A method, system and medium for constructing a three-dimensional temperature field of a waste heat boiler

Abstract

The present invention relates to the technical field of boiler temperature measurement, and in particular to a method, system and medium for constructing a stereoscopic three-dimensional temperature field of a waste heat boiler. The method comprises the following steps: deploying a fiber optic sensor grid on the waste heat boiler to obtain a distributed temperature acquisition network; sampling the distributed temperature acquisition network to obtain a real-time temperature sequence of the boiler; performing data preprocessing on the real-time temperature sequence of the boiler to obtain an initial temperature field of the boiler; constructing fluid dynamics model parameters for the waste heat boiler based on the initial temperature field of the boiler to obtain a set of fluid characteristic parameters of the boiler; performing multi-physical field coupling on the set of fluid characteristic parameters of the boiler to obtain a theoretical temperature field of the boiler; and deploying an acoustic sensor array and collecting data on the waste heat boiler to obtain acoustic spectrum collection data. The present invention can complement the respective advantages and limitations by fusing multiple temperature field data to obtain comprehensive and accurate temperature information of the waste heat boiler.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler temperature measurement, and in particular to a method, system and medium for constructing a three-dimensional temperature field of a waste heat boiler. Background Art

[0002] In the application of glass melting furnaces, the role of waste heat boilers is particularly critical. Glass melting furnaces produce a large amount of high-temperature waste gas during the production process. These waste gases contain a large amount of heat energy. If they can be effectively recycled and utilized, it will greatly reduce energy consumption and production costs. However, the existing waste heat boiler technology has significant technical difficulties in constructing and controlling the three-dimensional temperature field, which limits the efficiency and stability of waste heat utilization.

[0003] In a glass melting furnace, the temperature and flow of the exhaust gas will fluctuate with the working cycle of the melting furnace and the change of glass formula. This change puts higher requirements on the temperature control of the waste heat boiler. Traditional temperature monitoring technologies, such as thermocouples or infrared sensors, can usually only provide one-dimensional or two-dimensional temperature data. These technologies measure the temperature at certain specific locations or surfaces of the boiler, but cannot provide the overall temperature distribution inside the boiler. Since the temperature distribution inside the boiler is three-dimensional, this local temperature measurement cannot fully reflect the temperature conditions inside the entire boiler. For example, there are hot spots inside the boiler, and the temperature of these areas is much higher than other areas, but traditional monitoring technologies cannot detect these local high-temperature areas. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a method, system and medium for constructing a three-dimensional temperature field of a waste heat boiler to solve at least one of the above technical problems.

[0005] To achieve the above purpose, a method for constructing a three-dimensional temperature field of a waste heat boiler comprises the following steps:

[0006] Step S1: deploying a fiber optic sensor grid on the waste heat boiler to obtain a distributed temperature acquisition network; sampling the distributed temperature acquisition network to obtain a real-time temperature sequence of the boiler; performing data preprocessing on the real-time temperature sequence of the boiler to obtain an initial temperature field of the boiler;

[0007] Step S2: constructing the fluid dynamics model parameters of the waste heat boiler based on the initial temperature field of the boiler to obtain a boiler fluid characteristic parameter set; performing multi-physical field coupling on the boiler fluid characteristic parameter set to obtain a boiler theoretical temperature field;

[0008] Step S3: Arrange an acoustic sensor array and collect data on the waste heat boiler to obtain acoustic spectrum collection data; perform sound velocity characteristic analysis on the acoustic spectrum collection data to obtain acoustic-temperature mapping data; perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field;

[0009] Step S4: Use an infrared thermal imager to perform multi-angle scanning on the waste heat boiler to obtain a boiler thermal image sequence; perform spatial registration and fusion on the boiler thermal image sequence to obtain the global thermal distribution data of the boiler; perform heat conduction inversion based on the global thermal distribution data of the boiler to obtain the infrared temperature field of the boiler;

[0010] Step S5: performing temperature field fusion on the boiler initial temperature field, the boiler theoretical temperature field, the boiler acoustic temperature field and the boiler infrared temperature field to obtain the final three-dimensional temperature field of the boiler.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: performing three-dimensional modeling on the waste heat boiler to obtain a boiler structure grid model, and performing optical fiber sensor point distribution calculation on the waste heat boiler based on the boiler structure grid model to obtain an optical fiber sensor point distribution plan;

[0013] Step S12: deploying optical fiber sensors on the waste heat boiler according to the optical fiber sensor deployment plan to obtain a distributed temperature acquisition network;

[0014] Step S13: collecting the temperature time series of the waste heat boiler to obtain a boiler historical temperature sequence set, and identifying the temperature gradient rate of the boiler historical temperature sequence set to obtain a boiler temperature dynamic feature set;

[0015] Step S14: setting the sampling frequency of the distributed temperature acquisition network according to the boiler temperature dynamic feature set to obtain a sampling parameter configuration table;

[0016] Step S15: performing data acquisition on the distributed temperature acquisition network according to the sampling parameter configuration table to obtain the boiler original temperature data stream, and performing abnormal value detection on the boiler original temperature data stream to obtain a boiler temperature abnormality mark sequence;

[0017] Step S16: performing data repair on the boiler original temperature data stream according to the boiler temperature abnormality mark sequence to obtain the boiler real-time temperature sequence;

[0018] Step S17: Calculate the temperature gradient of the boiler real-time temperature sequence to obtain the boiler temperature gradient field, and identify the heat flux distribution of the waste heat boiler based on the boiler temperature gradient field to obtain the heat flux density distribution data inside the boiler; perform data fusion on the boiler real-time temperature sequence and the heat flux density distribution data inside the boiler to obtain the boiler initial temperature field.

[0019] Preferably, step S2 comprises the following steps:

[0020] Step S21: evaluating the density of the medium in the waste heat boiler based on the initial temperature field of the boiler to obtain the density distribution data of the medium in the furnace;

[0021] Step S22: evaluating the fluid viscosity of the density distribution data of the medium in the furnace to obtain boiler temperature-viscosity correlation data;

[0022] Step S23: Calculate the Reynolds number of the waste heat boiler according to the boiler temperature-viscosity correlation data to obtain boiler fluid turbulence characteristic parameters;

[0023] Step S24: selecting a turbulence model for the waste heat boiler based on the boiler fluid turbulence characteristic parameters to obtain a boiler turbulence calculation model;

[0024] Step S25: performing numerical simulation of the flow field of the waste heat boiler based on the boiler turbulence calculation model to obtain a boiler fluid characteristic parameter set;

[0025] Step S26: Perform temperature field iterative calculation according to the boiler fluid characteristic parameter set to obtain the boiler theoretical temperature field.

[0026] Preferably, step S25 comprises the following steps:

[0027] Step S251: setting boundary conditions for the waste heat boiler to obtain a boiler boundary constraint parameter set;

[0028] Step S252: meshing the boiler turbulence calculation model according to the boiler boundary constraint parameter set to obtain a boiler turbulence calculation grid model;

[0029] Step S253: discretizing the control equations of the boiler turbulence calculation grid model to obtain a boiler discrete equation group;

[0030] Step S254: performing SIMPLE algorithm solution configuration on the boiler discrete equation group to obtain boiler numerical solver configuration data;

[0031] Step S255: performing parallel calculation and solving of the boiler discrete equations according to the boiler numerical solver configuration data to obtain an initial solution of the boiler flow field;

[0032] Step S256: performing radiation heat transfer evaluation on the initial solution of the boiler flow field to obtain the boiler radiation heat flux field;

[0033] Step S257: performing convective heat transfer evaluation on the waste heat boiler according to the boiler radiation heat flux field to obtain a boiler heat transfer efficiency diagram;

[0034] Step S258: performing coupling calculation on the boiler radiation heat flux field and the boiler heat transfer efficiency diagram to obtain a boiler fluid characteristic parameter set.

[0035] Preferably, step S3 comprises the following steps:

[0036] Step S31: configuring acoustic sensors for the waste heat boiler to obtain a distribution map of acoustic measurement points of the boiler;

[0037] Step S32: formulating an acoustic sensor array for the waste heat boiler according to the boiler acoustic measurement point distribution map to obtain an acoustic sensor array layout plan;

[0038] Step S33: performing signal channel planning for the acoustic sensor array arrangement scheme to obtain an acoustic signal channel configuration table;

[0039] Step S34: formulating an acoustic excitation signal for the waste heat boiler according to the acoustic signal channel configuration table to obtain a boiler acoustic wave excitation pulse sequence;

[0040] Step S35: performing acoustic wave emission control on the acoustic sensor array according to the boiler acoustic wave excitation pulse sequence to obtain a boiler acoustic wave emission synchronization sequence, and performing synchronous sampling control on the boiler acoustic wave emission synchronization sequence to obtain a boiler original acoustic signal set;

[0041] Step S36: performing signal preprocessing on the original acoustic signal set of the boiler to obtain acoustic spectrum acquisition data;

[0042] Step S37: Perform sound velocity characteristic analysis on the acoustic spectrum acquisition data to obtain acoustic-temperature mapping data, and perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field.

[0043] Preferably, step S37 includes the following steps:

[0044] Step S371: extracting the sound velocity from the acoustic spectrum acquisition data to obtain the boiler sound wave propagation velocity distribution data;

[0045] Step S372: constructing a temperature mapping relationship for the boiler sound wave propagation velocity distribution data to obtain acoustic-temperature mapping data;

[0046] Step S373: performing temperature inversion on the acoustic-temperature mapping data to obtain initial acoustic temperature data;

[0047] Step S374: performing spatial interpolation on the initial acoustic temperature data to obtain an initial boiler acoustic temperature field;

[0048] Step S375: identifying the acoustic wave refraction effect of the initial boiler acoustic temperature field to obtain boiler acoustic wave propagation path distortion data;

[0049] Step S376: Perform acoustic wave path compensation on the initial boiler acoustic temperature field according to the boiler acoustic wave propagation path distortion data to obtain the boiler acoustic compensated temperature field, and perform acoustic imaging reconstruction on the boiler acoustic compensated temperature field to obtain the boiler acoustic temperature field.

[0050] Preferably, step S4 comprises the following steps:

[0051] Step S41: dividing the waste heat boiler into infrared imaging areas to obtain a boiler scanning area plan, and configuring thermal imager points on the waste heat boiler according to the boiler scanning area plan to obtain a thermal imager point position set;

[0052] Step S42: setting thermal imager calibration parameters for the thermal imager point location set to obtain a boiler thermal radiation calibration parameter table;

[0053] Step S43: performing multi-angle synchronous scanning control on the thermal imager array based on the boiler thermal radiation calibration parameter table to obtain a sequence of original infrared images of the boiler;

[0054] Step S44: performing spatial transformation and resampling on the original infrared image sequence of the boiler to obtain a boiler infrared registration image set;

[0055] Step S45: performing multi-resolution image fusion on the boiler infrared registration image set to obtain the global thermal distribution data of the boiler, and performing temperature quantitative conversion on the global thermal distribution data of the boiler to obtain a surface temperature distribution map of the boiler;

[0056] Step S46: constructing a heat conduction equation for the waste heat boiler based on the boiler surface temperature distribution diagram to obtain a boiler heat conduction mathematical model;

[0057] Step S47: Perform temperature field inversion on the waste heat boiler according to the mathematical model of boiler heat conduction to obtain the infrared temperature field of the boiler.

