Main distributing valve working condition abnormity monitoring method based on sound recognition

Abstract

The invention discloses a main distributing valve working condition abnormity monitoring method based on sound recognition, which comprises the following steps: acquiring voiceprint recognition related data including oil level data, audio data, vibration signals, temperature measurement image data, oil pressure data, oil temperature data and guide vane opening through a comprehensive data acquisition 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; fault judgment is carried out through a fault comprehensive judgment module; the method is used for solving the problem that in the prior art, main distributing valve fault recognition cannot be carried out through other parameters of a system, and consequently the efficiency and accuracy of fault recognition are low, and through deep learning and fault judgment process logic design on all data parameters, comprehensive monitoring of main distributing valve faults is achieved, and the fault recognition efficiency and accuracy are improved. And the recognition rate and the accuracy of main distributing valve equipment working condition abnormity or fault judgment 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 working conditions of a main pressure distribution valve based on sound recognition. Background Art

[0002] The speed governor system of a hydropower station unit is a key control part in a hydro-turbine generator unit. It is responsible for automatically adjusting the opening of the turbine guide vanes according to the change of the grid load to control the input of water flow energy, thereby maintaining the unit speed within the set range and ensuring the stability of the output frequency. The speed governor system usually includes an electrical control part, an oil pressure device, and a mechanical hydraulic part. It can realize the control of various links such as the start-up, power generation, pumping, phase adjustment, no-load, grid connection, load increase and decrease, and shutdown of the unit to adapt to the increase and decrease of the grid load. The main pressure regulating valve is a key hydraulic control element in the turbine speed regulating system. Its function is to adjust the opening of the turbine guide vanes under the control of the speed governor, thereby controlling the flow of water through the guide vanes, and then controlling the speed of the turbine and the output power of the generator set; the main pressure regulating valve is usually composed of one or more hydraulic valves, which realize precise control of the guide vane opening through the pressure change of the hydraulic oil. In the speed governor system, the main pressure regulating valve receives signals from the speed governor, which may be based on feedback from the unit speed, load or other operating parameters. The main pressure regulating valve adjusts its opening according to these signals to ensure that the unit operates in the predetermined working state. In the hydro-turbine generator set, the stable and reliable operation of the main pressure regulating valve is crucial to ensure the stable power supply of the power system and the safe operation of the unit. If the main pressure regulating valve fails, it may cause the unit to fail to adjust the speed normally, and may even cause the unit to overspeed or other safety accidents. Therefore, the maintenance and fault diagnosis of the main pressure regulating valve is an important part of the operation and maintenance of the hydropower station.

[0003] The prior art generally installs temperature, pressure and other sensors on the oil pipeline of the governor system, judges the system fault by abnormal changes in temperature and pressure, and then analyzes the cause of the fault as possible abnormality of the main pressure distribution valve by combining parameters such as the guide vane opening, active power and oil level change of the governor system. The conclusion is drawn by analyzing the changes in other system parameters. The basis for fault judgment is indirect and not timely. Therefore, it is necessary to design a main pressure distribution valve operating condition abnormality monitoring method based on sound recognition to solve the above problems. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method for monitoring the abnormal operating condition of a main pressure-distributing valve based on sound recognition. The method is used to solve the problem that the prior art cannot identify main pressure-distributing valve faults through other system parameters, resulting in low efficiency and accuracy in fault identification. By performing deep learning on various data parameters and logical design of the fault judgment process, comprehensive monitoring of main pressure-distributing valve faults is achieved, greatly improving the recognition rate and accuracy of abnormal operating condition or fault judgment of the main pressure-distributing valve equipment.

[0005] In order to achieve the above technical effects, the technical solution adopted by the present invention is: A method for monitoring abnormal working condition of a main pressure distribution valve based on sound recognition comprises the following steps: S1, collect voiceprint recognition related data through the comprehensive data acquisition module, the comprehensive data acquisition module includes an oil level sensor, a bone conduction pickup, a vibration sensor, an infrared temperature measurement camera, an oil pressure sensor and an oil temperature sensor; the voiceprint recognition related data includes oil level data, audio data, vibration signal, temperature measurement image data, oil pressure data, oil temperature data and guide vane opening; 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, perform fault judgment through the comprehensive fault judgment module: 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 main pressure distribution 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, determine whether it is a known fault or an unknown fault, if it is compared with the historical data record in the database, it is determined to be a known fault and the determination result is output; if it is not compared with the historical data record in the database, it is determined to be an unknown fault, and enter step S403 to perform unknown fault type determination; S403, executing unknown fault type determination: The oil level data, vibration signal, guide vane opening data, oil pressure data and oil temperature 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, a temperature measurement camera is used to conduct an inspection to locate the fault location, and the professional team is notified to conduct on-site inspection to confirm the fault; S404, after confirming the fault, the fault is marked and sent down, and updated into the scoring database of step S402.

