Pressure reducing valve multi-mode working condition abnormity monitoring method based on sound recognition

Abstract

The invention discloses a pressure reducing valve multi-mode working condition abnormity monitoring method based on voice recognition. The method comprises the steps that pressure data, flow data, audio data, position switch switching value data, vibration analog quantity data and image data are collected through a comprehensive data collection module; voiceprint recognition related data acquired by the comprehensive data acquisition module is transmitted to a cloud server platform through a network transmission module; identifying and scoring the audio data through a voiceprint scoring module; the fault comprehensive judgment module is used for carrying out fault judgment, the problems that in the prior art, a means for carrying out comprehensive judgment on multiple types of defects is lacked, manual on-site checking is relied on, and the efficiency and accuracy of fault recognition are low are solved, and the advantages that comprehensive fault judgment can be carried out through combination of multi-modal data and voiceprint scoring, and the efficiency and accuracy of fault recognition are improved are achieved. And the recognition rate and the accuracy rate of working condition abnormity or fault judgment of the pressure reducing valve equipment are greatly improved.

Description

Technical Field

[0001] The invention belongs to the technical field of fault diagnosis of hydropower stations, and in particular relates to a method for monitoring abnormal multi-modal working conditions of pressure reducing valves based on sound recognition. Background Art

[0002] The pressure reducing valve is a valve that reduces the inlet pressure to a required outlet pressure by throttling the valve disc, and can use the energy of the medium itself to keep the outlet pressure basically unchanged when the inlet pressure and flow rate change; common failures of the pressure reducing valve include water leakage, the outlet pressure is almost equal to the inlet pressure, large pressure swing, increased noise, blockage, etc. The use of pressure reducing valves in hydropower stations is very common. In the water diversion system of a hydropower station, the pressure reducing valve can be used to reduce the water pressure to protect the pipeline and equipment from the influence of high pressure. Once the pressure reducing valve fails, the pressure reducing valve leaks at the least and the equipment is at risk of water ingress; in severe cases, the equipment may be damaged by high pressure, or even major accidents such as flooding of the plant may occur; traditional abnormal monitoring methods often directly detect faults, such as appearance quality inspection: check whether there are cracks, cold shuts, sand holes, pores, slag holes, shrinkage and oxidation inclusions on the surface of the pressure reducing valve; pressure regulating performance inspection: within the given pressure regulating range, the outlet pressure should be able to be adjusted continuously between the maximum and minimum values, and there should be no jamming Resistance and abnormal vibration; Flow characteristic inspection: When the outlet flow changes, the pressure reducing valve shall not have abnormal action, and the negative deviation value of its outlet pressure shall not exceed 20% of the outlet pressure for the direct-acting pressure reducing valve and 10% of the outlet pressure for the pilot-operated pressure reducing valve; Pressure characteristic inspection: When the inlet pressure changes, the pressure reducing valve shall not have abnormal vibration, and the deviation value of its outlet pressure shall not exceed 10% of the outlet pressure for the direct-acting pressure reducing valve and 5% of the outlet pressure for the pilot-operated pressure reducing valve; Sealing performance inspection: For elastic sealing structure, the leakage shall be in accordance with Table 1; for metal-metal sealing structure, the allowable leakage shall not exceed 0.5% of the maximum flow. The elastic seal of the outlet pressure gauge should be zero, and the metal-metal seal shall not exceed 0.2MPa / min; However, the existing technology has the following disadvantages: Existing technologies often have various defects and have not formed an overall multi-modal system to achieve comprehensive judgment, resulting in only partial fault identification, and the fault identification rate and accuracy are very low. For example, pressure characteristic inspection, pressure regulation performance inspection, and sealing performance inspection require manual on-site testing and inspection, which is time-consuming and labor-intensive. In addition, some external faults such as water leakage in the pressure reducing valve cannot be monitored online in real time, and maintenance personnel are required to take the initiative to patrol and observe; therefore, it is necessary to design a multi-modal abnormal operating condition monitoring method of the pressure reducing valve based on sound recognition to solve the above problems. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method for monitoring the multimodal abnormal operating conditions of a pressure reducing valve based on sound recognition. The method is used to solve the problem that the prior art lacks the means for comprehensive judgment of multiple types of defects, relies on manual on-site verification, and has low efficiency and accuracy in fault identification. The method has the characteristics of being able to perform comprehensive fault judgment through multimodal data combined with voiceprint scoring, thereby greatly improving the recognition rate and accuracy of abnormal operating conditions or fault judgment of pressure reducing valve equipment.