[0058] Preferably, step S5 comprises the following steps:

[0059] Step S51: performing data reliability evaluation on the initial temperature field of the boiler to obtain an initial temperature credibility distribution map, and performing confidence correction on the initial temperature field of the boiler according to the initial temperature credibility distribution map to obtain an initial weighted temperature field of the boiler;

[0060] Step S52: performing model error statistics on the boiler theoretical temperature field according to the boiler infrared temperature field to obtain a predicted temperature error distribution map, and performing error compensation on the boiler theoretical temperature field according to the predicted temperature error distribution map to obtain a boiler theoretical corrected temperature field;

[0061] Step S53: performing inversion accuracy evaluation on the infrared temperature field of the boiler according to the initial temperature field of the boiler to obtain an infrared temperature accuracy distribution map, and performing accuracy calibration on the infrared temperature field of the boiler according to the infrared temperature accuracy distribution map to obtain an infrared calibration temperature field of the boiler;

[0062] Step S54: spatially registering and fusing the boiler acoustic temperature field, the boiler initial weighted temperature field, the boiler theoretically corrected temperature field and the boiler infrared calibration temperature field to obtain the boiler multi-source registered temperature field;

[0063] Step S55: Perform physical constraint verification on the boiler multi-source registration temperature field to obtain a set of physical constraint conditions for boiler temperature;

[0064] Step S56: performing constraint optimization on the boiler multi-source registration temperature field according to the boiler temperature physical constraint condition set to obtain the final three-dimensional temperature field of the boiler.

[0065] The present invention realizes all-round and continuous temperature monitoring of the waste heat boiler through the deployment of the optical fiber sensor grid, provides high-resolution temperature data, and enhances the perception of temperature changes inside the boiler. The distributed temperature acquisition network can collect the boiler temperature sequence in real time, provide instant feedback for the operating status of the boiler, and help to timely discover and deal with potential temperature anomalies. Based on the initial temperature field of the boiler, a more accurate fluid dynamics model can be constructed to obtain a set of boiler fluid characteristic parameters, thereby improving the accuracy of theoretical temperature field calculation. Through multi-physical field coupling, the factors such as fluid dynamics, heat conduction and radiation inside the boiler are comprehensively considered to obtain a theoretical temperature field that is closer to the actual working state. The layout of the acoustic sensor array provides a non-invasive temperature monitoring method, which can detect temperature changes in difficult-to-reach areas inside the boiler, increasing the flexibility and safety of monitoring. By analyzing the acoustic spectrum data, a mapping relationship between the sound velocity characteristics and the temperature is established, and the boiler acoustic temperature field is further obtained by the inversion technology, providing an additional data dimension for the temperature distribution inside the boiler. The infrared thermal imager is used for multi-angle scanning to fully capture the heat distribution on the boiler surface, providing intuitive image data for boiler surface temperature monitoring. By inverting the heat conduction of the boiler's global heat distribution data, the boiler's infrared temperature field is obtained, providing accurate data for temperature analysis of the boiler surface and near-surface areas. Since a single temperature monitoring method cannot fully reflect the three-dimensional temperature distribution inside the boiler, by fusing multiple temperature field data, their respective advantages and limitations can be complemented to obtain more comprehensive and accurate temperature information. The fused three-dimensional temperature field can solve the problem of hot spots inside the boiler that cannot be detected by traditional monitoring technology, provide more accurate temperature control, reduce equipment losses caused by temperature fluctuations, and improve the efficiency and stability of waste heat utilization.

[0066] Preferably, the present invention further provides a three-dimensional temperature field construction system for a waste heat boiler, which is used to execute the three-dimensional temperature field construction method for a waste heat boiler as described above. The three-dimensional temperature field construction system for a waste heat boiler comprises:

[0067] The temperature acquisition module is used to deploy a fiber optic sensor grid on the waste heat boiler to obtain a distributed temperature acquisition network; sample the distributed temperature acquisition network to obtain a real-time temperature sequence of the boiler; perform data preprocessing on the real-time temperature sequence of the boiler to obtain the initial temperature field of the boiler;

[0068] Theoretical temperature calculation module: construct the fluid dynamics model parameters of the waste heat boiler based on the initial temperature field of the boiler to obtain the boiler fluid characteristic parameter set; perform multi-physics field coupling on the boiler fluid characteristic parameter set to obtain the boiler theoretical temperature field;

[0069] Acoustic analysis module: Arrange the acoustic sensor array and collect data on the waste heat boiler to obtain the acoustic spectrum collection data; perform sound velocity characteristic analysis on the acoustic spectrum collection data to obtain the acoustic-temperature mapping data; perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field;

[0070] Infrared scanning module: Use infrared thermal imager to scan the waste heat boiler at multiple angles to obtain boiler thermal image sequence; perform spatial registration and fusion on the boiler thermal image sequence to obtain the boiler global thermal distribution data; perform heat conduction inversion based on the boiler global thermal distribution data to obtain the boiler infrared temperature field;

[0071] Temperature field fusion module: The temperature field of the boiler initial temperature field, the boiler theoretical temperature field, the boiler acoustic temperature field and the boiler infrared temperature field are fused to obtain the final three-dimensional temperature field of the boiler.

[0072] The present invention can fully capture the temperature distribution inside the boiler and provide more comprehensive thermal energy management by integrating multiple temperature monitoring technologies. The system can monitor the boiler temperature in real time, detect abnormalities in time, and prevent boiler overheating and potential safety accidents.

[0073] Preferably, a computer-readable storage medium stores a method for constructing a three-dimensional temperature field of a waste heat boiler as described above, which can be loaded and executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Other features, objects and advantages of the present invention will become more apparent from reading the detailed description with reference to the following drawings:

[0075] Figure 1 A schematic diagram of the steps of a method for constructing a three-dimensional temperature field of a waste heat boiler according to an embodiment is shown;

[0076] Figure 2 A detailed flow chart of step S25 of an embodiment is shown. DETAILED DESCRIPTION

[0077] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0078] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0079] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0080] To achieve this, please refer to Figure 1 to Figure 2 The present invention provides a method for constructing a three-dimensional temperature field of a waste heat boiler, comprising the following steps:

[0081] Step S1: deploying a fiber optic sensor grid on the waste heat boiler to obtain a distributed temperature acquisition network; sampling the distributed temperature acquisition network to obtain a real-time temperature sequence of the boiler; performing data preprocessing on the real-time temperature sequence of the boiler to obtain an initial temperature field of the boiler;

[0082] Step S2: constructing the fluid dynamics model parameters of the waste heat boiler based on the initial temperature field of the boiler to obtain a boiler fluid characteristic parameter set; performing multi-physical field coupling on the boiler fluid characteristic parameter set to obtain a boiler theoretical temperature field;

[0083] Step S3: Arrange an acoustic sensor array and collect data on the waste heat boiler to obtain acoustic spectrum collection data; perform sound velocity characteristic analysis on the acoustic spectrum collection data to obtain acoustic-temperature mapping data; perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field;

[0084] Step S4: Use an infrared thermal imager to perform multi-angle scanning on the waste heat boiler to obtain a boiler thermal image sequence; perform spatial registration and fusion on the boiler thermal image sequence to obtain the global thermal distribution data of the boiler; perform heat conduction inversion based on the global thermal distribution data of the boiler to obtain the infrared temperature field of the boiler;

[0085] Step S5: performing temperature field fusion on the boiler initial temperature field, the boiler theoretical temperature field, the boiler acoustic temperature field and the boiler infrared temperature field to obtain the final three-dimensional temperature field of the boiler.

[0086] In this embodiment, the fiber optic sensor grid is first deployed. The boiler is modeled using three-dimensional modeling software, such as Autodesk 3dsMax, and the optimal placement of the fiber optic sensor is calculated based on the model. The Brilliance series fiber optic sensors are used because they are suitable for the boiler environment due to their high temperature resistance. Sensors are deployed at 5-meter intervals at key parts of the boiler, such as the flue gas inlet and outlet, the furnace wall and the furnace roof, to form a distributed temperature acquisition network. Then, the network is sampled through the National Instruments DAQ system, and the sampling frequency is set to 10 times per second to obtain the real-time temperature sequence of the boiler. Subsequently, Python is used in conjunction with the NumPy and Pandas libraries to perform data preprocessing on the real-time temperature sequence, including outlier processing, to obtain the initial temperature field of the boiler. Based on this temperature field, a fluid dynamics model is constructed in ANSYS Fluent, and a set of boiler fluid characteristic parameters, such as thermal conductivity and specific heat capacity, is input to perform multi-physics field coupling simulation to obtain the theoretical temperature field of the boiler. At the same time, an acoustic sensor array is arranged around the boiler, using PCB brand Piezoelectric Microphone, arranged at intervals of 2 meters, and the acoustic spectrum data is collected at a frequency of 10,000 times per second through the NI USB-4431 data collector. Then, the Fourier transform is used in MATLAB to analyze the sound velocity characteristics, establish the acoustic-temperature mapping data, and obtain the boiler acoustic temperature field through inversion. In addition, the boiler is scanned from multiple angles using a FLIR T640 infrared thermal imager to capture a sequence of boiler thermal images. The images are spatially registered and fused in the ENVI software to obtain the global thermal distribution data of the boiler. Based on these data, the heat conduction equation is used for inversion calculation to obtain the infrared temperature field of the boiler. Finally, the initial temperature field, theoretical temperature field, acoustic temperature field and infrared temperature field of the boiler are spatially registered and fused in ArcGIS, and the weighted average method is used to integrate the temperature field data to obtain the final three-dimensional temperature field of the boiler.

[0087] Preferably, step S1 comprises the following steps:

[0088] Step S11: performing three-dimensional modeling on the waste heat boiler to obtain a boiler structure grid model, and performing optical fiber sensor point distribution calculation on the waste heat boiler based on the boiler structure grid model to obtain an optical fiber sensor point distribution plan;

[0089] Specifically, a 3D modeling software such as Autodesk 3ds Max or SolidWorks can be used to perform 3D modeling of the waste heat boiler. First, the detailed size and shape parameters of the boiler, including diameter, length, wall thickness, etc., are collected. Then, based on these parameters, a 3D geometric model of the boiler is created in the modeling software. During the model creation process, a mesh refinement tool is used to ensure the accuracy of the boiler structure mesh model. For example, the mesh size can be set to 5mm×5mm×5mm to ensure the fineness of the model. A professional fiber optic sensor layout calculation tool, such as COMSOL Multiphysics software, is used to perform fiber optic sensor point calculation based on the boiler structure mesh model. In COMSOL, the parameters of sensor point distribution, such as sensor spacing, coverage, etc., are defined. For example, the sensor spacing is set to 10cm to ensure that there is no blind spot in the temperature monitoring of the key areas inside the boiler. Through simulation calculation, a fiber optic sensor point distribution scheme is obtained, which details the specific location and number of each sensor. For example, 10 sensors are arranged in the front, back, left, right and top of the boiler, for a total of 50 sensors.

