Voiceprint diagnosis method and device for boiler water-cooled wall coking

Abstract

The invention relates to the technical field of boiler coking diagnosis, in particular to a boiler water wall coking voiceprint diagnosis method and device, and the method comprises the steps: collecting a voiceprint signal in a boiler; generating voiceprint data in the boiler according to the voiceprint signal in the boiler; at least one voiceprint signal feature is extracted according to the voiceprint data in the boiler, and the boiler water wall coking position and degree of the boiler are diagnosed according to the at least one voiceprint signal feature. Therefore, the problem that the coking condition of the boiler water-cooled wall cannot be comprehensively tracked in real time due to the fact that an existing boiler water-cooled wall coking detection technology is limited by factors such as equipment service life and safety in the related technology is solved.

Description

Technical Field

[0001] This application relates to the field of boiler coking diagnosis technology, and in particular to an acoustic signature diagnosis method and apparatus for coking on boiler water-cooled walls. Background Technology

[0002] Coal-fired power, as a fundamental and system-regulating power source, plays a vital role in the power energy structure. During the operation of coal-fired power units, severe coking on the boiler water-cooled walls seriously restricts the safe, stable, and economical operation of the unit. Coking on the water-cooled walls affects the heat exchange between the high-temperature flue gas and the working fluid inside the tubes, leading to reduced boiler efficiency and potentially causing problems such as localized overheating of the heating surfaces.

[0003] In related technologies, the main methods for preventing coking on boiler water-cooled walls are still to improve the combustion conditions in the furnace through operational adjustments and fuel blending, and to remove coking from the water-cooled walls in a timely manner when necessary, in conjunction with soot blowing equipment. In order to clearly understand the coking situation on the boiler water-cooled walls and thus efficiently adjust the boiler operation or remove coking, it is necessary to detect the location and extent of coking on the boiler water-cooled walls in a timely manner. Existing boiler water-cooled wall coking detection often relies on direct observation through observation holes, while some invasive detection devices can enter the furnace through observation holes to collect data, such as laser ranging and high-temperature camera imaging.

[0004] However, the aforementioned detection methods are limited by factors such as equipment lifespan and safety, and cannot comprehensively and in real-time track the coking situation of boiler water-cooled walls. Acoustic thermometry has received widespread attention in recent years, and some researchers have used this technology to measure the temperature distribution inside the furnace to predict the coking situation of water-cooled walls. However, the coking problem of water-cooled walls is not solely related to temperature, and the accuracy of coking detection results obtained through this indirect measurement method remains to be verified. Summary of the Invention

[0005] This application provides an acoustic signature diagnostic method and apparatus for coking on boiler water-cooled walls, in order to solve the problems in related technologies, such as the inability to comprehensively and in real time track the coking status of boiler water-cooled walls due to limitations in equipment lifespan and safety factors in existing boiler water-cooled wall coking detection technologies.

[0006] The first aspect of this application provides an acoustic signature diagnostic method for coking on the boiler water-cooled wall, comprising the following steps: acquiring acoustic signature signals inside the boiler; generating acoustic signature data inside the boiler based on the acoustic signature signals inside the boiler; extracting at least one acoustic signature signal feature based on the acoustic signature data inside the boiler; and diagnosing the location and extent of coking on the boiler water-cooled wall based on the at least one acoustic signature signal feature.

[0007] Through the aforementioned technical means, the embodiments of this application can accurately diagnose the location and degree of coking on the water-cooled wall by collecting acoustic signals inside the boiler, generating structured acoustic data, and extracting feature parameters. This eliminates the need for traditional methods that rely on manual inspections or lagging parameter monitoring. By using the correlation analysis between acoustic features and coking status, it can capture acoustic anomalies in the early stages of coking, significantly improving the timeliness and accuracy of coking fault diagnosis. This effectively avoids safety hazards such as overheating of the heating surface and tube rupture caused by missed or misdiagnosed coking, providing real-time and reliable technical support for the safe and efficient operation of the boiler.

[0008] Optionally, in one embodiment of this application, the step of extracting at least one acoustic signature feature based on the acoustic signature data inside the boiler includes: inputting the acoustic signature data inside the boiler into a pre-trained acoustic signature diagnostic model to output the at least one acoustic signature feature generated by the real-time coking situation of the water-cooled wall inside the boiler.

[0009] Through the above-mentioned technical means, the embodiments of this application can extract the acoustic signal features corresponding to coking of the water-cooled wall by inputting the acoustic data inside the boiler into a pre-trained acoustic diagnostic model. This replaces the tedious process of manually extracting features in the traditional way. With the help of the model's autonomous learning and recognition capabilities, the characteristic acoustic information generated by coking can be accurately captured, which greatly improves the efficiency and accuracy of feature extraction. At the same time, it can realize the real-time perception of the coking situation of the water-cooled wall, and provide efficient and reliable core data support for subsequent diagnosis of coking location and degree and matching of measures.

[0010] Optionally, in one embodiment of this application, the method further includes: obtaining the current operating status of the boiler; matching diagnostic measures for the boiler based on the current operating status and the location and extent of coking on the boiler water-cooled wall, wherein the diagnostic measures include at least one of soot blowing, boiler shutdown, and inspection.

[0011] Through the aforementioned technical means, the embodiments of this application can accurately match targeted diagnostic measures such as soot blowing, shutdown, and inspection by combining the current operating status of the boiler with the specific location and degree of coking on the water-cooled wall. This breaks away from the traditional "one-size-fits-all" boiler maintenance model, avoiding energy waste and damage to the heating surface caused by blind soot blowing. It can also trigger shutdown and inspection actions in a timely manner when the degree of coking is severe, effectively preventing safety accidents such as overheating of the heating surface and tube rupture caused by coking. This significantly improves the safety and economy of boiler operation and achieves refined and intelligent management and control of coking faults.

