A valve internal leakage detection and early warning method and system

Abstract

The present application provides a valve internal leakage detection and early warning method and system, which relates to the field of valve internal leakage detection technology. The method includes: performing acoustic wave signal detection on the valve to obtain a detection acoustic wave signal set; performing filtering and reconstruction to obtain a reconstructed acoustic wave signal set; performing fluid feature detection on the valve to determine a detection fluid feature data set; loading a valve flaw detection data set; based on a valve internal leakage joint detection algorithm, performing valve internal leakage feature analysis according to the reconstructed acoustic wave signal set, the detection fluid feature data set and the valve flaw detection data set to determine the valve internal leakage generation coefficient; if the valve internal leakage generation coefficient is greater than / equal to the valve internal leakage generation threshold, generating a valve internal leakage alarm instruction; performing internal leakage tracing detection on the valve to generate a valve internal leakage tracing early warning signal. The present application can solve the technical problem of low accuracy of valve internal leakage detection in the prior art, and achieve the technical effect of improving the accuracy of valve internal leakage detection.

Description

Technical Field

[0001] The present application relates to the technical field of valve internal leakage detection, and in particular to a valve internal leakage detection and early warning method and system. Background Art

[0002] Valve internal leakage detection refers to the detection of the flow of internal fluid media after the valve is closed to determine whether the valve has leakage due to failure to completely close.

[0003] At present, most of the existing valve internal leakage detection relies on a single detection data or characteristic data for internal leakage detection, which may cause errors such as abnormal detection equipment to affect the internal leakage detection results, resulting in the limitation of a single method for internal leakage detection, resulting in reduced accuracy and timeliness of valve internal leakage detection. Therefore, a method is needed to solve the above problems.

[0004] In summary, there is a technical problem in the prior art that most of the internal leakage detection of valves relies on only a single method, which may result in the limitation of the single method and further affects the accuracy of the internal leakage detection of valves. Summary of the invention

[0005] The purpose of this application is to provide a valve internal leakage detection and early warning method and system to solve the technical problem that the existing technology mostly relies on a single method for internal leakage detection, which may lead to the limitation of a single method and further affects the accuracy of valve internal leakage detection.

[0006] In view of the above problems, the present application provides a valve internal leakage detection and early warning method and system.

[0007] In a first aspect, the present application provides a valve internal leakage detection and early warning method, which is implemented by a valve internal leakage detection and early warning system, wherein the method includes: performing acoustic wave signal detection on the valve according to a valve internal leakage detection end to obtain a detection acoustic wave signal set; performing filtering and reconstruction according to the detection acoustic wave signal set to obtain a reconstructed acoustic wave signal set; performing fluid characteristic detection on the valve according to a fluid detection end in the valve to determine a detection fluid characteristic data set; loading a valve flaw detection data set; based on a valve internal leakage joint detection algorithm, performing valve internal leakage characteristic analysis according to the reconstructed acoustic wave signal set, the detection fluid characteristic data set and the valve flaw detection data set to determine a valve internal leakage generation coefficient; if the valve internal leakage generation coefficient is greater than / equal to a valve internal leakage generation threshold, generating a valve internal leakage alarm instruction; performing internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction to generate a valve internal leakage tracing early warning signal.

[0008] In a second aspect, the present application further provides a valve internal leakage detection and early warning system, which is used to execute a valve internal leakage detection and early warning method as described in the first aspect, wherein the system includes: a detection acoustic wave signal set acquisition module, the detection acoustic wave signal set acquisition module is used to perform acoustic wave signal detection on the valve according to the valve internal leakage detection end, and obtain a detection acoustic wave signal set; a reconstruction acoustic wave signal set acquisition module, the reconstruction acoustic wave signal set acquisition module is used to perform filtering and reconstruction according to the detection acoustic wave signal set, and obtain a reconstructed acoustic wave signal set; a detection fluid characteristic data set determination module, the detection fluid characteristic data set determination module is used to perform fluid characteristic detection on the valve according to the valve internal fluid detection end, and determine the detection fluid characteristic data set; a valve flaw detection detection data set loading module, the valve detection The flaw detection data set loading module is used to load the valve flaw detection data set; the valve internal leakage generation coefficient determination module is used to perform valve internal leakage feature analysis based on the valve internal leakage joint detection algorithm, according to the reconstructed acoustic wave signal set, the detected fluid characteristic data set and the valve flaw detection data set, and determine the valve internal leakage generation coefficient; the valve internal leakage alarm instruction generation module is used to generate a valve internal leakage alarm instruction if the valve internal leakage generation coefficient is greater than / equal to the valve internal leakage generation threshold; the valve internal leakage tracing warning signal generation module is used to perform internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction, and generate a valve internal leakage tracing warning signal.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The method comprises the following steps: performing acoustic wave signal detection on the valve according to the valve internal leakage detection end to obtain a detection acoustic wave signal set; performing filtering and reconstruction according to the detection acoustic wave signal set to obtain a reconstructed acoustic wave signal set; performing fluid feature detection on the valve according to the valve internal fluid detection end to determine a detection fluid feature data set; loading a valve flaw detection data set; based on a valve internal leakage joint detection algorithm, performing valve internal leakage feature analysis according to the reconstructed acoustic wave signal set, the detection fluid feature data set and the valve flaw detection data set to determine a valve internal leakage generation coefficient; generating a valve internal leakage alarm instruction if the valve internal leakage generation coefficient is greater than / equal to a valve internal leakage generation threshold; performing internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction to generate a valve internal leakage tracing early warning signal. That is to say, by using a variety of valve internal leakage detection methods, complementing and verifying each other, the technical goal of improving the accuracy and reliability of valve internal leakage detection is finally achieved, and the technical effect of fully understanding the working status and performance changes of the valve is achieved.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0013] Figure 1 A flow chart of a valve internal leakage detection and early warning method for this application;

[0014] Figure 2 This is a structural schematic diagram of a valve internal leakage detection and early warning system for this application.