[0004] In order to achieve the above technical effects, the technical solution adopted by the present invention is: A method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition comprises the following steps: S1, collecting voiceprint recognition related data through a comprehensive data acquisition module, the comprehensive data acquisition module includes a pressure sensor, a flow sensor, a bone conduction pickup, a valve position relay, a vibration sensor and a camera; the voiceprint recognition related data includes pressure data, flow data, audio data, position switch switching data, vibration analog data and image data; S2, transmits the voiceprint recognition related data collected by the comprehensive data collection module to the cloud server platform through the network transmission module; S3, identifying and scoring the audio data through the voiceprint scoring module; S4, performing fault determination through a comprehensive fault determination module.

[0005] Preferably, in step S1, a pressure sensor is arranged at the inlet and outlet of the pressure reducing valve to collect pressure data of the inlet and outlet of the pressure reducing valve; a flow sensor is arranged at the inlet and outlet of the pressure reducing valve to collect flow data of the inlet and outlet of the pressure reducing valve; a bone conduction pickup is arranged on the pressure reducing valve housing and the wall on one side of the pressure reducing valve to collect audio data; a valve position relay is arranged at the inlet and outlet of the pressure reducing valve to collect position switch switching data of the inlet and outlet of the pressure reducing valve; a vibration sensor is arranged at the pressure reducing valve housing to collect vibration analog data of the pressure reducing valve housing; a camera is arranged on the wall on one side of the pressure reducing valve to collect images of the operation process of the pressure reducing valve.

[0006] Furthermore, two pressure sensors are set, two flow sensors are set, two bone conduction pickups are set, and two valve position relays are set; the bone conduction pickup adopts ZKHY-IVOD-001 industrial diagnostic pickup with a frequency response of 20-20kHz; the vibration sensor adopts DSP vibration sensor with a measurement range of 0-50g.

[0007] Preferably, in step S2, the network transmission module includes a switch module consisting of a photoelectric switch and a convergence switch; the pressure data, flow data, position switch switching data and vibration analog data are respectively connected to the convergence switch through the photoelectric switch, the image data and audio data are extracted with characteristic values ​​through the on-site control unit, and the image data and audio data are connected to the convergence switch through the photoelectric switch; the convergence switch is connected to the cloud server platform.

[0008] Furthermore, the PC device is connected to the cloud server platform after passing through the network isolation device, and the cloud server platform communicates with the mobile terminal through the router; the aggregation switch also communicates with the local server for data storage.

[0009] Preferably, in step S3, the audio data is identified and scored by the voiceprint scoring module, including in the local control unit, and the audio data is weighted dimension reduction optimized by the MFCC algorithm: First, the audio data is pre-emphasized, and then the pre-emphasized audio data is framed; the framed audio data is subjected to fast Fourier transform to convert the waveform data into frequency, and then the Mel coefficient filtering is performed; finally, endpoint detection is performed and the final MFCC result is tested to ensure the availability of the data and process the invalid data.

[0010] Furthermore, in the Mel coefficient filtering process, the Mel scale is a cepstrum parameter extracted from the frequency domain, which describes the nonlinear characteristics of the human ear frequency. Its relationship with the frequency is: Mel(f)=2585×log(1+f / 700); Mel(f) represents the Mel scale.