[0012] Optionally, in one embodiment of this application, before extracting at least one of the acoustic signature signal features based on the acoustic signature data inside the boiler, the method further includes: preprocessing the acoustic signature data inside the boiler to extract acoustic signature data inside the boiler that meets preset usage conditions based on the time domain, frequency domain, and time-frequency domain.

[0013] Through the above-mentioned technical means, the embodiments of this application can provide more accurate and reliable input for subsequent feature extraction and machine learning model diagnosis by preprocessing the acoustic fingerprint data inside the boiler, thereby improving the accuracy of coking state diagnosis.

[0014] Optionally, in one embodiment of this application, diagnosing the location and extent of coking on the boiler water-cooled wall based on the at least one acoustic signature feature includes: obtaining the actual type and / or operating parameters of the boiler; and determining the location and extent of coking on the boiler water-cooled wall by combining the actual type and / or operating parameters and the at least one acoustic signature feature.

[0015] Through the aforementioned technical means, the embodiments of this application can diagnose the location and extent of coking on water-cooled walls by combining the actual boiler type and operating parameters with acoustic signature characteristics. This overcomes the limitations of traditional monitoring methods that rely on lagging parameters such as temperature and pressure, enabling early and accurate location and quantitative assessment of coking faults. It can capture key acoustic signature signals such as flow-induced noise when high-temperature flue gas flows around the heated surface of a single-sided hot water-cooled wall, quickly matching coking acoustic signature spectra under different boiler types and operating conditions. This significantly improves the pertinence and accuracy of the diagnosis, effectively avoiding problems such as corrosion of the heated surface and decreased thermal efficiency caused by coking expansion, and providing real-time and reliable technical support for the safe and efficient operation of the boiler.

[0016] Optionally, in one embodiment of this application, the method further includes: predicting the coking trend of the boiler based on the location and degree of coking on the boiler water-cooled wall; and generating a warning signal for the boiler based on the coking trend.

[0017] Through the aforementioned technical means, the embodiments of this application can predict the coking development trend based on the specific location and degree of coking on the boiler water-cooled wall, and generate targeted reminder signals. This can overcome the limitations of traditional post-event handling and achieve proactive early warning of coking faults. Furthermore, it can predict the risk of coking spread in advance and promptly push out graded reminders such as soot blowing, inspection, or boiler shutdown, avoiding serious accidents such as heat deviation of heating surfaces and tube rupture caused by aggravated coking, reducing the number of unplanned shutdowns, significantly improving the safety and continuity of boiler operation, and providing a decision-making basis for intelligent operation and maintenance throughout the boiler's entire life cycle.

[0018] A second aspect of this application provides an acoustic signature diagnostic device for coking on the boiler water-cooled wall, comprising: a acquisition module for acquiring acoustic signature signals inside the boiler; a generation module for generating acoustic signature data inside the boiler based on the acoustic signature signals inside the boiler; and a diagnostic module for extracting at least one acoustic signature signal feature based on the acoustic signature data inside the boiler, and diagnosing the location and extent of coking on the boiler water-cooled wall based on the at least one acoustic signature signal feature.

[0019] Through the aforementioned technical means, the embodiments of this application can accurately diagnose the location and degree of coking on the water-cooled wall by collecting acoustic signals inside the boiler, generating structured acoustic data, and extracting feature parameters. This eliminates the need for traditional methods that rely on manual inspections or lagging parameter monitoring. By using the correlation analysis between acoustic features and coking status, it can capture acoustic anomalies in the early stages of coking, significantly improving the timeliness and accuracy of coking fault diagnosis. This effectively avoids safety hazards such as overheating of the heating surface and tube rupture caused by missed or misdiagnosed coking, providing real-time and reliable technical support for the safe and efficient operation of the boiler.

[0020] Optionally, in one embodiment of this application, the diagnostic module includes: an output unit, used to input the acoustic fingerprint data inside the boiler into a pre-trained acoustic fingerprint diagnostic model, so as to output at least one acoustic fingerprint signal feature generated by the real-time coking situation of the water-cooled wall inside the boiler.

[0021] Through the above-mentioned technical means, the embodiments of this application can extract the acoustic signal features corresponding to coking of the water-cooled wall by inputting the acoustic data inside the boiler into a pre-trained acoustic diagnostic model. This replaces the tedious process of manually extracting features in the traditional way. With the help of the model's autonomous learning and recognition capabilities, the characteristic acoustic information generated by coking can be accurately captured, which greatly improves the efficiency and accuracy of feature extraction. At the same time, it can realize the real-time perception of the coking situation of the water-cooled wall, and provide efficient and reliable core data support for subsequent diagnosis of coking location and degree and matching of measures.

[0022] Optionally, in one embodiment of this application, it further includes: an acquisition module for acquiring the current operating status of the boiler; and a matching module for matching diagnostic measures of the boiler based on the current operating status and the location and degree of coking on the boiler water-cooled wall, wherein the diagnostic measures include at least one of soot blowing, boiler shutdown, and inspection.

[0023] Through the aforementioned technical means, the embodiments of this application can accurately match targeted diagnostic measures such as soot blowing, shutdown, and inspection by combining the current operating status of the boiler with the specific location and degree of coking on the water-cooled wall. This breaks away from the traditional "one-size-fits-all" boiler maintenance model, avoiding energy waste and damage to the heating surface caused by blind soot blowing. It can also trigger shutdown and inspection actions in a timely manner when the degree of coking is severe, effectively preventing safety accidents such as overheating of the heating surface and tube rupture caused by coking. This significantly improves the safety and economy of boiler operation and achieves refined and intelligent management and control of coking faults.

