Volute disc valve working condition abnormity monitoring method based on sound recognition

Abstract

According to the volute disc valve working condition abnormity monitoring method based on sound recognition, the working condition of a volute disc valve is evaluated by adopting multi-feature data; wherein the characteristic information comprises voiceprint signals collected on a volute disc valve hydraulic device, the water flow size of a disc valve drainage pipe, vibration data of the hydraulic device, the device surface water leakage amount, temperature data and image monitoring data, and the voiceprint signals, the disc valve drainage pipe water flow size, the hydraulic device vibration data, the device surface water leakage amount, the temperature data and the image monitoring data serve as the characteristics of fault state recognition, a state evaluation model based on linear probability judgment is constructed, and the characteristic information is trained; and integrating to form a database under different fault types of the volute disc valve, and inputting real-time data to obtain the fault type with the highest PLDA ratio, namely the fault type of the to-be-diagnosed feature sequence. Fault features and fault diagnosis of the volute disc valve of the water turbine can be extracted in real time.

Description

Technical Field

[0001] The invention relates to the technical field of hydraulic turbine fault detection, and in particular to a method for monitoring abnormal working conditions of a volute disc valve based on sound recognition. Background Art

[0002] The water diversion chamber of the turbine is the first part where water flows into the turbine. In order to adapt to different flow and head conditions, the shapes and materials of the diversion chambers used in various types of reaction turbines are different. In summary, there are three major categories: open diversion chamber, tank diversion chamber and volute diversion chamber. The appearance of the volute diversion chamber is very similar to a snail shell, so it is usually referred to as a volute. In order to ensure uniform water supply to the water guide mechanism, the cross-section of the volute is gradually reduced. At the same time, it can form the necessary circulation flow in front of the water guide mechanism to reduce the working intensity of the water guide mechanism. As a maintenance drain valve, the volute disc valve refers to a drain valve set to remove the accumulated water in the volute, pressure steel pipe and tailwater pipe during maintenance. The volute drainage is completed by the volute disc valve, which is installed at the lowest point of the volute to drain the water to the tailwater pipe. The drain valve is driven by a hydraulic servo, and the oil supply of the hydraulic servo is completed by a manual oil pump.

[0003] Common faults of the volute disc valve include the volute disc valve not closing tightly, causing the water in the volute to leak into the tailwater pipe through the gap between the volute disc valve disc, causing abnormal vibration and sound in the volute disc valve drain pipe; defects in the weld of the volute disc valve drain pipe, causing water leakage in the volute drain pipe; failure of the volute disc valve shaft seal, causing water in the tailwater pipe to leak through the volute disc valve shaft, etc.

[0004] The existing industry has a relatively blank monitoring method for the working condition of the volute disc valve. Most hydropower stations only upload the switching signal of the volute disc valve to the monitoring system. The monitoring of abnormal working conditions of the disc valve is almost zero. It is only discovered on-site by operation and maintenance personnel and related routine inspections are carried out during unit maintenance.

[0005] As the existing technology cannot realize real-time online monitoring of the volute disc valve, the operation and maintenance personnel can only conduct inspections based on their experience. Therefore, before the volute disc valve is opened for drainage during unit maintenance, it is impossible to accurately determine whether the volute disc valve is operating normally, and the only way to determine whether the operating condition is normal is by the drainage volume. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a method for monitoring the abnormal working condition of a volute disc valve based on sound recognition, which monitors all internal faults of the volute disc valve through voiceprint recognition technology and realizes real-time online monitoring of abnormal sounds; by arranging bone conduction sensors, thermal infrared cameras, acceleration sensors, water immersion sensors, etc. on the hydraulic operating device, and integrating multimodal data into the same system, real-time online monitoring of the water flow size, leakage, vibration amplitude and abnormal working condition of the equipment in the disc valve drain pipe is realized, solving the problem that the existing monitoring method cannot conduct comprehensive analysis; by deep learning of various data parameters and logical design of fault judgment process, comprehensive monitoring of abnormal working condition of turbine fast gate is realized, which greatly improves the recognition rate and accuracy of abnormal working condition or fault judgment of fast gate equipment.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is: A method for monitoring abnormal working condition of a volute disc valve based on sound recognition, comprising: A comprehensive monitoring system for multi-modal abnormal working conditions of volute disc valves based on voiceprint recognition is established. The monitoring system includes an acquisition unit, a data transmission unit, a voiceprint analysis unit and a comprehensive fault diagnosis unit. The sensor hardware constitutes the acquisition unit, which collects data online in real time and transmits it to the background server through the data transmission unit for comprehensive analysis and judgment. The voiceprint analysis unit is used to determine whether there are known faults or unknown abnormalities in the working condition of the turbine fast gate. The specific type of fault is comprehensively judged based on the multi-modal sensor data, and finally the result is output.

