High-precision coal-fired boiler ammonia mixing combustion engineering quantification analysis method and system

Abstract

The application discloses a high-precision coal-fired power generation boiler mixed-ammonia combustion engineering quantification analysis method and system, relates to the technical field of boiler combustion, and simultaneously relates to the field of electric digital processing. A specific implementation scheme is as follows: model construction and calibration are performed through an ammonia-coal mixed combustion model group, a NOx prediction model and a thermal calculation model used for realizing mixed-ammonia combustion engineering quantification are obtained, online simulation is performed in a specified simulation software, boiler ammonia-mixed combustion characteristic parameters, visual results and thermal calculation results are output, whether to perform ammonia-mixed ratio adjustment optimization is determined through the combustion characteristic parameter results and the thermal calculation results, if yes, the online simulation is continuously performed according to the optimized results until the NOx prediction model and the thermal calculation model converge and corresponding models are output, otherwise, the NOx prediction model and the thermal calculation model are taken as convergence models and output, and the problem of insufficient ammonia-mixed ratio control precision in the mixed-ammonia combustion process is solved through the technology of the application.

Description

Technical Field

[0001] This invention relates to the field of boiler combustion, and also to the field of digital electrical processing technology, particularly to a high-precision quantitative analysis method and system for mixed ammonia combustion in coal-fired power generation boilers. Background Technology

[0002] Existing engineering quantitative analysis methods for boiler combustion employ a multi-level, multi-scale technical system, ranging from macroscopic empirical estimation to mesoscopic network simulation and then to microscopic fine simulation. The core challenge of ammonia-infused combustion in boilers lies in the extreme complexity of the physicochemical processes involved, including the quantification of NOx (nitrogen oxides) generation and reduction, N2O (nitrous oxide, a potent greenhouse gas), and NH3 escape from incomplete combustion. Existing quantitative methods use regression analysis to correlate key indicators such as NOx emissions and combustion efficiency with core operating parameters, forming empirical formulas or lookup tables. Furthermore, by solving the three-dimensional Navier-Stokes equations, they comprehensively describe the turbulent flow, heat transfer, mass transfer, and chemical reaction processes within the combustion chamber, providing comprehensive, visualized information. This helps to more accurately capture the physical details crucial to combustion efficiency and pollutant generation, such as local high-temperature zones and poorly mixed zones.

[0003] For example, the simplified modeling method, equipment, and computer program product for a complex mixed fuel combustion reaction kinetic model disclosed in Chinese invention patent announcement number CN120277990B includes: model coupling, a first simplification of the model using the directed relation graph method based on error propagation and species sensitivity analysis, optimization of the simplified model using the optimal pre-exponential factor, temperature index, and activation energy of the three-parameter modified Arenius equation as optimization objects, and further simplification of the optimized model.

[0004] For example, Chinese invention patent CN114239430B discloses a method and system for predicting NOx at the furnace outlet based on numerical simulation, including a prediction model establishment part and a real-time prediction part; it also involves a prediction method, including: processing DCS data to obtain the boundary conditions required for numerical simulation calculation, establishing a three-dimensional geometric model of the furnace, calculating the NOx concentration data at the furnace inlet under different operating conditions, and determining the database; based on the database, establishing a NOx prediction model at the furnace outlet using the support vector method; and finally, predicting the NOx concentration at the furnace outlet in real time based on the real-time furnace inlet parameters under actual operating conditions.

[0005] The current engineering quantitative analysis of ammonia combustion technology is achieved through a multi-level, iterative approach. Chemical reactor networks, with their perfect balance between accuracy and cost, are the dominant force in current engineering analysis. Computational fluid dynamics (CFD) is the ultimate tool for solving deep technical problems. Furthermore, CFD can help achieve higher-precision engineering quantitative analysis of ammonia combustion, especially in coal-fired power plant boilers. The implementation process typically includes three core stages: basic foundation and verification, high-fidelity mechanistic analysis, and engineering simplification and optimization.

[0006] The foundational foundation and validation phase primarily establishes a reliable benchmark model for pure coal combustion. This typically involves using boiler drawings, coal quality data, and operational data as input to create a benchmark CFD (Computational Fluid Dynamics) model that accurately reproduces the boiler's performance under pure coal combustion conditions. Simulations are then performed, and the results are compared with boiler design values ​​or actual field measurements, including boiler outlet oxygen and carbon dioxide concentrations, boiler efficiency, furnace outlet flue gas temperature, main steam temperature, reheat steam temperature, and NOx emission concentration (serving as a benchmark for subsequent ammonia blending comparisons). High-fidelity mechanistic analysis reveals the microscopic mechanisms and macroscopic effects of ammonia blending combustion. This involves using the calibrated benchmark model, ammonia blending parameters, and detailed chemical reaction mechanisms as input. The key inputs are ammonia combustion introduced into the validated benchmark model, along with a kinetic model and modified radiative heat transfer. At the microscopic level, this outputs ammonia-nitrogen conversion rate, coal-nitrogen conversion rate, instantaneous NOx generation or reduction rate distribution, and key free radical distribution. At the macroscopic level, it outputs the three-dimensional temperature field within the furnace, heat flux density distribution, flue gas composition distribution, and overall boiler performance (efficiency, steam temperature). The engineering simplification and optimization application transforms high-fidelity knowledge into practical engineering tools and achieves optimization. The inputs are high-fidelity data output from the high-fidelity mechanistic analysis stage. A simplified NOx prediction model is calibrated, and modified radiative heat transfer correlations are solidified. These are integrated into a calibrated CFD model with a faster solution setting, ultimately yielding a corresponding fast model that enables the engineering quantification of ammonia-mixed combustion in coal-fired power boilers.

[0007] The above-mentioned technology has at least the following technical problems:

[0008] Because ammonia is a liquid or gaseous fuel, its flow rate is easily affected by temperature and pressure fluctuations, leading to unstable ammonia blending ratios and fluctuating combustion characteristics. If the ammonia blending ratio is too high, it may result in a reducing atmosphere in the furnace, flame destabilization, or even deflagration. Therefore, the control of the ammonia blending flow rate requires high precision and rapid response. Current systems utilize a closed-loop ammonia blending ratio control system to proportionally control the total ammonia injection flow rate, achieving feedforward regulation, and also controlling NO2 in the tail gas. xAlternatively, furnace temperature feedback can be used to adjust the ammonia injection flow rate. However, due to fluctuations in coal flow signals and changes in fuel characteristics, the calculation error of the ammonia blending ratio is large, failing to achieve uniform control of multi-point ammonia injection. Only the total amount is controlled, resulting in insufficient accuracy in the ammonia blending ratio control during the ammonia blending combustion process. Summary of the Invention

[0009] This invention provides a high-precision method and system for quantitative analysis of ammonia blending combustion in coal-fired power boilers. It enables precise control of the ammonia blending ratio during combustion, thereby achieving more accurate quantitative analysis of ammonia blending combustion processes. The technical solution provided in this application is as follows:

[0010] According to the first aspect of this application, a high-precision quantitative analysis method for ammonia-blended combustion engineering in coal-fired power generation boilers is provided. This method includes: constructing and calibrating a NOx prediction model and a thermodynamic calculation model for realizing the quantitative analysis of ammonia-blended combustion engineering through an ammonia-coal co-combustion model set; performing online simulation in a specified simulation software; and outputting corresponding combustion characteristic parameter results and thermodynamic calculation results. The ammonia-coal co-combustion model set includes an ammonia-coal co-combustion NOx prediction model, a kinetic model, and a modified radiative heat transfer coefficient. The thermodynamic calculation results are the data set output by the NOx prediction model and the thermodynamic calculation model. The performance of the NOx prediction model and the thermodynamic calculation model is judged based on the combustion characteristic parameter results and the thermodynamic calculation results. Based on the judgment results, it is determined whether to perform ammonia blending ratio adjustment and optimization. If ammonia blending ratio adjustment and optimization are performed, online simulation continues in the specified simulation software based on the results of the ammonia blending ratio adjustment and optimization until the NOx prediction model and the thermodynamic calculation model converge. The corresponding converged NOx prediction model and converged thermodynamic calculation model are then output; otherwise, the NOx prediction model and the thermodynamic calculation model are output as the corresponding converged models.

[0011] According to another aspect of this application, a high-precision quantitative analysis system for ammonia-blended combustion engineering in coal-fired power generation boilers is provided, comprising: a model construction and calibration module, an ammonia blending ratio adjustment and optimization module, and a model convergence module. The model construction and calibration module is used to construct and calibrate a NOx prediction model and a thermodynamic calculation model for realizing the quantitative analysis of ammonia-blended combustion engineering using an ammonia-coal co-combustion model set. It then performs online simulation in a specified simulation software, outputting the corresponding combustion characteristic parameter results and thermodynamic calculation results. These results constitute the data set output by the NOx prediction model and the thermodynamic calculation model. The ammonia blending ratio adjustment and optimization module is used to determine whether to perform ammonia blending ratio adjustment and optimization based on the thermodynamic calculation results. The model convergence module, if ammonia blending ratio adjustment and optimization is performed, continues online simulation in the specified simulation software based on the optimized results until the NOx prediction model and the thermodynamic calculation model converge, outputting the corresponding converged NOx prediction model and converged thermodynamic calculation model. Otherwise, it outputs the NOx prediction model and the thermodynamic calculation model as the corresponding converged model.