[0024] Optionally, in one embodiment of this application, it further includes: a preprocessing module, used to preprocess the boiler internal acoustic data before extracting at least one of the acoustic signal features based on the boiler internal acoustic data, so as to extract boiler internal acoustic data that meets preset usage conditions based on the time domain, frequency domain, and time-frequency domain.

[0025] Through the above-mentioned technical means, the embodiments of this application can provide more accurate and reliable input for subsequent feature extraction and machine learning model diagnosis by preprocessing the acoustic fingerprint data inside the boiler, thereby improving the accuracy of coking state diagnosis.

[0026] Optionally, in one embodiment of this application, the diagnostic module includes: an acquisition unit for acquiring the actual type and / or operating parameters of the boiler; and a determination unit for determining the location and extent of coking on the boiler water-cooled wall by combining the actual type and / or operating parameters and the at least one acoustic signature feature.

[0027] Through the aforementioned technical means, the embodiments of this application can diagnose the location and extent of coking on water-cooled walls by combining the actual boiler type and operating parameters with acoustic signature characteristics. This overcomes the limitations of traditional monitoring methods that rely on lagging parameters such as temperature and pressure, enabling early and accurate location and quantitative assessment of coking faults. It can capture key acoustic signature signals such as flow-induced noise when high-temperature flue gas flows around the heated surface of a single-sided hot water-cooled wall, quickly matching coking acoustic signature spectra under different boiler types and operating conditions. This significantly improves the pertinence and accuracy of the diagnosis, effectively avoiding problems such as corrosion of the heated surface and decreased thermal efficiency caused by coking expansion, and providing real-time and reliable technical support for the safe and efficient operation of the boiler.

[0028] Optionally, in one embodiment of this application, it is further configured to: predict the coking trend of the boiler based on the location and degree of coking on the boiler water-cooled wall; and generate a warning signal for the boiler based on the coking trend.

[0029] Through the aforementioned technical means, the embodiments of this application can predict the coking development trend based on the specific location and degree of coking on the boiler water-cooled wall, and generate targeted reminder signals. This can overcome the limitations of traditional post-event handling and achieve proactive early warning of coking faults. Furthermore, it can predict the risk of coking spread in advance and promptly push out graded reminders such as soot blowing, inspection, or boiler shutdown, avoiding serious accidents such as heat deviation of heating surfaces and tube rupture caused by aggravated coking, reducing the number of unplanned shutdowns, significantly improving the safety and continuity of boiler operation, and providing a decision-making basis for intelligent operation and maintenance throughout the boiler's entire life cycle.

[0030] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the acoustic signature diagnosis method for coking of boiler water-cooled walls as described in the above embodiments.

[0031] A fourth aspect of this application provides a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described acoustic signature diagnostic method for coking on boiler water-cooled walls.

[0032] A fifth aspect of this application provides a computer program product that stores a computer program that, when executed by a processor, implements the above-described acoustic signature diagnosis method for coking on boiler water-cooled walls.

[0033] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0034] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of an acoustic signature diagnostic method for coking on a boiler water-cooled wall according to an embodiment of this application; Figure 2 This is a schematic diagram of the sensor array arrangement in a boiler water-cooled wall coking acoustic sound detection method according to a specific embodiment of this application; Figure 3 This is a schematic diagram of the model pre-training process in a boiler water-cooled wall coking acoustic sound pattern diagnosis method according to a specific embodiment of this application; Figure 4 This is a schematic diagram illustrating the diagnosis of coking status in a boiler water-cooled wall coking acoustic sound diagnostic method according to a specific embodiment of this application; Figure 5 This is a schematic flowchart of an acoustic signature diagnostic method for coking on a boiler water-cooled wall according to a specific embodiment of this application. Figure 6 This is a schematic diagram of the acoustic signature diagnostic device for coking of boiler water-cooled walls according to an embodiment of this application; Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application.

[0035] Appendix Figure 3 and 4 In the diagram, 2 is the waveguide rod, 3 is the sensor, 4 is the connecting line, 5 is the data acquisition and analysis equipment, 6 is the computer, 11 is the non-heated wall of the boiler's single-sided hot water cold wall, and 12 is the heated wall of the boiler's single-sided hot water cold wall. Detailed Implementation

[0036] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0037] Existing technologies for detecting coking on boiler water-cooled walls primarily rely on invasive methods. For example, patent CN111380460B, a method for determining the location and thickness of coking on boiler water-cooled walls, employs laser ranging technology. A laser rangefinder is inserted through an observation hole to obtain the actual distance between the observation hole and the coking surface, thereby measuring the location and thickness of the coking within the furnace. Similarly, patent CN112763490B, a probe and method for detecting coking on high-temperature water-cooled walls, utilizes stereoscopic vision technology. A laser probe is inserted at designated measurement points. After obtaining the reflected and scattered light from the coking surface, a three-dimensional reconstruction method is used to obtain the three-dimensional morphology of the water-cooled wall surface, thus determining the coking condition of the high-temperature water-cooled walls in the furnace. Furthermore, utility model patent CN219243628U, a device for detecting coking conditions on the water-cooled walls of a power plant flame boiler, proposes a high-temperature camera device with a protective gas system. Inserted through an observation hole, it allows direct observation of the coking condition of the water-cooled walls within the furnace.