[0008] The above-mentioned collection unit includes a thermal infrared camera installed on the wall opposite to the hydraulic operating device of the volute disc valve, a bone conduction pickup installed on the flange of the hydraulic device of the volute disc valve, an acceleration sensor installed at the connection between the flange and the drain pipe, and a water immersion sensor installed at the connection between the flange and the drain pipe.

[0009] The above-mentioned data transmission unit includes an edge computing gateway, a switch and a server platform, which transmits the sensor data to the cloud data platform of the background server through a data cable for further analysis.

[0010] The above-mentioned voiceprint analysis unit has a built-in voiceprint recognition algorithm engine. The voiceprint recognition algorithm engine performs weighted dimensionality reduction optimization based on the MFCC feature vector, and applies a vector quantization algorithm to identify the mechanical noise signal outside the cable line, including voiceprint MFCC extraction, UBM calculation, i-vector calculation, and linear probability discriminant analysis.

[0011] The above voiceprint MFCC extraction steps are: Step 1: Pre-emphasis: The medium is the carrier of energy. When the sound source size is constant, the higher the frequency, the more serious the loss of sound energy by the medium. Pre-emphasis can compensate for the loss of high-frequency part to a certain extent and protect the integrity of the sound channel information. Step 2: Framing: Divide the continuous voiceprint signal into multiple short time periods, i.e. frames, for subsequent processing; Step 3: FFT transformation: Perform fast Fourier transform (FFT) on each frame of signal. Through Fourier transform, the waveform data is transformed into the frequency domain. The specific formula of FFT is: ; in, is the time domain signal, is the frequency domain signal, N is the number of sampling points, and j is the imaginary unit; Step 4: Mel coefficient filtering; Mel scale is the cepstrum parameter extracted from the frequency domain, which describes the nonlinear characteristics of the human ear frequency. Its relationship with frequency can be approximately expressed by the following formula: Mel(f)=2585×log(1+f / 700); Step 5: Endpoint detection: Endpoint detection will detect the final result to ensure the availability of data and process invalid data.

[0012] In the above-mentioned calculation of UBM, the universal background model UBM, namely the Universal Background Model, first collects speech irrelevant to the target, trains a UBM, and then uses the target speech data to adjust the parameters of the UBM through an adaptive algorithm to obtain the target model parameters; the parameters adjusted by the adaptive algorithm are only the mean.

[0013] The above calculation process of i-vector is: The process of solving i-vector is the MAP maximum a posteriori probability estimation process, that is, given the speaker data, the posterior probability distribution P (w|X) P(w|X)P(w|X) of i-vector is estimated, and the mean of the distribution is the i-vector of the speaker. In this process, the EM algorithm process is included to solve the model parameter T matrix, and w is the hidden variable. We get the calculation formula of i-vector: ; Where N(u) is a diagonal matrix of dimension CF × CF, and the diagonal blocks are N c I , ( c = 1 , . . . , C ) , is a CF × 1 supervector consisting of all first-order BW statistics F~c.

