An online monitoring and early warning method and system for coking of an organic heat carrier boiler tube

Abstract

The application discloses an online monitoring and early warning method and system for furnace tube coking of an organic heat carrier boiler, and belongs to the technical field of industrial boiler safe operation and maintenance, and comprises the following steps: S1, collecting multi-source operation data of the organic heat carrier boiler; S2, pre-processing and feature extraction of the data; S3, construction of a coking degree-temperature inversion model; S4, dynamic evaluation of coking degree quantization; and S5, hierarchical early warning and differential operation response. Based on a large number of high-temperature deterioration experiments and CFD numerical simulation results, a multi-physical field coupling coking quantization evaluation model is constructed, real-time dynamic monitoring and accurate hierarchical quantization evaluation of the furnace tube coking state are realized, a multi-level early warning response mechanism and full-dimensional data traceability analysis function are matched, coking safety hazards are effectively identified in advance, fine operation and maintenance of equipment are guided, the incidence of boiler coking accidents is reduced from the source, and the intelligent safety supervision level of special equipment is improved.

Description

Technical Field

[0001] This invention belongs to the fields of intelligent equipment monitoring, safe operation and maintenance of industrial boilers, and Internet of Things big data technology, specifically relating to an online monitoring and early warning method and system for coking of boiler tubes in organic heat carrier boilers. Background Technology

[0002] Organic heat carrier boilers are special heat energy conversion devices that use heat transfer oil as the heat transfer medium. Due to their advantages of high temperature and low pressure, uniform heat transfer, and high temperature control accuracy, they are widely used in industrial production fields such as chemical, textile, food, and building materials. During long-term circulation at high temperatures of 350℃~500℃, heat transfer oil is highly susceptible to chain chemical reactions such as thermal decomposition, oxidation, and polymerization. This leads to a decrease in the flash point, an increase in kinematic viscosity, an increase in acid value, and an increase in residual carbon content, resulting in continuous deterioration of the oil's physical properties. The deteriorated heat transfer oil generates coking precursors such as gums, asphaltenes, and carbonaceous particles, which continuously adhere to the inner wall of the furnace tubes, gradually accumulating to form a dense coking layer.

[0003] Coking of furnace tubes is a core hidden danger restricting the safe operation of organic heat carrier boilers: on the one hand, the low thermal conductivity of the coking layer will significantly reduce the heat exchange efficiency of the furnace tubes, resulting in a large amount of energy waste and increasing the energy consumption cost of enterprise production; on the other hand, the coking layer will hinder the forced convection heat exchange of heat transfer oil, leading to local heat accumulation in the furnace tubes and abnormal temperature rise on the wall surface, which can easily cause major safety accidents such as overheating deformation, bulging, cracking or even tube rupture of the furnace tubes, seriously threatening the safety of industrial production and the safety of personnel and property.

[0004] Currently, there are significant shortcomings in the industry's detection and prevention technologies for coking in boiler tubes of organic heat carrier boilers. The specific deficiencies of existing technical solutions are as follows:

[0005] 1. The monitoring method is outdated and cannot achieve dynamic sensing. Current coking detection mainly relies on manual periodic sampling to conduct offline analysis of indicators such as viscosity, acid value, and flash point of the heat transfer oil, and on human experience to indirectly determine whether coking has occurred. This method has a long detection cycle, is cumbersome to operate, and can only obtain static data at a certain moment. It cannot track the dynamic evolution of heat transfer oil deterioration and furnace tube coking in real time, making it difficult to detect early coking risks.

[0006] 2. Limited monitoring dimensions and low assessment accuracy. Traditional monitoring equipment can only collect basic boiler operating parameters (temperature, pressure, flow rate), and cannot establish a coupled correlation model between furnace tube wall temperature, heat transfer oil flow field characteristics, and coking degree. It cannot accurately quantify coking thickness and coking risk, and relies solely on manual experience for judgment, which is highly subjective and prone to errors.

[0007] 3. Lack of targeted tiered early warning and maintenance guidance mechanism. The existing early warning system only sets a single over-temperature alarm threshold, which cannot distinguish between mild, moderate and severe coking conditions, and cannot match corresponding maintenance strategies. This easily leads to over-maintenance or missed hazard detection, resulting in high maintenance costs and poor safety control.

[0008] 4. Lack of data traceability and trend analysis capabilities. Traditional monitoring methods lack a systematic data storage and analysis system, making it impossible to trace the entire process of coking development, predict coking evolution trends, and support predictive maintenance of equipment and optimization of boiler operating parameters.

[0009] In summary, existing technologies cannot meet the industry's needs for real-time monitoring, accurate assessment, graded early warning, and intelligent operation and maintenance of coking in boiler tubes of organic heat carrier boilers. There is an urgent need to develop an intelligent online monitoring and early warning technology solution for coking based on the coupling of experimental data and numerical simulation. Summary of the Invention

[0010] The technical problem solved by this invention: The purpose of this invention is to address the shortcomings of existing coking monitoring technologies for organic heat carrier boilers, such as lagging performance, inaccurate assessments, limited early warning systems, and inefficient operation and maintenance, by providing an online monitoring and early warning method and system for coking in boiler tubes of organic heat carrier boilers.

[0011] Technical Solution: To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0012] A method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers includes the following steps:

[0013] S1: Collect multi-source operation data of organic heat carrier boilers;

[0014] S2: Data preprocessing and feature extraction;

[0015] S3: Construct a coking degree-temperature inversion model;

[0016] S4: Quantitative dynamic assessment of coking degree;

[0017] S5: Tiered early warning and differentiated operation and maintenance response;

[0018] S6: Full data storage and coking source analysis.

[0019] Furthermore, in step S1, a multi-dimensional sensor group is installed in the furnace, coiled furnace tubes, and heat transfer oil circulation pipeline of the organic heat carrier boiler to collect multi-source parameters of boiler operation in real time. The multi-source parameters include the internal temperature of the furnace, the outer wall temperature of the inner / middle / outer coils, the inlet and outlet temperatures of the heat transfer oil, the inlet and outlet pressures of the heat transfer oil, and the real-time flow rate of the heat transfer oil.

[0020] Furthermore, in step S2, the collected data undergoes outlier removal, filtering and noise reduction, and data calibration preprocessing to extract core characteristic parameters such as the average temperature of the outer wall of the furnace tube, temperature fluctuation amplitude, temperature difference between the inlet and outlet of the heat transfer oil, and flow stability.