[0038] However, existing boiler water-cooled wall coking detection technologies have the following main drawbacks: (1) Existing detection technologies have a short lifespan, and frequent equipment replacements result in high costs. Because existing detection technologies primarily rely on invasive methods, both high-temperature cameras and laser rangefinders require entry into the furnace through observation holes or additional measuring points to collect valid data. Invasive detection equipment needs to be frequently inserted into the furnace, directly exposed to the radiation heat transfer and flow erosion of the high-temperature flue gas, which easily leads to equipment damage and short service life. Frequent equipment replacement, on the other hand, results in high detection costs.

[0039] (2) Existing detection technologies have limitations in the layout of measurement points. Existing invasive detection equipment relies heavily on the limited number of pre-determined observation holes in the boiler. The location and number of these observation holes vary depending on the boiler type, potentially leading to insufficient measurement points and an inability to comprehensively detect coking information on the water-cooled walls inside the furnace.

[0040] (3) Existing detection technologies lack continuity in data acquisition. Existing detection technologies are limited by the equipment's tolerance to the high temperatures inside the furnace. To prevent equipment damage, each measurement can only involve brief, instantaneous sampling, making long-term, continuous signal acquisition impossible. This intermittent working mode results in significant data gaps in the timeline, failing to provide real-time guidance for predicting early coking formation and subsequent development.

[0041] Therefore, this application proposes a new detection method.

[0042] The following describes, with reference to the accompanying drawings, an acoustic signature diagnostic method and apparatus for boiler water-cooled wall coking according to embodiments of this application. Addressing the problem mentioned in the background section of the related technologies that, due to limitations in equipment lifespan and safety, existing boiler water-cooled wall coking detection technologies cannot comprehensively and in real-time track the coking status of boiler water-cooled walls, this application provides an acoustic signature diagnostic method for boiler water-cooled wall coking. In this method, acoustic signature detection equipment arranged outside the boiler can continuously collect acoustic signature data and analyze its time-frequency domain characteristics. Through data analysis using a diagnostic model, predictions can be made in the early stages of coking, before it seriously affects operation, guiding timely soot blowing and other coking removal operations, thus minimizing the impact of water-cooled wall coking on boiler operation. This solves the problem in the related technologies where existing boiler water-cooled wall coking detection technologies, limited by equipment lifespan and safety, cannot comprehensively and in real-time track the coking status of boiler water-cooled walls.

[0043] Specifically, Figure 1 This is a schematic flowchart illustrating an acoustic signature diagnostic method for coking on a boiler water-cooled wall, provided in an embodiment of this application.

[0044] like Figure 1 As shown, the acoustic signature diagnostic method for coking on the boiler water-cooled wall includes the following steps: In step S101, acoustic fingerprint signals inside the boiler are collected.

[0045] Boiler acoustic fingerprints can be understood as sound signals generated by various physical and chemical processes inside the boiler during operation. These signals are collected, converted, and analyzed to form characteristic acoustic data, which are essentially the "acoustic fingerprints" of the boiler's internal state.

[0046] The generation of acoustic signals inside the boiler is directly related to the core internal operating components, and the main sources may include: fluid dynamic noise, combustion noise, mechanical vibration, etc.

[0047] In actual implementation, the waveguide rod of this application embodiment can be installed on the non-heated wall surface of the single-sided hot water-cooled wall of the boiler. The waveguide rod has high temperature resistance and the maximum working temperature can reach more than 500 ℃. It can isolate the direct influence of the temperature of the non-heated wall surface of the single-sided hot water-cooled wall on the sensor. The bottom is magnetic and can be adsorbed onto the surface of the non-heated wall surface of the single-sided hot water-cooled wall.

[0048] Furthermore, the sensor can be installed on the other side of the waveguide rod away from the non-heated wall of the single-sided hot water cold wall to achieve non-invasive detection of acoustic signals inside the boiler. The measurable frequency range of the sensor can be set to 0~20 kHz, and the measurable sound pressure range can be set to 0~170 dB. The measurable frequency and measurable sound pressure range of the sensor can cover the acquisition range of various acoustic signals inside the boiler, meet the acoustic signal acquisition requirements inside the boiler, and can further import the acoustic signals received by the sensor into the data acquisition and analysis equipment through the connecting cable to obtain acoustic data inside the boiler in real time and continuously.

[0049] Specifically, both sides of the waveguide rod are magnetic. One side can be adsorbed onto the non-heated surface of the single-sided hot water-cooled wall, while the other side is connected to the sensor. The acoustic signature signal is transmitted through a connecting rod in the middle, which can effectively reduce the damage to the sensor caused by the temperature of the non-heated surface of the single-sided hot water-cooled wall and extend the sensor's lifespan. The sensor is connected to the waveguide rod and receives the acoustic signature signal data transmitted by the waveguide rod, realizing non-invasive detection of acoustic signature signal data in the furnace. The sensor is small in size and can be arranged in an array on the non-heated surface of the single-sided hot water-cooled wall. It can be densely arranged in high-coking-risk areas such as the high-temperature combustion zone, and relatively sparsely arranged in other low-coking-risk areas. The limited number of sensors can be used to spatially diagnose and locate coking in a large area of ​​water-cooled wall. According to the detection needs of different types of boilers, different numbers and distributions of sensors can be arranged to achieve a comprehensive diagnosis of the location and severity of coking in the boiler water-cooled wall.