[0012] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0013] 1. A NOx prediction model and a thermodynamic calculation model for quantifying ammonia-coal co-fired combustion were constructed and calibrated using an ammonia-coal co-fired model group. These models were then used for online simulation in designated software, realizing a numerical simulation system for ammonia-coal co-fired combustion that reflects the actual operating characteristics of boilers. The system outputs corresponding combustion characteristic parameters and thermodynamic calculation results, which helps to more accurately quantify the impact of ammonia blending ratio on the combustion temperature field, nitrogen conversion pathway, and NOx emissions. This overcomes the shortcomings of traditional ammonia-coal co-fired simulation methods in accurately predicting NOx and nitrogen conversion pathways, and solves the problem that traditional thermodynamic calculations cannot adapt to the operating conditions of high-moisture flue gas and ammonia combustion. It also corrects for radiative heat transfer in high-moisture flue gas and improves the accuracy of thermodynamic calculations for heating surfaces. Finally, the performance of the NOx prediction model and the thermodynamic calculation model was evaluated based on the combustion characteristic parameters and thermodynamic calculation results, and the final determination was made based on the evaluation results. Whether or not ammonia blending ratio adjustment and optimization are performed, a closed-loop optimization-based ammonia blending combustion control system is formed, along with a basis for real-time adjustment of the ammonia blending ratio. This fills the gap in the lack of engineering-based control basis for the ammonia blending ratio. If ammonia blending ratio adjustment and optimization are performed, online simulation continues in the designated simulation software based on the results of the optimized ammonia blending ratio adjustment until the NOx prediction model and the thermodynamic calculation model converge, thereby enhancing parameter convergence and achieving long-term stable online prediction. The converged NOx prediction model and converged thermodynamic calculation model are output, which helps improve the accuracy of ammonia blending ratio matching and avoids problems caused by insufficient or excessive ammonia. Otherwise, the NOx prediction model and thermodynamic calculation model are output as the corresponding converged models, forming a general numerical simulation method for ammonia blending combustion that can be used for different loads, different coal types, and different ammonia blending methods. At the same time, it can be directly used as a guidance model for boiler operation, realizing refined control.

[0014] 2. The acquired boiler combustion data is compared with the corresponding set thresholds for compliance judgment. If the compliance judgment result meets the preset conditions, it indicates compliance; otherwise, it indicates non-compliance, and the ammonia blending ratio is adjusted and optimized. The threshold judgment enables real-time assessment of whether the current ammonia blending conditions meet emission and efficiency requirements, supports continuous operation and automatic optimization, avoids manual experience-based judgment, and improves the consistency and repeatability of the judgment results. Next, the deviation between the measured ammonia flow rate and the set ammonia flow rate for the current preset simulation period is acquired in real time and recorded as the ammonia blending deviation. This deviation is matched with the ammonia blending deviation tolerance range, which represents the maximum tolerance for deviation in the ammonia blending ratio. By comparing the measured ammonia flow rate and the set flow rate in real time, not only... This method facilitates the rapid detection of insufficient or excessive ammonia injection and improves ammonia blending accuracy. It solves the problem of lagging ammonia flow regulation and control, which fails to detect and correct ammonia injection deviations in a timely manner. If the ammonia blending deviation falls within the tolerance range, it indicates that the ammonia blending ratio for the current simulation period is qualified; otherwise, it indicates that the ammonia blending ratio for the current simulation period is unqualified. The method also determines the degree of ammonia blending ratio abnormality and analyzes the changes in coal flow signals and fuel characteristics during the simulation period to determine whether to adjust the ammonia blending ratio. This allows for advance adjustment of the ammonia injection amount, thereby improving control efficiency and avoiding the problem of lagging or inaccurate ammonia injection regulation caused by traditional ammonia blending control not considering the dynamic changes in coal flow and coal quality.

[0015] 3. By acquiring coal flow signal fluctuation data and fuel characteristic change data within the simulated time period, quantified values ​​of coal flow signal fluctuation and fuel characteristic change are obtained respectively. This enables precise measurement of boiler fuel-side disturbances, visualization and calculable capture of transient fluctuations, and characterization of complex dynamic combustion behavior using multiple indicators. This overcomes the shortcomings of traditional boiler control, which only monitors the average coal flow rate and cannot capture the true fluctuation characteristics. Then, the quantified values ​​of coal flow signal fluctuation and fuel characteristic change are compared with the maximum limit of coal flow fluctuation and the threshold of fuel characteristic change, respectively, to obtain the corresponding difference values, which are recorded as the coal flow signal fluctuation difference and the fuel characteristic change difference, respectively. This solves the problem of the lack of existing technologies for... The quantitative criteria for triggering ammonia blending adjustments can easily lead to problems such as adjustment lag or misadjustment. At the same time, the magnitude of the difference can be used to determine the degree of fluctuation, the source of disturbance, and the impact on combustion stability, providing stronger data support for subsequent control. Finally, the total combustion characteristic deviation is obtained by fusing the difference in coal flow signal fluctuation and the difference in fuel characteristic change with the corresponding weights. This helps to improve the overall robustness of the control system, which is driven by factors that cause instability in regulation. It also reduces the probability of existing ammonia blending control relying solely on a single variable (such as NOx) and ignoring the multi-factor coupling of fuel-side disturbances. This not only helps to improve the adaptability of boiler ammonia blending combustion under dynamic operating conditions and coal quality fluctuation conditions, but also improves the NOx control effect.

[0016] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0017] The accompanying drawings are provided for a better understanding of this solution and do not constitute a limitation of this application. Wherein:

[0018] Figure 1 This is a flowchart illustrating the high-precision quantitative analysis method for mixed ammonia combustion in coal-fired power boilers provided in this embodiment of the invention.

[0019] Figure 2 This is a flowchart illustrating the output thermal calculation results provided in an embodiment of the present invention;

[0020] Figure 3 This is a schematic diagram of the process for determining the degree of ammonia doping ratio abnormality provided in an embodiment of the present invention;

[0021] Figure 4 This is a schematic diagram of the high-precision quantitative analysis system for mixed ammonia combustion in coal-fired power boilers provided in this embodiment of the invention.

[0022] Figure 5 This is a convergence plot of the NOx prediction model;

[0023] Figure 6 This is the convergence graph of the thermal calculation model;

[0024] Figure 7 This is a temperature field distribution diagram of the cross-section of the ammonia combustion furnace.

[0025] Figure 8 This is a numerical simulation diagram of ammonia combustion;

[0026] Figure 9 Comparison of core characteristic parameters of ammonia-mixed combustion Figure 1 ;

[0027] Figure 10 Comparison of core characteristic parameters of ammonia-mixed combustion Figure 2 ;

[0028] Figure 11 This is a graph showing the accuracy analysis of the NOx prediction model for coal-fired ammonia combustion. Detailed Implementation

[0029] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0030] This invention provides a high-precision method and system for quantitative analysis of ammonia combustion in coal-fired power boilers. For example... Figure 1 The flowchart shown is a high-precision quantitative analysis method for ammonia combustion in coal-fired power plants provided by an embodiment of the present invention. The processing flow of this method may include the following steps:

[0031] A NOx prediction model and a thermodynamic calculation model for realizing the engineering quantification of ammonia-coal co-combustion were obtained by constructing and calibrating the ammonia-coal co-combustion model group. The model was then simulated online in a specified simulation software, and the corresponding combustion characteristic parameters, visualization results and thermodynamic calculation results were output. The ammonia-coal co-combustion model group includes an ammonia-coal co-combustion NOx prediction model, a kinetic model and a modified radiative heat transfer coefficient. The combustion characteristic parameter results and thermodynamic calculation results are the data set output by the NOx prediction model and the thermodynamic calculation model.

[0032] pass Figure 5 as well as Figure 6 It can display the specific convergence trend of the model separately. Figure 5 The convergence plot of the NOx prediction model is presented to verify the effectiveness of its training. The change in loss (prediction error) with the number of iterations is also shown. The plot shows that the loss value rapidly decreases from 3, eventually falling below the convergence threshold (0.2), indicating model convergence. This ultimately proves that the training process of the NOx prediction model is effective and the model performance is stable. Furthermore, through... Figure 6 The graph displays the convergence of the thermal calculation model, verifying its stability and showing the change in residual value (calculation error) with the number of iterations. The trend in the graph shows that the residual value rapidly decreases from 4, eventually falling below the convergence threshold (0.2), indicating model convergence. This graph ultimately proves the reliability of the thermal calculation model's results, making it suitable for engineering analysis, particularly applicable to the thermal calculations used in this invention.

[0033] It should be added that the specified simulation software refers to CFD software and custom thermal calculation software programs. CFD software is a type of specialized software tool that uses numerical calculation methods to simulate fluid (gas or liquid) flow, heat transfer, chemical reactions, and other processes. It discretizes the continuous medium mechanics equations (such as the Navier-Stokes equations) and solves them by computer, thereby enabling visual analysis and performance prediction of complex flow phenomena. Its core functions include, but are not limited to, fluid flow simulation, heat transfer analysis, multiphase flow modeling, chemical reaction and combustion simulation, and simulation of complex processes such as particle transport, gas-solid reactions, and spraying. For example, the most widely used general-purpose CFD software in industry is ANSYS Fluent, which supports complex physical models.

[0034] In one optional implementation, the online simulation of the CFD software may include: a) establishing a three-dimensional geometric model based on the boiler furnace, burner, and ammonia injection arrangement, and performing mesh generation; b) setting boundary conditions and initial conditions, including coal quality parameters, load, primary air volume, secondary air volume, ammonia blending ratio, inlet temperature and pressure, etc.; c) selecting and enabling turbulence model, combustion or reaction kinetic model, radiation heat transfer model, and NOx generation or reduction sub-model; d) using discrete solution methods such as finite volume and finite difference to iteratively solve the control equations and perform convergence judgment based on the dual criteria of residual and monitored quantity (e.g., the change of key monitored quantity is less than a preset threshold); e) outputting cloud maps of temperature field, velocity field, component field, NOx distribution, etc., and key cross-section or measuring point data, and using the key data for NOx prediction model calibration, thermodynamic calculation verification, and performance evaluation.

[0035] Specifically, visualization results refer to the ability to intuitively display cloud maps of the flame temperature field, velocity field, component field, and NOx pollutant distribution inside the furnace through simulation using specified simulation software. This helps to determine the flame morphology, temperature level, and NOx concentration generation based on the cloud maps. Combustion characteristic parameters include furnace combustion temperature, composition of combustion products, NOx emission concentration, unburned ammonia content, boiler efficiency, and unburned carbon content.