[0090] Step S12: deploying optical fiber sensors on the waste heat boiler according to the optical fiber sensor deployment plan to obtain a distributed temperature acquisition network;

[0091] Specifically, the actual deployment of optical fiber sensors on the waste heat boiler can be carried out according to the optical fiber sensor deployment plan. Select optical fiber sensors with high temperature tolerance and high sensitivity. For example, use Bragg grating sensors, whose measurement range can reach -50 to +900°C, which is suitable for the high temperature environment inside the boiler. During the deployment process, use a laser marking machine to mark the specific installation position of the sensor on the surface of the boiler. Then, use high-temperature glue to fix the optical fiber sensor at the marked position to ensure that the sensor is in close contact with the boiler surface to obtain accurate temperature readings. For areas that are difficult to directly contact, high-temperature resistant stainless steel cables are used to introduce the sensors, and high-temperature resistant sealing materials are used to ensure that heat does not leak from the holes through which the cables pass. Finally, all optical fiber sensors are connected to a central data acquisition system, for example, using a National Instruments data acquisition card and LabVIEW software. Through the above steps, a distributed temperature acquisition network can be obtained.

[0092] Step S13: collecting the temperature time series of the waste heat boiler to obtain a boiler historical temperature sequence set, and identifying the temperature gradient rate of the boiler historical temperature sequence set to obtain a boiler temperature dynamic feature set;

[0093] Specifically, a distributed temperature acquisition network can be used to perform temperature time series acquisition on the waste heat boiler. Specifically, through a central data acquisition system, such as National Instruments' PXIe-4081 module, which is a high-performance multi-channel data acquisition device, the temperature readings of each fiber optic sensor are continuously recorded at set time intervals (for example, every 5 seconds). These readings are stored in a database to form a boiler historical temperature series set, which contains temperature data of various parts of the boiler in the past month. Using data analysis software, such as MATLAB or Python with NumPy and SciPy libraries, the temperature gradient rate of the boiler historical temperature series set is identified. By writing a script, the temperature change rate of each sensor in different time periods is calculated, for example, the average temperature change rate per hour is calculated. By analyzing these change rates, time periods with significant temperature changes are identified, such as boiler startup, steady-state operation, and shutdown stages, and areas with large temperature gradients are marked, for example, areas with temperature gradients greater than 0.5°C / minute. After summarizing this information, a boiler temperature dynamic feature set is obtained.

[0094] Step S14: setting the sampling frequency of the distributed temperature acquisition network according to the boiler temperature dynamic feature set to obtain a sampling parameter configuration table;

[0095] Specifically, LabVIEW software can be used to configure the sampling parameters of the data acquisition card. According to the high-temperature gradient rate areas and time periods identified by the boiler temperature dynamic feature set, the sampling frequency of the sensors in these areas is adjusted. For example, for areas with high temperature gradient rate during the boiler startup phase, the sampling frequency is increased from once every 5 seconds to once every 1 second to capture rapid temperature changes. For areas with slower temperature changes during the steady-state operation phase, the sampling frequency is reduced to once every 10 seconds to reduce the amount of data and save storage space. Finally, a sampling parameter configuration table is generated, which lists in detail the ID, location, set sampling frequency, and reasons for adjustment of each sensor. For example, the sensor with sensor ID 101 is located at the front of the boiler. Due to the high temperature gradient rate during the startup phase, the sampling frequency is set to once every 1 second.

[0096] Step S15: performing data acquisition on the distributed temperature acquisition network according to the sampling parameter configuration table to obtain the boiler original temperature data stream, and performing abnormal value detection on the boiler original temperature data stream to obtain a boiler temperature abnormality mark sequence;

[0097] Specifically, according to the sampling parameter configuration table, the National Instruments PXIe-4081 data acquisition card can be used to automatically collect boiler temperature data from the distributed temperature acquisition network according to the ID of each sensor and the set sampling frequency. For example, the sensor with sensor ID 101 is located at the front of the boiler, and according to the configuration table, its sampling frequency is set to once every 1 second. The data acquisition card transmits these temperature readings to the server of the monitoring center in real time to form a boiler raw temperature data stream. The Python programming language is used with the Pandas and NumPy libraries to perform outlier detection on the boiler raw temperature data stream. The parameters for outlier detection are set. For example, if the reading of a sensor exceeds ±3 standard deviations or the difference with the reading of the adjacent sensor exceeds 5°C within 5 consecutive sampling cycles, the reading is considered to be an outlier. Once an outlier is detected, the system automatically records the sensor ID and timestamp to form a boiler temperature abnormality mark sequence.

[0098] Step S16: performing data repair on the boiler original temperature data stream according to the boiler temperature abnormality mark sequence to obtain the boiler real-time temperature sequence;

[0099] Specifically, MATLAB software can be used to interpolate or smooth abnormal data based on the boiler temperature abnormality mark sequence. For example, if a sensor with sensor ID 101 detects an abnormal value at a certain point in time, the average value of the sensor in the first 10 normal sampling cycles is selected as a substitute for the abnormal value. If the abnormal value continues to appear, the average temperature value of the adjacent sensors can also be used for interpolation to repair the abnormal data. During the data repair process, the physical rationality of the data is checked. For example, if the repaired temperature value still exceeds the normal operating range of the boiler, further analyze the potential causes, such as sensor failure or boiler abnormality, and promptly notify the maintenance personnel for inspection. After data repair, the real-time temperature sequence of the boiler is obtained.

[0100] Step S17: Calculate the temperature gradient of the boiler real-time temperature sequence to obtain the boiler temperature gradient field, and identify the heat flux distribution of the waste heat boiler based on the boiler temperature gradient field to obtain the heat flux density distribution data inside the boiler; perform data fusion on the boiler real-time temperature sequence and the heat flux density distribution data inside the boiler to obtain the boiler initial temperature field.

[0101] Specifically, the data processing software MATLAB can be used to calculate the temperature gradient of the real-time temperature sequence of the boiler. Specifically, the numerical differentiation toolbox of MATLAB is used to numerically differentiate the time-temperature data of each sensor to calculate the rate of change of temperature over time, that is, the temperature gradient. For example, the temperature change of the sensor with sensor ID 101 in the past 10 minutes is calculated, and its temperature gradient is 0.2°C / s. The calculated temperature gradient data is used to identify the heat flow distribution inside the boiler. Finite element analysis software such as ANSYS Fluent is used to input the temperature gradient field data, and the heat flow simulation is performed in combination with the fluid dynamics model of the boiler. During the simulation process, the physical parameters of the fluid, such as specific heat capacity, thermal conductivity, etc., are set, and a suitable turbulence model, such as the k-ε model, is selected to improve the accuracy of the simulation. Through simulation, the heat flux density distribution data inside the boiler is obtained. For example, in the outlet area of ​​the boiler, the heat flux density is 500 W / m². Finally, Python is used with the SciPy library to match the temperature data and the heat flux density data according to the spatial position, and a three-dimensional temperature field model is generated in combination with the geometric model of the boiler. In this model, each grid point contains not only real-time temperature data but also corresponding heat flux density data. Through data fusion, an initial temperature field that comprehensively reflects the temperature distribution and heat flux dynamics inside the boiler can be obtained.

[0102] Preferably, step S2 comprises the following steps:

[0103] Step S21: evaluating the density of the medium in the waste heat boiler based on the initial temperature field of the boiler to obtain the density distribution data of the medium in the furnace;

[0104] Specifically, the data of the initial temperature field of the boiler can be used, which contains the temperature information of various positions inside the boiler. Use thermodynamic software, such as Chemical Engineering Suite, to evaluate the density of the medium in the furnace. In the specific operation, the temperature value in the initial temperature field of the boiler is used as an input parameter. Combined with the chemical composition and pressure data of the boiler medium, the software can calculate the density of the medium at various temperatures according to the thermodynamic equation. For example, the boiler medium is set to flue gas, the main components of which are nitrogen, oxygen and carbon dioxide, and the pressure is 1 atmosphere. According to the initial temperature field of the boiler, it is found that the temperature range inside the boiler is 800°C to 1200°C. Input these data into the Chemical Engineering Suite software, and the software calculates the density of the medium at different temperatures based on the ideal gas state equation and the actual gas state equation. For example, at 1000°C, the density of the flue gas is 0.3 kg / m³. In this way, the density distribution data of the medium in the furnace can be obtained, which describes in detail the density changes of the medium inside the boiler at different positions and temperatures.

[0105] Step S22: evaluating the fluid viscosity of the density distribution data of the medium in the furnace to obtain boiler temperature-viscosity correlation data;

[0106] Specifically, the fluid viscosity can be evaluated based on the density distribution data of the medium in the furnace. Use fluid mechanics simulation software, such as ANSYS Fluent, to evaluate the viscosity of the medium. In actual operation, the density data of the medium is used as an input parameter, combined with the chemical composition and temperature data of the medium, and the software can calculate the viscosity of the medium at various temperatures based on the fluid mechanics equation. For example, the boiler medium is set to flue gas, and the main components are nitrogen, oxygen and carbon dioxide. According to the density distribution data of the medium in the furnace, it is known that the density and temperature of the medium are different in different areas of the boiler. Enter these data into the ANSYS Fluent software, and select a suitable fluid viscosity model, such as the Sutherland formula, to calculate the dynamic viscosity at different temperatures. For example, at 1000°C, the dynamic viscosity of the flue gas is 1.8×10 -5 Pa·s. Through simulation, the boiler temperature-viscosity correlation data can be obtained, which describes in detail the viscosity changes of the medium inside the boiler at different temperatures and densities.

[0107] Step S23: Calculate the Reynolds number of the waste heat boiler according to the boiler temperature-viscosity correlation data to obtain boiler fluid turbulence characteristic parameters;

[0108] Specifically, when implementing step S23, it is necessary to calculate the Reynolds number (Re) of the fluid flow in the boiler, which is a key dimensionless number that characterizes the flow state in fluid dynamics. The calculation formula of the Reynolds number is: , where Re is the Reynolds number, which is used to characterize the flow state; ρ is the density of the fluid, in kg / m³, which indicates the mass of the fluid per unit volume; v is the flow velocity of the fluid, in m / s, which indicates the distance the fluid moves per unit time; D is the characteristic length, in m, which is usually the pipe diameter or the characteristic length of the fluid flow; is the dynamic viscosity of the fluid, in Pa·s, which indicates the magnitude of the internal resistance of the fluid. ANSYS Fluent software is used to perform this calculation. First, the relevant parameters of the medium in the boiler are collected. For example, in a specific area of ​​the boiler, the density ρ of the medium is measured to be 0.8 kg / m³, and the dynamic viscosity 2.5×10 -5 Pa·s, flow velocity v is 3 m / s, and characteristic length D (assumed to be the pipe diameter) is 0.5 m. Enter these parameters into ANSYS Fluent software, and the software will calculate the Reynolds number of the area according to the above formula: The calculation result assumes Re=12000, which indicates that the flow state in this area is in a turbulent state. By performing similar calculations in different areas of the boiler, the turbulent characteristic parameters of the entire boiler fluid can be obtained, including the Reynolds number in different areas.