[0021] Further, in step S3, firstly, a three-dimensional numerical model is established, including the combustion chamber, flue gas flow, radiative heat transfer, and flow and heat transfer within the heat transfer oil pipe, to simulate the temperature response characteristics of the furnace tube outer wall under different coking degrees; secondly, through high-temperature dynamic coking experiments, the variation law of coke generation in the heat transfer oil with temperature and time is obtained, and the kinetic parameters of the coking reaction are calibrated; the experimental data are correlated with the simulation results to correct the numerical model; then, based on the calibrated simulation data, a quantitative mathematical relationship model between the coking degree and the online monitorable parameter, the outer wall temperature, is fitted and established.

[0022] Furthermore, the quantitative mathematical relationship model between the degree of coking and the online monitorable parameter of outer wall temperature is specifically an outer wall temperature inversion model:

[0023] ;

[0024] in, Indicates the degree of coking (%). Indicates the temperature of the outer wall (°C). A univariate inversion mapping function representing the temperature of the outer wall to the degree of coking.

[0025] Furthermore, the model is an exponential function, and the furnace tube outer wall temperature inversion model is expressed as:

[0026] ;

[0027] in, The degree of coking in the furnace tubes (%, ranging from 0 to 100%). The average temperature of the outer wall of the furnace tube after pretreatment (°C); It is a dimensionless amplitude coefficient; The temperature sensitivity coefficient (1 / ℃); The reference temperature (°C, corresponding to) (Temperature intercept obtained by extrapolating the fitted curve when =0%).

[0028] Furthermore, in step S4, the coking degree assessment compares the calculated coking degree ω value with a preset safety threshold, classifying the coking state of the furnace tube into four levels:

[0029] (1) Normal state: When When the percentage is less than 5%, it is considered a safe state, and the equipment operating parameters are stable.

[0030] (2) Mild coking: When 5% ≤ When the concentration is less than 15%, coking tends to occur on the inner wall of the furnace tube, and the heat exchange efficiency shows a slight downward trend.

[0031] (3) Moderate coking: When 15% ≤ When the heat exchange efficiency is less than 35%, the coking layer on the inner wall of the furnace tube thickens significantly, the heat exchange efficiency decreases significantly, and there is a risk of local overheating.

[0032] (4) Severe coking: When When the concentration is ≥35%, severe coking occurs on the inner wall of the furnace tube, resulting in a significant decrease in heat exchange efficiency and an extremely high risk of high-temperature tube rupture.

[0033] Furthermore, in step S5, based on the coking level calculated by the model, the corresponding early warning level and operation and maintenance handling strategy are matched, and early warning signals are pushed synchronously through multiple terminals to achieve differentiated safety prevention and control.

[0034] Furthermore, in step S6, the cloud platform stores all raw monitoring data, preprocessed characteristic parameters, coking assessment results, early warning records and operation and maintenance records in real time, and automatically generates furnace tube temperature change curves, coking trend evolution reports, and equipment operation energy consumption reports.

[0035] An online monitoring and early warning system for coking in boiler tubes of an organic heat carrier boiler, used in the above method, includes a data acquisition layer, a cloud platform layer, and an application terminal layer;

[0036] The data acquisition layer includes a multi-dimensional sensor array and a data acquisition adapter, providing the system with raw operational data in all dimensions;

[0037] The cloud platform layer includes a data processing module, a coking assessment module, an early warning management module, and a data storage module;

[0038] At the application terminal level, it provides operation and maintenance personnel with visual operation and remote operation and maintenance services, including web management terminal, on-site monitoring screen, and mobile phone APP.

[0039] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0040] 1. Breaking through the limitations of traditional static monitoring, achieving dynamic monitoring around the clock. This invention uses a high-frequency, multi-dimensional sensor network for data acquisition, combined with real-time data processing in the cloud, completely solving the problems of lag and data dispersion in traditional offline sampling and monitoring. It can track the dynamic evolution of coking in furnace tubes 24 hours a day without interruption.

[0041] 2. The model is accurate and reliable, with high precision in coking quantification assessment. The coking assessment model of this invention is constructed based on a large amount of data from high-temperature degradation experiments at 400℃~600℃, CFD combustion heat transfer simulation, and coil flow coking simulation. It accurately couples the correlation mechanism between temperature, flow rate, flow field, and coking, abandoning the manual experience-based judgment mode, and realizing the quantitative and accurate calculation of the degree of coking. The assessment accuracy is improved by more than 90% compared with traditional technology.

[0042] 3. Tiered early warning and precise operation and maintenance provide highly targeted prevention and control. This invention divides the coking level into four levels and a three-level early warning mechanism, matching differentiated operation and maintenance solutions. It can accurately distinguish between normal operation, potential hidden dangers, and high-risk fault conditions, avoiding resource waste caused by over-maintenance and preventing safety accidents caused by missed hidden dangers, thus greatly improving the level of precision in boiler operation and maintenance.

[0043] 4. Multi-terminal collaborative operation and maintenance with a high degree of intelligence. The system relies on an IoT cloud platform to achieve collaborative management and control of multiple terminals such as web, large screen, and mobile devices. It supports remote real-time monitoring, early warning push, and data traceability, which significantly reduces the cost of manual operation and maintenance and improves the efficiency of intelligent supervision of special equipment.

[0044] 5. End-to-end data accumulation supports long-term optimization. The system realizes data storage throughout the entire process of operation, evaluation, early warning, and maintenance, and can trace the evolution of coking. It provides big data support for boiler structure optimization, heat transfer oil selection, and operating parameter tuning, helping the industrial boiler industry to upgrade to intelligent and energy-saving levels.

[0045] 6. The model is highly interpretable, and the physical meaning of the parameters is clear. The coking degree-temperature inversion model constructed in this invention is based on the heat transfer mechanism and Arrhenius-type coking reaction kinetics. The parameters in the model (coking sensitivity coefficient)... Temperature sensitivity coefficient Reference temperature All of these (etc.) have clear physical meanings, avoiding the lack of interpretability of pure black box models, and making them easier for engineers to understand and apply.

[0046] 7. Multi-dimensional cross-validation ensures high reliability. Based on the temperature inversion model, this invention introduces a heat transfer coefficient inversion module and an online coking thermal resistance identification method, using the increment of coking thermal resistance... With temperature inversion results Cross-validation effectively avoids misjudgments caused by a single monitoring dimension, making the evaluation results more reliable.

[0047] 8. Possesses the ability to predict coking trends and achieve proactive early warning. This invention introduces a coking time factor to construct a time evolution model of coking degree, which can predict the coking development trend in the future. It can issue trend warnings in advance before the coking degree reaches the warning threshold, leaving sufficient time window for operation and maintenance personnel to intervene, and realizing the leap from passive response to proactive prevention.