[0036] like Figure 7 The figure shown is a temperature field distribution diagram of the cross section of a 30% ammonia-mixed combustion furnace. This figure shows the temperature distribution inside the furnace when there is 30% ammonia-mixed combustion. The key features include the central red-orange zone (near the burner) which is a high-temperature zone, gradually cooling outward to the blue low-temperature zone. This can visually verify that the temperature field of ammonia-mixed combustion meets the engineering expectations and that the high-temperature zone is located in a reasonable position.

[0037] Through the above Figure 7 This can further illustrate the numerical simulation of ammonia combustion, namely the following... Figure 8The numerical simulation diagram of ammonia combustion shows the changes in furnace temperature and NOx emissions as the proportion of ammonia increases. The core trend is that the higher the proportion of ammonia, the slightly lower the furnace temperature and the significantly lower the NOx emissions. Therefore, this diagram can directly prove the core environmental advantages of ammonia combustion, while the temperature fluctuation is within a reasonable range.

[0038] It should be noted that the custom thermal calculation software program, specifically the thermal calculation software program for ammonia combustion in coal-fired boilers, is a dedicated thermal calculation tool developed for ammonia combustion in coal-fired boilers. It mainly includes the following three functions:

[0039] (1) Basic data management function. Supports inputting or calling basic physical property data of coal and ammonia fuel, inputting boiler structure and design parameters, setting basic operating parameters such as rated evaporation capacity, rated steam parameters (pressure, temperature), and feedwater temperature of boiler, and flexibly setting the mixing ratio of coal and ammonia combustion. The ratio adjustment range (0-100%) can be set to meet the calculation needs of different mixed ammonia combustion conditions.

[0040] (2) Thermal calculation function. Includes: 1) Combustion reaction calculation, which can automatically calculate the theoretical air demand and actual air consumption of ammonia fuel based on fuel element analysis data and ammonia ratio; accurately calculate the composition, volume and mass of combustion products (flue gas), and calculate the theoretical and actual volume of flue gas; calculate the theoretical and actual combustion temperature of ammonia combustion, and correct the temperature calculation results by combining the furnace heat loss; 2) Boiler heat balance calculation, which calculates the thermal efficiency of the boiler under ammonia combustion conditions and generates a heat balance table and a heat loss ratio analysis report; 3) Heat transfer calculation for heating surfaces: For the radiant heating surfaces in the furnace, the radiation characteristics (emissivity, radiation intensity) of the flue gas from the ammonia-mixed combustion are calculated, and the radiative heat absorption is verified in conjunction with the furnace structural parameters; simultaneously, for the convective heating surfaces (convective tube bundles, economizers, air preheaters, etc.), based on the flue gas flow rate, velocity, temperature, and heating surface parameters, the convective heat transfer and the temperature changes along the flue gas and working fluid are calculated; and the heat absorption distribution data of each heating surface is output to verify whether the matching of the heating surfaces meets the steam parameter requirements under the ammonia-mixed combustion conditions; 4) Operating condition comparison analysis calculation. Supports the comparison of thermal calculation results of the same boiler under pure coal combustion and different ammonia-mixed ratio conditions, and automatically generates comparison reports of indicators such as thermal efficiency, flue gas temperature, various heat losses, and steam parameters.

[0041] Based on the above thermal calculations, and combined with Figure 9 as well as Figure 10 It can display a comparison chart of the core combustion characteristic parameters for different ammonia-mixing ratios. Figure 9The impact of ammonia blending ratio on burnout rate (fuel utilization efficiency) and NOx emissions was compared. The core trend was that a higher ammonia blending ratio resulted in a slightly lower burnout rate (smaller efficiency loss) but a significant reduction in NOx emissions. Ultimately, the trade-off between environmental benefits and efficiency losses was quantified, demonstrating the significant overall benefits of ammonia blending combustion. Figure 10 The graph shows the impact of the ammonia blending ratio on boiler thermal efficiency and flue gas temperature. A higher ammonia blending ratio leads to a slight decrease in thermal efficiency, but a slight increase in flue gas temperature. Ultimately, this graph reflects the effect of ammonia blending combustion on boiler thermal performance, with the efficiency loss remaining within an acceptable range for engineering applications.

[0042] In one optional embodiment, the combustion reaction calculation, boiler heat balance calculation, and heat transfer calculation of the heating surface can be implemented according to a family of commonly used thermodynamic calculation formulas in the art, for example:

[0043] a) Theoretical air volume or actual air volume: The theoretical oxygen demand is obtained by stoichiometric calculation based on fuel element analysis and the theoretical air volume L0 is converted. The actual air volume L = λ * L0, where λ is the excess air coefficient.

[0044] b) Flue gas composition and flue gas volume: Based on the conservation of elements and the excess air coefficient, calculate the mole fraction and volume or mass flow rate of CO2, H2O, N2, O2 and nitrogen-containing components;

[0045] c) Theoretical or actual combustion temperature: The adiabatic flame temperature Tad is solved iteratively using the energy conservation principle (reaction exothermic = product sensible heat + heat loss), and then corrected by considering furnace heat loss, incomplete combustion loss, etc. to obtain the actual combustion temperature.

[0046] d) Boiler efficiency and heat balance: According to the heat balance relationship, input heat = effective heat absorption + various heat losses (flue gas heat loss, mechanical or chemical incomplete combustion loss, heat dissipation loss, etc.), calculate the boiler efficiency and heat loss ratio, and generate a heat balance table;

[0047] e) Radiative heat transfer on the heated surface: using the Stefan-Boltzmann form Q rad =σ*ε*A*(T g 4 -T w 4 The radiative heat absorption Q is calculated by combining the smoke opacity or effective radiative heat transfer coefficient. rad Where σ is the Stefan-Boltzmann constant, ε is the emissivity, A is the effective area involved in radiative heat transfer, and T g 4 The characteristic temperature of flue gas (or radiant gas) is usually taken as the flue gas temperature near a certain cross section of the furnace or the heating surface, T. w 4The surface temperature of the heated surface (pipe wall temperature / wall temperature); the heat transfer rate of the heated surface is Q. conv Press Q conv =h*A*ΔTlm calculates the convective heat transfer, where ΔTlm is the logarithmic mean temperature difference, obtained by subtracting the end temperature difference at both ends of the heated surface and performing logarithmic processing, and then dividing the two results. The heat transfer coefficient h can be determined by the correlation Nu=f(Re,Pr), where Nu is the Nusselt number, Re is the Reynolds number, and Pr is the Prandtl number.

[0048] The formulas and parameters mentioned above are common implementation methods in the field of boiler thermal calculation. They can be implemented by custom thermal calculation software programs through iterative calculation and report output. The above examples are for illustration only and do not constitute a limitation.

[0049] (3) Results output and analysis functions. Visual display of calculation results: output fuel physical property parameters, combustion calculation results, heat balance data, and heat transfer data of heating surface in tabular form; display the correlation between the ammonia mixing ratio and key indicators such as thermal efficiency, flue gas temperature, and combustion temperature in line graphs and bar graphs; Report generation and export: support one-click generation of standardized thermal calculation reports, which include calculation basis, parameter settings, detailed calculation process, result analysis and conclusions; Over-limit warning function: automatically warn users of abnormal operating conditions such as excessively high or low combustion temperature, overheating of heating surface, and sudden drop in thermal efficiency that may occur during ammonia mixing combustion, prompting users to adjust the ammonia mixing ratio or operating parameters.

[0050] Furthermore, the result output and analysis function can also output model calibration and prediction accuracy evaluation images and statistical tables, including: scatter plots of predicted NOx and simulated or measured NOx, error distribution plots, statistical tables of indicators such as RMSE, MAE or R2, and convergence curves of the ammonia blending ratio iteration process; and can output comparative reports of NOx average value or exceedance time under different control strategies, unburned ammonia (NH3 escape characterization index) and boiler efficiency, so as to quantitatively verify the improvement of NOx control effect and ammonia utilization efficiency.

[0051] It should be explained that the ammonia-coal co-combustion NOx prediction model is used to calculate the formation and reduction trends of NOx during the co-combustion of ammonia and coal. Based on key factors such as the combustion chemical reaction mechanism, temperature field distribution, mixing state, and ammonia injection method, this model achieves rapid prediction of NOx emission concentrations through coupled analysis of thermodynamic parameters, ammonia cracking reactions, and fuel-type and thermodynamic NOx. The model can be used for online control of the ammonia injection ratio, enabling the system to reduce NH3 slip while ensuring low NOx emissions and improving ammonia utilization efficiency.

[0052] The kinetic model is primarily used to describe the chemical reaction rate process between ammonia and coal under high-temperature conditions, including ammonia decomposition, ammonia reduction to NO, coal combustion, volatile matter release, and carbon combustion. This model characterizes the stepwise reaction pathway between ammonia and coal through reaction kinetic equations and rate constants, and can simulate the thermal decomposition rate of ammonia, the conversion patterns of nitrogen-based intermediates (NH4+, NH2), and the changes in nitrogen-oxygen reaction rates. The kinetic model provides fundamental data for CFD or rapid simulations, making predictions of the ammonia-coal combustion process more accurate and faster.

[0053] Correcting the radiative heat transfer coefficient alters the composition of the flue gas within the furnace (e.g., changes in H2O, N2, and NOx content), thus affecting the flue gas radiation characteristics. The corrected radiative heat transfer coefficient model adjusts the coefficient by calculating the impact of gas composition changes on radiation absorptivity and emissivity, making the heat transfer simulation more closely resemble real-world conditions. The corrected radiative coefficient can improve the accuracy of thermal calculation models, enhance furnace temperature prediction and combustion distribution simulation, and provide a reliable basis for thermal calculations under ammonia injection control and load variations.

[0054] The performance of the NOx prediction model and the thermal calculation model is evaluated based on the results of thermodynamic calculations. Based on the evaluation results, it is determined whether to adjust and optimize the ammonia blending ratio to improve the accuracy of the ammonia blending ratio during the combustion of ammonia.