[0011] Preferably, in step S3, the audio data is identified and scored by the voiceprint scoring module, including extracting i-vector processing from the audio data after weighted dimensionality reduction optimization in the local control unit, and the specific method is: Use the GMM-UBM model to obtain the Baum-Welch statistic UBM; The obtained MFCC vector is input into the GMM-UBM for training, and the target MFCC vector is input into the GMM-UBM model. The GMM parameter M of the target device is adaptively obtained through the MAP algorithm based on the original UBM. Process the baum-welch statistic through the DNN neural network and extract the i-vector: Determine whether the audio sample Y comes from device S: Assume that H0 indicates that Y comes from device S, and H1 indicates that Y does not come from device S, then the likelihood formula is: LR = p(Y|H0) / p(Y|H1); log(LR)=log(p(Y|H0))-log(p(Y|H1)); Where p represents the likelihood function; when LR>θ, the result H1 is retained, otherwise the result H0 is retained; θ is a predetermined threshold.

[0012] Preferably, in step S3, the audio data is identified and scored by the voiceprint scoring module, including performing PLDA discriminant analysis in the cloud server platform, and the specific method is: The sound data is used as training audio and divided into audio of I devices. Each device has J different audio segments. The jth audio of the i-th device is defined as X ij ; According to factor analysis, define X ij The generative model is: ; Where μ represents the mean of all training data; F is the identity space, which contains information that can be used to represent various devices; h i is the identity of a specific device or the position of the device in the identity space; G is the error space, which contains information that can be used to represent different audio changes of the same device; w ij represents the position of the device in the G space; ϵij is the final residual noise term; this term is a zero-mean Gaussian distribution with a variance of Σ; Calculate whether two audios are composed of the feature h in the device space i Generate or calculate two audio frequencies i The generated likelihood is used to obtain the score value; the specific formula is: ; In the formula, score represents the score value; n1 and n2 are the I-Vector vectors of the two audios respectively. The assumption that the two audios come from the same space is Hs, and the assumption that they come from different spaces is Hd. Among them, p(n1,n2|hs) is the likelihood function of the two audios coming from the same space; p(n1|hd) and p(n2|hd) are the likelihood functions of n1 and n2 coming from different spaces respectively; the similarity of the two audios is calculated by calculating the log-likelihood ratio: The ratio, the score, and the possibility that the two audios belong to the same device are positively correlated. The higher the ratio, the higher the score, and the greater the possibility that the two audios belong to the same device. The lower the ratio, the lower the score, and the less likely the two audios belong to the same device. A normal value interval is set for the score value, and the score value falling within the normal value interval is judged as normal, otherwise it is judged as abnormal.

[0013] Furthermore, in step S3, the collected audio data is divided into a training set and a test set, and the test set is further divided into a registration set and a verification set. i-vectors are extracted from the three sets respectively, and the data of the training set is used to train the PLDA discriminant analysis model, and the data of the test set is used for scoring.

[0014] Preferably, in step S4, the method for performing fault determination by the comprehensive fault determination module is: S401, scoring the real-time audio data collected by the bone conduction pickup through the voiceprint scoring module, and making a judgment based on the scoring value. If the scoring value is normal, the pressure reducing valve device is judged to be normal; if the scoring value is abnormal, the fault type is judged; S402, perform fault type determination, compare the score value with the database, and determine whether it is a known fault or an unknown fault; if it is a known fault, proceed to step S403 to perform known fault type determination; if it is an unknown fault, proceed to step S404 to perform unknown fault type determination; S403, performing known fault type determination: Obtain the switch data of the position switch of the valve position relay and the vibration analog data of the vibration sensor to determine whether the position relay is actuated and whether the displacement time of the vibration data corresponds: If the position relay does not operate, the output is a known fault of a general pressure reducing valve device; If the position relay is actuated and the position change time corresponds, the output is a pressure reducing valve equipment operation event; If the position relay is actuated and the shifting time does not correspond, the output is a known fault of a general pressure reducing valve device; S404, executing unknown fault type determination: The position switch data, flow data, vibration analog data and pressure data are used to jointly determine whether the fault is an internal fault or an external fault. If it is determined to be an internal fault, the professional team is directly notified to conduct on-site inspection to confirm the fault. If it is determined to be an external fault, the camera is used to conduct inspections to locate the fault location, and the professional team is notified to conduct on-site inspections to confirm the fault. S405, after confirming the fault, the fault is marked and sent down, and updated into the scoring database of step S402.