[0006] Preferably, in step S1, an oil level sensor is arranged at the oil pressure tank to collect oil level data of the oil pressure tank; a bone conduction pickup is arranged at the main pressure regulating valve housing to collect audio data; a vibration sensor is arranged at the main pressure regulating valve housing to collect vibration signals of the main pressure regulating valve housing; an infrared temperature measuring camera is arranged on a wall on one side of the main pressure regulating valve to collect temperature measurement image data of the main pressure regulating valve operation process; an oil pressure sensor and an oil temperature sensor are arranged on the main oil supply pipe to collect oil pressure and oil temperature data of the main oil supply pipe; and the guide vane opening data is collected by connecting to the turbine unit operation management system.

[0007] Furthermore, two oil level sensors, one bone conduction pickup, and one vibration sensor are provided; the bone conduction pickup adopts the ZKHY-IVOD-001 industrial diagnostic pickup with a frequency response of 20-20kHz; the vibration sensor adopts a DSP vibration sensor with a measurement range of 0-50g.

[0008] Preferably, in step S2, the network transmission module includes a switch module consisting of a photoelectric switch and a convergence switch; the oil level data, vibration signal, oil pressure data, oil temperature data and guide vane opening data are respectively connected to the convergence switch through the photoelectric switch, the temperature measurement image data and audio data are extracted with characteristic values ​​through the local control unit, and the temperature measurement 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.

[0009] 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.

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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 by h 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.

[0014] 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.

[0015] The beneficial effects of the present invention are as follows: 1. The present invention arranges a bone conduction voiceprint sensor on the end cover or base of the main pressure distribution valve to collect the sound signals of the main pressure distribution valve under normal working conditions and abnormal working conditions. By comparing the decoded different signal data, an alarm is given to the operation and maintenance personnel of the hydropower station, so that they can quickly and accurately determine whether it is a main pressure distribution valve failure, providing a basis and time guarantee for the stable operation of the unit and fault handling.

[0016] 2. Through the installation of various types of acquisition sensors, multi-modal online real-time monitoring of the main pressure distribution valve equipment area is realized, and unmanned operation is truly realized; the present invention adopts bone conduction pickup to collect 50Hz-4KHz audio data, which can collect almost all abnormal conditions in the main pressure distribution valve equipment area, and is also assisted by pressure, vibration, flow, switch quantity and other data. Therefore, this solution can identify all faults in the main pressure distribution valve equipment area of ​​the equipment, and greatly improve the accuracy of abnormal monitoring of the main pressure distribution valve equipment operating condition; through voiceprint recognition technology, and through comprehensive judgment and analysis of multi-modal data such as flow, vibration, pressure, image, switch quantity, etc., the accuracy of abnormal monitoring of the main pressure distribution valve equipment operating condition can be greatly improved, and human resources can be saved 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

[0017] 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 Schematic diagram of the process of the voiceprint scoring module in an embodiment of the present invention; Figure 4 The figure is a flow chart of comprehensive fault determination in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] Embodiment 1: like Figure 1 As shown, a method for monitoring the abnormal working condition of a main pressure regulating valve based on sound recognition comprises the following steps: S1, collect voiceprint recognition related data through the comprehensive data acquisition module, the comprehensive data acquisition module includes an oil level sensor, a bone conduction pickup, a vibration sensor, an infrared temperature measurement camera, an oil pressure sensor and an oil temperature sensor; the voiceprint recognition related data includes oil level data, audio data, vibration signal, temperature measurement image data, oil pressure data, oil temperature data and guide vane opening; 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, perform fault judgment through the comprehensive fault judgment module: 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 main pressure distribution 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, determine whether it is a known fault or an unknown fault, if it is compared with the historical data record in the database, it is determined to be a known fault and the determination result is output; if it is not compared with the historical data record in the database, it is determined to be an unknown fault, and enter step S403 to perform unknown fault type determination; S403, executing unknown fault type determination: The oil level data, vibration signal, guide vane opening data, oil pressure data and oil temperature 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, a temperature measurement camera is used to conduct an inspection to locate the fault location, and the professional team is notified to conduct on-site inspection to confirm the fault; S404, after confirming the fault, the fault is marked and sent down, and updated into the scoring database of step S402.

[0019] like Figure 2As shown, preferably, in step S1, the oil level sensor is arranged at the oil pressure tank to collect the oil level data of the oil pressure tank; the bone conduction pickup is arranged on the main pressure regulating valve housing to collect the audio data; the vibration sensor is arranged on the main pressure regulating valve housing to collect the vibration signal of the main pressure regulating valve housing; the infrared temperature measuring camera is arranged on the wall on one side of the main pressure regulating valve to collect the temperature measurement image data of the main pressure regulating valve operation process; the oil pressure sensor and the oil temperature sensor are arranged on the main oil supply pipe to collect the oil pressure and oil temperature data of the main oil supply pipe; the guide vane opening data is collected by connecting to the turbine unit operation management system.

[0020] Furthermore, two oil level sensors, one bone conduction pickup, and one vibration sensor are provided; the bone conduction pickup adopts the ZKHY-IVOD-001 industrial diagnostic pickup with a frequency response of 20-20kHz; the vibration sensor adopts a DSP vibration sensor with a measurement range of 0-50g.

[0021] Figure 2 In the process, the bone conduction pickup is installed on the end cap or base of the main pressure regulating valve.

