the machine learning model can be trained according to the actual operating state of the water pump and the state represented by the operating state score during long-term use. The parameters that can be trained are, for example, a specified number, -100, 100, 10, a, c, Etc., during the long-term use of this method, the internal parameters are continuously trained, so that the operating status score obtained by analysis can more accurately reflect the operating status of the target water pump.

[0084] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0085] In a second aspect, the present application provides a method for detecting the operating status of a water pump based on voiceprint analysis.

[0086] Figure 2A block diagram of a water pump operation status detection system based on voiceprint analysis in an embodiment of the present application is shown. The system can be implemented as the status analysis device of the first aspect, or included in the status analysis device of the first aspect.

[0087] Reference Figure 2 , the device specifically comprises:

[0088] An acquisition module 210 is used to acquire sound monitoring data of a target water pump under an operating state;

[0089] An extraction module 220, for substituting the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information; and

[0090] An analysis module 230, configured to determine an operating status score of a target water pump according to the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information;

[0091] The analysis module 230 is further configured to:

[0092] Determine a first state score according to the sound amplitude data and preset amplitude data;

[0093] Determine a second state score according to the signal-to-noise ratio data of each selected frequency point information;

[0094] The operating state score of the target water pump is determined according to the first state score and the second state score.

[0095] Furthermore, the extraction module 220 is further configured to:

[0096] Performing time integration on the sound monitoring data to obtain the sound amplitude data;

[0097] Performing a fast Fourier transform on the sound monitoring data to obtain sound spectrum data, wherein the sound spectrum data includes multiple frequency point information;

[0098] The signal-to-noise ratio data of each frequency point information is analyzed, and a specified number of frequency point information with the highest signal-to-noise ratio is determined as the selected frequency point information.

[0099] Furthermore, the analysis module 230 is further configured to:

[0100] Assume the sound amplitude data is A, and the preset amplitude data is , the first state score is ,but

[0101] ;

[0102] In the formula, a is a preset constant and .

[0103] Furthermore, the analysis module 230 is further configured to:

[0104] Assume that there are n selected frequency information, and the i-th frequency information is , the signal-to-noise ratio data of the i-th frequency point information is , the preset signal-to-noise ratio is ;

[0105] Compare the signal-to-noise ratio data of the selected frequency point information with the preset signal-to-noise ratio, determine the selected frequency point information whose signal-to-noise ratio data is higher than the preset signal-to-noise ratio, suppose there are m selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio, and the jth selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio is , The signal-to-noise ratio data is , the second state score is ,but

[0106] ;

[0107] Where c is a preset constant and , for The actual signal-to-noise ratio rank is, for The preset signal-to-noise ratio order is For the corresponding The preset weight factor.

[0108] Furthermore, the analysis module 230 is further configured to:

[0109] Assume the running status score is W, then .

[0110] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described device can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0111] In summary, this application at least has the following beneficial effects:

[0112] A method and system for detecting the operating status of a water pump based on voiceprint analysis are provided, which can intelligently select frequency points representing the operating status of the water pump and analyze the signal-to-noise ratio data of these frequency points to determine the operating status data of the water pump.

[0113] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A method for detecting the operation status of a water pump based on voiceprint analysis, characterized in that: include: Acquire the sound monitoring data of the target water pump under the operating state; Substituting the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information; Determine the operating status score of the target water pump according to the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information; Determining the operating status score of the target water pump according to the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information includes: Determine a first state score according to the sound amplitude data and preset amplitude data; Determine a second state score according to the signal-to-noise ratio data of each selected frequency point information; Determine the operating state score of the target water pump according to the first state score and the second state score; Determining the second state score according to the signal-to-noise ratio data of each selected frequency point information includes: Assume that there are n selected frequency information, and the i-th frequency information is , the signal-to-noise ratio data of the i-th frequency point information is , the preset signal-to-noise ratio is ; Compare the signal-to-noise ratio data of the selected frequency point information with the preset signal-to-noise ratio, determine the selected frequency point information whose signal-to-noise ratio data is higher than the preset signal-to-noise ratio, suppose there are m selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio, and the jth selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio is , The signal-to-noise ratio data is , the second state score is ,but ； Where c is a preset constant and , for The actual signal-to-noise ratio rank is for The preset signal-to-noise ratio order is For the corresponding The preset weight factor.

2. The detection method according to claim 1, characterized in that: Substituting the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information includes: Performing time integration on the sound monitoring data to obtain the sound amplitude data; Performing a fast Fourier transform on the sound monitoring data to obtain sound spectrum data, wherein the sound spectrum data includes multiple frequency point information; The signal-to-noise ratio data of each frequency point information is analyzed, and a specified number of frequency point information with the highest signal-to-noise ratio is determined as the selected frequency point information.

3. The detection method according to claim 1, characterized in that: Determining the first state score according to the sound amplitude data and the preset amplitude data includes: Assume the sound amplitude data is A, and the preset amplitude data is , the first state score is ,but ； In the formula, a is a preset constant and .

4. The detection method according to claim 3, characterized in that: Determining the operating status score of the target water pump according to the first status score and the second status score includes: Assume that the running status score is ,but .

5. A water pump operation status detection system based on voiceprint analysis, characterized in that: include: An acquisition module (210), used to acquire sound monitoring data of a target water pump under an operating state; An extraction module (220) is used to substitute the sound monitoring data into a pre-built feature extraction model to obtain sound amplitude data and a specified number of selected frequency point information and signal-to-noise ratio data of each selected frequency point information; as well as An analysis module (230), configured to determine an operating status score of a target water pump based on the sound amplitude data and the signal-to-noise ratio data of each selected frequency point information; The analysis module (230) is further configured to: Determine a first state score according to the sound amplitude data and preset amplitude data; Determine a second state score according to the signal-to-noise ratio data of each selected frequency point information; Determine the operating state score of the target water pump according to the first state score and the second state score; Determining the second state score according to the signal-to-noise ratio data of each selected frequency point information includes: Assume that there are n selected frequency information, and the i-th frequency information is , the signal-to-noise ratio data of the i-th frequency point information is , the preset signal-to-noise ratio is ; Compare the signal-to-noise ratio data of the selected frequency point information with the preset signal-to-noise ratio, determine the selected frequency point information whose signal-to-noise ratio data is higher than the preset signal-to-noise ratio, suppose there are m selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio, and the jth selected frequency point information whose signal-to-noise ratio data is greater than the preset signal-to-noise ratio is , The signal-to-noise ratio data is , the second state score is ,but ； Where c is a preset constant and , for The actual signal-to-noise ratio rank is for The preset signal-to-noise ratio order is For the corresponding The preset weight factor.

6. The detection system according to claim 5, characterized in that: The extraction module (220) is further configured to: Performing time integration on the sound monitoring data to obtain the sound amplitude data; Performing a fast Fourier transform on the sound monitoring data to obtain sound spectrum data, wherein the sound spectrum data includes multiple frequency point information; The signal-to-noise ratio data of each frequency point information is analyzed, and a specified number of frequency point information with the highest signal-to-noise ratio is determined as the selected frequency point information.

7. The detection system according to claim 6, characterized in that: The analysis module (230) is further configured to: Assume the sound amplitude data is A, and the preset amplitude data is , the first state score is ,but ； In the formula, a is a preset constant and .

8. The detection system according to claim 7, characterized in that: The analysis module (230) is further configured to: Assume the running status score is W, then .

Images