[0048] In summary, based on extensive high-temperature degradation experiments and CFD numerical simulation results, this invention constructs a multi-physics field coupled coking quantitative assessment model to achieve real-time dynamic monitoring and precise graded quantitative assessment of the coking status of furnace tubes. It is equipped with a multi-level early warning response mechanism and a full-dimensional data traceability analysis function, which can effectively identify coking safety hazards in advance, guide the refined operation and maintenance of equipment, reduce the incidence of boiler coking accidents from the source, and improve the level of intelligent safety supervision of special equipment. Attached Figure Description

[0049] Figure 1 This is an overall flowchart of the method for online monitoring and early warning of coking in organic heat carrier boilers according to the present invention;

[0050] Figure 2 This is a schematic diagram of the three-layer architecture of the organic heat carrier boiler coking online monitoring and early warning system of the present invention;

[0051] Figure 3 This is a schematic diagram of the layout structure of the temperature sensor on the outer wall of the furnace tube according to the present invention;

[0052] Figure 4 The numerical simulation curve of the furnace tube outer wall temperature changing with the degree of coking;

[0053] Figure 5 This is a block diagram of the computational logic of the multiphysics coupling cohesion evaluation model of the present invention.

[0054] Among them, 1. Furnace tube, 2. Temperature sensor. Detailed Implementation

[0055] The present invention will be further illustrated below with reference to specific embodiments. These embodiments are implemented based on the technical solutions of the present invention, and it should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0056] Example 1

[0057] This embodiment of a method for online monitoring and early warning of coking in boiler tubes of an organic heat carrier boiler includes the following steps:

[0058] S1: Collect multi-source operation data of organic heat carrier boilers.

[0059] Multi-dimensional sensor groups are installed in the furnace, coils, furnace tubes, and heat transfer oil circulation pipelines of the organic heat carrier boiler. The sensor groups collect multi-source parameters of boiler operation in real time at a high frequency of 4 to 6 seconds / time. Specifically, these parameters include: furnace internal temperature, outer wall temperature of inner / middle / outer coils, heat transfer oil inlet and outlet temperature, heat transfer oil inlet and outlet pressure, and real-time flow rate of heat transfer oil.

[0060] Among them, 3 to 5 temperature sensors on the outer wall of the furnace tube are evenly distributed around the back-fire side of the outer coil (high-risk area for coking) to cover the key heating area of ​​the boiler and avoid local monitoring blind spots; flow sensors and pressure sensors are installed at the inlet and outlet of the main circulation pipeline of heat transfer oil to ensure that the collected data can accurately reflect the overall operating status of the boiler; all raw data collected by the sensors are uploaded to the cloud monitoring platform in real time through the data acquisition adapter.

[0061] S2: Data preprocessing and feature extraction

[0062] The cloud platform receives the collected data, completes outlier removal, filtering and noise reduction, data calibration and preprocessing, and extracts core characteristic parameters such as the average temperature of the outer wall of the furnace tube, temperature fluctuation range, temperature difference between the inlet and outlet of the heat transfer oil, and flow stability.

[0063] In this embodiment, during equipment operation, the sensor array collects real-time data on furnace temperature, outer coil outer wall temperature, heat transfer oil inlet and outlet temperatures, heat transfer oil flow rate, and furnace pressure. All signals are connected to a data acquisition adapter supporting the Modbus TCP protocol. After initial noise reduction at the edge, the adapter sends a data packet containing a timestamp, device ID, and outer wall temperature at each point to the cloud platform every 5 seconds via a 4G network. The cloud platform filters out sensor faulty values ​​using threshold filtering, performs data filtering and noise reduction using a 5-point moving average method, and completes data calibration based on the equipment's factory calibration parameters to eliminate system errors.

[0064] The specific steps of threshold filtering and the 5-point moving average method are as follows:

[0065] Assume the sensor is at time... The collected raw temperature sequence is First, threshold filtering is performed:

[0066] ;

[0067] in, The temperature sequence (°C) after threshold filtering. For the sensor at any time The original temperature (°C) was collected. For a moment The effective temperature (°C) of the adjacent previous moment. and These are the lower and upper temperature thresholds set according to the rated operating conditions of the equipment (in this embodiment) =345℃, =750℃), temperature values ​​outside this range are considered abnormal values ​​caused by sensor malfunction or signal interference, and are replaced by the valid value of the adjacent previous moment.

[0068] Subsequently, a 5-point moving average method was used for filtering and noise reduction.

[0069] ;

[0070] in, It is the smoothed temperature value output after 5-point moving average smoothing.

[0071] For regions with fewer than 5 points at either end of the sequence, a 2- or 3-point local averaging is used. The filtered temperature data is then combined with the equipment's factory calibration parameters for linear calibration.

[0072] ;

[0073] in, For temperature data after linear calibration, b and b are the calibration slope and intercept of each sensor channel, respectively, which are determined by multi-point temperature calibration experiments before the equipment leaves the factory.

[0074] S3. Construct a coking degree-temperature inversion model.

[0075] S31: First, a three-dimensional numerical model is established, including the combustion chamber, flue gas flow, radiative heat transfer, and flow and heat transfer within the heat transfer oil pipe, to simulate different degrees of coking. (Expressed as a percentage of the coke layer thickness to the inner radius of the furnace tube) the outer wall temperature T of the furnace tube. outwall Response characteristics.

[0076] The three-dimensional numerical model includes the differential equations for heat conduction in the furnace tubes, the boundary conditions for the coke layer on the inner wall of the furnace tubes, the radiation-convection coupled heat transfer boundary on the outer wall of the furnace tubes, and the continuity and momentum equations for the flow within the heat transfer oil pipes. The specific details are as follows:

[0077] (1) The differential equation for heat conduction of the furnace tube (cylindrical coordinate system) is:

[0078] ;

[0079] in: Density of furnace tube material (kg / m) 3 ), The specific heat capacity at constant pressure of the furnace tube material (J / (kg·K)). Temperature (K) For time (s), For radial coordinates (m). Circumferential coordinates (rad) y is the axial coordinate (m). The thermal conductivity of the furnace tube material is (W / (m·K)). Internal heat source term (W / m) 3 ).

