Online monitoring and tracing method and system for defects in large storage tanks based on equivalent sound source

Abstract

The embodiment of the present invention discloses a method and system for online monitoring and tracing of defects of large storage tanks based on equivalent sound sources. The method includes: receiving original audio signals collected by multiple sound sensors installed on the storage tank, and preprocessing the original audio signals; matching based on the preprocessed audio features and local features in the audio feature library to determine whether the original audio signal has environmental noise; based on whether the original audio signal has environmental noise, reconstructing the original audio signal to obtain audio feature models corresponding to each original audio signal; clustering each audio feature model to obtain multiple categories of audio feature model sets; obtaining equivalent sound sources of each category of audio feature model sets, and determining the sound source type of the storage tank based on the equivalent sound sources. The embodiment of the present invention can detect corrosion anomalies in time and issue an alarm by real-time monitoring of the sound signals inside the storage tank, thereby improving the safety and monitoring efficiency of the storage tank.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of storage tank detection, and in particular to a method and system for online monitoring and tracing of defects in large storage tanks based on an equivalent sound source. Background Art

[0002] Large storage tanks are widely used in the industrial field to store various liquids and gases. However, due to long-term use and the influence of the external environment, storage tanks have certain safety hazards, such as corrosion, cracks, leakage and other problems.

[0003] Traditional tank monitoring methods mainly rely on visual inspection and regular testing, which have certain limitations in real-time and accuracy. In recent years, acoustic emission methods have been widely used in equipment safety monitoring, non-destructive monitoring and other fields. Acoustic emission is a non-destructive testing technology that detects and analyzes materials or structures by monitoring sound signals inside or on the surface of an object. This technology can be used to detect defects, cracks, fatigue and other abnormalities in materials.

[0004] However, acoustic emission technology is seriously affected by environmental noise, and it is difficult to interpret complex sound signals. Especially when there are multiple environmental noises, it is usually difficult to accurately distinguish the specific type and location of device failure in these noises, which poses a huge challenge to the safe operation of storage tanks.

[0005] Therefore, there is a need for an acoustic emission analysis method that can accurately monitor the safety status of storage tanks in real time under complex background noise and trace the source of abnormal signals. Summary of the invention

[0006] In order to overcome the shortcomings of the prior art, the purpose of the embodiments of the present invention is to provide a large storage tank defect online monitoring and tracing method and system based on equivalent sound sources, which can detect corrosion anomalies in time and issue an alarm by real-time monitoring of the sound signals inside the storage tank, thereby improving the safety and monitoring efficiency of the storage tank.

[0007] To solve the above problems, the first aspect of the embodiment of the present invention discloses a method for online monitoring and tracing of defects in large storage tanks based on an equivalent sound source, comprising:

[0008] Receiving original audio signals collected by a plurality of sound sensors installed on the storage tank, and preprocessing the original audio signals;

[0009] Matching is performed based on the preprocessed audio features and the local features in the audio feature library to determine whether the original audio signal contains environmental noise;

[0010] Based on whether the original audio signal has environmental noise, the original audio signal is reconstructed to obtain an audio feature model corresponding to each original audio signal;

[0011] Clustering each audio feature model to obtain audio feature model sets of multiple categories;

[0012] An equivalent sound source of the audio feature model set of each category is obtained, and the sound source type of the storage tank is determined based on the equivalent sound source.

[0013] As a preferred embodiment, in a first aspect of the embodiment of the present invention, preprocessing the original audio signal includes:

[0014] Performing a first preprocessing on the original audio signal, wherein the first preprocessing is one or more of filtering, denoising, and data cleaning;

[0015] The original audio signal after the first preprocessing is subjected to a second preprocessing, wherein the second preprocessing is wavelet decomposition.

[0016] As a preferred embodiment, in the first aspect of the embodiment of the present invention, matching is performed based on the preprocessed audio features and the local features in the audio feature library to determine whether the audio features contain environmental noise, including:

[0017] Based on the high-pass filter, low-pass filter, scaling function and wavelet function involved in the wavelet decomposition as audio features, matching is performed with local features preset in the audio feature library;

[0018] The original audio signal is classified into two categories according to the result of feature matching, namely, one with environmental noise and one without environmental noise.