[0015] Description of reference numerals:

[0016] Detection acoustic wave signal set acquisition module 11, reconstruction acoustic wave signal set acquisition module 12, detection fluid characteristic data set determination module 13, valve flaw detection data set loading module 14, valve internal leakage generation coefficient determination module 15, valve internal leakage alarm instruction generation module 16, valve internal leakage tracing warning signal generation module 17. DETAILED DESCRIPTION

[0017] This application provides a valve internal leakage detection early warning method and system, which solves the technical problem that the existing technology mostly relies on a single method for internal leakage detection, resulting in the limitation of a single method for valve internal leakage detection, which further affects the accuracy of valve internal leakage detection. The technical goal of improving the accuracy and reliability of valve internal leakage detection is achieved, and the technical effect of fully understanding the working status and performance changes of the valve is achieved.

[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0019] Embodiment 1

[0020] Please see attached Figure 1 The present application provides a valve internal leakage detection and early warning method, wherein the method is applied to a valve internal leakage detection and early warning system, and the method specifically comprises the following steps:

[0021] Step 1: Perform acoustic wave signal detection on the valve according to the valve internal leakage detection end to obtain a detection acoustic wave signal set;

[0022] Specifically, the valve internal leakage detector is a device used to detect internal leakage of valves. The valve internal leakage detector is based on the fact that gas will generate turbulence when leaking, thereby generating sound waves of a specific frequency. The valve internal leakage detector is installed on the valve internal leakage detection end to detect the sound wave signal of the valve and obtain the detection sound wave signal set.

[0023] Step 2: Perform filtering and reconstruction according to the detected sound wave signal set to obtain a reconstructed sound wave signal set;

[0024] Specifically, by identifying the complexity of the sound source of the detected sound wave signal set, classifying it according to the complexity, performing noise reduction processing on high-complexity signals, and integrating all signals to obtain the reconstructed sound wave signal set, it helps to improve the quality and signal-to-noise ratio of the sound wave signal, and provide a better data basis for subsequent signal analysis or application.

[0025] Step 3: Detect the fluid characteristics of the valve according to the fluid detection end in the valve to determine a data set of detected fluid characteristics;

[0026] Specifically, a sensor or other equipment is provided at the fluid detection end in the valve to detect fluid characteristics, such as the flow rate and pressure of the fluid, and then determine a data set of detected fluid characteristics.

[0027] Step 4: Load the valve flaw detection data set;

[0028] Specifically, valve flaw detection is used to detect internal or external defects of valves, which may include cracks, corrosion, pores, inclusions, etc. Through valve flaw detection, safety hazards of valves can be discovered in time to ensure the normal operation and safety of valves. For example, valve flaw detection equipment is used to perform flaw detection on valves to obtain valve flaw detection data sets.

[0029] Step 5: Based on the valve internal leakage joint detection algorithm, perform valve internal leakage feature analysis according to the reconstructed acoustic wave signal set, the detected fluid feature data set and the valve flaw detection data set to determine the valve internal leakage generation coefficient;

[0030] Specifically, the reconstructed acoustic signal set, the detected fluid feature data set, and the valve flaw detection data set are integrated to form a comprehensive data set for subsequent valve internal leakage feature analysis. Key features related to valve internal leakage are extracted from the comprehensive data set. The comprehensive data set may include the frequency, amplitude, and waveform characteristics of the acoustic signal, the flow rate changes and pressure fluctuations of the fluid features, and the abnormal signals or damage detected in the valve detection. The extracted features are analyzed using the valve internal leakage joint detection algorithm to identify patterns or laws related to valve internal leakage. Based on the results of the feature analysis, the valve internal leakage generation coefficient is output.

[0031] Step 6: If the valve internal leakage generation coefficient is greater than / equal to the valve internal leakage generation threshold, a valve internal leakage alarm instruction is generated;

[0032] Specifically, the calculated valve internal leakage generation coefficient is compared with the valve internal leakage generation threshold. The valve internal leakage generation threshold is set according to actual operating experience, safety standards or relevant regulations to determine whether the valve has an internal leakage risk. If the valve internal leakage generation coefficient is greater than or equal to the valve internal leakage generation threshold, it indicates that the valve has an internal leakage risk and a valve internal leakage alarm instruction is generated.

[0033] Step seven: performing internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction, and generating a valve internal leakage tracing early warning signal.

[0034] Specifically, according to the valve internal leakage alarm instruction, the valve is subjected to internal leakage tracer detection. Tracer detection includes pressurizing the tracer gas inside the valve under test and detecting the leakage of the tracer gas externally, which can detect the leakage location and the size of the leakage. Based on the results of the tracer detection, a valve internal leakage tracer warning signal is generated, which contains specific information about the internal leakage, such as the location, degree or potential risk, for taking further measures, such as emergency maintenance, valve replacement or adjustment of operating parameters.

[0035] The valve internal leakage detection and early warning method is applied to a valve internal leakage detection and early warning system, which can achieve the technical goal of improving the accuracy and reliability of valve internal leakage detection and achieve the technical effect of fully understanding the working status and performance changes of the valve.

[0036] Furthermore, the present application also includes the following steps:

[0037] Performing sound source complexity identification according to each detection sound wave signal in the detection sound wave signal set to determine the sound source complexity of multiple signals;

[0038] Determine whether the sound source complexity of the multiple signals is less than a sound source complexity threshold, and obtain multiple sound source complexity determination results;

[0039] Classifying the detection sound wave signal set according to the multiple sound source complexity judgment results to obtain a first detection sound wave signal group and a second detection sound wave signal group;

[0040] Perform adaptive filtering and noise reduction according to the second detection sound wave signal group to obtain a second noise-reduced detection sound wave signal group;

[0041] The first detection sound wave signal group and the second noise reduction detection sound wave signal group are integrated according to the timing characteristics of sound wave signal generation to obtain the reconstructed sound wave signal set.

[0042] Specifically, the sound source complexity is identified based on each detection sound wave signal in the detection sound wave signal set. The sound source complexity may refer to the number and degree of variation of different frequency, amplitude or phase components contained in the detection sound wave signal, thereby determining the sound source complexity of each of the multiple signals.