[0014] The above linear probability discriminant analysis process is: Probabilistic Linear Discriminant Analysis (PLDA) is a channel compensation algorithm, also known as the probabilistic LDA algorithm. PLDA is usually based on I-Vector features and provides channel compensation for them. Assume that the training data speech consists of I speakers’ speech, where each speaker has J different speech segments. Then, we define i Personal j The voice is ; Based on factor analysis, define The generative model is: ; in, μ represents the mean of all training data; F can be regarded as the identity space, which contains information that can be used to represent various speakers; It can be regarded as the identity of a specific speaker, or the position of the speaker in the identity space; G can be regarded as the error space, which contains information that can be used to represent different voice changes of the same speaker; It represents the position in G space; is the final residual noise term, which is used to represent the other parts that have not been explained; this term is a zero-mean Gaussian distribution with a variance of Σ; The goal of PLDA model training is to input a bunch of data , the output can best represent the parameters of the data set θ = [μ,F,G,Σ].

[0015] In PLDA, it is necessary to calculate whether two speech sounds are related by features in the speaker space. generated by The likelihood of the generated class is not affected by the difference in the class space. The scoring formula is given below: ; As in the above formula, and are the I-Vector vectors of the two voices. The assumption that the two voices come from the same space is Hs, and the assumption that they come from different spaces is Hd. Among them, p( , | hs) is the likelihood function of the two speech sounds coming from the same space; p( |hd),p |hd) are and By calculating the log-likelihood ratio, we can measure the similarity between two speech sounds. The higher the ratio, the higher the score, and the greater the possibility that the two voices belong to the same speaker; The lower the ratio, the lower the score, and the less likely it is that the two voices belong to the same speaker.

[0016] After obtaining the sound print model of the volute disc valve, the above-mentioned comprehensive fault diagnosis unit determines whether leakage occurs through the parameters transmitted by the bone conduction pickup, thermal infrared thermometer, and vibration sensor installed on the volute disc valve hydraulic device, outputs the results and takes corresponding measures; When an abnormality is found in the voiceprint score of the real-time analysis device, it is compared with the known fault model issued by the edge computing gateway. If it is a known fault, an alarm of the known fault is directly output; if it does not match the known fault model library, an unknown abnormality alarm is output.

[0017] Once an unknown abnormal alarm is reported, the surveillance camera will carefully observe and analyze the alarm point, and at the same time, the operation personnel will be arranged to go to the site to confirm the situation and make corresponding treatment. After the treatment is completed, they will come back to report and then improve the fault model database. Next time, the same fault will be reported as a known fault; 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, the known fault will be added for judgment.

[0018] The present invention provides a method for monitoring abnormal working conditions of a volute disc valve based on sound recognition, which has the following beneficial effects: 1. Through the installation of various types of data collection sensors, multi-modal online real-time monitoring of the volute disc valve equipment area is realized to fill the gap in the industry; 2. The bone conduction pickup collects 50Hz-4KHz audio data, which can collect almost all abnormal conditions in the volute disc valve equipment area. At the same time, it is assisted by temperature, vibration, image monitoring and other data. Therefore, this solution can identify all faults in the volute disc valve equipment area of ​​the equipment, greatly improving the accuracy of abnormal monitoring of the volute disc valve equipment working condition; 3. Through voiceprint recognition technology, and through comprehensive judgment and analysis of multimodal data such as temperature, vibration, image, and gas monitoring, the accuracy of abnormal operating condition monitoring of volute disc valve equipment can be greatly improved, while saving human resources and reducing the time required for fault identification, ensuring the safe and stable operation of the unit and avoiding accidents during startup and shutdown. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 This is a schematic diagram of the installation position of the collection unit of the present invention; Figure 2 It is a structural schematic diagram of the data transmission module of the present invention; Figure 3 Schematic diagram of UBM training of the present invention. DETAILED DESCRIPTION

[0020] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and embodiments.