[0019] As a preferred embodiment, in the first aspect of the embodiment of the present invention, based on whether the original audio signal has environmental noise, the original audio signal is reconstructed to obtain an audio feature model corresponding to each original audio signal, including:

[0020] Adding a preset environmental noise item to both the original audio signal with and without environmental noise, wherein the weight of the environmental noise item added to the original audio signal with environmental noise is greater than the weight of the environmental noise item added to the original audio signal without environmental noise, thereby completing reconstruction of the original audio signal;

[0021] A signal feature extraction algorithm is used to extract features from the reconstructed original audio signals to generate audio feature models corresponding to each original audio signal, wherein the audio feature models include time domain features and frequency domain features.

[0022] As a preferred embodiment, in the first aspect of the embodiment of the present invention, a signal feature extraction algorithm is used to extract features from the reconstructed original audio signal to generate an audio feature model corresponding to each original audio signal, including:

[0023] The reconstructed original audio signal is converted into a frequency domain signal through Fourier transform;

[0024] The time domain features and frequency domain features of the reconstructed original audio signal are extracted using a system identification method to obtain an audio feature model corresponding to each original audio signal.

[0025] As a preferred embodiment, in the first aspect of the embodiment of the present invention, obtaining an equivalent sound source of an audio feature model set of each category includes:

[0026] The features of the center of the class to which the audio feature model set of each category belongs are used as the equivalent sound source of the audio feature model set of each category, or the arithmetic mean of the frequency domain features of the audio feature model set of each category is used as the equivalent sound source of the audio feature model set of each category.

[0027] As a preferred embodiment, in the first aspect of the embodiment of the present invention, the features of the class center of each category of the audio feature model set are used as the equivalent sound source of each category of the audio feature model set, including:

[0028] The equivalent sound source is determined by the minimum cost function J of the clustering algorithm, which specifically includes:

[0029] The audio feature model consists of a sensitivity coefficient vector a jk Description, based on the sensitivity coefficient vector a jk Different sound source features are generated by the target optimization algorithm until the cost function J k is the minimum:

[0030]

[0031] Among them, J k is the cost function of the audio feature model set of the kth category, and the sensitivity coefficient vector between the jth sound source and the m sound sensors of the audio feature model set of the kth category is a jk =[a 1jk ,a 2jk ,…,a mjk ] T , a ijk is the sensitivity coefficient between the jth sound source and the ith sound sensor of the audio feature model set of the kth category, 1≤i≤m, 1≤j≤s, and the equivalent sound source coefficient vector between the kth equivalent sound source and the mth sound sensor is the equivalent sound source coefficient between the kth equivalent sound source and the i-th sensor, Calculated by the following formula:

[0032]

[0033] The cost function J of the audio feature model set of the kth category k The target sound source feature corresponding to the minimum is used as the equivalent sound source of the audio feature model set of the kth category. The target sound source feature is based on J k Minimum sensitivity coefficient vector a jk The corresponding sound source characteristics.

[0034] As a preferred embodiment, in a first aspect of the embodiment of the present invention, determining the sound source type of the storage tank based on the equivalent sound source includes:

[0035] Inputting the equivalent sound source into the audio feature library for matching to determine the sound source type corresponding to the equivalent sound source;

[0036] The amplitude of the target equivalent sound source is compared with the abnormal threshold of the corresponding sound source type. If the amplitude of the target equivalent sound source is greater than the abnormal threshold of the corresponding sound source type, the audio feature model set of the corresponding category of the target equivalent sound source is defined as an abnormal sound source.

[0037] As a preferred embodiment, in the first aspect of the embodiments of the present invention, based on the positions of each sound sensor corresponding to the abnormal sound source and the time difference received by each sound sensor corresponding to the abnormal sound source, the sound source position corresponding to the abnormal sound source is determined for positioning.

[0038] A first aspect of an embodiment of the present invention discloses an online monitoring and tracing system for defects in large storage tanks based on an equivalent sound source, which includes: an electronic device and a plurality of sound sensors installed on the storage tank, wherein the electronic device is used to execute the online monitoring and tracing method for defects in large storage tanks based on an equivalent sound source as described in the first aspect of the embodiment of the present invention.

[0039] A third aspect of an embodiment of the present invention discloses an electronic device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the computer program, the steps of the method for online monitoring and tracing of defects in large storage tanks based on an equivalent sound source disclosed in the first aspect of the embodiment of the present invention are implemented.