[0109] Step S24: selecting a turbulence model for the waste heat boiler based on the boiler fluid turbulence characteristic parameters to obtain a boiler turbulence calculation model;

[0110] Specifically, based on the boiler fluid turbulence characteristic parameters, a suitable turbulence model can be selected to carry out numerical simulation of the boiler flow field. Use ANSYS Fluent software and select a turbulence model according to the size of the Reynolds number and other fluid characteristic parameters. For example, for flows with Reynolds numbers greater than 4000, the k-ε model or the k-ω model can be selected. Assume that the Reynolds number distribution varies from 8000 to 20000 in different areas of the boiler. It is decided to use the k-ε model in areas with lower Reynolds numbers (e.g., 8000-15000) because it is suitable for turbulence simulation with medium Reynolds numbers; and to use the k-ω model in areas with higher Reynolds numbers (e.g., 15000-20000) because it is more suitable for turbulence simulation with high Reynolds numbers. In ANSYS Fluent, set the corresponding turbulence model parameters, such as constants such as Cμ, C1 and C2 in the k-ε model, and constants such as σk and σω in the k-ω model. By selecting a suitable turbulence model and setting relevant parameters, a boiler turbulence calculation model is obtained.

[0111] Step S25: performing numerical simulation of the flow field of the waste heat boiler based on the boiler turbulence calculation model to obtain a boiler fluid characteristic parameter set;

[0112] Specifically, please refer to the sub-steps of step S25 for the detailed implementation process of this embodiment.

[0113] Step S26: Perform temperature field iterative calculation according to the boiler fluid characteristic parameter set to obtain the boiler theoretical temperature field.

[0114] Specifically, the boiler fluid characteristic parameter set can be imported into Fluent. These parameters include the temperature, velocity, pressure, and contributions of radiation and convection heat transfer. Set the iterative calculation parameters of Fluent, for example, set the residual value to be less than 1×10 -5As the criterion for iterative convergence. In Fluent, select the Energy Equation to simulate the temperature field inside the boiler. Enter the physical properties of the boiler, such as specific heat capacity, thermal conductivity, and density. Assume that the specific heat capacity of the boiler medium is 1200 J / (kg·K), the thermal conductivity is 0.05 W / (m·K), and the density is 0.9 kg / m³. It is also necessary to define the boundary conditions of the boiler wall, for example, setting the wall heat flux to 1000 W / m². Next, start the iterative calculation. Fluent will discretize the energy equation based on the input fluid characteristic parameters and boundary conditions through numerical methods (such as the finite volume method), and then solve it iteratively. During the iteration process, Fluent will update the temperature field in real time until the convergence criteria are met. For example, after 500 iterations, Fluent shows that the residual value has dropped to 1×10 -5 Below, it shows that the temperature field has converged. Finally, the theoretical temperature field of the boiler is obtained, which is a three-dimensional temperature distribution diagram showing the temperature value of each grid cell inside the boiler. For example, in the outlet area of ​​the boiler, the theoretical temperature field shows a temperature of 1100°C, while in the inlet area of ​​the boiler, the temperature is 900°C.

[0115] Preferably, step S25 comprises the following steps:

[0116] Step S251: setting boundary conditions for the waste heat boiler to obtain a boiler boundary constraint parameter set;

[0117] Specifically, computational fluid dynamics (CFD) software, such as ANSYS Fluent, can be used to define the boundary conditions of the boiler. Boundary conditions include parameters such as inlet velocity, outlet pressure, and wall temperature. For example, the flow velocity at the boiler inlet is set to 10 m / s, the pressure at the outlet is set to 1 atm (101325 Pa), and the wall temperature is set to 500°C. These parameters constitute the boiler boundary constraint parameter set. Select appropriate boundary conditions based on the actual size of the boiler and the fluid properties. For example, if the flow at the boiler inlet is developing, a development segment needs to be set to simulate the process of the flow accelerating from a stationary state to a specified speed. For the outlet, select the pressure outlet boundary condition and set a reference pressure. The wall is set to be adiabatic or set to a specific temperature value. Through these specific parameter settings, the boiler boundary constraint parameter set can be obtained.

[0118] Step S252: meshing the boiler turbulence calculation model according to the boiler boundary constraint parameter set to obtain a boiler turbulence calculation grid model;

[0119] Specifically, the boiler turbulence calculation model can be meshed using ANSYS Fluent software according to the boiler boundary constraint parameter set. First, import the three-dimensional geometric model of the boiler, and then select the appropriate mesh type. For example, select a structured mesh for better quality, or select an unstructured mesh for greater flexibility. Set the mesh size, for example, use a finer mesh (e.g., a mesh size of 10 mm) in critical areas of the boiler (such as near the inlet and outlet and wall), and a coarser mesh (e.g., a mesh size of 50 mm) in areas where the flow is more uniform. The growth rate of the mesh can also be set to ensure a smooth transition from fine to coarse meshes. With these settings, a boiler turbulence calculation mesh model can be obtained, which contains all the necessary geometric details and boundary conditions inside the boiler. Through the above steps, a boiler turbulence calculation mesh model can be generated.

[0120] Step S253: discretizing the control equations of the boiler turbulence calculation grid model to obtain a boiler discrete equation group;

[0121] Specifically, ANSYS Fluent software can be used to discretize the control equations in the boiler turbulence calculation grid model. The control equations mainly include mass conservation equations, momentum conservation equations, and energy conservation equations. First, select a suitable discretization method in Fluent, such as a high-resolution scheme for processing the momentum equation to reduce numerical diffusion and improve accuracy. In the specific operation, a specific discretization scheme is selected for each control equation. For example, for the momentum equation, a second-order upwind scheme is selected, which is suitable for capturing shock wave surfaces in high-speed flows. For the pressure-velocity coupling term, the PISO (Pressure-Implicit with Splitting of Operators) algorithm is used, which is a pressure-velocity coupling algorithm that can effectively handle transient flow problems. After the settings in Fluent are completed, the software will automatically discretize the control equations and convert the continuous differential equations into discrete algebraic equations, namely the boiler discrete equations. These equations contain all the physical information of the flow inside the boiler.

[0122] Step S254: performing SIMPLE algorithm solution configuration on the boiler discrete equation group to obtain boiler numerical solver configuration data;

[0123] Specifically, the SIMPLE (Semi-Implicit Method for Pressure-Linked Equations) algorithm can be configured to solve the boiler discrete equations. The SIMPLE algorithm is a pressure-velocity coupling algorithm suitable for solving incompressible flows. In ANSYS Fluent, first enter the solver configuration panel and select the SIMPLE algorithm as the pressure-velocity coupling method. Next, set the parameters of the SIMPLE algorithm. For example, set the relaxation factor of the pressure equation to 0.3, which is a parameter that controls the iterative convergence of the pressure field. For the momentum equation, set the relaxation factor to 0.7 to ensure the stability and convergence of the momentum equation. It is also necessary to set the maximum number of iterations to 1000 to ensure that the solution process can converge within a reasonable number of iterations. After the settings are completed, Fluent will generate the boiler numerical solver configuration data based on these parameter configuration data. These data guide the solver how to perform iterative calculations and how to balance the stability and convergence of the calculations.

[0124] Step S255: performing parallel calculation and solving of the boiler discrete equations according to the boiler numerical solver configuration data to obtain an initial solution of the boiler flow field;

[0125] Specifically, according to the boiler numerical solver configuration data, ANSYS Fluent software can be used to perform parallel calculations and solve the boiler discrete equations. First, configure Fluent's parallel computing environment, which involves setting the number of parallel processors, for example, choosing to use 16 processor cores for calculation. In Fluent, enter the "Run Calculation" menu, select the "Parallel" option, and set the parallel environment parameters, such as the number of processors and load balancing strategy. Next, start the solution process. Fluent will automatically assign tasks to each processor core to process the calculation of the boiler flow field in parallel. For example, set each time step to 0.01 seconds and the total calculation time to 10 seconds, and Fluent will perform parallel iterative calculations for these 1000 time steps. During the calculation process, Fluent will monitor the calculation progress and resource usage of each processor in real time to ensure the efficiency and stability of the calculation. After the calculation is completed, the initial solution of the boiler flow field is obtained, including the distribution of physical quantities such as velocity, pressure and temperature.

[0126] Step S256: performing radiation heat transfer evaluation on the initial solution of the boiler flow field to obtain the boiler radiation heat flux field;

[0127] Specifically, you can use ANSYS Fluent software and enable its radiation heat transfer model to evaluate the radiation heat transfer of the initial solution of the boiler flow field to obtain the boiler radiation heat flux field. Fluent provides a variety of radiation models, such as the P-1 model, DO model, and Monte Carlo model. In this example, the DO model is selected. First, define the radiation characteristics of the medium inside the boiler in Fluent. Considering the typical environment in a glass melting furnace, set the radiation characteristics of the medium to reflect the conditions in the glass melting furnace. For example, the volume fraction of silicate particles commonly found in glass melting furnaces can be set to 0.1%, and its radiation absorption coefficient can be specified to be 0.5m -1 . These silicate particles are typical components in the exhaust gas of glass melting furnaces, and they have a significant effect on radiation heat transfer. Then, set the number of iterations of the radiation model, for example, set it to 100 times to ensure the convergence of the radiation heat transfer calculation. Next, run the radiation heat transfer evaluation. Fluent will calculate the radiation heat flux field inside the boiler based on the initial solution of the flow field and the defined radiation characteristics. After the calculation is completed, the data of the boiler radiation heat flux field is obtained, including the magnitude and direction of the radiation heat flux of each grid cell.

[0128] Step S257: performing convective heat transfer evaluation on the waste heat boiler according to the boiler radiation heat flux field to obtain a boiler heat transfer efficiency diagram;

[0129] Specifically, ANSYS Fluent software can be used to import the data of the boiler radiation heat flux field and evaluate the convective heat transfer of the waste heat boiler to obtain the boiler heat transfer efficiency diagram. In Fluent, activate the convection heat transfer model and set the relevant physical parameters, such as the thermal conductivity, specific heat capacity and density of the fluid. Assume that the medium flowing inside the boiler is glass melting furnace exhaust gas, which has a thermal conductivity of 0.02 W / (m·K), a specific heat capacity of 1100 J / (kg·K) and a density of 0.8 kg / m³. These parameters reflect the typical thermophysical properties of glass melting furnace exhaust gas. Use Fluent's boundary conditions to define the temperature of the boiler wall, for example, set the wall temperature to 800°C. Then, select a suitable convection heat transfer model, such as the k-ε model, to evaluate the convection heat transfer between the fluid and the wall. After setting up in Fluent, run the simulation, the software will calculate the convection heat transfer inside the boiler and generate a heat transfer efficiency diagram. This diagram shows the heat transfer efficiency of each part of the boiler, for example, in the outlet area of ​​the boiler, the convective heat transfer efficiency is 80%. In this way, the convective heat transfer in the waste heat boiler of the glass melting furnace can be accurately evaluated.

[0130] Step S258: performing coupling calculation on the boiler radiation heat flux field and the boiler heat transfer efficiency diagram to obtain a boiler fluid characteristic parameter set.