[0055] If the ammonia blending ratio is adjusted and optimized, online simulation will continue in the specified simulation software based on the results of the ammonia blending ratio adjustment and optimization until the NOx prediction model and the thermodynamic calculation model converge. The corresponding converged NOx prediction model and converged thermodynamic calculation model will be output. Otherwise, the NOx prediction model and the thermodynamic calculation model will be output as the corresponding converged models.

[0056] In this embodiment, by constructing and precisely calibrating an ammonia-coal co-combustion model group, a unified modeling of the NOx formation mechanism, combustion dynamics, and high-moisture flue gas radiative heat transfer effect during ammonia-coal co-combustion was achieved. This transformed the complex reaction process into quantifiable data output, improving the accuracy of numerical simulations in relation to real boiler operating conditions. Furthermore, based on the thermodynamic calculation results generated from the online simulation, the performance of the NOx prediction model and the thermodynamic calculation model was evaluated, allowing for timely identification of model deviation sources and changes in operating status, thus ensuring the reliability and consistency of the prediction results. Building upon this, by introducing ammonia blending ratio adjustment optimization and re-inputting the optimization results into the simulation to form a closed-loop iteration, automated correction of the ammonia blending amount was achieved. This enabled the model to continuously approach real operating conditions and achieve convergence results. The final converged model output possesses high stability and accuracy, which can be used to guide actual ammonia-coal co-combustion operation in boilers, achieving more precise NOx control, a more reasonable ammonia blending ratio configuration, and better combustion thermodynamic performance. The solution provided in this embodiment not only improves the engineering applicability of ammonia-coal co-combustion simulation but also enhances the precision of ammonia blending control.

[0057] Furthermore, a rapid NOx prediction model and a thermodynamic calculation model were obtained through model construction and calibration using an ammonia-coal co-combustion model set. The specific process is as follows:

[0058] The researchers will run a kinetic model and correct the radiative heat transfer coefficient in a designated simulation software based on reference combustion data. The reference combustion data includes, but is not limited to, different coal qualities, ammonia blending ratios, and load ranges.

[0059] During the simulation process using the specified simulation software, key data is extracted to train and calibrate the NOx prediction model. Key data includes, but is not limited to, the NOx concentration at the boiler outlet, the temperature of key cross sections inside the furnace, and the component concentration.

[0060] The accuracy of the corrected radiative heat transfer coefficient is verified by simulating the temperature field and heat flow distribution output by specified simulation software, thereby ensuring the reliability of the thermal calculation module. This leads to the acquisition of a fast NOx prediction model and a thermal calculation model. Both the fast NOx prediction model and the thermal calculation model are trained and optimized by a high-precision kinetic model, enabling them to predict actual operating conditions more reliably.

[0061] In this embodiment, by running a kinetic model and correcting the radiative heat transfer coefficient in a specified simulation software, and by extracting key operational data such as the NOx concentration at the boiler outlet, the temperature of key cross-sections in the furnace, and the concentration of components during the simulation process, the NOx prediction model is trained and calibrated. This allows the model to more accurately reflect the actual combustion chemical reaction path and heat transfer coupling changes, thereby improving the engineering reliability of the prediction results. At the same time, by constructing a rapid NOx prediction model and a thermodynamic calculation model through the temperature field and heat flow distribution output by the simulation, the originally computationally intensive and time-consuming three-dimensional detailed simulation is transformed into a lightweight model, achieving combustion performance prediction with both higher accuracy and higher speed. Furthermore, this method not only helps to improve the model generation efficiency and real-time calculation, but also enhances the model's adaptability to different loads, different coal qualities, and different ammonia blending ratios, providing a more engineering-practical high-precision tool for ammonia blending combustion optimization, NOx emission control, and boiler thermal efficiency improvement.

[0062] like Figure 2 The diagram shown is a flowchart illustrating the output of thermal calculation results provided in an embodiment of the present invention. The specific logic is as follows: First, the ammonia-mixed combustion data is input into the specified simulation software; then, the obtained rapid NOx prediction model and thermal calculation model are used to perform the simulation in the specified simulation software; next, the thermal calculation results are output; finally, the performance is judged based on the output thermal calculation results to determine whether the ammonia blending ratio needs to be adjusted and optimized. If the performance does not meet the standards, the ammonia blending ratio in the input parameters is adjusted, and step two is continued until the optimal ammonia blending ratio is reached. Through the above process, the simulation accuracy of the ammonia-mixed combustion process is improved, and the boiler combustion thermal performance is also improved.

[0063] Furthermore, an online simulation is run in the specified simulation software to output the corresponding thermodynamic calculation results. The specific steps are as follows:

[0064] Step 1: Input the mixed ammonia combustion data into the specified simulation software. The mixed ammonia combustion data includes, but is not limited to, coal quality, ammonia blending ratio, and load.

[0065] Step two: Perform simulations using the obtained fast NOx prediction model and thermal calculation model in the specified simulation software.

[0066] Step 3: Output the thermal calculation results, which include, but are not limited to, NOx emission characteristics, furnace temperature field, heat flow distribution, and boiler combustion data.

[0067] Step 4: Determine whether the performance meets the standard based on the output thermodynamic calculation results to determine whether to adjust and optimize the ammonia doping ratio. If it does not meet the standard, adjust the ammonia doping ratio in the input parameters, and then continue to execute Step 2 to perform a fast simulation again until the optimal ammonia doping ratio is reached, that is, the ammonia doping ratio corresponding to zero error. If it meets the standard, output the NOx prediction model and the thermodynamic calculation model as the corresponding convergence model.

[0068] In one optional implementation, zero error can be defined as follows: within a preset iteration window, all performance constraints are met and the predicted or calculated output tends to be stable: NOx emission concentration ≤ maximum NOx emission concentration, unburned ammonia content ≤ upper limit of unburned ammonia content, boiler efficiency ≥ lower limit of boiler efficiency, and unburned carbon content ≤ maximum threshold of unburned carbon content; simultaneously, in K consecutive iterations, the changes in key output quantities satisfy |ΔNOx|<ε1, |ΔNH3|<ε2, and |Δη|<ε3 (e.g., K=3, ε1=5mg or Nm3, ε2=2ppm, ε3=0.1%), where ε1, ε2, and ε3 are the corresponding thresholds or limits, then the NOx prediction model and the thermal calculation model are determined to have converged, and the converged model is output; if the maximum number of iterations is reached but the conditions are not met, an anomaly warning is triggered and a rollback strategy is executed.

[0069] In this embodiment, by inputting ammonia-blended combustion data (including coal quality, ammonia blending ratio, and load) into designated simulation software and conducting simulations using a rapid NOx prediction model and a thermodynamic calculation model, efficient and accurate analysis of the ammonia-blended combustion process is achieved. The output of thermodynamic calculation results, including NOx emission characteristics, furnace temperature field, heat flow distribution, and boiler combustion data, helps to comprehensively reflect the combustion state and emission performance, providing a complete basis for operation control. Simultaneously, performance compliance judgment based on the output results helps to identify in real time whether the combustion effect and environmental indicators meet requirements, thereby deciding whether to adjust and optimize the ammonia blending ratio, thus improving the scientific nature and timeliness of control decisions. Furthermore, if the requirements are not met, the system can automatically adjust the ammonia blending ratio and re-simulate, achieving closed-loop optimization and iterative convergence of the ammonia blending strategy until the optimal ammonia blending ratio is obtained. This scheme is beneficial for improving the modeling efficiency and simulation accuracy of the ammonia-blended combustion process, avoiding reliance on manual adjustments based on experience, and improving NOx emission control effectiveness, ammonia utilization efficiency, and boiler combustion thermodynamic performance.

[0070] Furthermore, the specific process for determining whether to optimize the ammonia blending ratio based on the judgment results is as follows:

[0071] Obtain boiler combustion data from the output thermodynamic calculation results. Boiler combustion data includes NOx emission concentration, unburned ammonia content, boiler efficiency, and unburned carbon content.

[0072] The boiler combustion data is compared with the corresponding set thresholds to determine compliance. If the compliance result meets the preset conditions, it means that the standard is met; otherwise, it means that the standard is not met, and the ammonia blending ratio is adjusted and optimized.

[0073] Figure 11 This chart describes the accuracy analysis of the NOx prediction model for coal-fired ammonia combustion. It was used for model validation by comparing the model's predicted NOx emissions with actual measured values. The blue dashed line (ideal fit line) and the red solid line (actual fit line) in the chart show that the blue data points closely follow the ideal fit line (y=x), indicating a high degree of agreement between the predicted and actual values. Significance: This validates the accuracy of the NOx prediction model and also indirectly reflects the difference between the actual and ideal values.

[0074] The thresholds include the maximum limit for NOx emission concentration, the upper limit for unburned ammonia content, the lower limit for boiler efficiency, and the maximum threshold for unburned carbon content. When the non-compliance criteria are NOx emission concentration greater than the maximum limit for NOx emission concentration, unburned ammonia content greater than the upper limit for unburned ammonia content, boiler efficiency less than the lower limit for boiler efficiency, and unburned carbon content greater than the maximum threshold for unburned carbon content, if there are more boiler combustion data than the preset number set by researchers that meet the non-compliance criteria, it means that the non-compliance is not met; otherwise, it means that the non-compliance is met.

[0075] The specific process for adjusting and optimizing the ammonia blending ratio is as follows:

[0076] The deviation between the measured ammonia flow rate and the set ammonia flow rate during the current preset simulation period is obtained in real time by a specified simulation software. This deviation is denoted as the ammonia blending deviation, which is the ratio of the difference between the measured ammonia flow rate and the set ammonia flow rate to the set ammonia flow rate. This deviation is then matched with the ammonia blending deviation tolerance range, which represents the maximum deviation of the ammonia blending ratio that can be tolerated. The ammonia blending deviation tolerance range is obtained from a preset database and is preset by researchers based on empirical rules.