[0040] A fourth aspect of an embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the steps of the method for online monitoring and tracing of defects in large storage tanks based on an equivalent sound source disclosed in the first aspect of the embodiment of the present invention.

[0041] The fifth aspect of the embodiment of the present invention discloses a computer program product. When the computer program product runs on a computer, the computer executes the steps of the method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources disclosed in the first aspect of the embodiment of the present invention.

[0042] A sixth aspect of an embodiment of the present invention discloses an application publishing platform, which is used to publish a computer program product. When the computer program product runs on a computer, the computer executes the steps of the online monitoring and tracing method for large storage tank defects based on equivalent sound sources disclosed in the first aspect of the embodiment of the present invention.

[0043] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0044] 1. The embodiment of the present invention can accurately identify various failure modes of storage tanks such as leakage, cracking, deformation, impact, corrosion defects, etc. through audio collection sensors.

[0045] 2. The present invention can monitor multiple failure behaviors simultaneously and accurately locate multiple concurrent failure behaviors.

[0046] 3. The present invention can reduce false alarms caused by environmental factors, for example, by identifying the noise signal interference caused by mechanical vibration, thunderstorm, dust, etc. acting on storage tanks, thereby reducing the false alarm rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of a method for online monitoring and tracing of defects of large storage tanks based on equivalent sound sources provided in the first embodiment of the present invention;

[0048] Figure 2 It is a structural schematic diagram of a large storage tank defect online monitoring and tracing system based on an equivalent sound source disclosed in the second embodiment of the present invention;

[0049] Figure 3 It is a structural schematic diagram of an electronic device disclosed by the present invention. DETAILED DESCRIPTION

[0050] This specific implementation manner is merely an explanation of an embodiment of the present invention, and it is not a limitation of the embodiment of the present invention. After reading this specification, a person skilled in the art may make non-creative modifications to the embodiment as needed, but as long as it is within the scope of the claims of the embodiment of the present invention, it is protected by the patent law.

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments in the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present invention.

[0052] The term "comprise" and any variations thereof in the specification and claims of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0053] In the embodiments of the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0054] In the embodiment of the present invention, a multi-channel sound sensor installed on the storage tank collects the audio signal emitted when the tank defect is generated, and analyzes the collected audio signal through the technology based on the equivalent sound source, so as to realize the real-time monitoring and tracing of the failure modes such as leakage, corrosion, impact, cracking, defects, etc. inside the storage tank, which is described in detail in conjunction with the accompanying drawings below.

[0055] Embodiment 1

[0056] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for online monitoring and tracing defects of large storage tanks based on equivalent sound sources disclosed in an embodiment of the present invention. Figure 1 As shown in the figure, the online monitoring and tracing method of large storage tank defects based on equivalent sound source includes:

[0057] S110, receiving original audio signals collected by multiple sound sensors installed on the storage tank, and preprocessing the original audio signals.

[0058] The executor of the embodiment of the present invention is an electronic device, which can be a server, a desktop computer, a laptop computer, a mobile phone, a tablet computer, a dedicated terminal and other devices with certain processing functions. It can communicate with the sound sensor in a wired or wireless manner, and can send collected audio signals to the sound sensor, and receive audio signals collected by the sound sensor.

[0059] The sound sensor can be installed on the outside of the bottom plate of the tank or the outside of the edge plate of the bottom of the tank. Of course, it can also be installed in other positions as needed, such as the outer wall of the tank. The sound sensor can be implemented by a sensor array such as a microphone array, which can monitor at least two or more characteristic sound wave frequencies, that is, it can use m (m is greater than or equal to 2) channels of original audio signals, and the audio sensitivity coefficient of each channel is different, or at least two or more channels have different audio sensitivity coefficients.

[0060] Defect monitoring technology based on acoustic emission usually relies on an ideal monitoring environment. This is because when the environmental noise is large, it will interfere with the analysis of the sound waves generated by the defects. Since the audio signal generated by the defect is weak, it is usually necessary to amplify the audio signal before observation. However, if the environmental noise and the signal generated by the defect cannot be effectively distinguished, the defect type and location cannot be accurately identified. In addition, the audio monitoring equipment receives many sound sources, and how to effectively distinguish various sound sources is crucial for tank defect monitoring.

[0061] After receiving the original audio signal, the original audio signal may be preprocessed, and the preprocessing may include any one or more of filtering, denoising, and data cleaning, as well as wavelet decomposition. Wavelet decomposition may include signal decomposition, signal processing, and signal reconstruction.