[0131] Specifically, operations can be performed in ANSYS Fluent to integrate the boiler radiation heat flux field with the boiler heat transfer efficiency map. Use Fluent's post-processing tools to superimpose the two heat transfer effects to obtain a comprehensive boiler fluid characteristic parameter set. In this process, some parameters, such as the weight factors of radiation and convection heat transfer, need to be adjusted to ensure the accuracy of the coupled calculation. For example, according to experimental data or empirical formulas, it is set that radiation heat transfer accounts for 40% of the total heat transfer and convection heat transfer accounts for 60%. Fluent will calculate the total heat transfer of each grid unit based on these proportions and the respective heat transfer data. After the calculation is completed, the boiler fluid characteristic parameter set is obtained, which includes the temperature, velocity, pressure of the fluid, and the contribution of radiation and convection heat transfer. For example, it can be concluded that in a specific area of ​​the boiler, the temperature of the fluid is 1200°C, the velocity is 5m / s, the pressure is 2 atm, the radiation heat transfer is 1000 W / m², and the convection heat transfer is 1500 W / m².

[0132] Preferably, step S3 comprises the following steps:

[0133] Step S31: configuring acoustic sensors for the waste heat boiler to obtain a distribution map of acoustic measurement points of the boiler;

[0134] Specifically, acoustic analysis software, such as LMS Test.Lab, can be used to assist in determining the optimal placement of sensors. First, a three-dimensional model of the boiler is imported, and the propagation path of sound waves inside the boiler is simulated in the software. Considering the shape and size of the boiler, it is decided to arrange acoustic sensors in four directions, front, back, left, right, and top of the boiler. Three sensors are selected to be arranged in the front and back directions of the boiler, two sensors are arranged in the left and right directions, and one sensor is arranged on the top, for a total of 10 sensors. These locations are selected because they can cover the key areas of the boiler and can capture the most representative acoustic data. The acoustic simulation results are used to determine the exact location of each sensor. For example, the first sensor is arranged in the center of the front of the boiler, 0.5 meters away from the boiler wall and 1.5 meters high; the other two sensors are arranged 1 meter on each side of the center of the front. Sensors in other directions are arranged in a similar way. In this way, the distribution map of the acoustic measurement points of the boiler can be obtained.

[0135] Step S32: formulating an acoustic sensor array for the waste heat boiler according to the boiler acoustic measurement point distribution map to obtain an acoustic sensor array layout plan;

[0136] Specifically, the acoustic sensor array layout plan can be formulated according to the distribution map of the acoustic measurement points of the boiler. Choose to use high-performance acoustic sensors, such as the PCB brand Piezoelectric Accelerometer (piezoelectric acceleration sensor), model 375A04. According to the distribution map of the measurement points, the installation position of the sensor is marked on the surface of the boiler, and the sensor is fixed with a high-temperature resistant stainless steel bracket. Each sensor is connected to the central data acquisition device through a data acquisition system, such as the NI USB-4431 of National Instruments. After all sensors are arranged, a field test is carried out to ensure that each sensor can work properly and the data acquisition system can collect data stably. Through the arrangement and testing steps, the acoustic sensor array layout plan can be obtained.

[0137] Step S33: performing signal channel planning for the acoustic sensor array arrangement scheme to obtain an acoustic signal channel configuration table;

[0138] Specifically, signal processing software, such as LabVIEW from National Instruments, can be used to plan and manage the signal channels of acoustic sensors. First, according to the 10 acoustic sensors arranged in step S32, a unique signal channel is assigned to each sensor. Use the NI PXIe-4431 dynamic signal acquisition module from National Instruments, which supports synchronous acquisition of up to 16 channels. In LabVIEW, create a channel configuration for each sensor, including parameters such as sampling rate, resolution, and input range. For example, set the sampling rate to 100 kHz, the resolution to 16 bits, and the input range to ±5 V to ensure that the acoustic signals inside the boiler can be captured. Next, generate an acoustic signal channel configuration table to record the channel number, sampling settings, and physical location of each sensor in detail. For example, the channel configuration of sensor 1 is channel 1, the sampling rate is 100 kHz, the resolution is 16 bits, the input range is ±5 V, and the location is at the center of the front of the boiler. The configurations of other sensors are recorded in the same way.

[0139] Step S34: formulating an acoustic excitation signal for the waste heat boiler according to the acoustic signal channel configuration table to obtain a boiler acoustic wave excitation pulse sequence;

[0140] Specifically, an acoustic excitation signal generator, such as Rohde & Schwarz's SMBV100A, can be used to generate a boiler acoustic wave excitation pulse train. A broadband pulse is selected as the excitation signal because it can excite multiple vibration modes of the boiler structure, which helps to obtain more comprehensive acoustic characteristic data. A program is written in LabVIEW to control the Rohde & Schwarz SMBV100A signal generator and set the parameters of the pulse train, such as a pulse width of 0.1 ms, a pulse interval of 1 s, and an amplitude of 5 V. The selection of these parameters is based on the acoustic properties of the boiler material and the expected detection sensitivity. The excitation signal is then amplified to a sufficient power level through a power amplifier, such as Crest Audio's PA-300, to ensure that the signal can penetrate the boiler wall and excite the internal structure. The amplified excitation signal is connected to the acoustic sensor array so that it synchronously receives the excitation signal and the boiler's response signal. In this way, a boiler acoustic wave excitation pulse train can be obtained.

[0141] Step S35: performing acoustic wave emission control on the acoustic sensor array according to the boiler acoustic wave excitation pulse sequence to obtain a boiler acoustic wave emission synchronization sequence, and performing synchronous sampling control on the boiler acoustic wave emission synchronization sequence to obtain a boiler original acoustic signal set;

[0142] Specifically, the acoustic sensor array can be controlled to emit acoustic waves according to the boiler acoustic wave excitation pulse sequence. A multi-channel acoustic wave emission controller, such as a system composed of a National Instruments PXI platform and corresponding LabVIEW software, is used to control the emission timing of each acoustic sensor. The system is set to ensure that the emission of all sensors is synchronized. In the specific operation, an independent emission channel is configured for each sensor, and the emission controller is set to generate a synchronization signal that matches the excitation pulse sequence defined in step S34. For example, the emission interval of each sensor is set to 1 second, the pulse width is 0.1 milliseconds, and the amplitude is 5 volts. Use LabVIEW software to write a program to control the digital signal generator module (such as NI PXI-4402) of the PXI platform to generate these synchronous emission signals. Next, the boiler acoustic wave emission synchronization sequence is synchronously sampled and controlled. The same PXI platform and LabVIEW software are used to control the data acquisition of the acoustic sensor. The dynamic signal acquisition module (such as NI PXIe-4431) of the PXI platform is configured to synchronously acquire the response signals of all sensors. The sampling rate is set to 200 kHz, and the acquisition time is set to 10 milliseconds. Through the above steps, the original acoustic signal sets of the boiler can be obtained. These signal sets contain the acoustic wave emission and response data collected from each sensor.

[0143] Step S36: performing signal preprocessing on the original acoustic signal set of the boiler to obtain acoustic spectrum acquisition data;

[0144] Specifically, signal processing software, such as MATLAB or LabVIEW, can be used to perform signal preprocessing. First, the raw acoustic signal of each sensor is detrended to remove any DC bias. Next, the signal is processed by a window function, such as a Hamming window, to reduce spectral leakage. The signal after window function processing is subjected to a fast Fourier transform (FFT) to convert the time domain signal into a frequency domain signal. The number of FFT points is selected to be 2048 to ensure that the spectral resolution is high enough, for example, the spectral resolution is 9.765625 Hz. After FFT, the spectrum data is filtered to remove noise and non-correlated frequency components. For example, only the frequency range of 300 Hz to 10 kHz is retained because this range contains the main information of the boiler sound waves. Through the above steps, acoustic spectrum acquisition data can be obtained, which describes the frequency distribution and intensity of the sound waves inside the boiler.

[0145] Step S37: Perform sound velocity characteristic analysis on the acoustic spectrum acquisition data to obtain acoustic-temperature mapping data, and perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field.

[0146] Specifically, please refer to the sub-steps of step S37 for the detailed implementation process of this embodiment.

[0147] Preferably, step S37 includes the following steps:

[0148] Step S371: extracting the sound velocity from the acoustic spectrum acquisition data to obtain the boiler sound wave propagation velocity distribution data;

[0149] Specifically, acoustic analysis software, such as LMS Test.Lab, can be used to process acoustic spectrum acquisition data. Import the acoustic spectrum acquisition data into the Test.Lab software. These data include the frequency distribution of sound waves at different times and different locations. Choose to use the waveform fitting method to extract the sound velocity. Specifically, focus on specific peaks in the spectrum, which correspond to standing wave patterns on the propagation path of sound waves in the boiler. For example, it is observed that there are obvious peaks at 1000 Hz and 2000 Hz, which are formed by the reflection of sound waves between the walls of the boiler. Use the peak detection tool in the Test.Lab software to identify these peaks and measure the frequency difference between them. According to the path length and frequency difference of the sound wave propagating in the boiler, the sound velocity formula is used: v=λf, where v is the sound velocity, λ is the wavelength, and f is the frequency. For example, if the frequency difference is 1000 Hz and the path length corresponding to the wavelength difference is 1.5 meters, then the sound velocity v is 1.5 m / 1000 Hz=1.5 m / s. Through the above operations, the boiler sound wave propagation velocity distribution data can be obtained, which describe the propagation velocity of the sound wave at different positions in the boiler.

[0150] Step S372: constructing a temperature mapping relationship for the boiler sound wave propagation velocity distribution data to obtain acoustic-temperature mapping data;

[0151] Specifically, a mapping relationship between the acoustic wave propagation velocity and the temperature can be constructed based on the boiler acoustic wave propagation velocity distribution data. Use statistical analysis software, such as MATLAB, to process the boiler acoustic wave propagation velocity distribution data. Compare and analyze the acoustic velocity data with the known temperature data inside the boiler, which comes from the optical fiber sensor. Assume that there is a linear relationship between the acoustic velocity and the temperature, and use the least squares method to fit the relationship between the acoustic velocity and the temperature. For example, there are the following data points: at 500°C, the acoustic velocity is 340 m / s; at 600°C, the acoustic velocity is 345 m / s; at 700°C, the acoustic velocity is 350 m / s. Use MATLAB's polyfit function to fit these data points to obtain a linear equation, such as v=0.5T+300, where v is the acoustic velocity (unit: m / s) and T is the temperature (unit: °C). Through the above steps, acoustic-temperature mapping data can be obtained, which describes the relationship between the acoustic wave propagation velocity and the temperature inside the boiler.

[0152] Step S373: performing temperature inversion on the acoustic-temperature mapping data to obtain initial acoustic temperature data;

[0153] Specifically, data processing software, such as MATLAB, can be used to perform temperature inversion. First, import the acoustic-temperature mapping relationship obtained in step S372, which is a linear equation describing the relationship between sound velocity and temperature, for example, v=0.5T+300. Use this mapping relationship to invert the sound wave propagation velocity distribution data. Specifically, apply the mapping equation to each measured sound velocity value and solve the corresponding temperature value. For example, if the sound velocity measurement value at a certain location is 350 m / s, substitute this value into the equation To calculate the temperature, we get Through the above steps, all measured sound velocity values ​​are inverted to obtain the initial acoustic temperature data at each location.