[0077] In one optional embodiment, the ammonia blending deviation can be further expressed as a relative deviation δ=(Q_NH3^meas-Q_NH3^set) or Q_NH3^set, where Q_NH3^meas is the measured ammonia flow rate and Q_NH3^set is the set ammonia flow rate; the ammonia blending deviation tolerance range can be set to [δmin, δmax], for example, δmin=-0.03, δmax=+0.03 or ±0.02 under stricter control; its determination method can be calculated based on the flowmeter accuracy, the steady-state deviation of the control loop and the quantile of historical stable operating data (e.g., covering 95% of stable samples), and written into a preset database in the form of configuration items for online reading.

[0078] Furthermore, the abnormal ammonia blending ratio threshold can be used to distinguish between strong adjustment and fine adjustment scenarios. For example, an abnormal ammonia blending ratio greater than 0.5 can be judged as a significant abnormality to trigger the first adjustment strategy. The threshold can be determined and stored based on the statistical inflection point in historical data where exceeding the limit leads to NOx exceeding the standard or a significant increase in NH3 escape. The above examples are for illustration only and do not constitute a limitation.

[0079] In one optional implementation, the maximum limit for coal flow fluctuation, the threshold for fuel characteristic change, and the limit for combustion characteristic determination can be set using a normalized threshold setting method: when the quantized value of coal flow signal fluctuation and the quantized value of fuel characteristic change have both been mapped to [0, 1], the maximum limit for coal flow fluctuation can be set to 0.2, the threshold for fuel characteristic change to 0.2, and the limit for combustion characteristic determination to 0.25, for example; or it can be adaptively updated based on historical compliance data according to different boilers or coal types. The above examples are for illustration only and do not constitute a limitation.

[0080] If the ammonia blending deviation falls within the ammonia blending deviation tolerance range, it indicates that the ammonia blending ratio for the current simulation period is acceptable; otherwise, it indicates that the ammonia blending ratio for the current simulation period is unacceptable, and an ammonia blending ratio anomaly degree judgment is performed. Specifically, the judgment involves analyzing the changes in coal flow signals and fuel characteristics during the simulation period to determine whether to adjust the ammonia blending ratio. If the ammonia blending ratio is acceptable, the NOx prediction model and the thermodynamic calculation model are output as the corresponding convergence models.

[0081] In this embodiment, by extracting and judging key boiler combustion data from the thermal calculation results, a digital and closed-loop mechanism for evaluating the combustion performance of ammonia blending and optimizing ammonia blending is formed. By comparing NOx emission concentration, unburned ammonia content, boiler efficiency, and unburned carbon content with set thresholds, this method not only comprehensively assesses the environmental friendliness, economy, and combustion completeness of the current ammonia blending combustion operation, but also accurately determines whether the combustion state meets the standards. Furthermore, in cases where standards are not met, the deviation between the measured ammonia flow rate and the set ammonia flow rate is used as the core adjustment basis. Combined with the ammonia blending deviation tolerance range, this effectively identifies whether the ammonia injection rate is reasonable, avoiding unnecessary adjustments due to slight fluctuations. When the deviation exceeds the tolerance range, by analyzing coal flow signal fluctuations and fuel characteristic changes, the source of the abnormal ammonia blending ratio is further determined, thus distinguishing between ammonia injection anomalies and deviations caused by fuel disturbances, making ammonia blending control more targeted and stable. The method provided in this embodiment achieves refined monitoring, high-precision judgment, and intelligent adjustment of the ammonia blending combustion process, which helps improve NOx control, reduce NH3 escape, increase boiler efficiency, and enhance the system's adaptability and operational reliability under varying operating conditions.

[0082] like Figure 3The diagram shows a flowchart for determining the degree of ammonia blending abnormality according to an embodiment of the present invention. The specific logic is as follows: if the ammonia blending abnormality ratio is greater than the ammonia blending abnormality threshold value used to classify the degree of ammonia blending abnormality, the total ammonia flow rate adjustment ratio is obtained based on the ammonia blending abnormality difference mapping to obtain an optimized ammonia blending ratio, which is then input into the designated simulation software for simulation again using the fast NOx prediction model and the thermodynamic calculation model. Otherwise, the fine-tuning control ratio is obtained based on the projection of the ammonia blending abnormality ratio, and the optimized ammonia blending ratio is obtained through the fine-tuning control ratio, which is then input into the designated simulation software for simulation again. Through the above process, not only is the efficiency of ammonia blending adjustment improved, but the accuracy of the ammonia blending ratio is also helped to ensure the accuracy of the thermodynamic calculation results during the ammonia blending combustion process.

[0083] As a further embodiment, the specific procedure for determining the degree of ammonia blending abnormality is as follows:

[0084] If the abnormal ammonia blending ratio is greater than the ammonia blending ratio abnormality threshold used to classify the degree of ammonia blending ratio abnormality, the first ammonia blending ratio adjustment strategy is adopted; otherwise, the second ammonia blending ratio adjustment strategy is adopted. The abnormal ammonia blending ratio is the deviation between the ammonia blending deviation and the ammonia blending deviation tolerance range. That is, when the ammonia blending deviation is greater than the maximum value of the ammonia blending deviation tolerance range, the abnormal ammonia blending ratio is the ratio between the difference between the ammonia blending deviation and the maximum value of the ammonia blending deviation tolerance range, and the maximum value of the ammonia blending deviation tolerance range; or when the ammonia blending deviation is less than the minimum value of the ammonia blending deviation tolerance range, the abnormal ammonia blending ratio is the ratio between the difference between the ammonia blending deviation and the minimum value of the ammonia blending deviation tolerance range, and the minimum value of the ammonia blending deviation tolerance range.

[0085] While monitoring the ammonia blending ratio, the coal flow signals and fuel characteristic changes during simulated periods when the ammonia blending ratio was not up to standard were analyzed:

[0086] The first adjustment strategy for the ammonia blending ratio is specifically as follows: the total ammonia flow rate adjustment ratio is obtained based on the mapping of the ammonia blending anomaly difference to obtain the optimized ammonia blending ratio. This means that the optimized total ammonia flow rate is obtained by multiplying the total ammonia flow rate adjustment ratio with the initial total ammonia flow rate. The optimized ammonia blending ratio is updated based on the optimized total ammonia flow rate. This means that the optimized ammonia blending ratio is recalculated based on the obtained optimized total ammonia flow rate and then input into the specified simulation software to re-simulate using the fast NOx prediction model and the thermodynamic calculation model. The ammonia blending anomaly difference represents the difference between the ammonia blending anomaly ratio and the ammonia blending ratio anomaly boundary value.

[0087] Specifically, the ammonia blending anomaly difference is input into the ammonia flow rate regulation mapping table, and the corresponding ammonia total flow rate regulation ratio is output. The ammonia flow rate regulation mapping table is used to reflect the mapping relationship between the ammonia blending anomaly difference and the ammonia total flow rate regulation ratio. Typically, researchers input the ammonia blending anomaly difference over a historical period, along with the ammonia total flow rate regulation ratio set based on empirical rules, into an initial data table constructed using a logistic regression algorithm. The cross-entropy loss function is used as the optimization criterion, and the table is trained using the scikit-learn framework to obtain the ammonia flow rate regulation mapping table.

[0088] In one alternative implementation, logistic regression training can employ L2 regularization and minimize cross-entropy loss through iterative optimization. Training convergence is determined when the loss decreases below a preset threshold or reaches the maximum number of iterations after several consecutive iterations. Simultaneously, input features can be standardized and outliers removed to improve training stability. The above examples are for illustrative purposes only and do not constitute limitations.

[0089] The second ammonia blending ratio adjustment strategy is as follows: a fine-tuning ratio is obtained based on the projection of the abnormal ammonia blending ratio, and an optimized ammonia blending ratio is obtained by fine-tuning the control ratio. Specifically, the fine-tuning control ratio is multiplied by the initial ammonia blending ratio to obtain the fine-tuned optimized ammonia blending ratio, which is then input into the specified simulation software for simulation using the fast NOx prediction model and the thermodynamic calculation model. The first ammonia blending ratio adjustment strategy is used to adjust and optimize the ammonia blending ratio, and its adjustment and optimization strength is higher than that of the second ammonia blending ratio adjustment strategy.

[0090] Specifically, the ammonia doping anomaly ratio is input into the fine-tuning projection sequence to output the corresponding fine-tuning control ratio. The fine-tuning projection sequence is obtained by researchers through training on the ammonia doping anomaly ratio obtained in historical time periods and the fine-tuning control ratio set according to empirical rules. This sequence is then input into the initial sequence constructed by the logistic regression algorithm and trained using the least squares criterion and the statsmodels framework. This allows for fitting the correlation between the ammonia doping anomaly ratio and the fine-tuning control ratio.

[0091] To ensure the reproducibility of the mapping table, projection sequence, or adjustment set in specific applications, historical samples can be trained, validated, or cross-validated during the training phase, and evaluation metrics such as R2, RMSE, and MAE can be output. For example, an acceptance criterion can be set as R2 ≥ 0.90 and RMSE below a preset threshold on the test set. After training, the model parameters, version number, applicable operating conditions, and corresponding thresholds can be fixed and stored together to support version management and backtracking under different coal types or load switching. The above examples are for illustration only and do not constitute a limitation.

[0092] When using the least squares criterion for fitting, the fitted residual distribution can be output and robust processing can be performed on outlier residual samples (such as weighted least squares or optional implementations of Huber robust regression) to improve the fit robustness under operating condition disturbances. The above examples are for illustration only and do not constitute a limitation.