[0080] (2) Boundary conditions for coke layer on the inner wall of furnace tube (third type of boundary condition):

[0081] ;

[0082] in: The thermal conductivity of the furnace tube wall is (W / (m·K)). The inner radius of the furnace tube is (m). The convective heat transfer coefficient of the heat transfer oil (W / (m)) 2 ·K)), The temperature of the inner wall of the furnace tube (K). The mainstream temperature (K) of the heat transfer oil. Heat release of the coking layer (W / m) 2 ), which is generated by the exothermic effect of the coking reaction.

[0083] (3) Radiation-convection coupled heat transfer boundary between the outer wall of the furnace tube and the flue gas:

[0084] ;

[0085] in: The outer radius of the furnace tube is (m). The flue gas emissivity of the outer wall of the furnace tube (dimensionless). The Stefan-Boltzmann constant is 5.67 × 10⁻⁶. -8 W / (m 2 ·K 4 )), The temperature of the outer wall of the furnace tube (K). The flue gas temperature is (K). The flue gas convective heat transfer coefficient (W / (m)) 2 ·K)).

[0086] (4) The continuity equation and momentum equation for the flow in the heat transfer oil pipe (using...) (Turbulence model closed)

[0087] ;

[0088] in: Density of heat transfer oil (kg / m³) 3 ), is the flow velocity vector (m / s). Pressure (Pa). Effective viscosity (Pa·s). Acceleration due to gravity (m / s²) 2 ).

[0089] The four three-dimensional numerical models described above constitute the complete set of governing equations for the multiphysics coupled simulation in this application: the furnace tube heat conduction differential equation serves as the main governing equation, solving for the three-dimensional temperature field of the tube wall and providing simulation data of the outer wall temperature for the inversion of the coking degree; the boundary conditions of the coking layer on the inner wall of the furnace tube bidirectionally couple the coking reaction kinetics with the heat transfer process, realizing the quantitative expression of the coking thermal resistance effect; the radiation-convective coupled heat transfer boundary of the outer wall of the furnace tube accurately characterizes the radiation-dominated heat input mechanism in the furnace of the organic heat carrier boiler, ensuring the accuracy of the temperature field simulation and providing a simulation basis for the equivalent layout of sensors; the continuity equation and momentum equation of the flow in the heat transfer oil pipe represent the convective heat transfer coefficient. The calculations provide flow parameters that reflect the regulatory effect of flow velocity changes on coking behavior, providing a quantitative basis for flow velocity optimization decisions in graded early warning. These four elements, when connected in series, form a complete mathematical description of the entire heat transfer process: flue gas heating, furnace tube heat conduction, coking layer heat resistance, and heat transfer oil heat absorption. This provides a parameterized correction framework for the correlation correction between experimental data and simulation results in step S32, and provides a source of simulation data with complete thermophysical meaning for fitting the coking degree-temperature inversion model in step S33. All four are indispensable and together constitute the core engine of digital twin simulation.

[0090] S32: Experimental Correlation and Model Calibration: Through high-temperature dynamic coking experiments, the variation of heat transfer oil coke formation with temperature and time was obtained, and the Arrhenius-type coking reaction kinetic parameters were calibrated. These kinetic parameters include the pre-exponential factor. and activation energy The experimental data were correlated with the simulation results, and the least squares method was used to correct the numerical model to ensure its accuracy.

[0091] 1. In this embodiment, the specific method of the high-temperature dynamic coking experiment is as follows:

[0092] (1) Experimental setup: A self-built high-temperature dynamic coking experimental rig for heat transfer oil was used, including a heat transfer oil storage tank, a high-temperature circulating pump, an electrically heated tubular test section (20# carbon steel pipe with an inner diameter of 25mm and a length of 1000mm), a temperature control system, a pressure monitoring system, and a sampling and analysis system. Six K-type thermocouples were evenly arranged along the axial direction on the outer wall of the test section to monitor the temperature distribution of the pipe wall in real time.

[0093] (2) Experimental conditions: Lianyuan L-QB300 heat transfer oil was selected as the experimental medium. The experimental temperature range was 400℃~600℃ (5 temperature conditions were set at 50℃ intervals). The heat transfer oil flow rate was set to 2~6 m / s (5 flow rate conditions were set at 1 m / s intervals). The experimental duration was 0~500 hours (sampling was performed every 50 hours).

[0094] (3) Coking amount measurement method: After each experimental condition is completed, the test section of the pipe is cut and sampled, and the coking amount is measured by weighing:

[0095]

[0096] in, The increment in coking mass (g). The mass (g) of the pipe section after coking. The mass (g) of the cleaned pipe section before the experiment, and the average thickness of the coking layer. Calculated from coking quality increment:

[0097]

[0098] in, The density of the coke layer (taken as 1.2~1.5 g / cm³) 3 (determined by actual measurement) The inner surface area of ​​the pipe (m²) 2 ), The length of the pipe section is (m).

[0099] (4) Calibration of coking reaction kinetic parameters: Based on experimental data, the coking rate was described using the Arrhenius-type coking reaction kinetic equation:

[0100]

[0101] in, Coking rate (mm / h). The pre-exponential factor is (mm / h). R is the activation energy of the coking reaction (J / mol), and R is the universal gas constant (8.314 J / (mol·K)). The temperature of the inner wall of the furnace tube (K). The correction function reflects the influence of flow state on coking, where Re is the Reynolds number and Pr is the Prandtl number.

[0102] Nonlinear regression fitting was performed on the experimental data to calibrate the pre-exponential factor. and activation energy The specific value (in this embodiment, the activation energy of the coking reaction of Lianyuan L-QB300 heat transfer oil) ≈85~120 kJ / mol, pre-exponential factor ≈10 5 ~10 8 mm / h, the specific value depends on the heat transfer oil grade and operating conditions.

[0103] (5) Correction of the correlation between experimental data and simulation results: The coking layer thickness measured in the experiment was corrected. and its corresponding outer wall temperature Comparison with CFD simulation results Compare the simulation results and calculate the deviations under each working condition:

[0104]

[0105] in, It is the simulation deviation (°C) of the outer wall temperature under the i-th working condition; For the first The outer wall temperature (°C) measured under each working condition. For the first The outer wall temperature (°C) obtained by CFD simulation under various operating conditions.

[0106] The least squares method was used to correct the key parameters in the numerical model (thermal conductivity of the coking layer). Flue gas emissivity of the outer wall of the furnace tube (etc.), so that the root mean square error between the simulation results and experimental data of the corrected model is minimized:

[0107] .

[0108] The error of the corrected numerical model is controlled within ±5%.

[0109] S33: Establishing a quantitative relationship: Based on the calibrated simulation data, a quantitative mathematical relationship model is fitted to establish the relationship between the degree of coking and the online monitorable parameter, outer wall temperature. Specifically, the model is an outer wall temperature inversion model.