[0050] For example, taking a tangentially rounded boiler as an example, one arrangement of the sensor array is as follows: Figure 2 As shown, based on existing operational experience, dense sensor deployment is used in high-temperature, high-coking-risk areas such as the four corner burners and the main combustion zone of the boiler, enabling accurate and timely coking diagnosis. In low-coking-risk areas such as the burnout zone, sensor deployment is relatively sparse, primarily used for judging the overall trend of coking within the furnace. This deployment strategy allows for spatial diagnosis and localization of coking on large-area water-cooled walls using a limited number of sensors.

[0051] The data acquisition and analysis equipment can use a high-speed data acquisition card with a sampling rate of >10kHz to achieve real-time, full-range capture of voiceprint signal data.

[0052] In step S102, boiler internal acoustic signature data is generated based on the boiler internal acoustic signature signal.

[0053] Boiler internal acoustic signature data can be understood as a storable and analyzable digital dataset formed by the collection, quantization, and preprocessing of internal acoustic signature signals during boiler operation; it is the "digital carrier" of acoustic signature signals. It is not a simple audio recording file, but rather structured and characteristic data extracted after professional processing of the original acoustic signature signals.

[0054] In actual implementation, the process of generating acoustic data from boiler acoustic signals can be achieved by first deploying high-temperature resistant acoustic sensors in areas with high boiler failure rates and concentrated sound sources to collect raw acoustic signals generated by furnace combustion, fluid flow, and mechanical vibration. Then, the data acquisition unit performs analog-to-digital conversion to obtain time-domain waveform digital data. After filtering, segmentation, and outlier removal to eliminate interference, algorithms such as Fourier transform and wavelet transform are used to extract time-domain (peak value, kurtosis, etc.), frequency-domain (dominant frequency, frequency band energy ratio, etc.) and time-frequency domain feature parameters. Finally, combined with the corresponding boiler load, pressure, temperature, and other operating parameters and status labels, the data is integrated and standardized into structured acoustic data that can be used for operating condition monitoring and fault diagnosis modeling.

[0055] In step S103, at least one acoustic signature feature is extracted based on the acoustic signature data inside the boiler, and the location and extent of coking on the boiler water-cooled wall are diagnosed based on the at least one acoustic signature feature.

[0056] Optionally, in one embodiment of this application, before extracting at least one acoustic signature feature based on the acoustic signature data inside the boiler, the method further includes: preprocessing the acoustic signature data inside the boiler to extract acoustic signature data inside the boiler that meets preset usage conditions based on the time domain, frequency domain, and time-frequency domain.

[0057] In actual implementation, the embodiments of this application can preprocess the original voiceprint data, filter out noise interference unrelated to coagulation, extract feature parameters from the time domain and frequency domain, construct a voiceprint diagnostic model, and perform pre-training.

[0058] Feature parameter extraction can include frequency domain features such as Mel frequency cepstral coefficients, spectral centroid, bandwidth, and subband energy ratio; and time domain features such as the mean and variance of amplitude. The extracted feature parameters can be combined into a feature vector.

[0059] The raw acoustic signature data collected contains background noise, which needs to be preprocessed to remove ultra-low frequency (mechanical vibration) and high frequency (electronic noise) interferences unrelated to coking, as well as combustion background noise, in order to obtain the acoustic signature feature signal caused by coking.

[0060] Preprocessing includes, but is not limited to, advanced algorithms such as wavelet transform and adaptive filtering. The original acoustic signature data covers a large amount of data under normal operating conditions without coking, as well as coking conditions under different loads and air distribution parameters. For the preprocessed acoustic signature data, feature parameters are extracted, such as spectral centroid and energy proportion in specific frequency bands. The extracted feature parameters are divided into training and testing samples according to a certain ratio. The training samples are used as input to the initial acoustic signature diagnostic model, pre-training parameters are set, and the model is pre-trained. The test samples are used to verify the predictive accuracy of the pre-trained diagnostic model.

[0061] Specifically, the preprocessing of the original voiceprint data can employ nonlinear wavelet threshold denoising, which effectively suppresses background noise such as combustion and fan operation. The non-stationary and abrupt voiceprint signals related to coking are preserved during the thresholding process because their corresponding wavelet coefficients have large amplitudes. They become clearer in the time-domain waveform after denoising, and the signal-to-noise ratio is significantly improved.

[0062] Through the above-mentioned technical means, the embodiments of this application can provide more accurate and reliable input for subsequent feature extraction and machine learning model diagnosis by preprocessing the acoustic fingerprint data inside the boiler, thereby improving the accuracy of coking state diagnosis.

[0063] Optionally, in one embodiment of this application, extracting at least one acoustic signature feature based on the acoustic signature data inside the boiler includes: inputting the acoustic signature data inside the boiler into a pre-trained acoustic signature diagnostic model to output at least one acoustic signature feature generated by the real-time coking situation of the water-cooled wall inside the boiler.

[0064] In practical implementation, the pre-training of the acoustic signature diagnostic model can employ an autoencoder algorithm to perform unsupervised training on acoustic signature data from boiler water-cooled walls under coking conditions. The core idea is to learn the acoustic signature data under normal boiler operation without coking. Specifically, the process can be set as follows: Under coking-free boiler conditions, acoustic signature data from water-cooled walls under various operating conditions, such as different loads and air distribution parameters, are collected. The different acoustic signature data generated by the non-uniform structure of the water-cooled wall itself (tube walls, fins) are treated as a whole, and their spectral feature parameters are extracted to form a pre-training database for the acoustic signature diagnostic model. This database contains the background acoustic features generated by the non-uniform structure of the water-cooled wall itself. The pre-training process is as follows: Figure 3 As shown.