[0093] In this embodiment, by introducing a graded control mechanism for abnormal ammonia blending ratios, refined and tiered adjustment of the ammonia blending ratio optimization is achieved, effectively improving the accuracy of ammonia blending combustion control and system stability. By calculating the abnormal ammonia blending ratio and comparing it with the ammonia blending ratio abnormality threshold, the severity of the abnormality can be distinguished, thereby selecting adjustment strategies of different intensities to avoid over-adjustment or under-adjustment, making ammonia blending control more rigorous and reasonable. Furthermore, for cases with a large degree of abnormality, the first adjustment strategy for the ammonia blending ratio is adopted. By mapping the ammonia blending abnormality difference to the total ammonia flow rate adjustment ratio, it is beneficial to quickly correct significantly deviated ammonia injection amounts, enabling the system to quickly return to a reasonable operating range under high deviation conditions. In cases of minor anomalies, a second ammonia blending ratio adjustment strategy is adopted. By projecting the anomaly ratio, a fine-tuning ratio is obtained, which can accurately correct minor deviations, reduce system oscillations, and improve control stability. At the same time, during the monitoring of the ammonia blending ratio, the fluctuations in coal flow and changes in fuel characteristics during all unqualified simulation periods are analyzed synchronously. This can identify the source of deviations and avoid misjudging ammonia injection anomalies due to fuel disturbances, making ammonia blending adjustment more targeted and reliable. The overall scheme of this embodiment improves the efficiency of ammonia blending adjustment, reduces the risk of NH3 escape, enhances NOx control, and strengthens the adaptability and robustness of the system under dynamic operating conditions, thereby improving the stability of ammonia-blended combustion in the boiler.

[0094] Furthermore, the changes in coal flow signals and fuel characteristics during the simulation period are analyzed. The specific process is as follows:

[0095] The first step is to acquire coal flow signal fluctuation data and fuel characteristic change data during the simulation period, respectively, and obtain the quantified values ​​of coal flow signal fluctuation and fuel characteristic change. The coal flow signal fluctuation data includes the coal flow fluctuation amplitude, relative fluctuation rate, control stability index, and signal delay. The fuel characteristic change data includes heat input stability, furnace temperature-coal flow correlation coefficient, coal mill power-flow ratio, and flame stability index. The above coal flow signal fluctuation data and fuel characteristic change data are read from the specified simulation software.

[0096] Specifically, the coal flow fluctuation amplitude is used to characterize the absolute magnitude of the instantaneous fluctuation of coal flow, and is obtained through the coal flow measurement signal at the coal mill outlet or coal feeder; the relative volatility is a dimensionless index used to measure the relative range of coal flow change, which is the ratio of the standard deviation to the mean of the coal flow time series signal; the control stability index is extracted from the control performance data of the coal feeder regulation loop and is used to describe the stability of the coal flow control loop; the signal delay is the grinding lag of the coal mill, the lag of the transmission pipeline, and the filtering delay of the sensor, which is the ratio of the time difference between the change of the control command and the change of the measured value.

[0097] Specifically, heat input stability is used to characterize the stability of heat release in the combustion system, and is the product of fuel flow rate and lower heating value; the furnace temperature-coal flow correlation coefficient is obtained by collecting furnace temperature measurement points (such as flue gas thermometers and radiation temperature measurement points) and corresponding coal flow signals, and calculating the Pearson correlation coefficient within a time window, which is used to measure the degree of matching between coal quantity changes and temperature response; the coal mill power-flow ratio characterizes whether the grindability of coal, grinding load and output capacity are matched, and is the ratio of coal mill motor power to coal flow rate; flame stability index is usually characterized by flame brightness fluctuation coefficient, which is used to measure whether the flame is continuous, uniform, concentrated, does not jump, and does not flicker.

[0098] It needs to be explained that the quantized value of coal flow signal fluctuation Where x represents the fluctuation data of each coal flow signal, i represents the number of each coal flow signal fluctuation data, and the maximum value of i is 4. All coal flow signal fluctuation data have been pre-normalized so that the value range is mapped between 0 and 1.

[0099] It should be added that signal delay is a fundamental system characteristic that directly affects control stability. Control stability is directly reflected in the fluctuation amplitude and relative volatility of coal flow. The greater the signal delay, the more difficult it is for the control system to adjust accurately and in a timely manner, resulting in poorer control stability. Poorer control stability means that the coal flow signal is more likely to deviate from the set value, resulting in large overshoot, undershoot, or continuous oscillation, thus leading to greater fluctuation amplitude and higher relative volatility. At the same time, fluctuation amplitude and relative volatility are both quantitative descriptions of the undesirable state of the coal flow signal, jointly reflecting the actual fluctuation of coal flow.

[0100] Similarly, the quantification value of fuel characteristic change can be obtained. , where y represents the change data of each fuel characteristic, j represents the number of each change data of each fuel characteristic, and the maximum value of j is 4. All change data of each fuel characteristic have been pre-normalized so that the numerical range is mapped between 0 and 1.

[0101] It should be added that there is a positive correlation between heat input stability and the correlation coefficient between furnace temperature and coal flow rate and the power-flow ratio of the coal mill. Stable heat input usually means a good correlation between furnace temperature and coal flow rate, as well as a reasonable coal mill power and coal flow rate ratio. The correlation coefficient between furnace temperature and coal flow rate is highly correlated with the flame stability index. At the same time, the stability of furnace temperature and the matching of coal flow rate directly affect the flame stability, while unstable coal flow rate and furnace temperature fluctuations often lead to flame instability. Furthermore, the power-flow ratio of the coal mill is also strongly correlated with the flame stability index. Appropriate coal mill power helps support a stable combustion process and improve flame stability.

[0102] The second step involves performing a difference calculation, i.e., a subtraction operation, between the quantified value of coal flow signal fluctuation and the quantified value of fuel characteristic change and the preset maximum limit of coal flow fluctuation and the fuel characteristic change threshold, respectively, to obtain the corresponding difference values, which are recorded as the difference between coal flow signal fluctuation and the difference between fuel characteristic change. The coal flow fluctuation threshold represents the maximum limit value for abnormal fluctuations in coal flow, and the fuel characteristic change threshold represents the maximum degree to which abnormal changes in fuel characteristics are limited. The maximum limit value for coal flow fluctuation and the fuel characteristic change threshold are both read from a preset database, which is generally preset and stored in the preset database by researchers for later retrieval.

[0103] The third step involves fusing the differences in coal flow signal fluctuations and fuel characteristic changes with their corresponding weights to obtain the total combustion characteristic deviation. This is achieved by multiplying the differences in coal flow signal fluctuations and fuel characteristic changes with their respective weights and then summing the results. The weights include the weights for coal flow signal fluctuations and fuel characteristic changes, both of which are preset values ​​from a pre-defined database. The quantified value for coal flow signal fluctuations is used to quantify the degree of fluctuation in the coal flow signal during the simulation period. The quantified value for fuel characteristic changes is used to quantify the degree of change in fuel characteristics during the simulation period. The total combustion characteristic deviation is used to comprehensively quantify the degree of deviation between the coal flow signal and the fluctuations in fuel characteristic changes.

[0104] In one optional implementation, the coal flow signal fluctuation weight w_c and the fuel characteristic change weight w_f can be w_c∈[0,1], w_f∈[0,1], and w_c+w_f=1. When the coal flow signal fluctuation contributes more significantly to the deviation, w_c=0.6 and w_f=0.4 can be exemplarily taken. When the fuel characteristic change is more significant, w_c=0.4 and w_f=0.6 can be exemplarily taken. The weights can be obtained through sensitivity analysis of historical data or parameter search with the goal of minimizing ammonia blending control error or exceeding the standard duration, and written into the preset database as the default configuration. The database is read and called during runtime. The above examples are for illustration only and do not constitute a limitation.

[0105] In this embodiment, a multi-dimensional, numerical combustion disturbance measurement system is established by systematically quantifying and fusing the fluctuations in coal flow signals and changes in fuel characteristics during the simulation period. Specifically, indicators such as the amplitude of coal flow fluctuations, relative volatility, control stability, and signal delay are quantified and compared with preset limits along with indicators such as heat input stability, furnace temperature-coal flow correlation coefficient, pulverizer power-flow ratio, and flame stability. This allows for a more accurate characterization of disturbance intensity and quantitative determination of the degree of exceeding limits. At the same time, the total combustion characteristic deviation obtained by weighted fusion compresses the complex multi-factor coupling influence into a single controllable indicator, facilitating automated decision-making and closed-loop regulation. Moreover, this method can distinguish the sources of deviation caused by fluctuations in the conveying system and changes in fuel properties, avoiding misjudgment and achieving targeted regulation. This improves the response speed and stability of ammonia blending and combustion control strategies, reduces NH3 escape and unburned carbon loss, and enhances boiler thermal efficiency.

[0106] Further, determine whether to adjust the ammonia blending ratio. The specific procedure is as follows:

[0107] The total combustion characteristic deviation is compared with the combustion characteristic judgment limit used to limit the maximum deviation of combustion characteristic changes. If the total combustion characteristic deviation is greater than the combustion characteristic judgment limit, the total combustion characteristic deviation is mapped to the ammonia combustion ratio adjustment set to obtain the corresponding coal flow adjustment ratio. The controlled ammonia blending ratio is obtained based on the coal flow adjustment ratio, that is, the coal flow adjustment ratio is multiplied by the initial coal flow to obtain the controlled coal flow. The controlled ammonia blending ratio is updated based on the controlled coal flow to obtain the controlled ammonia blending ratio. This means that the controlled ammonia blending ratio is recalculated based on the obtained controlled coal flow and input into the specified simulation software to re-simulate using the fast NOx prediction model and thermodynamic calculation model. Otherwise, no additional processing is performed.

[0108] It should be added that the combustion characteristic judgment limit is also a preset value, which is generally preset by researchers based on empirical rules and historical data; the ammonia combustion ratio adjustment set is used to reflect the mapping relationship between the total combustion characteristic deviation and the coal flow adjustment ratio. The researchers input the total combustion characteristic deviation obtained in the historical period and the coal flow adjustment ratio preset based on empirical rules into the initial dataset constructed by the linear regression algorithm, and train it based on the least squares criterion and the statsmodels framework. After the training is completed, the ammonia combustion ratio adjustment set is obtained.