[0110] ;

[0111] in, The degree of coking on the furnace tubes, The outer wall temperature of the furnace tube is represented by f, which is a univariate inversion mapping function from the outer wall temperature to the degree of coking. This inversion mapping function is constructed based on the heat transfer mechanism and the fitting of numerical simulation-experimental joint data. It is used to reverse the calculation of the degree of coking on the inner wall of the furnace tube, which cannot be directly measured, from the outer wall temperature that can be monitored online.

[0112] In this embodiment, the complete derivation process of the furnace tube outer wall temperature inversion model is as follows:

[0113] Based on the calibrated CFD simulation data, different coking degrees were extracted. (0%, 1%, 2%, 3%, 4%, 5%, ...) corresponding to the outer wall temperature of the furnace tube ,get( , Data pairs (i=1, 2, 3, 4, 5, ...).

[0114] Based on data fitting analysis, the degree of coking Temperature of the outer wall of the furnace tube The relationship between them is exponential, and the fitting model is as follows:

[0115]

[0116] in, The degree of coking in the furnace tubes (dimensionless, value range 0~100%). Temperature of the outer wall of the furnace tube (°C); It is a dimensionless amplitude coefficient; The temperature sensitivity coefficient (1 / ℃); The reference temperature (°C, corresponding to) (Temperature intercept obtained by extrapolating the fitted curve when =0%).

[0117] Logarithmic linearization of the above nonlinear model:

[0118]

[0119] The least squares method was used to fit the experimental-simulation joint dataset to determine the model parameters.

[0120] The specific fitting results for the YYW-7000Y(Q) type coil-type organic heat carrier boiler and Lianyuan L-QB300 heat transfer oil in this embodiment are as follows:

[0121]

[0122] in, =2.6806 (dimensionless) =0.0093 (1 / ℃), =344.08℃. The physical meanings of each parameter are as follows: =2.6806 is the coking sensitivity coefficient determined by the boiler structure (number of coils, pipe diameter, and arrangement); =0.0093 is the temperature sensitivity coefficient of the coking reaction of heat transfer oil, which reflects the sensitivity of the coking rate of heat transfer oil to the increase of temperature. Its value is related to the grade of heat transfer oil and the activation energy of the coking reaction. =344.08℃ is the theoretical extrapolated value of the starting temperature of coking for this type of boiler under the operating conditions of this heat transfer oil. When the outer wall temperature is lower than this temperature, the degree of coking approaches 0.

[0123] The coefficient of determination R of the model fit 2 =0.9976, and RMSE=0.87%, indicating that the model has extremely high fitting accuracy. In this application, R... 2 The calculation method for RMSE is as follows:

[0124] Coefficient of determination:

[0125] Coefficient of determination R 2 Model goodness of fit is used to measure how well a model can explain the variability of data. Its value is usually between 0 and 1, with the closer to 1 indicating that the model has a stronger ability to explain the observed data.

[0126]

[0127] in, This represents the total number of operating conditions used for verification. For the first Coking degree (%) obtained by CFD simulation under each working condition. For the first Under various working conditions, the inversion model of this invention ( The calculated predicted coking degree (%). The arithmetic mean of the simulated coking degree under all working conditions is calculated as follows:

[0128]

[0129] Root mean square error:

[0130] The root mean square error (RMSE) measures the average deviation between the model's predicted values ​​(inverted calculated values) and the actual values ​​(simulation or experimental values). The smaller the RMSE value, the higher the model's prediction accuracy.

[0131] .

[0132] This embodiment also performs real-time temperature correction of the heat transfer oil's physical properties:

[0133] Density of heat transfer oil Specific heat capacity thermal conductivity and kinematic viscosity Both change with temperature. To improve the accuracy of coking assessment, this invention introduces a temperature-based real-time property correction module:

[0134] (1) Density temperature correction:

[0135]

[0136] in, The density of the heat transfer oil based on temperature correction (kg / m³) 3 ), This represents the real-time temperature (K) of the heat transfer oil. Reference temperature (K). Reference temperature The density of the heat transfer oil below (kg / m³) 3 ), The volumetric expansion coefficient (1 / ℃) of the heat transfer oil is determined by the manufacturer's report or by experiment.

[0137] (2) Specific heat capacity temperature correction:

[0138] ;

[0139] in, The specific heat capacity of the heat transfer oil is based on temperature correction (J / (kg·K)). Reference temperature The specific heat capacity of the heat transfer oil (J / (kg·K)). The specific heat capacity temperature coefficient (J / (kg·K)) 2 )).

[0140] (3) Temperature correction for thermal conductivity:

[0141] ;

[0142] in: (The thermal conductivity is based on temperature correction, W / (m·K)). The thermal conductivity (W / (m·K)) is the thermal conductivity at the reference temperature. The thermal conductivity temperature coefficient (1 / ℃) is the coefficient of thermal conductivity.

[0143] (4) Viscosity temperature correction (Andrade formula):

[0144] ;

[0145] in: The temperature-corrected kinematic viscosity of the heat transfer oil (m) 2 / s), Reference temperature kinematic viscosity of heat transfer oil (m) 2 / s), The activation energy for viscous flow is (J / mol). This is the universal gas constant.

[0146] The aforementioned temperature correction parameters are linked to the heat transfer oil grade and stored in a cloud database. The system automatically retrieves the corresponding correction coefficients based on the real-time monitored heat transfer oil temperature, updates the physical property parameters, and recalculates the Reynolds number Re, Prandtl number Pr, and the heat transfer oil's convective heat transfer coefficient. This enables dynamic adaptive updating of model parameters.

[0147] Furthermore, in addition to enabling dynamic adaptive updates of model parameters, real-time temperature correction of heat transfer oil properties also offers the following benefits and technical advantages: It improves the adaptability of the coking initiation temperature threshold, avoiding judgment biases introduced by constant physical property parameters; it enhances the accuracy of heat transfer coefficient inversion and coking thermal resistance identification, eliminating accumulated heat transfer bias and making thermal resistance cross-validation results more accurate and reliable; it broadens the model's applicable temperature range across all operating conditions, maintaining good evaluation accuracy and robustness within the 350℃~600℃ range; it provides standardized adaptation interfaces for multiple grades of heat transfer oil, reducing deployment costs and promotion barriers in different application scenarios; and it suppresses the impact of physical property drift caused by heat transfer oil aging and deterioration, combining regular offline oil testing to periodically calibrate benchmark physical property parameters, ensuring the long-term evaluation reliability of the system.