[0015] The beneficial effects of the present invention are as follows: The present invention realizes multi-modal online real-time monitoring of the pressure reducing valve equipment area by installing various types of acquisition sensors, and truly realizes unmanned operation; the present invention adopts bone conduction microphone to collect 50Hz-4KHz audio data, which can collect almost all abnormal conditions in the pressure reducing valve equipment area, and is also assisted by pressure, vibration, flow, switch value and other data. Therefore, this scheme can identify all faults in the equipment pressure reducing valve equipment area, and greatly improve the accuracy of abnormal monitoring of the working condition of the pressure reducing valve equipment; through voiceprint recognition technology, and through comprehensive judgment and analysis of multi-modal data such as flow, vibration, pressure, image, switch value, etc., it can greatly improve the accuracy of abnormal monitoring of the working condition of the pressure reducing valve equipment, and save human resources while reducing the time required for fault identification, providing an opportunity for emergency handling of power system accidents, and greatly improving the stability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 It is a schematic diagram of the layout structure of each device in the comprehensive data acquisition module in an embodiment of the present invention; Figure 3 A schematic diagram of a data transmission path of a network transmission module in an embodiment of the present invention; Figure 4 Schematic diagram of the process of the voiceprint scoring module in an embodiment of the present invention; Figure 5 The figure is a schematic diagram of a flow chart of comprehensive fault determination in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] Embodiment 1: like Figure 1 As shown, a method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition comprises the following steps: S1, collecting voiceprint recognition related data through a comprehensive data acquisition module, the comprehensive data acquisition module includes a pressure sensor, a flow sensor, a bone conduction pickup, a valve position relay, a vibration sensor and a camera; the voiceprint recognition related data includes pressure data, flow data, audio data, position switch switching data, vibration analog data and image data; S2, transmits the voiceprint recognition related data collected by the comprehensive data collection module to the cloud server platform through the network transmission module; S3, identifying and scoring the audio data through the voiceprint scoring module; S4, performing fault determination through a comprehensive fault determination module.

[0018] Embodiment 2: like Figure 2As shown, in step S1, a pressure sensor is arranged at the inlet and outlet of the pressure reducing valve to collect pressure data of the inlet and outlet of the pressure reducing valve; a flow sensor is arranged at the inlet and outlet of the pressure reducing valve to collect flow data of the inlet and outlet of the pressure reducing valve; a bone conduction pickup is arranged on the housing of the pressure reducing valve and a wall on one side of the pressure reducing valve to collect audio data; a valve position relay is arranged at the inlet and outlet of the pressure reducing valve to collect position switch data of the inlet and outlet of the pressure reducing valve; a vibration sensor is arranged at the housing of the pressure reducing valve to collect vibration analog data of the housing of the pressure reducing valve; a camera is arranged on a wall on one side of the pressure reducing valve to collect images of the operation process of the pressure reducing valve; Figure 2 Where L is the flow sensor and Y is the pressure sensor.

[0019] Furthermore, two pressure sensors are set, two flow sensors are set, two bone conduction pickups are set, and two valve position relays are set; the bone conduction pickup adopts ZKHY-IVOD-001 industrial diagnostic pickup with a frequency response of 20-20kHz; the vibration sensor adopts DSP vibration sensor with a measurement range of 0-50g.

[0020] Embodiment three: like Figure 3 As shown, in step S2, the network transmission module includes a switch module composed of an optoelectronic switch and an aggregation switch; the pressure data, flow data, position switch switching data and vibration analog data are respectively connected to the aggregation switch through the optoelectronic switch, the image data and the audio data are extracted by the local control unit, and the image data and the audio data are connected to the aggregation switch through the optoelectronic switch; the aggregation switch is connected to the cloud server platform.

[0021] Furthermore, the PC device is connected to the cloud server platform after passing through the network isolation device, and the cloud server platform communicates with the mobile terminal through the router; the aggregation switch also communicates with the local server for data storage.