[0109] In this embodiment, by comparing the total deviation of combustion characteristics with the combustion characteristic judgment limit, intelligent and quantitative determination of whether ammonia blending ratio adjustment is needed during ammonia blending combustion is achieved. When the deviation exceeds the limit, the system automatically calculates the coal flow adjustment ratio based on the mapping relationship of the total deviation in the ammonia blending combustion ratio adjustment set, and obtains a new ammonia blending ratio accordingly, realizing real-time adaptive optimization of the ammonia blending strategy. If the deviation does not exceed the limit, the current operating condition is maintained to avoid unnecessary adjustment and system oscillation. At the same time, by quantifying complex combustion disturbances into a unified deviation index and combining it with a mapping mechanism to realize the automatic conversion from disturbance degree to control intensity, this method can ensure that the ammonia blending strategy is corrected in a timely manner when combustion stability decreases, coal quality fluctuates, or load changes, thereby improving NOx control effect and ammonia utilization efficiency, and avoiding the increase of NH3 escape or decrease of thermal efficiency caused by over-adjustment, so that the system maintains the best combustion performance while ensuring emission compliance.

[0110] Furthermore, in addition to determining whether to adjust the ammonia blending ratio, this also includes:

[0111] If both the optimization of the ammonia doping ratio and the control of the ammonia doping ratio change are triggered simultaneously within the same simulation period, the corresponding ammonia doping ratio is obtained and recorded as the adjusted ammonia doping ratio.

[0112] If the ratio of ammonia blending changes obtained based on adjusting the ammonia blending ratio falls within the tolerance range for ammonia blending ratio changes used to limit the range of ammonia blending ratio changes, then the simulation will continue to be performed again in the specified simulation software using the fast NOx prediction model and the thermodynamic calculation model based on adjusting the ammonia blending ratio.

[0113] If the ammonia blending ratio change ratio obtained based on adjusting the ammonia blending ratio does not fall within the tolerance range for ammonia blending ratio changes, then the adjusted ammonia blending ratio is optimized based on the obtained ammonia blending ratio change deviation value. This means multiplying the ammonia blending ratio change deviation value with the adjusted ammonia blending ratio to obtain the optimized adjusted ammonia blending ratio, which is then input into the specified simulation software for simulation using the fast NOx prediction model and the thermodynamic calculation model. The ammonia blending ratio change ratio is the ratio between the difference between the adjusted ammonia blending ratio and the initial ammonia blending ratio and the initial ammonia blending ratio. The ammonia blending ratio change deviation value represents the deviation between the ammonia blending ratio change ratio and the tolerance range for ammonia blending ratio changes. The tolerance range for ammonia blending ratio changes is a preset range that is pre-defined by researchers based on empirical rules and stored in a preset database.

[0114] Specifically, if the change ratio of ammonia blending ratio is greater than the maximum value of the tolerance range for ammonia blending ratio change, the difference between the change ratio of ammonia blending ratio and the maximum value of the tolerance range for ammonia blending ratio change is calculated by ratioing the difference to the maximum value of the tolerance range for ammonia blending ratio change, and the absolute value is taken to obtain the ammonia blending ratio deviation value. If the change ratio of ammonia blending ratio is less than the minimum value of the tolerance range for ammonia blending ratio change, the difference between the change ratio of ammonia blending ratio and the minimum value of the tolerance range for ammonia blending ratio change is calculated by ratioing the difference to the minimum value of the tolerance range for ammonia blending ratio change, and the absolute value is taken to obtain the ammonia blending ratio deviation value.

[0115] In this embodiment, by coordinating the optimization of ammonia blending ratio and the control of ammonia blending ratio changes within the same simulation period, a unified, dynamic, and stable comprehensive optimization mechanism for ammonia blending ratio is formed. When both types of control are triggered simultaneously, the system first integrates the results of the two types of control to obtain the final adjusted ammonia blending ratio, so that the control action achieves a balance between forced correction and combustion disturbance correction, thereby avoiding repeated control or control conflicts. Furthermore, by calculating the change ratio of the adjusted ammonia blending ratio relative to the initial ammonia blending ratio and comparing it with the tolerance range of ammonia blending ratio changes, it helps to limit the excessive amplitude of a single adjustment and prevent furnace temperature field disturbance, increased NOx fluctuations, or increased NH3 escape caused by sudden changes in ammonia blending amount. Among them, the ammonia blending change deviation value is used to quantify the degree of excessive adjustment and perform compensatory optimization accordingly, thereby ensuring the effectiveness of correction while maintaining the stability and controllability of system control. This method not only realizes the unified integration of multi-source control strategies, adaptive constraints on the adjustment amplitude, and automatic correction of over-adjustment, but also forms a dynamic ammonia blending control mechanism that can quickly respond to combustion deviations and ensure system stability.

[0116] Furthermore, it simultaneously triggered the optimization of ammonia blending ratio adjustment and the regulation of ammonia blending ratio changes, and subsequently included:

[0117] Obtain the ammonia blending deviation for the next simulation period and determine the ammonia blending ratio qualification. If the ammonia blending ratio is qualified, continue to perform ammonia blending combustion based on the current ammonia blending ratio. Otherwise, use the difference between the current ammonia blending ratio and the ammonia blending ratio of the previous simulation period as the ammonia blending abnormal ratio correction value.

[0118] The ammonia blending ratio is adjusted based on the ammonia blending abnormality ratio repair value to obtain the repaired ammonia blending ratio. This is achieved by multiplying the preset ratio set by the researchers with the ammonia blending abnormality ratio repair value, and then summing the product with the adjusted ammonia blending ratio. The repaired ammonia blending ratio is then input into the specified simulation software and simulated again using the fast NOx prediction model and the thermodynamic calculation model.

[0119] If the ammonia blending ratio is still not up to standard after repair, an ammonia blending ratio anomaly warning will be issued, and the initial ammonia blending ratio will be restored. Otherwise, it indicates that the rapid NOx prediction model and the thermodynamic calculation model have converged, and the converged NOx prediction model and the converged thermodynamic calculation model will be output.

[0120] In this embodiment, a closed-loop ammonia blending ratio control mechanism involving judgment, adjustment, repair, and convergence is constructed to achieve high stability, high precision, and high robustness control of the ammonia blending combustion process. Real-time compliance judgment of ammonia blending deviations in each simulation period helps to promptly identify ammonia injection deviations, preventing accumulated deviations from leading to NOx exceeding limits or increased NH3 escape. When the ammonia blending ratio is unqualified, the system uses the difference between the current and previous ammonia blending ratios to construct an abnormal ammonia blending ratio repair value, making the correction action directional and targeted, thus improving adjustment efficiency and enhancing the suppression of deviation trends. The repaired ammonia blending ratio is then re-inputted into the model for simulation to verify the repair. The effectiveness of the measures is assessed. If the repair fails, an anomaly warning is triggered and the system automatically reverts to the initial ammonia blending ratio, effectively preventing system oscillations or deterioration of combustion conditions due to continuous erroneous adjustments. This provides assurance from a control safety perspective. If the repair is successful, it indicates that the rapid NOx prediction model and the thermodynamic calculation model have reached a stable convergence state under this condition and can output reliable convergence results. Overall, this solution possesses adaptive repair capabilities, anomaly protection capabilities, and model convergence judgment capabilities. It can maintain the stability of ammonia injection adjustment under dynamic conditions, improve NOx control effects, reduce the risk of NH3 escape, and improve combustion efficiency, providing important technical support for the intelligent and controllable operation of ammonia blending combustion.

[0121] Figure 4 A schematic diagram of a high-precision quantitative analysis system for ammonia combustion in coal-fired power plants provided in this embodiment of the invention is shown. The high-precision quantitative analysis system for ammonia combustion in coal-fired power plants includes: a model construction and calibration module, an ammonia blending ratio adjustment and optimization module, and a model convergence module.

[0122] The model building and calibration module is used to build and calibrate the NOx prediction model and thermodynamic calculation model for realizing the engineering quantification of ammonia-coal combustion through the ammonia-coal co-combustion model group. The model group is then used to perform online simulation in the specified simulation software and output the corresponding thermodynamic calculation results. The ammonia-coal co-combustion model group includes the ammonia-coal co-combustion NOx prediction model, the kinetic model and the corrected radiative heat transfer coefficient. The thermodynamic calculation results are the data set output by the NOx prediction model and the thermodynamic calculation model.

[0123] The ammonia blending ratio adjustment and optimization module is used to determine the performance of the NOx prediction model and the thermal calculation model through thermal calculation results, and to determine whether to perform ammonia blending ratio adjustment and optimization based on the determination results, so as to improve the accuracy of the ammonia blending ratio in the ammonia combustion process.

[0124] The model convergence module is used to continue online simulation in the specified simulation software based on the results of the ammonia doping ratio adjustment and optimization, until the NOx prediction model and the thermodynamic calculation model converge, and output the corresponding converged NOx prediction model and converged thermodynamic calculation model. Otherwise, the NOx prediction model and the thermodynamic calculation model are output as the corresponding converged models.