[0043] Then, it is determined whether the complexity of the multiple signal sound sources is less than the sound source complexity threshold. For example, the sound source complexity threshold is set based on experience or a specific application scenario, and is used to distinguish whether the signal is complex enough to require further noise reduction processing. Multiple sound source complexity determination results are obtained through the determination. The multiple sound source complexity determination results include the result that the complexity of the multiple signal sound sources is less than the sound source complexity threshold and the result that the complexity of the multiple signal sound sources is greater than or equal to the sound source complexity threshold.

[0044] Next, the detection sound wave signal set is classified according to the results of multiple sound source complexity judgments, and the results of multiple signals with sound source complexity less than the sound source complexity threshold are classified into the first detection sound wave signal group, and the results of multiple signals with sound source complexity greater than or equal to the sound source complexity threshold are classified into the second detection sound wave signal group.

[0045] Next, the second detection sound wave signal group, that is, the signal group with higher sound source complexity, is subjected to adaptive filtering and noise reduction processing, and the filter parameters are dynamically adjusted according to the signal characteristics to remove noise, thereby obtaining the second noise reduction detection sound wave signal group, that is, the signal after noise reduction processing.

[0046] Then, the generation timing characteristics of the acoustic wave signal refer to the relative time relationship or time sequence between the signals. According to the generation timing characteristics of the acoustic wave signal, the first detection acoustic wave signal group and the second noise reduction detection acoustic wave signal group are integrated together, that is, the un-noised signal and the noise reduction processed signal are integrated to obtain a reconstructed acoustic wave signal set.

[0047] By identifying the complexity of the sound source, classifying it according to the complexity, performing noise reduction processing on high-complexity signals, and integrating all signals to obtain a reconstructed sound wave signal set, it helps to improve the quality and signal-to-noise ratio of the sound wave signal, and provide a better data basis for subsequent signal analysis or applications.

[0048] Furthermore, the present application also includes the following steps:

[0049] Perform interference sound wave signal feature recognition according to the second detection sound wave signal group to obtain an interference sound wave signal feature data set;

[0050] Clustering the second detection acoustic wave signal group according to the interference acoustic wave signal feature data set to obtain a plurality of acoustic wave signal areas;

[0051] activating an adaptive filter, wherein the adaptive filter performs filtering and noise reduction on the plurality of sound wave signal regions based on the interference sound wave signal feature data set to obtain a noise-reduced sound wave signal set;

[0052] Identify the influence of interference signals according to the noise reduction wave signal set to obtain a signal interference influence coefficient;

[0053] When the signal interference influence coefficient is less than the signal interference influence threshold, adding the noise reduction acoustic wave signal set to the second noise reduction detection acoustic wave signal group;

[0054] When the signal interference influence coefficient is greater than / equal to the signal interference influence threshold, the noise reduction sound wave signal set is deep filtered according to the adaptive filter to generate the second noise reduction detection sound wave signal group.

[0055] Specifically, the second detection sound wave signal group is subjected to feature recognition of interference sound wave signals, and features of possible interference components in the signal, such as frequency, amplitude, phase and other characteristics, are extracted, thereby forming an interference sound wave signal feature data set.

[0056] Then, the second detection sound wave signal group is clustered according to the interference sound wave signal feature data set, and similar second detection sound wave signals are grouped together to obtain multiple sound wave signal areas, which helps to classify signals with the same or similar interference characteristics into one category, providing convenience for subsequent processing.

[0057] Next, the adaptive filter is activated to filter and de-noise multiple sound wave signal areas based on the interfering sound wave signal feature data set to obtain a de-noised sound wave signal set. The adaptive filter can automatically adjust the filter parameters according to the characteristics of the signal to achieve the best noise reduction effect.

[0058] Next, after filtering and denoising, the interference signal impact is identified on the noise-reduced wave signal set, and the influence of the interference component in the noise-reduced signal is evaluated to obtain the signal interference influence coefficient. For example, the influence degree can be quantified by calculating a signal interference influence coefficient.

[0059] Then, the signal interference influence coefficient is compared with the signal interference influence threshold to determine whether the noise reduction effect meets the requirements. If the signal interference influence coefficient is less than the signal interference influence threshold, it means that the noise reduction effect is good, and the noise reduction sound wave signal set can be directly added to the second noise reduction detection sound wave signal group. If the signal interference influence coefficient is greater than or equal to the signal interference influence threshold, it means that interference still exists or the noise reduction effect is not ideal. At this time, it is necessary to perform deep filtering on the noise reduction sound wave signal set according to the adaptive filter to obtain better noise reduction effect and generate the final second noise reduction detection sound wave signal group.

[0060] By performing more sophisticated interference identification and processing on the signals in the second detection acoustic wave signal group, a cleaner and more accurate acoustic wave signal is obtained for subsequent signal analysis or feature extraction.

[0061] Furthermore, the present application also includes the following steps:

[0062] Predicting the probability of valve internal leakage according to the valve flaw detection data set to obtain a first valve internal leakage probability coefficient;

[0063] Predicting the probability of valve internal leakage according to the reconstructed acoustic wave signal set to obtain a second valve internal leakage probability coefficient;

[0064] Predicting the probability of valve internal leakage according to the detected fluid characteristic data set to obtain a third valve internal leakage probability coefficient;

[0065] Based on the expected valve internal leakage probability coefficient, respectively performing difference calculations on the first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient to obtain a first expected difference internal leakage probability coefficient, a second expected difference internal leakage probability coefficient and a third expected difference internal leakage probability coefficient;

[0066] Perform a proportion calculation according to the first expected difference inner leakage probability coefficient, the second expected difference inner leakage probability coefficient and the third expected difference inner leakage probability coefficient to generate a first inner leakage probability gain coefficient, a second inner leakage probability gain coefficient and a third inner leakage probability gain coefficient;

[0067] The first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient are gain optimized according to the first internal leakage probability gain coefficient, the second internal leakage probability gain coefficient and the third internal leakage probability gain coefficient to generate the valve internal leakage generation coefficient.

[0068] Specifically, a flaw detection feature internal leakage probability identification model is constructed, the valve flaw detection data set is processed, the probability of valve internal leakage is predicted, and the first valve internal leakage probability coefficient is obtained.

[0069] Then, an acoustic wave feature internal leakage probability recognition model is built, and the reconstructed acoustic wave signal set is processed to predict the probability of valve internal leakage and obtain the second valve internal leakage probability coefficient.