[0062] Based on the original audio signal after wavelet decomposition, the first classification can be performed. The first classification is mainly used to determine whether the original audio signal contains environmental noise, so that in subsequent operations, when reconstructing the original audio signal, the interference of environmental noise can be removed, thereby improving the ability to distinguish the sound source signals of corrosion abnormalities, cracks and other defects in the storage tank.

[0063] S120: Perform matching based on the preprocessed audio features and the local features in the audio feature library to determine whether the original audio signal contains environmental noise.

[0064] During wavelet decomposition, for the original audio signal of each channel, corresponding high-pass filter, low-pass filter, scaling function, wavelet function and other parameters are involved. These parameters are used as audio features and matched with the preset local features in the audio feature library to classify the original audio signals of each channel. The classification results include two categories: the presence of environmental noise and the absence of environmental noise.

[0065] The matching process is to match the audio features and local features (environmental noise features) for similarity. If the similarity is greater than a preset threshold, it means that the corresponding original audio signal has environmental noise.

[0066] The local characteristics are formed based on environmental noise such as mechanical vibration, thunderstorms and strong winds in the environment where the storage tank is located. The local characteristics corresponding to various environmental noises can be determined through experiments. When comparing, they can be compared with the local characteristics corresponding to various environmental noises, or with the local characteristics formed by the combination of various environmental noises.

[0067] After the comparison is completed, it can be determined whether each original audio signal has certain environmental noise, and the original audio signal can be reconstructed according to the environmental noise situation.

[0068] S130. Reconstruct the original audio signals based on whether there is environmental noise in the original audio signals to obtain audio feature models corresponding to the original audio signals.

[0069] The method for reconstructing the original audio signal is based on whether there is ambient noise. For example, a certain ambient noise item can be added to the original audio signal with ambient noise to suppress the ambient noise in the original audio signal. For the original audio signal without ambient noise, the ambient noise item may not be added, or the weight of the added ambient noise item may be much lower than the weight of the ambient noise item added to the original audio signal with ambient noise.

[0070] For the sake of distinction, the new audio signal obtained by reconstruction is recorded as the reconstructed audio signal.

[0071] The reconstructed audio signal is subjected to feature extraction to obtain an audio feature model of each reconstructed audio signal (corresponding to the original audio signal).

[0072] Feature extraction is mainly completed through time domain-frequency domain conversion. There are many ways to convert the time domain to frequency domain. For example, it can be achieved through Fourier transform. Then the time domain features and frequency domain features of the reconstructed audio signal are extracted using the system identification method to obtain the corresponding audio feature model.

[0073] S140: Cluster each audio feature model to obtain audio feature model sets of multiple categories.

[0074] The clustering of audio feature models can be achieved based on the frequency domain features of each audio feature model. There are many clustering methods. For example, it can be achieved through algorithms such as K-means and GNM, which are not specifically limited here.

[0075] It should be noted that the number of clusters p shall not exceed the number of sound sensors (or channels) m. After clustering, a final set of audio feature models of p categories is obtained, and each category includes audio signals of one or more sound sensors.

[0076] S150: Obtain equivalent sound sources of the audio feature model set of each category, and determine the sound source type of the storage tank based on the equivalent sound sources.

[0077] In order to further reduce the impact of environmental noise and enhance the ability to analyze sensitive signals, weaker acoustic emission signals can be specifically amplified and analyzed, thereby improving the performance and robustness of acoustic emission nondestructive testing technology, which is of great significance for ensuring the safe and stable operation of large storage tanks. In a preferred embodiment of the present invention, the technology of equivalent sound source is used to perform equivalent replacement of audio signals within a certain audio feature range.

[0078] The equivalent sound source can be based on the time domain characteristics of the audio feature model, or can also be based on the frequency domain characteristics of the audio feature model.

[0079] Exemplarily, the features of the class center to which the audio feature model set of each category belongs (referred to as sound source features, which can be time domain features or frequency domain features) can be used as the equivalent sound source of the audio feature model set of each category, or the arithmetic mean of the frequency domain features of the audio feature model set of each category can be used as the equivalent sound source of the audio feature model set of each category.