[0021] Example: The multi-modal working condition abnormality comprehensive monitoring method of the volute disc valve based on voiceprint recognition includes four major units: acquisition unit, data transmission unit, voiceprint analysis unit, and comprehensive fault diagnosis unit. As shown in the figure, the sensor hardware constitutes the acquisition unit, which collects data online in real time and transmits it to the background server through the data transmission unit for comprehensive analysis and judgment. The voiceprint analysis unit is used to determine whether there are known faults or unknown abnormalities in the working condition of the turbine fast gate, and the specific type of fault is comprehensively judged based on the multi-modal sensor data, and finally the result is output.

[0022] 1. Collection unit The main hardware of the multi-modal working condition abnormal comprehensive monitoring method of the volute disc valve based on voiceprint recognition includes: 1. A thermal infrared camera, installed on the wall opposite to the hydraulic operating device of the volute disc valve; 2. A bone conduction pickup, installed on the flange of the hydraulic device of the volute disc valve to better collect sound; 3. An acceleration sensor, installed at the connection between the flange and the drain pipe; 4. A water immersion sensor, installed at the connection between the flange and the drain pipe, the specific location is as follows: Figure 1 shown.

[0023] The data collection includes: Voiceprint signal: The sound signal generated by the volute disc valve hydraulic device during operation is collected through a bone conduction microphone. The signal contains rich information about the operating status of the equipment.

[0024] Water flow size: The water flow size in the drain pipe of the disc valve is measured by a flow sensor to reflect the opening state of the valve and the change in fluid pressure.

[0025] Vibration data: The vibration signal of the hydraulic device is collected through the acceleration sensor to analyze the mechanical operating status of the equipment.

[0026] Water leakage: The water leakage on the surface of the device is monitored by a water immersion sensor to determine whether there is a problem such as sealing failure.

[0027] Temperature data: The temperature distribution of the volute disc valve and its surroundings is measured by thermal infrared sensors to detect overheating or abnormal temperature points.

[0028] Image monitoring data: The image of the volute disc valve and its surrounding environment is captured by the camera to assist in judging changes in the appearance of the equipment and the surrounding environment.

[0029] 2. Data transmission module The data of the corresponding sensors are transmitted to the cloud data platform of the background server through the data line for further analysis. The data transmission system includes edge computing gateways, switches, server platforms, etc. Among them, the voiceprint feature value data is set with an edge computing gateway, which has the function of initial screening of audio data. Since the thermal infrared camera has two functions of temperature measurement and image monitoring, it is necessary to set up an edge computing gateway to transmit the temperature measurement data separately from the image and video data, which is convenient for the subsequent normalization of data; Figure 2 shown.

[0030] 3. Voiceprint model unit The voiceprint recognition algorithm engine performs weighted dimensionality reduction optimization based on MFCC feature vectors, and applies vector quantization algorithm to identify mechanical noise signals outside the cable line; including voiceprint MFCC extraction, UBM calculation, i-vector calculation, and linear probability discriminant analysis.

[0031] Voiceprint MFCC extraction process: The first step is pre-emphasis. As a carrier of energy, the medium loses more sound energy with higher frequencies when the size of the sound source is constant. Pre-emphasis can compensate for the loss of high-frequency parts to a certain extent and protect the integrity of the sound channel information. The second step is framing. During the generation process, the speech signal is affected by the state change of the sound-generating object, and the speed of state change is much slower than the speed of sound vibration. Therefore, it can be considered to be short-term stable. After framing, processing each frame signal is equivalent to processing a continuous signal with fixed features, which can reduce the impact of non-steady-state time variation. The third step is FFT transformation; FFT is the fast Fourier transform, which transforms the waveform data into the frequency domain through Fourier transform; the specific formula of FFT is: ; in, is the time domain signal, is the frequency domain signal, N is the number of sampling points, and j is the imaginary unit; The fourth step is Mel coefficient filtering; the Mel scale is the cepstrum parameter extracted from the frequency domain, which describes the nonlinear characteristics of the human ear frequency. Its relationship with the frequency can be approximately expressed by the following formula: Mel(f)=2585×log(1+f / 700); The fifth step of the MFCC algorithm is endpoint detection, which will detect the final result, ensure the availability of the data, and process the invalid data.