[0070] Next, a fluid characteristic internal leakage probability recognition model is constructed, the detected fluid characteristic data set is processed, the probability of valve internal leakage is predicted, and the third valve internal leakage probability coefficient is obtained.

[0071] Next, the expected valve internal leakage probability coefficient is infinitely close to 0. Based on the expected valve internal leakage probability coefficient, the differences between the first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient, and the third valve internal leakage probability coefficient and the expected values ​​are calculated respectively. The difference values ​​are respectively called the first expected difference internal leakage probability coefficient, the second expected difference internal leakage probability coefficient, and the third expected difference internal leakage probability coefficient.

[0072] Then, according to the magnitude of the first expected difference inner leakage probability coefficient, the second expected difference inner leakage probability coefficient and the third expected difference inner leakage probability coefficient, the proportion of each in the total difference is calculated to reflect the degree of deviation of different prediction methods relative to the expected valve inner leakage probability coefficient, and then determine the gain weight of each in the inner leakage probability prediction. The calculated proportion is the first inner leakage probability gain coefficient, the second inner leakage probability gain coefficient and the third inner leakage probability gain coefficient.

[0073] Next, the calculated first internal leakage probability gain coefficient, the second internal leakage probability gain coefficient and the third internal leakage probability gain coefficient are used to perform weighted calculation on the original first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient, and the results of different prediction methods are integrated to obtain a more accurate and reliable valve internal leakage generation coefficient as the valve internal leakage generation coefficient.

[0074] Through the differences between the results of various prediction methods and the expected values, the optimized valve internal leakage generation coefficient is finally obtained, which can more comprehensively reflect the internal leakage condition of the valve and provide strong support for the safe operation and maintenance of the valve.

[0075] Furthermore, the present application also includes the following steps:

[0076] Load the sample valve flaw detection data record and the sample flaw detection feature internal leakage probability identification coefficient record;

[0077] According to the sample valve flaw detection data record and the sample flaw detection feature internal leakage probability identification coefficient record, construct a flaw detection feature internal leakage probability identification model that meets the convergence condition;

[0078] Inputting the valve flaw detection data set into the flaw detection feature internal leakage probability identification model to obtain the flaw detection feature internal leakage probability identification coefficient;

[0079] Obtaining flaw detection equipment status data corresponding to the valve flaw detection data set;

[0080] Performing abnormality detection based on the flaw detection equipment status data to obtain abnormal characteristic data of the flaw detection equipment status;

[0081] Perform valve internal leakage probability interference identification based on the abnormal characteristic data of the flaw detection equipment state to obtain the abnormal internal leakage probability interference coefficient of the flaw detection equipment;

[0082] The detection feature internal leakage probability identification coefficient is corrected according to the abnormal internal leakage probability interference coefficient of the detection equipment to generate the first valve internal leakage probability coefficient.

[0083] Specifically, the sample valve flaw detection data record and the sample flaw detection feature internal leakage probability identification coefficient record are read from the storage medium to construct the flaw detection feature internal leakage probability identification model. For example, the sample valve flaw detection data record is the valve flaw detection data record of the historical time, and the sample flaw detection feature internal leakage probability identification coefficient record is the flaw detection feature internal leakage probability identification coefficient record of the historical time. The sample valve flaw detection data record may include the signal or image or measurement value of the valve flaw detection, and the sample flaw detection feature internal leakage probability identification coefficient record may include the internal leakage probability identification coefficient corresponding to the sample valve flaw detection data record.

[0084] Then, the flaw detection data records of the sample valves and the flaw detection feature internal leakage probability identification coefficient records are used to train the flaw detection feature internal leakage probability identification model through machine learning, which is used to predict the internal leakage probability of the valve according to the flaw detection features. During the training process, the parameters of the flaw detection feature internal leakage probability identification model are continuously adjusted to meet the convergence condition, that is, when the prediction results of the flaw detection feature internal leakage probability identification model gradually stabilize and the error reaches an acceptable range, the training of the flaw detection feature internal leakage probability identification model is completed. Among them, the convergence condition can be that the prediction results of the flaw detection feature internal leakage probability identification model no longer change significantly for at least two consecutive times.

[0085] Next, after the training of the flaw detection feature internal leakage probability recognition model is completed, the valve flaw detection data set, that is, the valve flaw detection data containing a series of unknown internal leakage probabilities, is input into the flaw detection feature internal leakage probability recognition model. The flaw detection feature internal leakage probability recognition model is used to predict each valve flaw detection data and output the corresponding flaw detection feature internal leakage probability recognition coefficient.

[0086] Next, the flaw detection equipment status data corresponding to the valve flaw detection data set is obtained. The flaw detection equipment status data may include the operating parameters, temperature, pressure, voltage, etc. of the equipment, which are used to reflect the status and performance of the flaw detection equipment during operation.

[0087] Then, by analyzing the state data of the flaw detection equipment, possible abnormal states of the equipment are identified, and abnormal characteristic data of the flaw detection equipment state is obtained, for example, by comparing with normal state data.

[0088] Next, since the abnormal characteristic data of the flaw detection equipment state may cause errors in the flaw detection data, thus affecting the prediction of the internal leakage probability, the interference degree of the abnormal data on the identification result of the valve internal leakage probability is evaluated based on the detected abnormal characteristic data of the flaw detection equipment state, and the interference coefficient of the abnormal internal leakage probability of the flaw detection equipment is obtained.

[0089] Next, based on the impact of equipment abnormality on the identification of internal leakage probability, the original internal leakage probability identification coefficient of the flaw detection feature is corrected using the internal leakage probability interference coefficient of the flaw detection equipment abnormality, thereby eliminating or reducing the impact of the equipment state abnormality on the prediction result and generating the first valve internal leakage probability coefficient.

[0090] By combining flaw detection data, equipment status data and machine learning technology, the accuracy and reliability of valve internal leakage probability identification are improved, which is of great significance for valve safety monitoring and fault prevention.