[0080] Taking the characteristics of the center of the class as an example, the method of obtaining the equivalent sound source is explained and illustrated. The equivalent sound source of each category can be determined by the minimum cost function J of the clustering algorithm. Taking the equivalent sound source of the kth category as an example, the audio feature model is composed of the sensitivity coefficient vector a jk Description, based on the sensitivity coefficient vector a jk Different sound source features are generated by the target optimization algorithm until the cost function J k is the minimum:

[0081]

[0082] Among them, J k is the cost function of the audio feature model set of the kth category, 1≤k≤p, and the sensitivity coefficient vector between the jth sound source and the m sound sensors of the audio feature model set of the kth category is a jk =[a 1jk ,a 2jk ,…,a mjk ] T , a ijkis the sensitivity coefficient between the jth sound source and the ith sound sensor of the audio feature model set of the kth category, 1≤i≤m, 1≤j≤s, and the equivalent sound source coefficient vector between the kth equivalent sound source and the mth sound sensor is the equivalent sound source coefficient between the kth equivalent sound source and the i-th sensor, Calculated by the following formula:

[0083]

[0084] The cost function J of the audio feature model set of the kth category k The target sound source feature corresponding to the minimum is used as the equivalent sound source of the audio feature model set of the kth category. The target sound source feature is based on J k Minimum sensitivity coefficient vector a jk The corresponding sound source characteristics.

[0085] From the above, it can be seen that the equivalent sound source can be a time domain feature or a frequency domain feature. When the sound source type is judged later, the equivalent sound source can be converted accordingly as needed. For example, if the baseline feature of the audio feature library is a time domain feature, and the equivalent sound source is a frequency domain feature, the frequency domain feature of the equivalent sound source can be converted from the frequency domain to the time domain to obtain the time domain feature corresponding to the equivalent sound source.

[0086] The equivalent sound source is input into the audio feature library for matching to determine the sound source type corresponding to the equivalent sound source. The matching process can also be performed through similarity comparison, and the sound source type corresponding to the benchmark feature whose similarity between the equivalent sound source and each benchmark feature is the highest and greater than a preset threshold is selected as the sound source type corresponding to the equivalent sound source.

[0087] The audio database can contain baseline features generated by leaks, cracks, deformations, impacts, and corrosion defects of large storage tanks.

[0088] The audio database may contain baseline characteristics of the ambient noise generated by large storage tanks during thunderstorms, high winds, hail, and dust storms.

[0089] The audio database can contain the baseline features of equipment vibration, liquid-wall collision, and friction under normal operation of large storage tanks.

[0090] Exemplarily, the type of sound signal is determined after matching the frequency spectrum characteristics of each equivalent sound source with the sound source feature library. If it is determined to be environmental noise, the equivalent sound source can be further determined after adding the noise source model on the basis of the original audio signal, and then the judgment is made. If it is determined to be non-environmental noise, the sound source type (that is, defect type) corresponding to the equivalent sound source of each non-environmental noise is determined, such as leakage, cracking, etc. The sound source type corresponding to each equivalent sound source is not necessarily the same. Therefore, the embodiment of the present invention can monitor multiple different defects of large storage tanks at the same time.

[0091] If a corresponding sound source type is not found for an equivalent sound source, the equivalent sound source is reconstructed and judged again until the corresponding sound source type is determined.

[0092] After determining the sound source type, it is determined again whether the equivalent sound source exceeds the abnormal threshold. The abnormal threshold is the maximum safe amplitude reached by the amplitude of the audio signal generated by the abnormal sound source. Therefore, when comparing the equivalent sound source and the abnormal threshold, it is necessary to determine the amplitude of the equivalent sound source. If the amplitude of the equivalent sound source is greater than the abnormal threshold corresponding to the type, it is recorded as the abnormal sound source type, and the corresponding equivalent sound source is recorded as the abnormal sound source.

[0093] In other embodiments, the sum of the amplitudes (total amplitude) or the sum of the energies (cumulative energy) of the equivalent sound sources of each defect type may be counted and compared with a preset total abnormal threshold. When the total amplitude or cumulative energy is greater than the preset total abnormal threshold, the equivalent sound source is compared with the abnormal threshold of the corresponding defect type. Otherwise, the total amplitude or cumulative energy of the audio signal is continuously recorded.

[0094] S160: Estimate the sound source position of the abnormal sound source.