[0022] Embodiment 4: like Figure 4 As shown, in step S3, the audio data is identified and scored by the voiceprint scoring module, including in the local control unit, and the audio data is weighted and optimized by MFCC algorithm: First, the audio data is pre-emphasized, and then the pre-emphasized audio data is framed; the framed audio data is subjected to fast Fourier transform to convert the waveform data into frequency, and then the Mel coefficient filtering is performed; finally, endpoint detection is performed and the final MFCC result is tested to ensure the availability of the data and process the invalid data.

[0023] Furthermore, as a carrier of energy, when the size of the sound source is constant, the higher the frequency, the more serious the loss of sound energy by the medium, and pre-emphasis can compensate for the loss of the high-frequency part to a certain extent, thereby protecting the integrity of the sound channel information; the audio signal is affected by the state changes of the sound-emitting object during the generation process, and the speed of state change is much slower than the speed of sound vibration, so it can be considered to be short-term stable. Processing each frame of the signal after framing is equivalent to processing a continuous signal with fixed features, thereby reducing the impact of non-steady-state time-varying signals.

[0024] Furthermore, in the Mel coefficient filtering process, the Mel scale is a cepstrum parameter extracted from the frequency domain, which describes the nonlinear characteristics of the human ear frequency. Its relationship with the frequency is: Mel(f)=2585×log(1+f / 700); Mel(f) represents the Mel scale.

[0025] Preferably, in step S3, the audio data is identified and scored by the voiceprint scoring module, including extracting i-vector processing from the audio data after weighted dimensionality reduction optimization in the local control unit, and the specific method is: Use the GMM-UBM model to obtain the Baum-Welch statistic UBM; The obtained MFCC vector is input into the GMM-UBM for training, and the target MFCC vector is input into the GMM-UBM model. The GMM parameter M of the target device is adaptively obtained through the MAP algorithm based on the original UBM. Process the baum-welch statistic through the DNN neural network and extract the i-vector: Determine whether the audio sample Y comes from device S: Assume that H0 indicates that Y comes from device S, and H1 indicates that Y does not come from device S, then the likelihood formula is: LR = p(Y|H0) / p(Y|H1); log(LR)=log(p(Y|H0))-log(p(Y|H1)); Where p represents the likelihood function; when LR>θ, the result H1 is retained, otherwise the result H0 is retained; θ is a predetermined threshold.

[0026] Preferably, in step S3, the audio data is identified and scored by the voiceprint scoring module, including performing PLDA discriminant analysis in the cloud server platform, and the specific method is: The sound data is used as training audio and divided into audio of I devices. Each device has J different audio segments. The jth audio of the i-th device is defined as X ij ; According to factor analysis, define X ijThe generative model is: ; Where μ represents the mean of all training data; F is the identity space, which contains information that can be used to represent various devices; h i is the identity of a specific device or the position of the device in the identity space; G is the error space, which contains information that can be used to represent different audio changes of the same device; w ij represents the position of the device in the G space; ϵij is the final residual noise term; this term is a zero-mean Gaussian distribution with a variance of Σ; Calculate whether two audios are composed of the feature h in the device space i Generate or calculate two audio frequencies i The generated likelihood is used to obtain the score value; the specific formula is: ; In the formula, score represents the score value; n1 and n2 are the I-Vector vectors of the two audios respectively. The assumption that the two audios come from the same space is Hs, and the assumption that they come from different spaces is Hd. Among them, p(n1,n2|hs) is the likelihood function of the two audios coming from the same space; p(n1|hd) and p(n2|hd) are the likelihood functions of n1 and n2 coming from different spaces respectively; the similarity of the two audios is calculated by calculating the log-likelihood ratio: The ratio, the score, and the possibility that the two audios belong to the same device are positively correlated. The higher the ratio, the higher the score, and the greater the possibility that the two audios belong to the same device. The lower the ratio, the lower the score, and the less likely the two audios belong to the same device. A normal value interval is set for the score value, and the score value falling within the normal value interval is judged as normal, otherwise it is judged as abnormal.

[0027] Furthermore, in step S3, the collected audio data is divided into a training set and a test set, and the test set is further divided into a registration set and a verification set. i-vectors are extracted from the three sets respectively, and the data of the training set is used to train the PLDA discriminant analysis model, and the data of the test set is used for scoring.