[0091] Furthermore, the present application also includes the following steps:

[0092] Build an acoustic wave feature endoleak probability recognition model;

[0093] Inputting the reconstructed acoustic wave signal set into the acoustic wave feature endoleak probability recognition model to obtain an acoustic wave feature endoleak probability recognition coefficient;

[0094] Obtaining detection end state data corresponding to the reconstructed acoustic wave signal set;

[0095] The acoustic wave characteristic internal leakage probability identification coefficient is corrected according to the detection end state data to obtain the second valve internal leakage probability coefficient.

[0096] Specifically, multiple valve acoustic wave detection data records are collected, including acoustic wave characteristics of different valve states, i.e., internal leakage and normal state, for training and verification of the acoustic wave characteristic internal leakage probability recognition model. The corresponding acoustic wave characteristic internal leakage probability recognition coefficient record is obtained. The acoustic wave characteristic internal leakage probability recognition model can be constructed using a deep learning network. The acoustic wave characteristic internal leakage probability recognition model is trained using multiple valve acoustic wave detection data records and acoustic wave characteristic internal leakage probability recognition coefficient records, and the parameters of the acoustic wave characteristic internal leakage probability recognition model are adjusted through an optimization algorithm, so that the acoustic wave characteristic internal leakage probability recognition model can accurately identify the acoustic wave characteristics of valve internal leakage. The performance of the acoustic wave characteristic internal leakage probability recognition model is evaluated by methods such as cross-validation to ensure that the acoustic wave characteristic internal leakage probability recognition model has good generalization ability.

[0097] Then, after the training of the acoustic wave feature internal leakage probability recognition model is completed, the reconstructed acoustic wave signal set, that is, the valve acoustic wave detection data containing multiple unknown internal leakage probabilities, is input into the acoustic wave feature internal leakage probability recognition model. Each reconstructed acoustic wave signal is predicted by the acoustic wave feature internal leakage probability recognition model, and the corresponding acoustic wave feature internal leakage probability recognition coefficient is output.

[0098] Next, the state data of the acoustic wave detection equipment corresponding to the reconstructed acoustic wave signal set is obtained, that is, the detection end state data corresponding to the reconstructed acoustic wave signal set. The state data of the acoustic wave detection equipment may include the operating parameters, temperature, pressure, voltage, etc. of the equipment, which are used to reflect the state and performance of the flaw detection equipment during operation.

[0099] Next, by analyzing the detection end state data corresponding to the reconstructed acoustic wave signal set, the possible abnormal state of the equipment is identified, and the abnormal characteristic data of the detection end state corresponding to the reconstructed acoustic wave signal set is obtained. For example, it is achieved by comparing with the normal state data. Since the abnormal characteristic data of the detection end state corresponding to the reconstructed acoustic wave signal set may cause errors in the acoustic wave data, thereby affecting the prediction of the internal leakage probability. Then, based on the detected abnormal characteristic data of the detection end state corresponding to the reconstructed acoustic wave signal set, the interference degree of the abnormal data on the valve internal leakage probability identification result is evaluated, and the abnormal internal leakage probability interference coefficient of the acoustic wave equipment is obtained. Based on the influence of the equipment abnormality on the internal leakage probability identification, the original acoustic wave characteristic internal leakage probability identification coefficient is corrected using the abnormal internal leakage probability interference coefficient of the acoustic wave equipment, thereby eliminating or reducing the influence of the abnormal equipment state on the prediction result, and generating a second valve internal leakage probability coefficient.

[0100] By building a valve internal leakage probability identification model based on acoustic wave characteristics and improving the accuracy of identification by correcting the detection end status data, it is helpful to achieve real-time monitoring of the valve status and fault warning, and improve the safety and reliability of valve operation.

[0101] Furthermore, the present application also includes the following steps:

[0102] Construct a fluid characteristic leakage probability identification model;

[0103] Inputting the detected fluid characteristic data set into the fluid characteristic internal leakage probability identification model to obtain the fluid characteristic internal leakage probability identification coefficient;

[0104] Obtaining detection end state data corresponding to the detection fluid characteristic data set;

[0105] The fluid characteristic internal leakage probability identification coefficient is corrected according to the detection end state data to obtain the third valve internal leakage probability coefficient.

[0106] Specifically, multiple valve fluid detection data records are collected. Corresponding fluid feature internal leakage probability recognition coefficient records are obtained. A fluid feature internal leakage probability recognition model can be constructed using a deep learning network. The fluid feature internal leakage probability recognition model is trained using multiple valve fluid detection data records and fluid feature internal leakage probability recognition coefficient records, and the parameters of the fluid feature internal leakage probability recognition model are adjusted through an optimization algorithm so that the fluid feature internal leakage probability recognition model can accurately identify the fluid features of valve internal leakage. The performance of the fluid feature internal leakage probability recognition model is evaluated through methods such as cross-validation to ensure that the fluid feature internal leakage probability recognition model has good generalization ability.

[0107] Then, after the training of the fluid characteristic leakage probability identification model is completed, the detected fluid characteristic data set, that is, the valve fluid detection data containing multiple unknown leakage probabilities, is input into the fluid characteristic leakage probability identification model. The fluid characteristic leakage probability identification model is used to predict each detected fluid characteristic data and output the corresponding fluid characteristic leakage probability identification coefficient.

[0108] Next, the detection end state data corresponding to the detected fluid characteristic data set is obtained. The fluid detection device state data may include the operating parameters, temperature, pressure, voltage, etc. of the device, which is used to reflect the state and performance of the fluid detection device during operation.

[0109] Next, by analyzing the detection end state data corresponding to the detection fluid characteristic data set, the possible abnormal state of the equipment is identified, and the detection end state abnormal characteristic data corresponding to the detection fluid characteristic data set is obtained. For example, it is achieved by comparing with the normal state data. Since the detection end state abnormal characteristic data corresponding to the detection fluid characteristic data set may cause errors in the fluid data, thereby affecting the prediction of the internal leakage probability. Then, based on the detected detection end state abnormal characteristic data corresponding to the detection fluid characteristic data set, the interference degree of the abnormal data on the valve internal leakage probability identification result is evaluated, and the abnormal internal leakage probability interference coefficient of the fluid detection equipment is obtained. Based on the influence of the equipment abnormality on the internal leakage probability identification, the abnormal internal leakage probability interference coefficient of the fluid detection equipment is used to correct the original fluid characteristic internal leakage probability identification coefficient, thereby eliminating or reducing the influence of the equipment state abnormality on the prediction result, and generating a third valve internal leakage probability coefficient.