[0065] Furthermore, in this embodiment, a pre-trained voiceprint diagnostic model can be used to process real-time voiceprint data, extract the voiceprint signal features generated by the real-time coking situation of the water-cooled wall inside the boiler, and diagnose the location and degree of coking. Through the above-mentioned technical means, the embodiments of this application can extract the acoustic signal features corresponding to coking of the water-cooled wall by inputting the acoustic data inside the boiler into a pre-trained acoustic diagnostic model. This replaces the tedious process of manually extracting features in the traditional way. With the help of the model's autonomous learning and recognition capabilities, the characteristic acoustic information generated by coking can be accurately captured, which greatly improves the efficiency and accuracy of feature extraction. At the same time, it can realize the real-time perception of the coking situation of the water-cooled wall, and provide efficient and reliable core data support for subsequent diagnosis of coking location and degree and matching of measures.

[0066] Optionally, in one embodiment of this application, diagnosing the location and extent of coking on the boiler water-cooled wall based on at least one acoustic signature feature includes: obtaining the actual type and / or operating parameters of the boiler; and determining the location and extent of coking on the boiler water-cooled wall by combining the actual type and / or operating parameters with at least one acoustic signature feature.

[0067] After training, when inputting acoustic fingerprint data from the early stages of coking or during the coking process, the model will be unable to reconstruct effectively due to the difference in patterns compared to the pre-training data, resulting in reconstruction error. This error can be used as an anomaly score for coking, thereby determining the degree of coking. This method identifies data that deviates abnormally from the background acoustic features as coking signals, thus ensuring applicability and accuracy on non-uniform structures. A large amount of acoustic fingerprint data from boilers in normal operation without coking is divided into training and validation sets. After training, the validation set is input into the trained diagnostic model, the reconstruction error of each sample is calculated, the distribution of the reconstruction error is statistically analyzed, and an appropriate threshold is selected.

[0068] The process of using a pre-trained voiceprint diagnostic model for coma diagnosis is as follows: Figure 4 As shown, after continuous operation under different loads, coking will occur on the heated wall surface of the single-sided hot water cold wall of the boiler, producing coke blocks of different shapes and locations. During boiler operation, acoustic fingerprint signals are collected in real time and preprocessed and feature extracted to obtain real-time feature vectors. The real-time feature vectors are input into the pre-trained diagnostic model to obtain reconstruction output and calculate reconstruction error. If the real-time reconstruction error is less than the threshold, it is judged that there is no obvious coking. If the real-time reconstruction error continues to be greater than the threshold, it is judged that there is a coking trend and an early warning of coking is issued. The location of coking can be determined according to the location of the sensor from which the warning signal originates. The severity and development trend of coking can be reflected by the magnitude and duration of the error value.

[0069] Through the aforementioned technical means, the embodiments of this application can diagnose the location and extent of coking on water-cooled walls by combining the actual boiler type and operating parameters with acoustic signature characteristics. This overcomes the limitations of traditional monitoring methods that rely on lagging parameters such as temperature and pressure, enabling early and accurate location and quantitative assessment of coking faults. It can capture key acoustic signature signals such as flow-induced noise when high-temperature flue gas flows around the heated surface of a single-sided hot water-cooled wall, quickly matching coking acoustic signature spectra under different boiler types and operating conditions. This significantly improves the pertinence and accuracy of the diagnosis, effectively avoiding problems such as corrosion of the heated surface and decreased thermal efficiency caused by coking expansion, and providing real-time and reliable technical support for the safe and efficient operation of the boiler.

[0070] Optionally, in one embodiment of this application, the method further includes: obtaining the current operating status of the boiler; matching diagnostic measures for the boiler based on the current operating status and the location and extent of coking on the boiler water-cooled wall, wherein the diagnostic measures include at least one of the following actions: soot blowing, boiler shutdown, and inspection.

[0071] Specifically, the embodiments of this application can determine whether soot blowing, boiler shutdown inspection, or other treatments are needed based on the actual operation of the boiler, and continuously optimize the diagnostic model by combining the accuracy of predictions. The model database is continuously improved by comprehensively considering the actual coking situation of the boiler under different loads, air distribution, and other parameter conditions, thus forming a closed-loop feedback.

[0072] Through the aforementioned technical means, the embodiments of this application can accurately match targeted diagnostic measures such as soot blowing, shutdown, and inspection by combining the current operating status of the boiler with the specific location and degree of coking on the water-cooled wall. This breaks away from the traditional "one-size-fits-all" boiler maintenance model, avoiding energy waste and damage to the heating surface caused by blind soot blowing. It can also trigger shutdown and inspection actions in a timely manner when the degree of coking is severe, effectively preventing safety accidents such as overheating of the heating surface and tube rupture caused by coking. This significantly improves the safety and economy of boiler operation and achieves refined and intelligent management and control of coking faults.

[0073] Optionally, in one embodiment of this application, the method further includes: predicting the coking trend of the boiler based on the location and degree of coking on the boiler water-cooled wall; and generating a warning signal for the boiler based on the coking trend.

[0074] Specifically, the embodiments of this application can reflect the severity and development trend of coking based on the magnitude and duration of the error value.

[0075] Through the aforementioned technical means, the embodiments of this application can predict the coking development trend based on the specific location and degree of coking on the boiler water-cooled wall, and generate targeted reminder signals. This can overcome the limitations of traditional post-event handling and achieve proactive early warning of coking faults. Furthermore, it can predict the risk of coking spread in advance and promptly push out graded reminders such as soot blowing, inspection, or boiler shutdown, avoiding serious accidents such as heat deviation of heating surfaces and tube rupture caused by aggravated coking, reducing the number of unplanned shutdowns, significantly improving the safety and continuity of boiler operation, and providing a decision-making basis for intelligent operation and maintenance throughout the boiler's entire life cycle.