[0032] Calculate UBM: Universal Background Model (UBM) is a model that first collects a large amount of speech irrelevant to the target, trains a UBM, and then uses a small amount of target speech data to adjust the parameters of the UBM through an adaptive algorithm to obtain the target model parameters. Generally, the only parameter adjusted by the adaptive algorithm is the mean. This approach can reduce the amount of data and parameters in actual use, and facilitates rapid training convergence and decoding calculations on mobile terminals.

[0033] The training of UBM is actually the training of Gaussian mixture model GMM. Assume that a UBM is trained using a large amount of background speech data. The next focus is how to adapt the parameters of UBM to obtain the GMM parameters of the target speaker. The schematic diagram is as follows: Figure 3 shown.

[0034] Calculating i-vector The process of solving i-vector is actually the MAP maximum a posteriori probability estimation process, that is, given the speaker data, the posterior probability distribution P ( w|X ) P(w|X)P(w|X) of i-vector is estimated, and the mean of the distribution is the i-vector of the speaker; this process includes the EM algorithm process - used to solve the model parameter T matrix, and w is actually the hidden variable; in order to obtain the mean of this posterior probability distribution, alignment statistics are needed, align means that some frames in the speech data of a certain speaker are attached to, generated from, and come from a certain component Gaussian in UBM. We can get a series of alignment statistics, including zero-order, first-order, and second-order statistics; using soft counting of the probability from a certain Gaussian component, we get Baum-Welch statistics, based on hard counting, the probability from a certain Gaussian component is the largest, that is, 1, and we get Viterbi statistics. As in the previous GMM-UBM blog post, we use Baum-Welch statistics; based on this, we get the calculation formula of i-vector: ; Where N(u) is a diagonal matrix of dimension CF × CF, and the diagonal blocks are N c I , ( c = 1 , . . . , C ) , is a CF × 1 supervector consisting of all first-order BW statistics F~c.

[0035] Linear Probability Discriminant Analysis Probabilistic Linear Discriminant Analysis (PLDA) is also a channel compensation algorithm, also known as the probabilistic form of the LDA algorithm. PLDA is usually based on I-Vector features and provides channel compensation for them; since the compensation ability of the PLDA algorithm is better than that of LDA, most people prefer PLDA.

[0036] In the field of voiceprint recognition, it is assumed that the training data speech consists of the speech of I speakers, where each speaker has J different segments of his own speech. Then, we define i Personal j The voice is ; Based on factor analysis, define The generative model is: ; in, μ represents the mean of all training data; F can be regarded as the identity space, which contains information that can be used to represent various speakers; It can be regarded as the identity of a specific speaker, or the position of the speaker in the identity space; G can be regarded as the error space, which contains information that can be used to represent different voice changes of the same speaker; It represents the position in G space; is the final residual noise term, which is used to represent the other parts that have not been explained; this term is a zero-mean Gaussian distribution with a variance of Σ; The goal of PLDA model training is to input a bunch of data , the output can best represent the parameters of the data set θ = [μ,F,G,Σ].

[0037] In PLDA, it is necessary to calculate whether two speech sounds are related by features in the speaker space. generated by The likelihood of the generated class is not affected by the difference in the class space. The scoring formula is given below: ; As in the above formula, and are the I-Vector vectors of the two voices. The assumption that the two voices come from the same space is Hs, and the assumption that they come from different spaces is Hd. Among them, p( , | hs) is the likelihood function of the two speech sounds coming from the same space; p( |hd),p |hd) are and By calculating the log-likelihood ratio, we can measure the similarity between two speech sounds. The higher the ratio, the higher the score, and the greater the possibility that the two voices belong to the same speaker; The lower the ratio, the lower the score, and the less likely it is that the two voices belong to the same speaker.