[0095] Since the equivalent sound source corresponds to a certain category of audio feature model set, it is possible to determine the specific sound sensor of the audio feature model set and the location of the sound sensor (record the location of each sound sensor during installation). Then, the start time of the original audio signal is searched in the real-time signal of the sound sensor, and the sound source location is estimated based on the time difference and the location of each sound sensor. The specific estimation process is an existing conventional technology and will not be repeated here.

[0096] Embodiment 2

[0097] See also Figure 2 , Figure 2 Schematic diagram of the structure of a large-scale storage tank defect online monitoring and tracing system based on equivalent sound source disclosed in an embodiment of the present invention. Figure 2 As shown, the large storage tank defect online monitoring and tracing system based on equivalent sound source includes a plurality of sound sensors 10 and an electronic device 20 installed on the outside of the storage tank bottom plate or the outside of the tank bottom edge plate, wherein the electronic device 20 is used for:

[0098] Receiving original audio signals collected by a plurality of sound sensors installed on the storage tank, and preprocessing the original audio signals;

[0099] Matching is performed based on the preprocessed audio features and the local features in the audio feature library to determine whether the original audio signal contains environmental noise;

[0100] Based on whether the original audio signal has environmental noise, the original audio signal is reconstructed to obtain an audio feature model corresponding to each original audio signal;

[0101] Clustering each audio feature model to obtain audio feature model sets of multiple categories;

[0102] An equivalent sound source of the audio feature model set of each category is obtained, and the sound source type of the storage tank is determined based on the equivalent sound source.

[0103] See also Figure 3 , Figure 3 A schematic diagram of the structure of an electronic device 20 that can be used to implement an embodiment of the present invention is shown. The electronic device 20 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present invention described herein or / and claimed.

[0104] like Figure 3 As shown, the electronic device includes at least one processor 210, and a memory connected to the at least one processor 210, such as a ROM (read-only memory) 220, a RAM (random access memory) 230, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 210 can perform various appropriate actions and processes according to the computer program stored in the ROM 220 or the computer program loaded from the storage unit 280 to the random access memory RAM 230. In the RAM 230, various programs and data required for the operation of the electronic device can also be stored. The processor 210, the ROM 220, and the RAM 230 are connected to each other via a bus 240. The I / O (input / output) interface 250 is also connected to the bus 240.

[0105] A number of components in the electronic device are connected to the I / O interface 250, including: an input unit 260, such as a keyboard, a mouse, etc.; an output unit 270, such as various types of displays, speakers, etc.; a storage unit 280, such as a disk, an optical disk, etc.; and a communication unit 290, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 290 allows the electronic device to exchange information / data with other devices through a computer network such as the Internet or / and various telecommunication networks.

[0106] The processor 210 may be a variety of general or / and special processing components with processing and computing capabilities. Some examples of the processor 210 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 210 executes one or more steps of the online monitoring and tracing method for defects of large storage tanks based on equivalent sound sources described in the first embodiment above.

[0107] In some embodiments, a method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 280. In some embodiments, part or all of the computer program may be loaded into or / and installed on an electronic device via ROM 220 and / or communication unit 290. When the computer program is loaded into RAM 230 and executed by processor 210, one or more steps of a method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources described in the first embodiment above may be executed. Alternatively, in other embodiments, the processor 210 may be configured to execute a method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources in any other appropriate manner (e.g., by means of firmware).

[0108] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, or / and combinations thereof. These various embodiments may include: being implemented in one or more computer programs that can be executed or / and interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0109] The computer programs for implementing the methods of the embodiments of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart or / and the block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0110] In the context of an embodiment of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0111] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0112] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0113] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0114] The above is a detailed introduction to a large storage tank defect online monitoring and tracing method and system based on an equivalent sound source disclosed in the present invention. Specific examples are used in this article to illustrate the principle and implementation mode of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation mode and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources, characterized in that: It includes: Receiving original audio signals collected by a plurality of sound sensors installed on the storage tank, and preprocessing the original audio signals; Matching is performed based on the preprocessed audio features and the local features in the audio feature library to determine whether the original audio signal has environmental noise; Based on whether the original audio signal has environmental noise, the original audio signal is reconstructed to obtain an audio feature model corresponding to each original audio signal; Clustering each audio feature model to obtain audio feature model sets of multiple categories; Obtaining an equivalent sound source of the audio feature model set of each category, and determining the sound source type of the storage tank based on the equivalent sound source; Among them, obtaining the equivalent sound source of the audio feature model set of each category includes: Taking the features of the center of the class to which each category of the audio feature model set belongs as the equivalent sound source of each category of the audio feature model set, or taking the arithmetic mean of the frequency domain features of each category of the audio feature model set as the equivalent sound source of each category of the audio feature model set; Determining the sound source type of the storage tank based on the equivalent sound source includes: Inputting the equivalent sound source into the audio feature library for matching to determine the sound source type corresponding to the equivalent sound source; Comparing the amplitude of the target equivalent sound source with the abnormal threshold of the corresponding sound source type, if the amplitude of the target equivalent sound source is greater than the abnormal threshold of the corresponding sound source type, defining the audio feature model set of the corresponding category of the target equivalent sound source as an abnormal sound source; Based on the positions of each sound sensor corresponding to the abnormal sound source and the time difference received by each sound sensor corresponding to the abnormal sound source, the sound source position corresponding to the abnormal sound source is determined for positioning.