[0028] Embodiment five: like Figure 5 As shown, in step S4, the method for performing fault determination by the comprehensive fault determination module is: S401, scoring the real-time audio data collected by the bone conduction pickup through the voiceprint scoring module, and making a judgment based on the scoring value. If the scoring value is normal, the pressure reducing valve device is judged to be normal; if the scoring value is abnormal, the fault type is judged; S402, perform fault type determination, compare the score value with the database, and determine whether it is a known fault or an unknown fault; if it is a known fault, proceed to step S403 to perform known fault type determination; if it is an unknown fault, proceed to step S404 to perform unknown fault type determination; S403, performing known fault type determination: Obtain the switch data of the position switch of the valve position relay and the vibration analog data of the vibration sensor to determine whether the position relay is actuated and whether the displacement time of the vibration data corresponds: If the position relay does not operate, the output is a known fault of a general pressure reducing valve device; If the position relay is actuated and the position change time corresponds, the output is a pressure reducing valve equipment operation event; If the position relay is actuated and the shifting time does not correspond, the output is a known fault of a general pressure reducing valve device; S404, executing unknown fault type determination: The position switch data, flow data, vibration analog data and pressure data are used to jointly determine whether the fault is an internal fault or an external fault. If it is determined to be an internal fault, the professional team is directly notified to conduct on-site inspection to confirm the fault. If it is determined to be an external fault, the camera is used to conduct inspections to locate the fault location, and the professional team is notified to conduct on-site inspections to confirm the fault. S405, after confirming the fault, the fault is marked and sent down, and updated into the scoring database of step S402.

Claims

1. A method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition, characterized in that: The following steps are involved: S1, collecting voiceprint recognition related data through a comprehensive data acquisition module, the comprehensive data acquisition module includes a pressure sensor, a flow sensor, a bone conduction pickup, a valve position relay, a vibration sensor and a camera; the voiceprint recognition related data includes pressure data, flow data, audio data, position switch switching data, vibration analog data and image data; S2, transmits the voiceprint recognition related data collected by the comprehensive data collection module to the cloud server platform through the network transmission module; S3, identifying and scoring the audio data through the voiceprint scoring module; S4, performing fault determination through a comprehensive fault determination module.

2. The method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition according to claim 1 is characterized in that: In step S1, a pressure sensor is arranged at the inlet and outlet of the pressure reducing valve to collect pressure data of the inlet and outlet of the pressure reducing valve; a flow sensor is arranged at the inlet and outlet of the pressure reducing valve to collect flow data of the inlet and outlet of the pressure reducing valve; a bone conduction pickup is arranged on the housing of the pressure reducing valve and the wall on one side of the pressure reducing valve to collect audio data; a valve position relay is arranged at the inlet and outlet of the pressure reducing valve to collect position switch data of the inlet and outlet of the pressure reducing valve; a vibration sensor is arranged on the housing of the pressure reducing valve to collect vibration analog data of the housing of the pressure reducing valve; a camera is arranged on the wall on one side of the pressure reducing valve to collect images of the operation process of the pressure reducing valve.

3. The method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition according to claim 1, characterized in that: In step S2, the network transmission module includes a switch module composed of an optoelectronic switch and an aggregation switch; the pressure data, flow data, position switch switching data and vibration analog data are respectively connected to the aggregation switch through the optoelectronic switch, the image data and audio data are extracted by the local control unit, and the image data and audio data are connected to the aggregation switch through the optoelectronic switch; the aggregation switch is connected to the cloud server platform.

4. The method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition according to claim 3 is characterized in that: In step S3, the audio data is identified and scored by the voiceprint scoring module, including in the local control unit, and the audio data is weighted and optimized by MFCC algorithm: First, the audio data is pre-emphasized, and then the pre-emphasized audio data is framed; the framed audio data is fast Fourier transformed to convert the waveform data into frequency, and then the Mel coefficient filtering is performed; Finally, endpoint detection is performed to detect the final MFCC results to ensure the availability of the data and process the invalid data.