[0038] 4. Comprehensive fault judgment module After obtaining the soundprint model of the volute disc valve, the parameters transmitted by the bone conduction pickup, thermal infrared thermometer and vibration sensor installed on the hydraulic device of the volute disc valve are used to determine whether leakage occurs, output the results and take corresponding measures.

[0039] When an abnormality is found in the voiceprint score of the real-time analysis device, it is compared with the known fault model issued by the edge computing gateway. If it is a known fault, an alarm of the known fault is directly output; if it does not match the known fault model library, an unknown abnormality alarm is output.

[0040] Once an unknown abnormal alarm is reported, the surveillance camera will first carefully observe and analyze the alarm point, and at the same time, the operating personnel will be arranged to go to the site to confirm the situation and make corresponding treatment. After the treatment is completed, they will come back to report and then improve the fault model database. Next time, the same fault will be alarmed as a known fault.

[0041] 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 volute disc valve based on sound recognition, characterized in that: include: A comprehensive monitoring system for multi-modal abnormal working conditions of volute disc valves based on voiceprint recognition is established. The monitoring system includes an acquisition unit, a data transmission unit, a voiceprint analysis unit and a comprehensive fault diagnosis unit. The sensor hardware constitutes the acquisition unit, which collects data online in real time and transmits it to the background server through the data transmission unit for comprehensive analysis and judgment. The voiceprint analysis unit is used to determine whether there are known faults or unknown abnormalities in the working condition of the turbine volute disc valve. The specific type of fault is comprehensively judged based on the multi-modal sensor data, and finally the result is output.

2. According to the method for monitoring abnormal working condition of a volute disc valve based on sound recognition as described in claim 1, it is characterized in that: The collection unit includes a thermal infrared camera installed on the wall opposite to the volute disc valve hydraulic operating device, a bone conduction pickup installed on the flange of the volute disc valve hydraulic device, an acceleration sensor installed at the connection between the flange and the drain pipe, and a water immersion sensor installed at the connection between the flange and the drain pipe.

3. According to the method for monitoring abnormal working condition of a volute disc valve based on sound recognition as described in claim 2, it is characterized in that: The data transmission unit includes an edge computing gateway, a switch and a server platform, which transmits the sensor data to the cloud data platform of the background server through a data cable for further analysis.

4. According to the method for monitoring abnormal working condition of a volute disc valve based on sound recognition as described in claim 3, it is characterized in that: The voiceprint analysis unit has a built-in voiceprint recognition algorithm engine. The voiceprint recognition algorithm engine performs weighted dimensionality reduction optimization based on MFCC feature vectors, and applies a vector quantization algorithm to identify mechanical noise signals outside the cable, including voiceprint MFCC extraction, UBM calculation, i-vector calculation, and linear probability discriminant analysis.

5. A method for monitoring abnormal working condition of a volute disc valve based on sound recognition according to claim 4, characterized in that: The voiceprint MFCC extraction steps are: Step 1: pre-emphasis; Step 2: Framing: Divide the continuous voiceprint signal into multiple short time periods, i.e. frames, for subsequent processing; Step 3: FFT transformation: Perform fast Fourier transform (FFT) on each frame of signal. Through Fourier transform, the waveform data is transformed into the frequency domain. The specific formula of FFT is: ； in, is the time domain signal, is the frequency domain signal, N is the number of sampling points, and j is the imaginary unit; Step 4: Mel coefficient filtering; Mel scale is the cepstrum parameter extracted from the frequency domain, which describes the nonlinear characteristics of the human ear frequency. Its relationship with frequency can be approximately expressed by the following formula: Mel(f)=2585×log(1+f / 700); Step 5: Endpoint detection: Endpoint detection will detect the final result to ensure the availability of data and process invalid data.