[0154] Step S374: performing spatial interpolation on the initial acoustic temperature data to obtain an initial boiler acoustic temperature field;

[0155] Specifically, spatial interpolation can be performed using geographic information system (GIS) software, such as ArcGIS, or scientific computing software, such as MATLAB. Assume that the initial acoustic temperature data at different locations of the boiler are obtained from the sound velocity measurement, and these data are discrete point data. Choose to use the Kriging interpolation method for spatial interpolation. In ArcGIS, use the "Kriging" tool, input the initial acoustic temperature data as point data, and set the interpolation parameters, such as the semivariogram model and the search radius. For example, setting the search radius to 1 meter means that the interpolation will consider other point data within 1 meter around each point. After running the Kriging interpolation, a continuous temperature field is obtained, which covers the entire boiler area and provides a temperature estimate for each location.

[0156] Step S375: identifying the acoustic wave refraction effect of the initial boiler acoustic temperature field to obtain boiler acoustic wave propagation path distortion data;

[0157] Specifically, acoustic simulation software, such as COMSOL Multiphysics, can be used to identify the acoustic refraction effect. First, the initial boiler acoustic temperature field is imported into COMSOL, and a three-dimensional model corresponding to the boiler geometry is established. In COMSOL, the speed of sound waves at different temperatures is set, and the Sutherland formula is used to describe the effect of temperature on the speed of sound, for example: , where c is the speed of sound, c0 is the speed of sound at a reference temperature, T is the temperature, T0 is the reference temperature, and β and γ are material parameters. Run the simulation, and the software will calculate the speed of sound waves in different temperature regions and predict the sound wave refraction effect caused by the temperature gradient. The simulation results will show the distortion of the sound wave propagation path, for example, the increase in the speed of sound waves in the high temperature area causes the sound wave propagation path to bend. Record these distortion data, including the degree and direction of the path bending, to obtain the boiler sound wave propagation path distortion data.

[0158] Step S376: Perform acoustic wave path compensation on the initial boiler acoustic temperature field according to the boiler acoustic wave propagation path distortion data to obtain the boiler acoustic compensated temperature field, and perform acoustic imaging reconstruction on the boiler acoustic compensated temperature field to obtain the boiler acoustic temperature field.

[0159] Specifically, the acoustic wave path compensation can be performed on the initial boiler acoustic temperature field according to the distortion data of the boiler acoustic wave propagation path. This compensation process is performed using COMSOL Multiphysics software. The acoustic wave propagation path is adjusted to compensate for the refraction effect caused by the temperature gradient. In the specific operation, the propagation path of the acoustic wave is adjusted according to the distortion data. For example, if the simulation shows that the acoustic wave path bends upward in a certain high temperature area, this bending can be compensated by adjusting the path or using a reverse temperature gradient. Rerun the simulation until the acoustic wave path distortion is effectively compensated. After the compensation is completed, the boiler acoustic compensated temperature field is obtained. Next, acoustic imaging software such as ImageJ or MATLAB is used to reconstruct the compensated temperature field. Select to use an image reconstruction algorithm, such as an inverse Fourier transform, to convert the compensated acoustic wave data into an image. For example, the ifft2 function in MATLAB is used to process the compensated data to obtain an image of the boiler acoustic temperature field. This image clearly shows the temperature distribution inside the boiler, including hot spots and temperature gradients. In this way, the boiler acoustic temperature field can be obtained.

[0160] Preferably, step S4 comprises the following steps:

[0161] Step S41: dividing the waste heat boiler into infrared imaging areas to obtain a boiler scanning area plan, and configuring thermal imager points on the waste heat boiler according to the boiler scanning area plan to obtain a thermal imager point position set;

[0162] Specifically, infrared imaging analysis software, such as FLIR Tools+, can be used in combination with the CAD drawings or actual dimensions of the boiler to define different areas of the boiler. The boiler is divided into several areas, for example, according to the length and width of the boiler, it is divided into 5 areas: top, bottom, front, back and sides. Next, the thermal imager points are configured according to the boiler scanning area plan. The FLIR T640 thermal imager is used, which has a high resolution of 640×480 and a temperature measurement accuracy of ±2°C or 2%. According to the size and shape of the boiler, it is decided to arrange 3 thermal imagers in each area of ​​the boiler to ensure comprehensive coverage. For example, in the front of the boiler, the thermal imagers are arranged at a height of 1 meter, 2 meters and 3 meters from the ground, and the horizontal angle of each thermal imager is perpendicular to the boiler surface to ensure the best line of sight. The specific position and angle of each thermal imager are recorded to form a thermal imager point location set. This location set includes the coordinates, height, angle and direction of each thermal imager.

[0163] Step S42: setting thermal imager calibration parameters for the thermal imager point location set to obtain a boiler thermal radiation calibration parameter table;

[0164] Specifically, the FLIR T640 thermal imager and its built-in calibration software and calibration equipment, such as a blackbody radiation source, can be used to set the thermal imager calibration parameters. First, the specific location of each thermal imager is determined based on the infrared imaging area division of the boiler and the thermal imager point location set. For example, three thermal imagers are arranged on the front, rear and two side walls of the boiler, and these locations are carefully selected based on the geometric shape and thermal distribution characteristics of the boiler. Then, the blackbody radiation source is set to a known temperature, such as 300°C, and placed in the center of the field of view of each thermal imager. Using the calibration function of the FLIR T640 thermal imager, the temperature measurement parameters of each thermal imager are adjusted through the software interface. Specifically, the gain, offset and linear coefficient are adjusted until the temperature displayed by the thermal imager is consistent with the actual temperature of the blackbody radiation source. For example, if the actual temperature of the blackbody radiation source is 300°C, adjust the parameters until the temperature displayed by the thermal imager is also 300°C. During the adjustment process, record the calibration parameters of each thermal imager, including gain (e.g. 1.0x), offset (e.g. 0°C), and linear coefficient (e.g. 0.01°C / mA), and associate these parameters with the corresponding thermal imager positions to form a boiler thermal radiation calibration parameter table. The response time, temperature measurement range, and spatial resolution of the thermal imager also need to be set. For example, set the response time to the maximum to capture rapid temperature changes in the boiler; set the temperature measurement range to 200°C to 1200°C to cover the temperature range of the boiler. Through the above calibration and setting steps, a boiler thermal radiation calibration parameter table can be obtained, which contains detailed calibration parameters for each thermal imager position.

[0165] Step S43: performing multi-angle synchronous scanning control on the thermal imager array based on the boiler thermal radiation calibration parameter table to obtain a sequence of original infrared images of the boiler;

[0166] Specifically, the FLIR Tools+ software can be used to manage the FLIR T640 thermal imager array, and the temperature measurement range and resolution of each thermal imager can be set according to the parameters in the boiler thermal radiation calibration parameter table. For example, the temperature measurement range of each thermal imager is set to 200°C to 1200°C, and the resolution is set to 640×480. Use an external synchronization controller, such as the PXIe-8135 timing and triggering module of National Instruments, to synchronize the scanning of all thermal imagers. Set the trigger signal so that all thermal imagers start scanning at the same time after receiving the trigger signal. For example, set the trigger interval to 10 seconds, so that every 10 seconds all thermal imagers will simultaneously capture an infrared image of the boiler. Through the above operations, the original infrared image sequence of the boiler can be obtained, which includes the thermal radiation data of the boiler at different time points and different angles.

[0167] Step S44: performing spatial transformation and resampling on the original infrared image sequence of the boiler to obtain a boiler infrared registration image set;

[0168] Specifically, image processing software, such as MATLAB or ENVI, can be used to perform spatial transformation resampling. First, all original infrared images are imported into the software. Next, the images are spatially transformed to correct image distortion caused by differences in the position and angle of the thermal imager. The perspective transformation method, which is an image correction technique, can correct perspective distortion in the image. For example, four known points on the boiler (such as corners) are selected as reference points, the positions of these points in the image are calculated, and these points are used to define the transformation matrix. Apply this transformation matrix to all images so that the boilers in all images are displayed at the same perspective and scale. Then, the transformed images are resampled to obtain a uniform resolution and pixel size. For example, the resolution of all images is resampled to 320×240. Through these steps, a boiler infrared registration image set can be obtained, which includes registered infrared images of the boiler at different time points.

[0169] Step S45: performing multi-resolution image fusion on the boiler infrared registration image set to obtain the global thermal distribution data of the boiler, and performing temperature quantitative conversion on the global thermal distribution data of the boiler to obtain a surface temperature distribution map of the boiler;

[0170] Specifically, image processing software ENVI and thermal analysis software, such as FLIR Tools+, can be used to perform multi-resolution image fusion. First, load the boiler infrared registration image set in ENVI, which has different spatial resolutions. Choose to use the Laplacian Pyramid method for image fusion. Set the parameters of the fusion algorithm, such as the number of pyramid layers is 3 and the fusion ratio is 0.5. After the fusion is completed, the global thermal distribution data of the boiler is obtained, which is a single image containing the overall thermal information of the boiler. Then, use FLIR Tools+ software to perform quantitative temperature conversion on the fused image. According to the parameters in the boiler thermal radiation calibration parameter table, the pixel values ​​of the thermal image are converted to actual temperature values. For example, the pixel value 1000 is converted to a temperature of 500°C, based on the linear relationship in the calibration parameter table. Through the above steps, the surface temperature distribution map of the boiler can be obtained.

[0171] Step S46: constructing a heat conduction equation for the waste heat boiler based on the boiler surface temperature distribution diagram to obtain a boiler heat conduction mathematical model;

[0172] Specifically, finite element analysis software, such as ANSYS Fluent, can be used to construct the heat conduction equation. Import the boiler surface temperature distribution map into ANSYS Fluent and use it as a boundary condition for heat conduction simulation. Build a three-dimensional geometric model of the boiler and define material properties such as thermal conductivity, specific heat capacity, and density. For example, set the thermal conductivity of the boiler material to 50 W / (m·K) and the specific heat capacity to 500 J / (kg·K). Next, set the heat conduction equation in Fluent, which is a partial differential equation that describes how heat propagates in the material. Choose to use a steady-state heat conduction model and set appropriate boundary conditions, such as ambient temperature and heat flow boundaries. Run the simulation, and Fluent will calculate the heat distribution and flow inside the boiler based on the input temperature distribution map and material properties. The simulation results provide a mathematical model of boiler heat conduction, which is a mathematical expression that describes the heat conduction characteristics inside the boiler. Through the above steps, the mathematical model of boiler heat conduction can be obtained.

[0173] Step S47: Perform temperature field inversion on the waste heat boiler according to the mathematical model of boiler heat conduction to obtain the infrared temperature field of the boiler.