[0076] like Figure 5 As shown, to enable those skilled in the art to more clearly understand this application, the following is an exemplary description of a method for diagnosing coking on boiler water-cooled walls using a specific embodiment: The acoustic signature method for diagnosing coking on the water-cooled wall of this boiler can be roughly summarized as follows: Step S1: Install one end of the waveguide rod onto the non-heated wall of the boiler's single-sided hot water cold wall via the bottom magnetic component to isolate the direct impact of the high temperature inside the furnace on the sensor; Step S2: Install a sensor on the other side of the waveguide rod to achieve non-invasive detection of acoustic signature data inside the boiler; Step S3: Import the acoustic signature signal data into the data acquisition and analysis equipment via the connecting cable to continuously acquire acoustic signature data inside the boiler in real time; Step S4: Preprocess the raw acoustic print data, extract feature parameters, construct a water-cooled wall coking acoustic print diagnostic model and perform pre-training; Step S5: Use a pre-trained voiceprint diagnostic model to process real-time voiceprint data, extract voiceprint signal features generated by real-time coking of boiler water-cooled walls, and diagnose the location and degree of coking. Step S6: Determine whether soot blowing or other treatments are needed based on the actual operation of the boiler, and continuously optimize the diagnostic model based on the accuracy of the prediction. Integrate the actual coking situation of the boiler under different operating conditions to continuously improve the model database and form a closed-loop feedback.

[0077] In summary, the acoustic signature diagnostic method for coking on boiler water-cooled walls proposed in this application can be summarized as follows: (1) Achieve non-invasive detection of coking on boiler water-cooled walls: Compared with invasive detection equipment such as probes and high-temperature cameras, acoustic fingerprint detection equipment can be installed on the outside of the non-heated wall of a single-sided hot water-cooled wall. It does not need to enter the high-temperature furnace and will not be affected by high-temperature flue gas radiation heat transfer, flow scouring, etc. It has extremely high safety and feasibility, greatly extends the life of monitoring equipment, and also reduces the operation and maintenance cost of detection equipment.

[0078] (2) Real-time online monitoring of coking on boiler water-cooled walls: The acoustic fingerprint detection equipment arranged outside the boiler can continuously collect acoustic fingerprint data and analyze the time-frequency domain characteristics of the acoustic fingerprint. Through data analysis of the diagnostic model, predictions can be made in the early stage of coking, before it has a serious impact on operation, to guide timely soot blowing and other coking removal operations, and to prevent the impact of coking on boiler operation of water-cooled walls to the greatest extent.

[0079] (3) Flexible detection layout: Different detection equipment layout schemes are adopted according to different furnace conditions. In areas with high coking risk, such as near the burner, relatively dense sensors can be arranged; while in areas with relatively low coking risk, a relatively sparse sensor array can be used, mainly for judging the overall trend of coking. Through this layout strategy, spatial diagnosis and localization of coking in large-area water-cooled walls can be achieved with a limited number of sensors. Through the flexible measurement point layout strategy, the system complexity and cost are controlled while ensuring the comprehensiveness of detection as much as possible.

[0080] The acoustic signature diagnostic method for boiler water-cooled wall coking proposed in this application can continuously collect acoustic signature data using acoustic signature detection equipment arranged outside the boiler, and analyze the time-frequency domain characteristics of the acoustic signature. Through data analysis of the diagnostic model, predictions can be made in the early stages of coking, before it seriously affects operation, guiding timely soot blowing and other coking removal operations, and minimizing the impact of water-cooled wall coking on boiler operation. This solves the problem in related technologies where existing boiler water-cooled wall coking detection technologies are limited by equipment lifespan and safety factors, resulting in the inability to comprehensively and in real-time track the coking status of boiler water-cooled walls.

[0081] Next, refer to the appendix. Figure 6 This application describes an acoustic signature diagnostic device for coking on boiler water-cooled walls, based on an embodiment of the present application.

[0082] Figure 6 This is a block diagram of the acoustic signature diagnostic device for coking on the boiler water-cooled wall according to an embodiment of this application.

[0083] like Figure 6 As shown, the acoustic signature diagnostic device 10 for coking on the boiler water-cooled wall includes: a data acquisition module 100, a data generation module 200, and a diagnostic module 300.

[0084] The acquisition module 100 is used to acquire acoustic signals inside the boiler.

[0085] The generation module 200 is used to generate boiler acoustic signature data based on the boiler acoustic signature signal.

[0086] The diagnostic module 300 is used to extract at least one acoustic signature feature based on the acoustic signature data inside the boiler, and to diagnose the location and extent of coking on the boiler water-cooled wall based on the at least one acoustic signature feature.

[0087] Optionally, in one embodiment of this application, the diagnostic module 300 includes an output unit for inputting acoustic fingerprint data inside the boiler into a pre-trained acoustic fingerprint diagnostic model to output at least one acoustic fingerprint signal feature generated by the real-time coking condition of the water-cooled wall inside the boiler.

[0088] Optionally, in one embodiment of this application, the acoustic signature diagnostic device 10 for boiler water-cooled wall coking further includes: an acquisition module and a matching module; wherein, the acquisition module is used to acquire the current operating status of the boiler; the matching module is used to match diagnostic measures for the boiler according to the current operating status, the location and degree of coking on the boiler water-cooled wall, wherein the diagnostic measures include at least one of the following actions: soot blowing, boiler shutdown, and inspection.