6. A method for monitoring abnormal working condition of a volute disc valve based on sound recognition according to claim 5, characterized in that: In the calculation of UBM, the universal background model UBM, namely the Universal Background Model, first collects speech irrelevant to the target, trains a UBM, and then uses the target speech data to adjust the parameters of the UBM through an adaptive algorithm to obtain the target model parameters; the parameters adjusted by the adaptive algorithm are only the mean.

7. A method for monitoring abnormal working condition of a volute disc valve based on sound recognition according to claim 6, characterized in that: The i-vector calculation process is as follows: The process of solving i-vector is the MAP maximum a posteriori probability estimation process, that is, given the speaker data, the posterior probability distribution P (w|X) P(w|X)P(w|X) of i-vector is estimated, and the mean of the distribution is the i-vector of the speaker. In this process, the EM algorithm process is included to solve the model parameter T matrix, and w is the hidden variable. The calculation formula of i-vector is obtained as follows: ； Where N(u) is a diagonal matrix of dimension CF × CF, and the diagonal blocks are N c I , ( c = 1 , . . . , C ) , is a CF × 1 supervector consisting of all first-order BW statistics F~c.

8. A method for monitoring abnormal working condition of a volute disc valve based on sound recognition according to claim 7, characterized in that: The linear probability discriminant analysis process is: Probabilistic Linear Discriminant Analysis (PLDA) is a channel compensation algorithm, also known as the probabilistic LDA algorithm. PLDA is usually based on I-Vector features and provides channel compensation for them. Assume that the training data speech consists of I speakers, each of whom has J different speech segments; then, we define i Personal j The voice is ; According to factor analysis, we define The generative model is: ； in, μ represents the mean of all training data; F can be regarded as the identity space, which contains information that can be used to represent various speakers; It can be regarded as the identity of a specific speaker, or the position of the speaker in the identity space; G can be regarded as the error space, which contains information that can be used to represent different voice changes of the same speaker; It represents the position in G space; is the final residual noise term, which is used to represent the other parts that have not been explained; this term is a zero-mean Gaussian distribution with a variance of Σ; The goal of PLDA model training is to input a bunch of data , the output can best represent the parameters of the data set θ = [μ,F,G,Σ]; In PLDA, it is necessary to calculate whether two speech sounds are related by features in the speaker space. generated by The likelihood of the generated class is not affected by the difference in the class space. The scoring formula is given below: ； As in the above formula, and are the I-Vector vectors of the two voices. The assumption that the two voices come from the same space is Hs, and the assumption that they come from different spaces is Hd. Among them, p( , | hs) is the likelihood function of the two speech sounds coming from the same space; p( |hd),p |hd) are and By calculating the log-likelihood ratio, we can measure the similarity between two speech sounds. The higher the ratio, the higher the score, and the greater the possibility that the two voices belong to the same speaker; The lower the ratio, the lower the score, and the less likely it is that the two voices belong to the same speaker.

9. A method for monitoring abnormal working condition of a volute disc valve based on sound recognition according to claim 8, characterized in that: After obtaining the sound print model of the volute disc valve, the fault comprehensive diagnosis unit determines whether leakage occurs through the parameters transmitted by the bone conduction pickup, thermal infrared thermometer and vibration sensor installed on the volute disc valve hydraulic device, outputs the result and takes corresponding measures; When the real-time analysis of the device voiceprint score shows an abnormality, it is compared with the known fault model issued by the edge computing gateway. If it is a known fault, the known fault alarm is directly output; if it does not match the known fault model library, an unknown abnormality alarm is output; Once an unknown abnormal alarm is reported, the surveillance camera will carefully observe and analyze the alarm point, and at the same time, the operation personnel will be arranged to go to the site to confirm the situation and make corresponding treatment. After the treatment is completed, they will come back to report and then improve the fault model database. Next time, the same fault will be reported as a known fault; 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, the known fault will be added for judgment.

Images