[0174] Specifically, finite element analysis software, such as COMSOL Multiphysics, can be used to perform temperature field inversion. Import the mathematical model of boiler heat conduction into COMSOL Multiphysics. This model includes the thermophysical properties of boiler materials, such as thermal conductivity, specific heat capacity and density, as well as boundary conditions, such as ambient temperature and heat flow boundary. Set the model parameters, for example, assuming that the thermal conductivity of the boiler material is 30 W / (m·K), the specific heat capacity is 600 J / (kg·K), and the density is 7850 kg / m³. Next, use COMSOL's inversion tools, such as parameter estimation or optimization modules, to adjust the model parameters so that the temperature distribution predicted by the model matches the actual measured infrared temperature field. Set the optimization goal, such as minimizing the mean square error between the model predicted temperature and the infrared temperature field. Run the inversion calculation, and COMSOL will automatically adjust the model parameters until the best fit solution is found. For example, it is found that the thermal conductivity initially set needs to be adjusted to 32 W / (m·K) to make the temperature distribution predicted by the model closest to the measured data of the infrared temperature field. In this way, the boiler infrared temperature field can be obtained, which is a temperature distribution map obtained through inversion calculation, which accurately reflects the temperature state inside the boiler.

[0175] Preferably, step S5 comprises the following steps:

[0176] Step S51: performing data reliability evaluation on the initial temperature field of the boiler to obtain an initial temperature credibility distribution map, and performing confidence correction on the initial temperature field of the boiler according to the initial temperature credibility distribution map to obtain an initial weighted temperature field of the boiler;

[0177] Specifically, data analysis software, such as Python with Pandas and NumPy libraries, can be used to process the data of the initial temperature field of the boiler. Import the data of the initial temperature field of the boiler, which comes from the fiber optic sensor grid and the acoustic sensor array. Perform statistical analysis on the data of each sensor, and calculate its standard deviation, mean, and maximum / minimum value to evaluate the stability and consistency of the data. For example, it is found that the standard deviation of sensor 1 is 0.5°C and the standard deviation of sensor 2 is 1.2°C, indicating that the data variability of sensor 2 is large. Based on these statistical indicators, create an initial temperature credibility distribution map. In this figure, the credibility of each sensor is determined by the stability of its data. For example, the sensor data with a standard deviation less than 1°C is set to high credibility, the standard deviation between 1°C and 2°C is set to medium credibility, and the standard deviation greater than 2°C is set to low credibility. Next, the confidence of the initial temperature field of the boiler is corrected according to the initial temperature credibility distribution map. For high-credibility data, its measured value is used directly; for medium-credibility data, the data of neighboring high-credibility sensors are used for weighted averaging; for low-credibility data, interpolation or model prediction methods are used for correction. For example, for sensor 2 with a standard deviation of 1.2°C, the data of neighboring sensors 1 and 3 are taken and weighted averaged according to the distance to correct the data of sensor 2. In this way, the initial weighted temperature field of the boiler can be obtained.

[0178] Step S52: performing model error statistics on the boiler theoretical temperature field according to the boiler infrared temperature field to obtain a predicted temperature error distribution map, and performing error compensation on the boiler theoretical temperature field according to the predicted temperature error distribution map to obtain a boiler theoretical corrected temperature field;

[0179] Specifically, the data of the boiler infrared temperature field and the data of the boiler theoretical temperature field can be imported into the ANSYS Fluent and Python environments. The two sets of data are compared and their temperature differences at each corresponding position are calculated. For example, it is found that in a certain area of ​​the boiler, the temperature shown by the infrared temperature field is 500°C, while the temperature predicted by the theoretical temperature field is 510°C, with a difference of 10°C. These differences are counted to create a predicted temperature error distribution map. In this figure, the error at each location is determined by the difference between the theoretical temperature field and the infrared temperature field. It is found that the error in some areas is large, which is caused by inaccurate model assumptions or improper boundary conditions. Next, the error compensation of the boiler theoretical temperature field is performed according to the predicted temperature error distribution map. Adjust the parameters of the theoretical temperature field, such as thermal conductivity, specific heat capacity, or boundary conditions, to reduce the prediction error. For example, adjust the thermal conductivity in the theoretical model from 50 W / (m·K) to 55 W / (m·K) to better match the data of the infrared temperature field. In this way, the boiler theoretical corrected temperature field can be obtained.

[0180] Step S53: performing inversion accuracy evaluation on the infrared temperature field of the boiler according to the initial temperature field of the boiler to obtain an infrared temperature accuracy distribution map, and performing accuracy calibration on the infrared temperature field of the boiler according to the infrared temperature accuracy distribution map to obtain an infrared calibration temperature field of the boiler;

[0181] Specifically, the data of the initial temperature field of the boiler and the data of the infrared temperature field of the boiler can be imported into MATLAB. These data include temperature readings obtained from different sensors and measuring devices. The two sets of data are compared and analyzed to calculate their temperature differences at each corresponding position. For example, it is found that in a certain area of ​​the boiler, the temperature displayed by the initial temperature field is 800°C, while the temperature displayed by the infrared temperature field is 820°C, with a difference of 20°C. Next, these difference data are used to create an infrared temperature accuracy distribution map. This distribution map shows the measurement accuracy of the infrared temperature field in different areas, with small differences in places with high accuracy and large differences in places with low accuracy. It is found that the infrared temperature measurement accuracy in some areas is low, which is caused by the location of the sensor, environmental factors or limitations of the measuring equipment. Then, the accuracy of the boiler infrared temperature field is calibrated according to the infrared temperature accuracy distribution map. For areas with low accuracy, the data of the initial temperature field is used to adjust the data of the infrared temperature field. For example, if the infrared temperature measurement values ​​in a region are generally 20°C higher, all infrared temperature measurements in the region are calibrated by subtracting 20°C. Through the above steps, the boiler infrared calibration temperature field can be obtained.

[0182] Step S54: spatially registering and fusing the boiler acoustic temperature field, the boiler initial weighted temperature field, the boiler theoretically corrected temperature field and the boiler infrared calibration temperature field to obtain the boiler multi-source registered temperature field;

[0183] Specifically, the data of the boiler acoustic temperature field, the boiler initial weighted temperature field, the boiler theoretically corrected temperature field and the boiler infrared calibration temperature field can be imported into ArcGIS. These data include temperature readings obtained from different measurement methods and equipment, which have different spatial resolutions and coordinate systems. Use ArcGIS's spatial registration tool to register these data to ensure that they are aligned in the same coordinate system. For example, convert the data of the acoustic temperature field from the coordinate system of the acoustic sensor to the coordinate system of the boiler's CAD model. Next, use ArcGIS's fusion tools, such as "Raster Data Fusion" or "Raster Calculator", to merge these temperature fields. Select a suitable fusion method, such as weighted average based on pixel values ​​or position-based interpolation methods, to fuse these data. For example, choose to take the average of four temperature field data at each location as the final temperature value. Through the above operations, the boiler multi-source registered temperature field can be obtained.

[0184] Step S55: Perform physical constraint verification on the boiler multi-source registration temperature field to obtain a set of physical constraint conditions for boiler temperature;

[0185] Specifically, the data of the multi-source registered temperature field of the boiler can be imported into ANSYS Fluent. This data includes temperature readings obtained from different sensors and measurement devices, which have been spatially registered and fused into a unified temperature field. Next, physical constraints are applied to verify this temperature field. These constraints include thermodynamic laws, such as the law of conservation of energy, and thermophysical properties of materials, such as thermal conductivity, specific heat capacity, and density. For example, check whether there are areas of heat accumulation or loss in the temperature field, which indicates errors or inconsistencies in the data. Set up a verification process that includes checking whether the temperature gradient is physically reasonable and whether the temperature change conforms to the basic laws of heat conduction and convection. For example, it is found that in a certain area of ​​the boiler, the temperature gradient is abnormally high, which indicates that the thermal conductivity of the area is set incorrectly or there is an unidentified heat source. Through the above steps, a set of physical constraints for boiler temperature can be obtained, which is a data set containing all physical constraints.

[0186] Step S56: performing constraint optimization on the boiler multi-source registration temperature field according to the boiler temperature physical constraint condition set to obtain the final three-dimensional temperature field of the boiler.

[0187] Specifically, ANSYS Fluent software can be used for constrained optimization. First, adjust the data of the boiler's multi-source registered temperature field according to the set of physical constraints on the boiler's temperature. For example, if the set of physical constraints indicates that the thermal conductivity of a certain area should be higher, adjust the temperature field data of that area accordingly. Use Fluent's optimization tools, such as parameter optimization or response surface methodology, to adjust key parameters in the temperature field. For example, set the objective function to minimize the difference between the temperature field and the physical constraints, and use Fluent's optimization algorithm to find the optimal temperature field parameters. Next, run the optimization calculation, and Fluent will automatically adjust the temperature field parameters until all physical constraints are met. For example, it is found that the temperature of a certain area of ​​the boiler needs to be increased by 5°C to meet the predictions of the heat conduction equation. Through the above operations, the final three-dimensional temperature field of the boiler can be obtained.

[0188] Preferably, the present invention further provides a three-dimensional temperature field construction system for a waste heat boiler, which is used to execute the three-dimensional temperature field construction method for a waste heat boiler as described above. The three-dimensional temperature field construction system for a waste heat boiler comprises:

[0189] The temperature acquisition module is used to deploy a fiber optic sensor grid on the waste heat boiler to obtain a distributed temperature acquisition network; sample the distributed temperature acquisition network to obtain a real-time temperature sequence of the boiler; perform data preprocessing on the real-time temperature sequence of the boiler to obtain the initial temperature field of the boiler;

[0190] Theoretical temperature calculation module: construct the fluid dynamics model parameters of the waste heat boiler based on the initial temperature field of the boiler to obtain the boiler fluid characteristic parameter set; perform multi-physics field coupling on the boiler fluid characteristic parameter set to obtain the boiler theoretical temperature field;

[0191] Acoustic analysis module: Arrange the acoustic sensor array and collect data on the waste heat boiler to obtain the acoustic spectrum collection data; perform sound velocity characteristic analysis on the acoustic spectrum collection data to obtain the acoustic-temperature mapping data; perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field;

[0192] Infrared scanning module: Use infrared thermal imager to scan the waste heat boiler at multiple angles to obtain boiler thermal image sequence; perform spatial registration and fusion on the boiler thermal image sequence to obtain the boiler global thermal distribution data; perform heat conduction inversion based on the boiler global thermal distribution data to obtain the boiler infrared temperature field;

[0193] Temperature field fusion module: The temperature field of the boiler initial temperature field, the boiler theoretical temperature field, the boiler acoustic temperature field and the boiler infrared temperature field are fused to obtain the final three-dimensional temperature field of the boiler.

[0194] Preferably, a computer-readable storage medium stores a method for constructing a three-dimensional temperature field of a waste heat boiler as described above, which can be loaded and executed by a processor.