[0110] By building a valve internal leakage probability identification model based on fluid characteristics and improving the accuracy of identification by correcting the detection end status data, it is helpful to achieve real-time monitoring of the valve status and fault warning, and improve the safety and reliability of valve operation.

[0111] In summary, the valve internal leakage detection and early warning method provided by the present application has the following technical effects:

[0112] The method comprises the following steps: performing acoustic wave signal detection on the valve according to the valve internal leakage detection end to obtain a detection acoustic wave signal set; performing filtering and reconstruction according to the detection acoustic wave signal set to obtain a reconstructed acoustic wave signal set; performing fluid feature detection on the valve according to the valve internal fluid detection end to determine a detection fluid feature data set; loading a valve flaw detection data set; based on a valve internal leakage joint detection algorithm, performing valve internal leakage feature analysis according to the reconstructed acoustic wave signal set, the detection fluid feature data set and the valve flaw detection data set to determine a valve internal leakage generation coefficient; generating a valve internal leakage alarm instruction if the valve internal leakage generation coefficient is greater than / equal to a valve internal leakage generation threshold; performing internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction to generate a valve internal leakage tracing early warning signal. That is to say, by using a variety of valve internal leakage detection methods, complementing and verifying each other, the technical goal of improving the accuracy and reliability of valve internal leakage detection is finally achieved, and the technical effect of fully understanding the working status and performance changes of the valve is achieved.

[0113] Embodiment 2

[0114] Based on the same inventive concept as the valve internal leakage detection and early warning method in the aforementioned embodiment, the present application also provides a valve internal leakage detection and early warning system, please refer to the attached Figure 2 , the system comprising:

[0115] A detection acoustic wave signal set acquisition module 11, wherein the detection acoustic wave signal set acquisition module 11 is used to perform acoustic wave signal detection on the valve according to the valve internal leakage detection end to obtain a detection acoustic wave signal set;

[0116] A reconstructed acoustic wave signal set acquisition module 12, wherein the reconstructed acoustic wave signal set acquisition module 12 is used to perform filtering and reconstruction according to the detected acoustic wave signal set to obtain a reconstructed acoustic wave signal set;

[0117] A detection fluid characteristic data set determination module 13, the detection fluid characteristic data set determination module 13 is used to detect the fluid characteristics of the valve according to the fluid detection end in the valve, and determine the detection fluid characteristic data set;

[0118] A valve flaw detection data set loading module 14, wherein the valve flaw detection data set loading module 14 is used to load a valve flaw detection data set;

[0119] A valve internal leakage generation coefficient determination module 15, the valve internal leakage generation coefficient determination module 15 is used to perform valve internal leakage feature analysis based on the reconstructed acoustic wave signal set, the detected fluid feature data set and the valve flaw detection data set based on the valve internal leakage joint detection algorithm, and determine the valve internal leakage generation coefficient;

[0120] A valve internal leakage alarm instruction generating module 16, wherein the valve internal leakage alarm instruction generating module 16 is used to generate a valve internal leakage alarm instruction if the valve internal leakage generation coefficient is greater than / equal to the valve internal leakage generation threshold value;

[0121] The valve internal leakage tracing warning signal generating module 17 is used to perform internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction and generate a valve internal leakage tracing warning signal.

[0122] Furthermore, the reconstructed acoustic wave signal set obtaining module 12 in the system is also used for:

[0123] Performing sound source complexity identification according to each detection sound wave signal in the detection sound wave signal set to determine the sound source complexity of multiple signals;

[0124] Determine whether the sound source complexity of the multiple signals is less than a sound source complexity threshold, and obtain multiple sound source complexity determination results;

[0125] Classifying the detection sound wave signal set according to the multiple sound source complexity judgment results to obtain a first detection sound wave signal group and a second detection sound wave signal group;

[0126] Perform adaptive filtering and noise reduction according to the second detection sound wave signal group to obtain a second noise-reduced detection sound wave signal group;

[0127] The first detection sound wave signal group and the second noise reduction detection sound wave signal group are integrated according to the timing characteristics of sound wave signal generation to obtain the reconstructed sound wave signal set.

[0128] Furthermore, the reconstructed acoustic wave signal set obtaining module 12 in the system is also used for:

[0129] Perform interference sound wave signal feature recognition according to the second detection sound wave signal group to obtain an interference sound wave signal feature data set;

[0130] Clustering the second detection acoustic wave signal group according to the interference acoustic wave signal feature data set to obtain a plurality of acoustic wave signal areas;

[0131] activating an adaptive filter, wherein the adaptive filter performs filtering and noise reduction on the plurality of sound wave signal regions based on the interference sound wave signal feature data set to obtain a noise-reduced sound wave signal set;

[0132] Identify the influence of interference signals according to the noise reduction wave signal set to obtain a signal interference influence coefficient;

[0133] When the signal interference influence coefficient is less than the signal interference influence threshold, adding the noise reduction acoustic wave signal set to the second noise reduction detection acoustic wave signal group;

[0134] When the signal interference influence coefficient is greater than / equal to the signal interference influence threshold, the noise reduction sound wave signal set is deep filtered according to the adaptive filter to generate the second noise reduction detection sound wave signal group.

[0135] Furthermore, the valve internal leakage generation coefficient determination module 15 in the system is also used for:

[0136] Predicting the probability of valve internal leakage according to the valve flaw detection data set to obtain a first valve internal leakage probability coefficient;

[0137] Predicting the probability of valve internal leakage according to the reconstructed acoustic wave signal set to obtain a second valve internal leakage probability coefficient;

[0138] Predicting the probability of valve internal leakage according to the detected fluid characteristic data set to obtain a third valve internal leakage probability coefficient;

[0139] Based on the expected valve internal leakage probability coefficient, respectively performing difference calculations on the first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient to obtain a first expected difference internal leakage probability coefficient, a second expected difference internal leakage probability coefficient and a third expected difference internal leakage probability coefficient;

[0140] Perform a proportion calculation according to the first expected difference inner leakage probability coefficient, the second expected difference inner leakage probability coefficient and the third expected difference inner leakage probability coefficient to generate a first inner leakage probability gain coefficient, a second inner leakage probability gain coefficient and a third inner leakage probability gain coefficient;

[0141] The first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient are gain optimized according to the first internal leakage probability gain coefficient, the second internal leakage probability gain coefficient and the third internal leakage probability gain coefficient to generate the valve internal leakage generation coefficient.