[0148] Based on the temperature inversion model, this invention further introduces a heat transfer coefficient inversion module, which identifies the thermal resistance of the coking layer online through real-time monitoring data, serving as an independent verification dimension for assessing the degree of coking.

[0149] The overall heat transfer coefficient K (based on the inner surface area of ​​the furnace tubes) is expressed by the following formula:

[0150] ;

[0151] in: The overall heat transfer coefficient based on the internal surface area (W / (m²)) 2 ·K)), The convective heat transfer coefficient on the heat transfer oil side (W / (m) 2 ·K)), The thermal conductivity of the furnace tube wall is (W / (m·K)). , These are the inner and outer radii (m) of the furnace tube, respectively. The thermal conductivity of the coking layer is (W / (m·K), taken as 0.15~0.3 W / (m·K)). The thickness of the coking layer is in meters (m). The side convective heat transfer coefficient of flue gas (W / (m)) 2 ·K)).

[0152] Based on energy balance, the overall heat transfer coefficient K can be derived from real-time monitoring data:

[0153] ;

[0154] in: Heat exchange for furnace tubes (W). The heat transfer oil flow rate (m) and the inlet / outlet temperature difference are determined by the heat transfer oil flow rate (m) and and specific heat capacity calculate: ; The inner surface area of ​​the furnace tube (m²) 2 ); The logarithmic mean temperature difference (K) The calculation method is as follows:

[0155] ;

[0156] in: , These are the flue gas inlet and outlet temperatures (K), respectively. , These are the inlet and outlet temperatures (K) of the heat transfer oil, respectively.

[0157] The inversion obtained Theoretical heat transfer coefficient under coking-free conditions The increase in coking thermal resistance was obtained by comparison:

[0158] ;

[0159] Coking thermal resistance increment With degree of coking They are positively correlated and can be used as a cross-validation index for temperature inversion model results. When the deviation between the two exceeds a preset threshold, the model self-calibration is triggered, which effectively improves the reliability of the evaluation.

[0160] This invention further introduces a coking time factor. A time evolution model of coking degree is constructed to predict coking trend.

[0161] The relationship between the rate of change in coking degree and the current coking degree and operating conditions:

[0162] ;

[0163] in: The change rate of coking degree (1 / h). ， Re) is the coking rate coefficient (related to temperature and flow rate), and m and n are the reaction orders (determined experimentally). This represents the maximum degree of coking (taken as 100%).

[0164] Coking rate coefficient Expressed in Arrhenius form:

[0165] ;

[0166] in: It is a pre-exponential factor (related to heat transfer oil and boiler structure). The velocity effect index (determined experimentally). The activation energy for the coking reaction is (J / mol).

[0167] Numerical integration of the above ordinary differential equations can predict the future. Coking degree after time:

[0168] ;

[0169] The system displays the coking trend prediction results together with the real-time evaluation results. When the predicted coking degree will reach the moderate or severe coking threshold within the preset time window, an early warning will be issued, providing maintenance personnel with sufficient time to intervene and achieving a leap from passive response to proactive prevention.

[0170] S4. Quantitative Dynamic Assessment of Coking Degree

[0171] Specifically, the process includes the following steps:

[0172] S41: Data Preprocessing. The cloud monitoring platform receives the raw data uploaded by the sensor group, performs standardized preprocessing operations, removes abnormal abrupt values ​​caused by sensor failure and signal interference, completes data noise reduction using threshold filtering and 5-point moving average method, and completes data calibration in conjunction with equipment calibration parameters to eliminate systematic and random errors;

[0173] S42: Coking degree calculation. This involves calculating the real-time monitored temperature of the furnace tube outer wall. Substitute into the quantitative relationship model established in step S3 Dynamically calculate the degree of coking at the current moment. .

[0174] S43: Coking Degree Assessment: Calculate the degree of coking. numerical values ​​and preset safety thresholds ( , , By comparing these data, the coking status of the furnace tubes was classified into four levels:

[0175] (1) Normal state: When When the percentage is less than 5%, it is considered a safe state, and the equipment operating parameters are stable.

[0176] (2) Mild coking: When 5% ≤ When the efficiency is less than 15%, coking tends to occur on the inner wall of the furnace tube, and the heat exchange efficiency shows a slight downward trend.

[0177] (3) Moderate coking: When 15% ≤ When the heat exchange efficiency is less than 35%, the coking layer on the inner wall of the furnace tube thickens significantly, the heat exchange efficiency decreases significantly, and there is a risk of local overheating.

[0178] (4) Severe coking: When When the concentration is ≥35%, severe coking occurs on the inner wall of the furnace tube, resulting in a significant decrease in heat exchange efficiency and an extremely high risk of high-temperature tube rupture.

[0179] S5, Tiered Early Warning and Differentiated Operation and Maintenance Response

[0180] Based on the coking level calculated by the model, the corresponding early warning level and operation and maintenance response strategy are matched, and early warning signals are pushed synchronously through multiple terminals to achieve differentiated safety prevention and control:

[0181] (1) Level 1 warning (mild coking): Only pushes routine monitoring reminders, prompting maintenance personnel to optimize the flow rate of heat transfer oil pipeline to the optimal range of 3~5m / s and continuously monitor the equipment operating status;

[0182] (2) Level II Early Warning (Moderate Coking): Push special early warning information and operation and maintenance optimization plan, indicating that there is a potential overheating risk, requiring operation and maintenance personnel to increase the frequency of heat transfer oil quality testing, shorten the special maintenance cycle of equipment, and closely track the coking development trend;

[0183] (3) Level 3 warning (severe coking): Triggers the sound and light alarm of the on-site equipment, and at the same time, pop-up alarm on the monitoring screen and emergency warning is pushed to the mobile APP, clearly indicating that there is a high risk of overheating and tube bursting, and instructing the operation and maintenance personnel to shut down the furnace for inspection, and carry out special operations such as furnace tube coking, heat transfer oil testing and replacement.

[0184] S6. Full Data Storage and Coking Source Tracing Analysis

[0185] The cloud platform stores all raw monitoring data, preprocessed characteristic parameters, coking assessment results, early warning records, and operation and maintenance records in real time. It automatically generates furnace tube temperature change curves, coking trend evolution reports, and equipment operation energy consumption reports. It supports operation and maintenance personnel to query historical data in multiple dimensions by time node, furnace tube layer, and operating condition, and fully trace the entire process of coking from its inception, development, and aggravation. It provides comprehensive data support for optimizing equipment operation and maintenance strategies, adjusting boiler operating parameters, and studying coking mechanisms.