[0195] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

Claims

1. A method for constructing a three-dimensional temperature field of a waste heat boiler, characterized in that: The following steps are involved: Step S1: deploying a fiber optic sensor grid on the waste heat boiler to obtain a distributed temperature collection network; Sampling the distributed temperature acquisition network to obtain the real-time boiler temperature series; Perform data preprocessing on the boiler's real-time temperature sequence to obtain the boiler's initial temperature field; Step S2: constructing the fluid dynamics model parameters of the waste heat boiler based on the initial temperature field of the boiler to obtain a boiler fluid characteristic parameter set; performing multi-physical field coupling on the boiler fluid characteristic parameter set to obtain a boiler theoretical temperature field; Step S3: deploying an acoustic sensor array and collecting data on the waste heat boiler to obtain acoustic spectrum collection data; Perform sound velocity characteristic analysis on the acoustic spectrum acquisition data to obtain acoustic-temperature mapping data; perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field; Step S4: Use an infrared thermal imager to perform multi-angle scanning on the waste heat boiler to obtain a boiler thermal image sequence; perform spatial registration and fusion on the boiler thermal image sequence to obtain the global thermal distribution data of the boiler; perform heat conduction inversion based on the global thermal distribution data of the boiler to obtain the infrared temperature field of the boiler; Step S5: performing temperature field fusion on the boiler initial temperature field, the boiler theoretical temperature field, the boiler acoustic temperature field and the boiler infrared temperature field to obtain the final three-dimensional temperature field of the boiler.

2. The method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: performing three-dimensional modeling on the waste heat boiler to obtain a boiler structure grid model, and performing optical fiber sensor point distribution calculation on the waste heat boiler based on the boiler structure grid model to obtain an optical fiber sensor point distribution plan; Step S12: deploying optical fiber sensors on the waste heat boiler according to the optical fiber sensor deployment plan to obtain a distributed temperature acquisition network; Step S13: collecting the temperature time series of the waste heat boiler to obtain a boiler historical temperature sequence set, and identifying the temperature gradient rate of the boiler historical temperature sequence set to obtain a boiler temperature dynamic feature set; Step S14: setting the sampling frequency of the distributed temperature acquisition network according to the boiler temperature dynamic feature set to obtain a sampling parameter configuration table; Step S15: performing data acquisition on the distributed temperature acquisition network according to the sampling parameter configuration table to obtain the boiler original temperature data stream, and performing abnormal value detection on the boiler original temperature data stream to obtain a boiler temperature abnormality mark sequence; Step S16: performing data repair on the boiler original temperature data stream according to the boiler temperature abnormality mark sequence to obtain the boiler real-time temperature sequence; Step S17: Calculate the temperature gradient of the boiler real-time temperature sequence to obtain the boiler temperature gradient field, and identify the heat flux distribution of the waste heat boiler based on the boiler temperature gradient field to obtain the heat flux density distribution data inside the boiler; perform data fusion on the boiler real-time temperature sequence and the heat flux density distribution data inside the boiler to obtain the boiler initial temperature field.

3. The method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: evaluating the density of the medium in the waste heat boiler based on the initial temperature field of the boiler to obtain the density distribution data of the medium in the furnace; Step S22: evaluating the fluid viscosity of the density distribution data of the medium in the furnace to obtain boiler temperature-viscosity correlation data; Step S23: Calculate the Reynolds number of the waste heat boiler according to the boiler temperature-viscosity correlation data to obtain boiler fluid turbulence characteristic parameters; Step S24: selecting a turbulence model for the waste heat boiler based on the boiler fluid turbulence characteristic parameters to obtain a boiler turbulence calculation model; Step S25: performing numerical simulation of the flow field of the waste heat boiler based on the boiler turbulence calculation model to obtain a boiler fluid characteristic parameter set; Step S26: Perform temperature field iterative calculation according to the boiler fluid characteristic parameter set to obtain the boiler theoretical temperature field.

4. The method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 3, characterized in that: Step S25 includes the following steps: Step S251: setting boundary conditions for the waste heat boiler to obtain a boiler boundary constraint parameter set; Step S252: meshing the boiler turbulence calculation model according to the boiler boundary constraint parameter set to obtain a boiler turbulence calculation grid model; Step S253: discretizing the control equations of the boiler turbulence calculation grid model to obtain a boiler discrete equation group; Step S254: performing SIMPLE algorithm solution configuration on the boiler discrete equation group to obtain boiler numerical solver configuration data; Step S255: performing parallel calculation and solving of the boiler discrete equations according to the boiler numerical solver configuration data to obtain an initial solution of the boiler flow field; Step S256: performing radiation heat transfer evaluation on the initial solution of the boiler flow field to obtain the boiler radiation heat flux field; Step S257: performing convective heat transfer evaluation on the waste heat boiler according to the boiler radiation heat flux field to obtain a boiler heat transfer efficiency diagram; Step S258: performing coupling calculation on the boiler radiation heat flux field and the boiler heat transfer efficiency diagram to obtain a boiler fluid characteristic parameter set.

5. The method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: configuring acoustic sensors for the waste heat boiler to obtain a distribution map of acoustic measurement points of the boiler; Step S32: formulating an acoustic sensor array for the waste heat boiler according to the boiler acoustic measurement point distribution map to obtain an acoustic sensor array layout plan; Step S33: performing signal channel planning for the acoustic sensor array arrangement scheme to obtain an acoustic signal channel configuration table; Step S34: formulating an acoustic excitation signal for the waste heat boiler according to the acoustic signal channel configuration table to obtain a boiler acoustic wave excitation pulse sequence; Step S35: performing acoustic wave emission control on the acoustic sensor array according to the boiler acoustic wave excitation pulse sequence to obtain a boiler acoustic wave emission synchronization sequence, and performing synchronous sampling control on the boiler acoustic wave emission synchronization sequence to obtain a boiler original acoustic signal set; Step S36: performing signal preprocessing on the original acoustic signal set of the boiler to obtain acoustic spectrum acquisition data; Step S37: Perform sound velocity characteristic analysis on the acoustic spectrum acquisition data to obtain acoustic-temperature mapping data, and perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field.

6. The method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 5, characterized in that: Step S37 The following steps are involved: Step S371: extracting the sound velocity from the acoustic spectrum acquisition data to obtain the boiler sound wave propagation velocity distribution data; Step S372: constructing a temperature mapping relationship for the boiler sound wave propagation velocity distribution data to obtain acoustic-temperature mapping data; Step S373: performing temperature inversion on the acoustic-temperature mapping data to obtain initial acoustic temperature data; Step S374: performing spatial interpolation on the initial acoustic temperature data to obtain an initial boiler acoustic temperature field; Step S375: identifying the acoustic wave refraction effect of the initial boiler acoustic temperature field to obtain boiler acoustic wave propagation path distortion data; Step S376: Perform acoustic wave path compensation on the initial boiler acoustic temperature field according to the boiler acoustic wave propagation path distortion data to obtain the boiler acoustic compensated temperature field, and perform acoustic imaging reconstruction on the boiler acoustic compensated temperature field to obtain the boiler acoustic temperature field.

7. The method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: dividing the waste heat boiler into infrared imaging areas to obtain a boiler scanning area plan, and configuring thermal imager points on the waste heat boiler according to the boiler scanning area plan to obtain a thermal imager point position set; Step S42: setting thermal imager calibration parameters for the thermal imager point location set to obtain a boiler thermal radiation calibration parameter table; Step S43: performing multi-angle synchronous scanning control on the thermal imager array based on the boiler thermal radiation calibration parameter table to obtain a sequence of original infrared images of the boiler; Step S44: performing spatial transformation and resampling on the original infrared image sequence of the boiler to obtain a boiler infrared registration image set; Step S45: performing multi-resolution image fusion on the boiler infrared registration image set to obtain the global thermal distribution data of the boiler, and performing temperature quantitative conversion on the global thermal distribution data of the boiler to obtain a surface temperature distribution map of the boiler; Step S46: constructing a heat conduction equation for the waste heat boiler based on the boiler surface temperature distribution diagram to obtain a boiler heat conduction mathematical model; Step S47: Perform temperature field inversion on the waste heat boiler according to the mathematical model of boiler heat conduction to obtain the infrared temperature field of the boiler.

8. The method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: performing data reliability evaluation on the initial temperature field of the boiler to obtain an initial temperature credibility distribution map, and performing confidence correction on the initial temperature field of the boiler according to the initial temperature credibility distribution map to obtain an initial weighted temperature field of the boiler; Step S52: performing model error statistics on the boiler theoretical temperature field according to the boiler infrared temperature field to obtain a predicted temperature error distribution map, and performing error compensation on the boiler theoretical temperature field according to the predicted temperature error distribution map to obtain a boiler theoretical corrected temperature field; Step S53: performing inversion accuracy evaluation on the infrared temperature field of the boiler according to the initial temperature field of the boiler to obtain an infrared temperature accuracy distribution map, and performing accuracy calibration on the infrared temperature field of the boiler according to the infrared temperature accuracy distribution map to obtain an infrared calibration temperature field of the boiler; Step S54: spatially registering and fusing the boiler acoustic temperature field, the boiler initial weighted temperature field, the boiler theoretically corrected temperature field and the boiler infrared calibration temperature field to obtain the boiler multi-source registered temperature field; Step S55: Perform physical constraint verification on the boiler multi-source registration temperature field to obtain a set of physical constraint conditions for boiler temperature; Step S56: performing constraint optimization on the boiler multi-source registration temperature field according to the boiler temperature physical constraint condition set to obtain the final three-dimensional temperature field of the boiler.

9. A three-dimensional temperature field construction system for a waste heat boiler, characterized in that: Used to execute the method for constructing a three-dimensional temperature field of a waste heat boiler according to claim 1, the three-dimensional temperature field construction system of the waste heat boiler comprises: The temperature acquisition module is used to deploy a fiber optic sensor grid on the waste heat boiler to obtain a distributed temperature acquisition network; sample the distributed temperature acquisition network to obtain a real-time temperature sequence of the boiler; perform data preprocessing on the real-time temperature sequence of the boiler to obtain the initial temperature field of the boiler; Theoretical temperature calculation module: construct the fluid dynamics model parameters of the waste heat boiler based on the initial temperature field of the boiler to obtain the boiler fluid characteristic parameter set; perform multi-physics field coupling on the boiler fluid characteristic parameter set to obtain the boiler theoretical temperature field; Acoustic analysis module: Arrange the acoustic sensor array and collect data on the waste heat boiler to obtain the acoustic spectrum collection data; perform sound velocity characteristic analysis on the acoustic spectrum collection data to obtain the acoustic-temperature mapping data; perform temperature field inversion on the acoustic-temperature mapping data to obtain the boiler acoustic temperature field; Infrared scanning module: Use infrared thermal imager to scan the waste heat boiler at multiple angles to obtain boiler thermal image sequence; perform spatial registration and fusion on the boiler thermal image sequence to obtain the boiler global thermal distribution data; perform heat conduction inversion based on the boiler global thermal distribution data to obtain the boiler infrared temperature field; Temperature field fusion module: The temperature field of the boiler initial temperature field, the boiler theoretical temperature field, the boiler acoustic temperature field and the boiler infrared temperature field are fused to obtain the final three-dimensional temperature field of the boiler.

10. A computer-readable storage medium, characterized in that: The method for constructing a three-dimensional temperature field of a waste heat boiler as claimed in any one of claims 1 to 8 is stored and can be loaded and executed by a processor.

Images