[0142] Furthermore, the valve internal leakage generation coefficient determination module 15 in the system is also used for:

[0143] Load the sample valve flaw detection data record and the sample flaw detection feature internal leakage probability identification coefficient record;

[0144] According to the sample valve flaw detection data record and the sample flaw detection feature internal leakage probability identification coefficient record, construct a flaw detection feature internal leakage probability identification model that meets the convergence condition;

[0145] Inputting the valve flaw detection data set into the flaw detection feature internal leakage probability identification model to obtain the flaw detection feature internal leakage probability identification coefficient;

[0146] Obtaining flaw detection equipment status data corresponding to the valve flaw detection data set;

[0147] Performing abnormality detection based on the flaw detection equipment status data to obtain abnormal characteristic data of the flaw detection equipment status;

[0148] Perform valve internal leakage probability interference identification based on the abnormal characteristic data of the flaw detection equipment state to obtain the abnormal internal leakage probability interference coefficient of the flaw detection equipment;

[0149] The detection feature internal leakage probability identification coefficient is corrected according to the abnormal internal leakage probability interference coefficient of the detection equipment to generate the first valve internal leakage probability coefficient.

[0150] Furthermore, the valve internal leakage generation coefficient determination module 15 in the system is also used for:

[0151] Build an acoustic wave feature endoleak probability recognition model;

[0152] Inputting the reconstructed acoustic wave signal set into the acoustic wave feature endoleak probability recognition model to obtain an acoustic wave feature endoleak probability recognition coefficient;

[0153] Obtaining detection end state data corresponding to the reconstructed acoustic wave signal set;

[0154] The acoustic wave characteristic internal leakage probability identification coefficient is corrected according to the detection end state data to obtain the second valve internal leakage probability coefficient.

[0155] Furthermore, the valve internal leakage generation coefficient determination module 15 in the system is also used for:

[0156] Construct a fluid characteristic leakage probability identification model;

[0157] Inputting the detected fluid characteristic data set into the fluid characteristic internal leakage probability identification model to obtain the fluid characteristic internal leakage probability identification coefficient;

[0158] Obtaining detection end state data corresponding to the detection fluid characteristic data set;

[0159] The fluid characteristic internal leakage probability identification coefficient is corrected according to the detection end state data to obtain the third valve internal leakage probability coefficient.

[0160] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The valve internal leakage detection and early warning method and specific examples in the aforementioned embodiment 1 are also applicable to a valve internal leakage detection and early warning system in this embodiment. Through the aforementioned detailed description of a valve internal leakage detection and early warning method, those skilled in the art can clearly understand a valve internal leakage detection and early warning system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0161] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0162] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A valve internal leakage detection and early warning method, characterized in that: The method comprises: Performing acoustic wave signal detection on the valve according to the valve internal leakage detection end to obtain a detection acoustic wave signal set; Perform filtering and reconstruction according to the detected acoustic wave signal set to obtain a reconstructed acoustic wave signal set; Perform fluid feature detection on the valve according to the fluid detection end in the valve to determine a data set of detected fluid features; Load valve flaw detection data set; Based on the valve internal leakage joint detection algorithm, the valve internal leakage feature analysis is performed according to the reconstructed acoustic wave signal set, the detected fluid feature data set and the valve flaw detection data set to determine the valve internal leakage generation coefficient; If the valve internal leakage generation coefficient is greater than / equal to the valve internal leakage generation threshold, a valve internal leakage alarm instruction is generated; Performing internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction to generate a valve internal leakage tracing early warning signal; Wherein, based on the valve internal leakage joint detection algorithm, the valve internal leakage feature analysis is performed according to the reconstructed acoustic wave signal set, the detected fluid feature data set and the valve flaw detection data set to determine the valve internal leakage generation coefficient, including: Predicting the probability of valve internal leakage according to the valve flaw detection data set to obtain a first valve internal leakage probability coefficient; Predicting the probability of valve internal leakage according to the reconstructed acoustic wave signal set to obtain a second valve internal leakage probability coefficient; Predicting the probability of valve internal leakage according to the detected fluid characteristic data set to obtain a third valve internal leakage probability coefficient; Based on the expected valve internal leakage probability coefficient, respectively performing difference calculations on the first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient to obtain a first expected difference internal leakage probability coefficient, a second expected difference internal leakage probability coefficient and a third expected difference internal leakage probability coefficient; Perform a proportion calculation according to the first expected difference inner leakage probability coefficient, the second expected difference inner leakage probability coefficient and the third expected difference inner leakage probability coefficient to generate a first inner leakage probability gain coefficient, a second inner leakage probability gain coefficient and a third inner leakage probability gain coefficient; The first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient are gain optimized according to the first internal leakage probability gain coefficient, the second internal leakage probability gain coefficient and the third internal leakage probability gain coefficient to generate the valve internal leakage generation coefficient.

2. The method according to claim 1, characterized in that Performing filtering and reconstruction according to the detected sound wave signal set to obtain a reconstructed sound wave signal set includes: Performing sound source complexity identification according to each detection sound wave signal in the detection sound wave signal set to determine the sound source complexity of multiple signals; Determine whether the sound source complexity of the multiple signals is less than a sound source complexity threshold, and obtain multiple sound source complexity determination results; Classifying the detection sound wave signal set according to the multiple sound source complexity judgment results to obtain a first detection sound wave signal group and a second detection sound wave signal group; Perform adaptive filtering and noise reduction according to the second detection sound wave signal group to obtain a second noise-reduced detection sound wave signal group; The first detection sound wave signal group and the second noise reduction detection sound wave signal group are integrated according to the timing characteristics of sound wave signal generation to obtain the reconstructed sound wave signal set.