[0186] Example 2

[0187] This embodiment provides an online monitoring and early warning system for coking in the boiler tubes of an organic heat carrier boiler, used to implement the online monitoring and early warning method for coking in the boiler tubes of the organic heat carrier boiler in Embodiment 1. The system adopts a three-layer architecture design, consisting of a data acquisition layer, a cloud platform layer, and an application terminal layer from top to bottom, specifically including:

[0188] (1) Data Acquisition Layer

[0189] The data acquisition layer includes a multi-dimensional sensor group and a data acquisition adapter, providing the system with raw operational data in all dimensions.

[0190] The multi-dimensional sensor group includes a high-precision furnace temperature sensor, a furnace tube outer wall temperature sensor, a thermal oil inlet and outlet temperature sensor, a thermal oil inlet and outlet pressure sensor, and a pipeline flow sensor, which fully covers the key monitoring points of the boiler, with high acquisition accuracy and fast response speed.

[0191] The data acquisition adapter supports dual communication interfaces of RS485 and RJ45, adopts the industrial-grade MQTT IoT communication protocol, and has edge computing, data noise reduction, and local caching functions. It can cache more than 7 days of monitoring data in offline mode and automatically upload data in batches via GPRS, 5G, 4G, Ethernet, etc. It is compatible with PLC control systems of various types of organic heat carrier boilers.

[0192] (2) Cloud platform layer

[0193] The cloud platform layer is the core processing hub of the system, including a data processing module, a coking assessment module, an early warning management module, and a data storage module;

[0194] The data processing module is used to receive raw data uploaded by the data acquisition layer, complete outlier removal, moving average filtering and noise reduction, sensor data calibration, and extract core feature parameters for coking assessment.

[0195] The coking assessment module has a built-in multi-physics coupled coking quantitative assessment model based on experiments and numerical simulations. It can automatically calculate the degree of coking in furnace tubes and determine the coking level based on the input core feature parameters, and supports online iterative updates of model parameters.

[0196] The early warning management module pre-stores early warning rules and handling strategies corresponding to four levels of coking. It can automatically match the coking status to generate corresponding early warning signals and supports simultaneous push notifications from multiple channels such as SMS, email, APP, and monitoring screen. It also supports user-defined early warning thresholds.

[0197] The data storage module uses a distributed non-relational database to classify and store raw monitoring data, coking assessment results, early warning records, and operation and maintenance logs. It has the ability to retrieve hundreds of millions of data points in milliseconds and supports hierarchical permission management, data traceability, and automatic report generation.

[0198] (3) Application terminal layer

[0199] At the application terminal level, it provides operation and maintenance personnel with visual operation and remote operation and maintenance services, including web management terminal, on-site monitoring screen, and mobile phone APP;

[0200] The web-based management interface features functions such as device information management, operation parameter configuration, historical data query, maintenance report export, and model parameter debugging.

[0201] The on-site monitoring screen provides a real-time visual display of the overall boiler distribution, boiler operating parameters, real-time coking level, and early warning status, enabling comprehensive visual monitoring of the equipment.

[0202] The mobile app is compatible with Android and iOS systems. It can receive real-time alert push messages, remotely view real-time device operation data, and query historical operation and maintenance records, supporting operation and maintenance management anytime, anywhere via mobile devices.

[0203] This system operates in a three-tier architecture. The data acquisition layer is responsible for all-dimensional data collection and transmission, the cloud platform layer is responsible for data processing, model calculation, early warning and control, and the application terminal layer is responsible for visualization and human-computer interaction. The system has a fast overall response speed and stable operation. It can be adapted to the long-term continuous operation of various industrial organic heat carrier boilers and has a significant intelligent prevention and control effect.

[0204] Example 3

[0205] This embodiment is based on the YYW-7000Y(Q) type coiled organic heat carrier boiler and Lianyuan L-QB300 heat transfer oil. It is the optimal implementation case of the present invention and is not intended to limit the scope of protection of the present invention.

[0206] (a) Equipment sensor deployment

[0207] For the YYW-7000Y(Q) type organic heat carrier boiler, a full-dimensional sensor network deployment was completed:

[0208] 1. Boiler tube temperature monitoring: On the backfire surface of the last three rings (the area with the most severe coking) of the outer ring of the boiler, four high-precision patch temperature sensors are evenly distributed in each ring to cover the high-temperature radiation area of ​​the coil.

[0209] 2. Furnace temperature monitoring: Install one high-temperature temperature sensor in the high-temperature core area at the boiler furnace burner outlet;

[0210] 3. Medium parameter monitoring: Temperature sensors, pressure sensors, and flow sensors are installed at the inlet and outlet of the heat transfer oil circulation pipeline, respectively;

[0211] All sensors are connected to the data acquisition adapter via shielded cables. The sampling frequency is set to 5 seconds per cycle, which is suitable for complex industrial conditions, resists electromagnetic interference, and ensures stable data transmission.

[0212] (II) Data Acquisition and Preprocessing

[0213] During equipment operation, the sensor array collects real-time data on furnace temperature, outer coil wall temperature, heat transfer oil inlet and outlet temperatures, pressure, and flow rate. All signals are connected to a data acquisition adapter supporting the Modbus TCP protocol. After initial noise reduction at the edge, the adapter sends a data packet containing a timestamp, equipment ID, and outer wall temperature at each point to the cloud platform every 5 seconds via a 4G network. The cloud platform uses threshold filtering to remove sensor malfunctions and abnormal values, employs a 5-point moving average method for data filtering and noise reduction, and combines this with the equipment's factory calibration parameters to complete data calibration and eliminate system errors.

[0214] (III) Model Construction

[0215] First, a three-dimensional geometric model was established based on the actual dimensions of the boiler. Then, combined with the collected operating parameters, numerical simulations of combustion and heat transfer were performed using Ansys Fluent software. In the simulation, coke layers of different thicknesses (e.g., 0 mm, 5 mm, 10 mm, corresponding to coking degrees of 0%, 10%, 20%, etc.) were sequentially placed on the inner wall of the furnace tubes (thermal conductivity set to 0.2 W / (m·K)). The temperature distribution of the outer wall of the furnace tubes under different coking degrees was obtained through simulation. The results show that the outer wall temperature increases significantly with increasing coking degree. Based on the simulation data, an exponential function model was used for fitting to obtain the coking degree. With outer wall temperature The specific inversion model is as follows:

[0216] .