5. The method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition according to claim 4 is characterized in that: In step S3, the audio data is identified and scored by the voiceprint scoring module, including extracting i-vector processing from the audio data after weighted dimensionality reduction optimization in the local control unit. The specific method is as follows: Use the GMM-UBM model to obtain the Baum-Welch statistic UBM; The obtained MFCC vector is input into the GMM-UBM for training, and the target MFCC vector is input into the GMM-UBM model. The GMM parameter M of the target device is adaptively obtained through the MAP algorithm based on the original UBM. Process the baum-welch statistic through the DNN neural network and extract the i-vector: Determine whether the audio sample Y comes from device S: Assume that H0 indicates that Y comes from device S, and H1 indicates that Y does not come from device S, then the likelihood formula is: LR = p(Y|H0) / p(Y|H1); log(LR)=log(p(Y|H0))-log(p(Y|H1)); Where p represents the likelihood function; when LR>θ, the result H1 is retained, otherwise the result H0 is retained; θ is a predetermined threshold.

6. The method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition according to claim 5, characterized in that: In step S3, the audio data is identified and scored by the voiceprint scoring module, including performing PLDA discriminant analysis in the cloud server platform. The specific method is as follows: The sound data is used as training audio and divided into audio of I devices. Each device has J different audio segments. The jth audio of the ith device is defined as X ij ; According to factor analysis, define X ij The generative model is: ； Where μ represents the mean of all training data; F is the identity space, which contains information that can be used to represent various devices; h i is the identity of a specific device or the position of the device in the identity space; G is the error space, which contains information that can be used to represent different audio changes of the same device; w ij represents the position of the device in the G space; ϵij is the final residual noise term; this term is a zero-mean Gaussian distribution with a variance of Σ; Calculate whether two audios are composed of the feature h in the device space i Generate or calculate two audio frequencies i The generated likelihood is used to obtain the score value; the specific formula is: ； In the formula, score represents the score value; n1 and n2 are the I-Vector vectors of the two audios respectively. The assumption that the two audios come from the same space is Hs, and the assumption that they come from different spaces is Hd. Among them, p(n1,n2|hs) is the likelihood function of the two audios coming from the same space; p(n1|hd) and p(n2|hd) are the likelihood functions of n1 and n2 coming from different spaces respectively; the similarity of the two audios is calculated by calculating the log-likelihood ratio: There is a positive correlation between the ratio, the score, and the possibility that the two audios belong to the same device; A normal value interval is set for the score value, and the score value falling within the normal value interval is judged as normal, otherwise it is judged as abnormal.

7. The method for monitoring abnormal multi-modal working conditions of a pressure reducing valve based on sound recognition according to claim 6 is characterized in that: In step S4, the method for performing fault determination by the comprehensive fault determination module is as follows: S401, scoring the real-time audio data collected by the bone conduction pickup through the voiceprint scoring module, and making a judgment based on the scoring value. If the scoring value is normal, it is determined that the pressure reducing valve device is normal; If the scoring value is abnormal, the fault type is determined; S402, performing fault type determination, comparing the score value with the database to determine whether it is a known fault or an unknown fault; If it is a known fault, proceed to step S403 to perform known fault type determination; If it is an unknown fault, the process proceeds to step S404 to determine the type of unknown fault; S403, performing known fault type determination: Obtain the switch data of the position switch of the valve position relay and the vibration analog data of the vibration sensor to determine whether the position relay is actuated and whether the displacement time of the vibration data corresponds: If the position relay does not operate, the output is a known fault of a general pressure reducing valve device; If the position relay is actuated and the position change time corresponds, the output is a pressure reducing valve equipment operation event; If the position relay is actuated and the shifting time does not correspond, the output is a known fault of a general pressure reducing valve device; S404, executing unknown fault type determination: The position switch data, flow data, vibration analog data and pressure data are used to jointly determine whether the fault is an internal fault or an external fault. If it is determined to be an internal fault, the professional team is directly notified to conduct on-site inspection to confirm the fault. If it is determined to be an external fault, the camera is used to conduct inspections to locate the fault location, and the professional team is notified to conduct on-site inspections to confirm the fault. S405, after confirming the fault, the fault is marked and sent down, and updated into the scoring database of step S402.

Images