[0089] Optionally, in one embodiment of this application, the acoustic signature diagnostic device 10 for coking of boiler water-cooled walls further includes: a preprocessing module, which is used to preprocess the acoustic signature data inside the boiler before extracting at least one acoustic signature signal feature based on the acoustic signature data inside the boiler, so as to extract the acoustic signature data inside the boiler that meets the preset usage conditions based on the time domain, frequency domain, and time-frequency domain.

[0090] Optionally, in one embodiment of this application, the diagnostic module 300 includes: an acquisition unit and a determination unit; wherein the acquisition unit is used to acquire the actual type and / or operating parameters of the boiler; and the determination unit is used to determine the location and extent of coking on the boiler water-cooled wall by combining the actual type and / or operating parameters and at least one acoustic signature feature.

[0091] Optionally, in one embodiment of this application, the acoustic signature diagnostic device 10 for boiler water-cooled wall coking is further used to: predict the coking trend of the boiler based on the location and degree of coking on the boiler water-cooled wall; and generate a reminder signal for the boiler based on the coking trend.

[0092] It should be noted that the explanation of the aforementioned embodiment of the acoustic fingerprint diagnosis method for coking of boiler water-cooled walls also applies to the acoustic fingerprint diagnosis device for coking of boiler water-cooled walls in this embodiment, and will not be repeated here.

[0093] The acoustic signature diagnostic device for boiler water-cooled wall coking proposed in this application can continuously collect acoustic signature data through an acoustic signature detection device arranged outside the boiler, and analyze the time-frequency domain characteristics of the acoustic signature. Through data analysis of the diagnostic model, it can make predictions in the early stage of coking, before it seriously affects operation, and guide timely soot blowing and other coking removal operations to minimize the impact of water-cooled wall coking on boiler operation. This solves the problem in related technologies where existing boiler water-cooled wall coking detection technologies are limited by equipment lifespan and safety factors, resulting in the inability to comprehensively and in real-time track the coking status of boiler water-cooled walls.

[0094] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 701, the processor 702, and the computer program stored on the memory 701 and executable on the processor 702.

[0095] When the processor 702 executes the program, it implements the acoustic fingerprint diagnosis method for coking of boiler water-cooled walls provided in the above embodiments.

[0096] Furthermore, electronic devices also include: Communication interface 703 is used for communication between memory 701 and processor 702.

[0097] The memory 701 is used to store computer programs that can run on the processor 702.

[0098] The memory 701 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0099] If the memory 701, processor 702, and communication interface 703 are implemented independently, then the communication interface 703, memory 701, and processor 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized into address buses, data buses, control buses, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0100] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0101] The processor 702 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0102] This application also provides a non-volatile computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described acoustic fingerprint diagnosis method for coking of boiler water-cooled walls.

[0103] This application also provides a computer program product storing a computer program that, when executed by a processor, implements the above-described acoustic signature diagnosis method for coking on boiler water-cooled walls.

[0104] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0105] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0106] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0107] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0108] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0109] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0110] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0111] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. An acoustic signature diagnostic method for coking on boiler water-cooled walls, characterized in that, Includes the following steps: Collect acoustic signals inside the boiler; Boiler internal acoustic signature data is generated based on the boiler internal acoustic signature signal. At least one acoustic signature feature is extracted from the acoustic signature data inside the boiler, and the location and extent of coking on the boiler water-cooled wall are diagnosed based on the at least one acoustic signature feature.

2. The method according to claim 1, characterized in that, The step of extracting at least one acoustic signature feature based on the acoustic signature data inside the boiler includes: The acoustic signature data inside the boiler is input into a pre-trained acoustic signature diagnostic model to output at least one acoustic signature signal feature generated by the real-time coking condition of the water-cooled wall inside the boiler.

3. The method according to claim 1, characterized in that, Also includes: Obtain the current operating status of the boiler; Based on the current operating conditions and the location and extent of coking on the boiler water-cooled wall, diagnostic measures for the boiler are matched, wherein the diagnostic measures include at least one of the following actions: soot blowing, boiler shutdown, and inspection.

4. The method according to claim 1, characterized in that, Before extracting at least one of the acoustic signature signal features based on the acoustic signature data inside the boiler, the method further includes: The boiler internal acoustic fingerprint data is preprocessed to extract boiler internal acoustic fingerprint data that meets preset usage conditions based on time domain, frequency domain, and time-frequency domain.

5. The method according to claim 1, characterized in that, The method of diagnosing the location and extent of coking on the boiler water-cooled wall based on at least one acoustic signature feature includes: Obtain the actual type and / or operating parameters of the boiler; Based on the actual type and / or operating condition parameters and the at least one acoustic signature feature, the location and extent of coking on the boiler water-cooled wall are determined.

6. The method according to claim 1, characterized in that, Also includes: The coking trend of the boiler is predicted based on the location and extent of coking on the boiler water-cooled wall. A warning signal for the boiler is generated based on the coking trend.

7. An acoustic signature diagnostic device for coking on boiler water-cooled walls, characterized in that, include: The acquisition module is used to acquire acoustic signals inside the boiler. The generation module is used to generate boiler acoustic signature data based on the boiler acoustic signature signal. The diagnostic module is used to extract at least one acoustic signature feature from the acoustic signature data inside the boiler, and to diagnose the location and extent of coking on the boiler water-cooled wall based on the at least one acoustic signature feature.

8. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the acoustic signature diagnostic method for coking of boiler water-cooled walls as described in any one of claims 1-6.

9. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the acoustic signature diagnostic method for coking of boiler water-cooled walls as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the acoustic signature diagnostic method for coking of boiler water-cooled walls as described in any one of claims 1-6.

Images