[0217] (iv) Quantitative assessment and early warning implementation of coking

[0218] 1. The average temperature of the outer wall of the furnace tube is 500℃. Substitute this temperature into the model to calculate the degree of coking. =8.75%, judged as mild coking, the system triggered a level one warning, the APP pushed a routine monitoring reminder, and the equipment continued to operate normally;

[0219] 2. The average temperature of the outer wall of the furnace tube is 550℃. Substituting this into the model, the degree of coking is calculated. =15.51%, which is judged as moderate coking. The system triggered a level 2 warning and pushed the warning information to the APP and email simultaneously, guiding the operation and maintenance personnel to optimize the heat transfer oil flow rate to 4m / s and shorten the maintenance cycle to 1 month.

[0220] 3. The average temperature of the furnace tube outer wall is 650℃. Substituting this into the model, the degree of coking is calculated. =43.44%, which is judged as severe coking. The system triggered a level 3 warning, with on-site audible and visual alarms, a pop-up alarm on the monitoring screen, and a forced emergency shutdown command pushed to the mobile APP. The maintenance personnel immediately stopped the machine to carry out coking removal and maintenance work.

[0221] (v) Data traceability and operation and maintenance optimization

[0222] The system stores all operational data in real time and automatically generates daily / weekly / monthly coking trend reports and energy consumption analysis reports. Maintenance personnel can query historical data through the web interface to trace the coking process and optimize boiler operating temperature and flow rate parameters based on data patterns, forming a closed-loop safe operation and maintenance system.

[0223] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for online monitoring and early warning of coking in boiler tubes of an organic heat carrier boiler, characterized in that: Includes the following steps: S1: Collect multi-source operation data of organic heat carrier boilers; S2: Data preprocessing and feature extraction; S3: Construct a coking degree-temperature inversion model; S4: Quantitative dynamic assessment of coking degree; S5: Tiered early warning and differentiated operation and maintenance response; S6: Full data storage and coking source analysis.

2. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 1, characterized in that: In step S1, a multi-dimensional sensor group is installed in the furnace, coiled furnace tubes, and heat transfer oil circulation pipeline of the organic heat carrier boiler to collect multi-source parameters of boiler operation in real time. The multi-source parameters include the internal temperature of the furnace, the outer wall temperature of the inner / middle / outer coils, the inlet and outlet temperatures of the heat transfer oil, the inlet and outlet pressures of the heat transfer oil, and the real-time flow rate of the heat transfer oil.

3. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 1, characterized in that: In step S2, the collected data undergoes outlier removal, filtering and noise reduction, and data calibration preprocessing. Core characteristic parameters such as the average temperature of the outer wall of the furnace tube, temperature fluctuation amplitude, temperature difference between the inlet and outlet of the heat transfer oil, and flow stability are extracted.

4. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 1, characterized in that: In step S3, firstly, a three-dimensional numerical model is established, encompassing the combustion chamber, flue gas flow, radiative heat transfer, and flow and heat transfer within the heat transfer oil pipes. This three-dimensional numerical model includes the differential equations for furnace tube heat conduction, boundary conditions for the coking layer on the inner wall of the furnace tubes, the radiative-convective coupled heat transfer boundary on the outer wall of the furnace tubes, and the continuity and momentum equations for the flow within the heat transfer oil pipes. Secondly, through high-temperature dynamic coking experiments, the variation of coke formation in the heat transfer oil with temperature and time is obtained, and the kinetic parameters of the Arrhenius-type coking reaction are calibrated. These kinetic parameters include the pre-exponential factor. and activation energy ; The experimental data and simulation results were correlated, and the numerical model was corrected using the least squares method. Then, based on the calibrated simulation data, a quantitative mathematical relationship model between the degree of coking and the online monitorable parameter of furnace tube outer wall temperature was established.

5. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 4, characterized in that: A quantitative mathematical model relating the degree of coking to the online-monitorable parameter of outer wall temperature, specifically an outer wall temperature inversion model: ； in, Indicates the degree of coking. Indicates the temperature of the outer wall of the furnace tube. A univariate inversion mapping function representing the temperature of the outer wall to the degree of coking.

6. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 5, characterized in that: The model is an exponential function, and the furnace tube outer wall temperature inversion model is expressed as: ； in, The degree of coking on the furnace tubes (%); This refers to the temperature of the outer wall of the furnace tube after pretreatment. It is a dimensionless amplitude coefficient; The temperature sensitivity coefficient (1 / ℃); The reference temperature (°C) corresponds to The temperature intercept obtained by extrapolating the fitted curve when =0%.

7. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 1, characterized in that: In step S4, the coking degree is assessed by comparing the calculated coking degree ω value with a preset safety threshold, and the coking state of the furnace tube is divided into four levels: (1) Normal state: When When the percentage is less than 5%, it is considered a safe state, and the equipment operating parameters are stable. (2) Mild coking: When 5% ≤ When the concentration is less than 15%, coking tends to occur on the inner wall of the furnace tube, and the heat exchange efficiency shows a slight downward trend. (3) Moderate coking: When 15% ≤ When the heat exchange efficiency is less than 35%, the coking layer on the inner wall of the furnace tube thickens significantly, the heat exchange efficiency decreases significantly, and there is a risk of local overheating. (4) Severe coking: When When the concentration is ≥35%, severe coking occurs on the inner wall of the furnace tube, resulting in a significant decrease in heat exchange efficiency and an extremely high risk of high-temperature tube rupture.

8. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 1, characterized in that: In step S5, based on the coking level calculated by the model, the corresponding early warning level and operation and maintenance handling strategy are matched, and early warning signals are pushed synchronously through multiple terminals to achieve differentiated safety prevention and control.

9. The method for online monitoring and early warning of coking in boiler tubes of organic heat carrier boilers according to claim 1, characterized in that: In step S5, the cloud platform stores all raw monitoring data, preprocessed characteristic parameters, coking assessment results, early warning records and operation and maintenance records in real time, and automatically generates furnace tube temperature change curves, coking trend evolution reports and equipment operation energy consumption reports.

10. An online monitoring and early warning system for coking in boiler tubes of an organic heat carrier boiler, used to implement the method described in any one of claims 1-9, characterized in that: It includes the data acquisition layer, cloud platform layer, and application terminal layer; The data acquisition layer includes a multi-dimensional sensor array and a data acquisition adapter, providing the system with raw operational data in all dimensions; The cloud platform layer includes a data processing module, a coking assessment module, an early warning management module, and a data storage module; At the application terminal level, it provides operation and maintenance personnel with visual operation and remote operation and maintenance services, including web management terminal, on-site monitoring screen, and mobile phone APP.

Images