3. The method according to claim 2, characterized in that Adaptively filtering and denoising the second detection sound wave signal group to obtain a second denoised detection sound wave signal group includes: Perform interference sound wave signal feature recognition according to the second detection sound wave signal group to obtain an interference sound wave signal feature data set; Clustering the second detection acoustic wave signal group according to the interference acoustic wave signal feature data set to obtain a plurality of acoustic wave signal areas; activating an adaptive filter, wherein the adaptive filter performs filtering and noise reduction on the plurality of sound wave signal regions based on the interference sound wave signal feature data set to obtain a noise-reduced sound wave signal set; Identify the influence of interference signals according to the noise reduction wave signal set to obtain a signal interference influence coefficient; When the signal interference influence coefficient is less than the signal interference influence threshold, adding the noise reduction acoustic wave signal set to the second noise reduction detection acoustic wave signal group; When the signal interference influence coefficient is greater than / equal to the signal interference influence threshold, the noise reduction sound wave signal set is deep filtered according to the adaptive filter to generate the second noise reduction detection sound wave signal group.

4. The method according to claim 1, characterized in that Predicting the probability of valve internal leakage according to the valve flaw detection data set to obtain a first valve internal leakage probability coefficient includes: Load the sample valve flaw detection data record and the sample flaw detection feature internal leakage probability identification coefficient record; According to the sample valve flaw detection data record and the sample flaw detection feature internal leakage probability identification coefficient record, construct a flaw detection feature internal leakage probability identification model that meets the convergence condition; Inputting the valve flaw detection data set into the flaw detection feature internal leakage probability identification model to obtain the flaw detection feature internal leakage probability identification coefficient; Obtaining flaw detection equipment status data corresponding to the valve flaw detection data set; Performing abnormality detection based on the flaw detection equipment status data to obtain abnormal characteristic data of the flaw detection equipment status; Perform valve internal leakage probability interference identification based on the abnormal characteristic data of the flaw detection equipment state to obtain the abnormal internal leakage probability interference coefficient of the flaw detection equipment; The detection feature internal leakage probability identification coefficient is corrected according to the abnormal internal leakage probability interference coefficient of the detection equipment to generate the first valve internal leakage probability coefficient.

5. The method according to claim 1, characterized in that Predicting the probability of valve internal leakage according to the reconstructed acoustic wave signal set to obtain a second valve internal leakage probability coefficient includes: Build an acoustic wave feature endoleak probability recognition model; Inputting the reconstructed acoustic wave signal set into the acoustic wave feature endoleak probability recognition model to obtain an acoustic wave feature endoleak probability recognition coefficient; Obtaining detection end state data corresponding to the reconstructed acoustic wave signal set; The acoustic wave characteristic internal leakage probability identification coefficient is corrected according to the detection end state data to obtain the second valve internal leakage probability coefficient.

6. The method according to claim 1, characterized in that Predicting the probability of valve internal leakage according to the detected fluid characteristic data set to obtain a third valve internal leakage probability coefficient includes: Construct a fluid characteristic leakage probability identification model; Inputting the detected fluid characteristic data set into the fluid characteristic internal leakage probability identification model to obtain the fluid characteristic internal leakage probability identification coefficient; Obtaining detection end state data corresponding to the detection fluid characteristic data set; The fluid characteristic internal leakage probability identification coefficient is corrected according to the detection end state data to obtain the third valve internal leakage probability coefficient.

7. A valve internal leakage detection and early warning system, characterized in that: For implementing the steps of the method according to any one of claims 1 to 6, the system comprises: A detection acoustic wave signal set acquisition module, wherein the detection acoustic wave signal set acquisition module is used to perform acoustic wave signal detection on the valve according to the valve internal leakage detection end to obtain a detection acoustic wave signal set; A reconstructed acoustic wave signal set acquisition module, wherein the reconstructed acoustic wave signal set acquisition module is used to perform filtering and reconstruction according to the detected acoustic wave signal set to obtain a reconstructed acoustic wave signal set; A detection fluid characteristic data set determination module, the detection fluid characteristic data set determination module is used to detect the fluid characteristics of the valve according to the fluid detection end in the valve, and determine the detection fluid characteristic data set; A valve flaw detection data set loading module, wherein the valve flaw detection data set loading module is used to load a valve flaw detection data set; A valve internal leakage generation coefficient determination module, the valve internal leakage generation coefficient determination module is used to perform valve internal leakage feature analysis based on the reconstructed acoustic wave signal set, the detected fluid feature data set and the valve flaw detection data set based on the valve internal leakage joint detection algorithm, and determine the valve internal leakage generation coefficient; A valve internal leakage alarm instruction generation module, wherein the valve internal leakage alarm instruction generation module is used to generate a valve internal leakage alarm instruction if the valve internal leakage generation coefficient is greater than / equal to the valve internal leakage generation threshold value; A valve internal leakage tracing warning signal generating module, the valve internal leakage tracing warning signal generating module is used to perform internal leakage tracing detection on the valve according to the valve internal leakage alarm instruction, and generate a valve internal leakage tracing warning signal; The valve internal leakage generation coefficient determination module is also used for: Predicting the probability of valve internal leakage according to the valve flaw detection data set to obtain a first valve internal leakage probability coefficient; Predicting the probability of valve internal leakage according to the reconstructed acoustic wave signal set to obtain a second valve internal leakage probability coefficient; Predicting the probability of valve internal leakage according to the detected fluid characteristic data set to obtain a third valve internal leakage probability coefficient; Based on the expected valve internal leakage probability coefficient, respectively performing difference calculations on the first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient to obtain a first expected difference internal leakage probability coefficient, a second expected difference internal leakage probability coefficient and a third expected difference internal leakage probability coefficient; Perform a proportion calculation according to the first expected difference inner leakage probability coefficient, the second expected difference inner leakage probability coefficient and the third expected difference inner leakage probability coefficient to generate a first inner leakage probability gain coefficient, a second inner leakage probability gain coefficient and a third inner leakage probability gain coefficient; The first valve internal leakage probability coefficient, the second valve internal leakage probability coefficient and the third valve internal leakage probability coefficient are gain optimized according to the first internal leakage probability gain coefficient, the second internal leakage probability gain coefficient and the third internal leakage probability gain coefficient to generate the valve internal leakage generation coefficient.

Images