[0125] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0126] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device, equipment, and storage medium embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0127] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0128] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A high-precision quantitative analysis method for ammonia combustion in coal-fired power boilers, characterized in that... The method includes: A model is constructed and calibrated using an ammonia-coal co-combustion model set to obtain a NOx prediction model and a thermodynamic calculation model for realizing the engineering quantification of ammonia-coal co-combustion. The model is then simulated online in a specified simulation software, and the corresponding combustion characteristic parameters, visualization results, and thermodynamic calculation results are output. The ammonia-coal co-combustion model set includes an ammonia-coal co-combustion NOx prediction model, a kinetic model, and a modified radiative heat transfer coefficient. The thermodynamic calculation results are the data set output by the NOx prediction model and the thermodynamic calculation model. The specific steps for performing online simulation in designated simulation software and outputting the corresponding thermodynamic calculation results are as follows: Step 1: Input the mixed ammonia combustion data into the designated simulation software. The mixed ammonia combustion data includes coal quality, ammonia blending ratio, and load. Step 2: Perform simulations using the obtained fast NOx prediction model and thermodynamic calculation model in the specified simulation software; Step 3: Output combustion characteristic parameters and thermodynamic calculation results, including NOx emission characteristics, furnace temperature field, heat flow distribution, and boiler combustion data; Step 4: Determine whether the performance meets the standard based on the output combustion characteristic parameters and thermodynamic calculation results, so as to determine whether to adjust and optimize the ammonia blending ratio. If it does not meet the standard, adjust the ammonia blending ratio in the input parameters and continue to execute Step 2 until the optimal ammonia blending ratio is reached. The performance of the NOx prediction model and the thermal calculation model is evaluated based on the results of thermal calculations, and it is determined whether to optimize the ammonia doping ratio based on the evaluation results. The specific process for determining whether to optimize the ammonia blending ratio based on the judgment result is as follows: Obtain the boiler combustion data from the output combustion characteristic parameters and thermodynamic calculation results. The boiler combustion data includes NOx emission concentration, unburned ammonia content, boiler efficiency, and unburned carbon content. The boiler combustion data is compared with the corresponding set thresholds to determine compliance. If the compliance result meets the preset conditions, it means that the standard is met; otherwise, it means that the standard is not met, and the ammonia blending ratio is adjusted and optimized. The specific process for optimizing the ammonia blending ratio is as follows: The deviation between the measured ammonia flow rate and the set ammonia flow rate for the current preset simulation period is obtained in real time and recorded as the ammonia blending deviation. This deviation is then matched with the ammonia blending deviation tolerance range, which represents the maximum deviation of the tolerated ammonia blending ratio. If the ammonia blending deviation falls within the ammonia blending deviation tolerance range, it indicates that the ammonia blending ratio for the current simulation period is qualified; otherwise, it indicates that the ammonia blending ratio for the current simulation period is unqualified, and the degree of ammonia blending ratio abnormality is determined. Specifically, the determination is made by analyzing the changes in coal flow signals and fuel characteristics during the simulation period to determine whether to adjust the ammonia blending ratio. If the ammonia blending ratio is adjusted and optimized, online simulation will continue in the specified simulation software based on the results of the ammonia blending ratio adjustment and optimization until the NOx prediction model and the thermodynamic calculation model converge. The corresponding converged NOx prediction model and converged thermodynamic calculation model will be output. Otherwise, the NOx prediction model and the thermodynamic calculation model will be output as the corresponding converged models.

2. The high-precision quantitative analysis method for mixed ammonia combustion in coal-fired power boilers as described in claim 1, characterized in that, The NOx prediction model and thermodynamic calculation model for realizing the engineering quantification of ammonia-coal co-combustion are obtained through model construction and calibration using an ammonia-coal co-combustion model set. The specific process is as follows: The dynamic model and the corrected radiative heat transfer coefficient are run in the specified simulation software based on reference combustion data. During the simulation process using the specified simulation software, key data is extracted to train and calibrate the NOx prediction model. The key data includes the NOx concentration at the boiler outlet, the temperature of key cross sections inside the furnace, and the component concentration. By simulating the temperature field and heat flow distribution output by specified simulation software, a fast NOx prediction model and a thermodynamic calculation model are obtained.

3. The high-precision quantitative analysis method for mixed ammonia combustion in coal-fired power boilers as described in claim 1, characterized in that, The specific procedure for determining the degree of abnormality in the ammonia blending ratio is as follows: If the abnormal ammonia ratio is greater than the ammonia ratio abnormality threshold value used to classify the degree of ammonia ratio abnormality, the first ammonia ratio adjustment strategy is adopted; otherwise, the second ammonia ratio adjustment strategy is adopted. The abnormal ammonia ratio is the deviation value between the ammonia ratio deviation and the ammonia ratio deviation tolerance range. While monitoring the ammonia blending ratio, the coal flow signals and fuel characteristic changes during simulated periods when the ammonia blending ratio was not up to standard were analyzed: The first adjustment strategy for the ammonia blending ratio specifically means: obtaining the total ammonia flow rate adjustment ratio based on the ammonia blending anomaly difference mapping to obtain the optimized ammonia blending ratio, and inputting it into the specified simulation software to re-simulate using the fast NOx prediction model and the thermodynamic calculation model. The ammonia blending anomaly difference represents the difference between the ammonia blending anomaly ratio and the ammonia blending ratio anomaly boundary value. The second ammonia doping ratio adjustment strategy is as follows: a fine-tuning ratio is obtained based on the projection of the ammonia doping abnormal ratio, an optimized ammonia doping ratio is obtained by fine-tuning the ratio, and the optimized ratio is input into the specified simulation software for re-simulation. The first ammonia blending ratio adjustment strategy has a higher optimization effect than the second ammonia blending ratio adjustment strategy.

4. The high-precision quantitative analysis method for ammonia combustion in coal-fired power boilers as described in claim 1, characterized in that, The specific process for analyzing the changes in coal flow signals and fuel characteristics during the simulated period is as follows: The coal flow signal fluctuation data and fuel characteristic change data during the simulation period are acquired respectively, and the quantized values ​​of coal flow signal fluctuation and fuel characteristic change are obtained respectively. The coal flow signal fluctuation data includes coal flow fluctuation amplitude, relative fluctuation rate, control stability index and signal delay. The fuel characteristic change data includes heat input stability, furnace temperature-coal flow correlation coefficient, coal mill power-flow ratio and flame stability index. The difference between the quantized value of coal flow signal fluctuation and the quantized value of fuel characteristic change are calculated with the preset maximum limit of coal flow fluctuation and the threshold of fuel characteristic change, respectively, and the corresponding difference values ​​are recorded as the difference between coal flow signal fluctuation and fuel characteristic change. The total combustion characteristic deviation is obtained by fusing the difference in coal flow signal fluctuations and the difference in fuel characteristic changes with their corresponding weights.

5. The high-precision quantitative analysis method for ammonia combustion in coal-fired power boilers as described in claim 1, characterized in that, The specific process for determining whether to adjust the ammonia blending ratio is as follows: The total deviation of combustion characteristics is compared with the combustion characteristic judgment limit used to limit the maximum deviation of combustion characteristic changes: If the total deviation of combustion characteristics exceeds the combustion characteristic judgment limit, the total deviation of combustion characteristics is mapped to the ammonia combustion ratio adjustment set to obtain the corresponding coal flow adjustment ratio. The ammonia blending ratio is then adjusted based on the coal flow adjustment ratio and input into the specified simulation software for re-simulation. Otherwise, no additional processing is performed.

6. The high-precision quantitative analysis method for ammonia combustion in coal-fired power boilers as described in claim 1, characterized in that, The determination of whether to adjust the ammonia blending ratio also includes: If the optimization of ammonia doping ratio and the control of ammonia doping ratio change are triggered simultaneously within the same simulation period, the corresponding ammonia doping ratio is obtained and recorded as the adjusted ammonia doping ratio. If the ammonia ratio change ratio obtained by adjusting the ammonia ratio falls within the ammonia ratio change tolerance range used to limit the range of ammonia ratio change, then continue to perform simulation in the specified simulation software based on adjusting the ammonia ratio. If the ratio of ammonia blending changes obtained based on adjusting the ammonia blending ratio does not fall within the tolerance range of ammonia blending ratio changes, then the ammonia blending ratio is optimized based on the obtained deviation value of ammonia blending changes to obtain the optimized ammonia blending ratio, which is then input into the specified simulation software for simulation. The ammonia blending ratio change ratio is the ratio between the difference between the adjusted ammonia blending ratio and the initial ammonia blending ratio and the initial ammonia blending ratio. The deviation value of ammonia doping change represents the deviation between the ammonia doping ratio change ratio and the tolerance range of ammonia doping ratio change.

7. The high-precision quantitative analysis method for mixed ammonia combustion in coal-fired power boilers as described in claim 6, characterized in that, The simultaneous triggering of ammonia blending ratio adjustment optimization and ammonia blending ratio change control also includes: Obtain the ammonia blending deviation for the next simulation period and determine the ammonia blending ratio qualification. If the ammonia blending ratio is qualified, continue to perform ammonia blending combustion based on the current ammonia blending ratio. Otherwise, use the difference between the current ammonia blending ratio and the ammonia blending ratio of the previous simulation period as the ammonia blending abnormal ratio repair value. The ammonia blending ratio is adjusted according to the ammonia blending abnormal ratio repair value to obtain the repaired ammonia blending ratio. The repaired ammonia blending ratio is then input into the specified simulation software for re-simulation. If the ammonia blending ratio is still not up to standard after repair, an ammonia blending ratio anomaly warning will be issued, and the initial ammonia blending ratio will be restored. Otherwise, it indicates that the rapid NOx prediction model and the thermodynamic calculation model have converged, and the converged NOx prediction model and the converged thermodynamic calculation model will be output.

8. A high-precision quantitative analysis system for ammonia combustion in coal-fired power plants, used to implement the high-precision quantitative analysis method for ammonia combustion in coal-fired power plants as described in any one of claims 1-7, characterized in that, include: The module includes model building and calibration, ammonia doping ratio adjustment and optimization, and model convergence. The model construction and calibration module is used to construct and calibrate a NOx prediction model and a thermodynamic calculation model for realizing the engineering quantification of ammonia-coal combustion through ammonia-coal co-combustion model group, and to perform online simulation in a specified simulation software, outputting the corresponding combustion characteristic parameters, visualization results and thermodynamic calculation results. The thermodynamic calculation results are the data set output by the NOx prediction model and the thermodynamic calculation model. The ammonia blending ratio adjustment and optimization module is used to make a judgment based on combustion characteristic parameters and thermodynamic calculation results, and to determine whether to perform ammonia blending ratio adjustment and optimization based on the judgment results. The model convergence module is used to continue online simulation in the specified simulation software based on the results of the ammonia doping ratio adjustment and optimization, until the NOx prediction model and the thermodynamic calculation model converge, and output the corresponding converged NOx prediction model and converged thermodynamic calculation model; otherwise, the NOx prediction model and the thermodynamic calculation model are output as the corresponding converged models.

Images