[0022] Preferably, in step S2, the network transmission module includes a switch module consisting of a photoelectric switch and a convergence switch; the oil level data, vibration signal, oil pressure data, oil temperature data and guide vane opening data are respectively connected to the convergence switch through the photoelectric switch, the temperature measurement image data and audio data are extracted with characteristic values ​​through the local control unit, and the temperature measurement 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.

[0023] 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.

[0024] Embodiment 2: like Figure 3 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.

[0025] 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.

[0026] 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.

[0027] 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 by h 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.

[0028] 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.

[0029] Embodiment three: like Figure 4 As shown, after the soundprint scoring value of the main pressure distribution valve equipment area is obtained, the main pressure distribution valve equipment operating status is obtained through the fault judgment logic of the main pressure distribution valve equipment area based on soundprint, temperature, image, vibration, oil level, active power, guide vane opening and other parameters, the result is output and corresponding measures are taken.

[0030] Audio data is collected through the installed bone conduction pickup, and the characteristic values ​​are extracted and input into the voiceprint model for judgment. The voiceprint model library contains voiceprint data of existing known faults in the main pressure distribution valve. The collected real-time voiceprint characteristic values ​​are scored. If normal, it means that the main equipment in the main pressure distribution valve area is normal.

[0031] The first case is that the voiceprint score value is abnormal, and the matching value with the known fault or event is high, so the known fault is output.

[0032] The second case is that the voiceprint score is abnormal. The background automatically analyzes whether there are abnormal mutations in the active power, guide vane opening, temperature, and liquid level data, and whether the vibration data is abnormal, to comprehensively determine whether the fault occurs inside or outside the main pressure regulating valve equipment. If it is external, it is necessary to operate the image monitoring camera in conjunction with the operating personnel to remotely locate the fault position at the first time, and promptly notify the equipment professional team to conduct on-site inspection and confirm the fault; if it is an internal fault, it is necessary to promptly notify the equipment professional team to conduct on-site inspection and clarify the fault after power outage and maintenance.

[0033] After confirming the fault, in addition to taking corresponding measures according to the fault, it is also necessary to register the identified fault and voiceprint data, mark it as a known fault, enter it into the fault model library, and send it to the local control unit. The next time the voiceprint data is abnormal, add the known fault for judgment. Through long-term operation accumulation, the accumulation of most common fault model libraries has been completed.

Claims

1. A method for monitoring abnormal working condition of a main pressure regulating valve based on sound recognition, characterized in that: The following steps are involved: S1, collect voiceprint recognition related data through the comprehensive data acquisition module, the comprehensive data acquisition module includes an oil level sensor, a bone conduction pickup, a vibration sensor, an infrared temperature measurement camera, an oil pressure sensor and an oil temperature sensor; the voiceprint recognition related data includes oil level data, audio data, vibration signal, temperature measurement image data, oil pressure data, oil temperature data and guide vane opening; 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, perform fault judgment through the comprehensive fault judgment module: 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 judged that the main pressure distribution 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 compared with the historical data record in the database, it is determined to be a known fault and the determination result is output; If there is no historical data record in the database, it is determined to be an unknown fault, and the process goes to step S403 to perform unknown fault type determination; S403, executing unknown fault type determination: The oil level data, vibration signal, guide vane opening data, oil pressure data and oil temperature 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, a temperature measurement camera is used to conduct an inspection to locate the fault location, and the professional team is notified to conduct on-site inspection to confirm the fault; S404, after confirming the fault, the fault is marked and sent down, and updated into the scoring database of step S402.

2. The method for monitoring the abnormal working condition of the main pressure regulating valve based on sound recognition according to claim 1 is characterized in that: In step S1, an oil level sensor is arranged at the oil pressure tank to collect the oil level data of the oil pressure tank; a bone conduction pickup is arranged at the main pressure regulating valve housing to collect audio data; a vibration sensor is arranged at the main pressure regulating valve housing to collect the vibration signal of the main pressure regulating valve housing; an infrared temperature measuring camera is arranged on the wall on one side of the main pressure regulating valve to collect the temperature measurement image data of the main pressure regulating valve operation process; an oil pressure sensor and an oil temperature sensor are arranged on the main oil supply pipe to collect the oil pressure and oil temperature data of the main oil supply pipe; and the guide vane opening data is collected by connecting to the turbine unit operation management system.

3. The method for monitoring the abnormal working condition of the main pressure regulating valve based on sound recognition according to claim 1 is characterized in that: In step S2, the network transmission module includes a switch module composed of a photoelectric switch and a convergence switch; the oil level data, vibration signal, oil pressure data, oil temperature data and guide vane opening data are respectively connected to the convergence switch through the photoelectric switch, the temperature measurement image data and audio data are extracted with characteristic values ​​through the local control unit, and the temperature measurement 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.

4. The method for monitoring the abnormal working condition of the main pressure regulating 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 the abnormal working condition of the main pressure regulating 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 the abnormal working condition of the main pressure regulating valve based on sound recognition according to claim 5 is 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 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 by h 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.

Images