2. The method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources according to claim 1 is characterized in that: Preprocessing the original audio signal includes: Performing a first preprocessing on the original audio signal, wherein the first preprocessing is one or more of filtering, denoising, and data cleaning; The original audio signal after the first preprocessing is subjected to a second preprocessing, wherein the second preprocessing is wavelet decomposition.

3. The method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources according to claim 2 is characterized in that: Matching is performed based on the preprocessed audio features and the local features in the audio feature library to determine whether the audio features contain environmental noise, including: Based on the high-pass filter, low-pass filter, scaling function and wavelet function involved in the wavelet decomposition as audio features, matching is performed with local features preset in the audio feature library; The original audio signal is classified into two categories according to the result of feature matching, namely, one with environmental noise and one without environmental noise.

4. The method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources according to claim 1 is characterized in that: Based on whether the original audio signal has environmental noise, the original audio signal is reconstructed to obtain an audio feature model corresponding to each original audio signal, including: Adding a preset environmental noise item to both the original audio signal with and without environmental noise, wherein the weight of the environmental noise item added to the original audio signal with environmental noise is greater than the weight of the environmental noise item added to the original audio signal without environmental noise, thereby completing reconstruction of the original audio signal; A signal feature extraction algorithm is used to extract features from the reconstructed original audio signals to generate audio feature models corresponding to each original audio signal, wherein the audio feature models include time domain features and frequency domain features.

5. The method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources according to claim 4 is characterized in that: The signal feature extraction algorithm is used to extract features from the reconstructed original audio signal to generate audio feature models corresponding to each original audio signal, including: The reconstructed original audio signal is converted into a frequency domain signal through Fourier transform; The time domain features and frequency domain features of the reconstructed original audio signal are extracted using a system identification method to obtain an audio feature model corresponding to each original audio signal.

6. The method for online monitoring and tracing of defects in large storage tanks based on equivalent sound sources according to claim 1 is characterized in that: The features of the center of the class to which each category of the audio feature model set belongs are used as the equivalent sound source of each category of the audio feature model set, including: Minimum cost function through clustering algorithm J Determine the equivalent sound source, which specifically includes: The audio feature model consists of a sensitivity coefficient vector Description, based on the sensitivity coefficient vector Different sound source features are generated by the target optimization algorithm until the cost function J k is the minimum: in, J k For the k The cost function of the audio feature model set of categories, k The first of the audio feature model sets of categories j Sound source and m The sensitivity coefficient vector between the sound sensors is , For the k The first of the audio feature model sets of categories j The sound source and i The sensitivity coefficient between the sound sensors is 1≤ i ≤ m , 1≤ j ≤ s , No. k Equivalent sound source and m The equivalent sound source coefficient vector between the sound sensors , For the k The equivalent sound source and i The equivalent sound source coefficient between sensors, Calculated by the following formula: The first k The cost function of the audio feature model set of categories J k The target sound source feature corresponding to the minimum is taken as the first k The equivalent sound source of the audio feature model set of categories, the target sound source feature is based on J k Minimum time sensitivity coefficient vector The corresponding sound source characteristics.

7. An online monitoring and tracing system for defects in large storage tanks based on equivalent sound sources, characterized in that: It includes: An electronic device and a plurality of sound sensors installed on a storage tank, wherein the electronic device is used to execute the online monitoring and tracing method for defects in a large storage tank based on an equivalent sound source as described in any one of claims 1 to 6.

Images