Coal blending combustion performance prediction method and device based on thermogravimetric curve analysis

Abstract

This invention relates to the field of coal blending combustion performance prediction and boiler adaptability assessment, and particularly to a method and apparatus for predicting the combustion performance of blended coal based on thermogravimetric curve analysis. The method first establishes a baseline thermogravimetric curve database containing the design coal type and optimal adaptability coal types, and calculates the theoretical coal quality parameters of the blended coal based on a coal quality parameter calculation module. Based on the theoretical coal quality parameters of the blended coal and the kinetic characteristics of each individual coal included in the blend, a weighted algorithm is used to simulate the mass loss behavior of the blended coal during the heating process to obtain the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the blended coal. Multi-dimensional similarity analysis is performed on the curves, and the kinetic parameters in the kinetic characteristics are iteratively adjusted through an optimization algorithm to improve prediction accuracy. This invention can accurately predict the nonlinear superposition effect of the coal blending combustion kinetic process, achieving optimization of the coal blending scheme and dynamic matching of the combustion state.

Description

Technical Field

[0001] This invention relates to the field of coal blending combustion performance prediction and boiler adaptability assessment technology, and in particular to a method, apparatus and computer-readable storage medium for predicting coal blending combustion performance based on thermogravimetric analysis. Background Technology

[0002] In thermal power plant combustion systems, boiler combustion performance is closely related to coal characteristics. Each boiler is designed with a specific coal type during the design phase, and its combustion characteristics (ignition temperature, burnout temperature, volatile matter release rate, etc.) are matched with the boiler structure and combustion method to achieve a better balance in terms of thermal efficiency, pollution emissions, and safety. However, due to the diversification of coal sources, fluctuations in coal quality, and fuel cost pressures, power plants generally adopt a blended coal combustion strategy to achieve fuel economy and combustion stability. The essence of blended coal combustion is the thermochemical coupling of physical mixing and combustion reactions between different coal types, and its results are influenced by coal quality parameters (ash content, volatile matter, fixed carbon, alkali metal content, etc.), particle size distribution, and kinetic characteristics. Traditional blended coal evaluation methods are usually based on weighted averages of coal quality or empirical criteria (such as ash fusion point, volatile matter, sulfur content, etc.), but these indicators often cannot reflect the nonlinear superposition effects of combustion kinetics, thus limiting their accuracy in predicting blended coal combustion performance and boiler adaptability.

[0003] Since its inception, thermogravimetric analysis (TGA) has been a crucial tool for studying the pyrolysis and combustion characteristics of coal. The TG curve records the weight loss of a coal sample during heating, characterizing the reaction stage, burnout rate, and thermal stability of the fuel. By analyzing the TG and its differential curve DTG (differential thermogravimetric curve), the combustion activity, kinetic parameters, and stage characteristics of coal can be obtained. However, currently, TGA is mainly used for single-coal sample studies and coal type comparisons; its application in simulating the combustion characteristics and adaptability evaluation of blended coals is still in the exploratory stage. Existing technologies have several shortcomings and deficiencies. First, traditional calculations and analyses of blended coal quality still heavily rely on linear weighted calculations, which cannot reflect the interactions between blended coal types in pyrolysis reaction rates, gasification characteristics, and oxidation reactions, leading to significant discrepancies between theoretical predictions and actual combustion. Second, current technologies lack dynamic thermal analysis to support simulations of blended coal combustion characteristics, making it impossible to quantitatively describe the blended coal combustion process using TGA curves, and lacking a closed-loop structure of "simulation-verification-optimization." Furthermore, previous thermogravimetric analyses of coal samples were limited to the laboratory, lacking a data sharing and feedback mechanism with on-site combustion conditions. Experimental results were not fully utilized and could not effectively guide combustion. In summary, existing coal blending methods still have significant shortcomings in terms of thermal analysis support, dynamic evaluation, and similarity determination. There is an urgent need for a coal blending evaluation method based on thermogravimetric curve analysis and simulation to achieve multi-faceted, optimizable, and quantitative evaluation of the combustion performance of blended coal. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a method for predicting the combustion performance of blended coal based on thermogravimetric analysis, comprising:

[0006] S1. Establish a database of reference thermogravimetric curves containing the design coal type and the optimal adaptability coal type, and calculate the theoretical coal quality parameters of the blended coal based on the blended coal quality parameter calculation module. S2, based on the theoretical coal quality parameters of the blended coal and the kinetic characteristics of each individual coal included in the blended coal, a weighted algorithm is used to simulate the mass loss behavior of the blended coal during the heating process, so as to obtain the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the blended coal. S3. A multi-dimensional similarity analysis is performed between the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the mixed coal and the actual thermogravimetric curve and the differential thermogravimetric curve of the mixed coal. Based on the differences in curve shape, characteristic temperature deviation and weight loss rate, a parameter correction mechanism is initiated. The kinetic parameters in the kinetic characteristics are adjusted iteratively through the optimization algorithm to improve the prediction accuracy.

[0007] In one embodiment of the present invention, S2 further includes: S21. A single coal kinetic model was constructed using the Arrhenius kinetic model to describe the coal pyrolysis and combustion process of a single coal and to obtain the single coal kinetic parameters. S22, weights are set based on the proportion of various types of coal in the blend, and the overall reaction rate of the blend is calculated using a weighted superposition model; S23, based on the overall reaction rate of the mixed coal, the conversion rate of the mixed coal with temperature is calculated using the numerical integration method. Based on this conversion rate, the theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal are calculated. The calculation formula is as follows: ; ; in, The conversion rate of the mixed coal as a function of temperature. This refers to the initial mass of the mixed coal.

[0008] In one embodiment of the present invention, the formula for calculating the overall reaction rate of the mixed coal in step S22 is as follows: ; in, For the first i The blending ratio of different types of coal, For the first i Frequency factor of coal cultivation For the first i The apparent activation energy of coal For the first i The reaction order of the coal. R This is the universal gas constant. T This is the current temperature.

[0009] In one embodiment of the present invention, S3 further includes: S31. The actual thermogravimetric curve and differential thermogravimetric curve of the mixed coal are obtained through experiments, and its experimental characteristic points are extracted. The experimental characteristic points include ignition temperature, burnout temperature, maximum weight loss rate and stage ratio. S32. Using machine learning, the method judges whether there is a significant deviation between the theoretical and actual thermogravimetric curves of the mixed coal based on the morphological similarity between the theoretical and actual thermogravimetric curves, the deviation of ignition temperature, the deviation of the peak of maximum weight loss rate, the deviation of burnout temperature, and the difference in reaction stages with temperature. It also judges whether there is a significant difference between the theoretical and actual differential thermogravimetric curves based on the structural similarity between the peak shape of the theoretical and actual differential thermogravimetric curves. S33. When there is a deviation between the theoretical thermogravimetric curve and the actual thermogravimetric curve of the mixed coal, and the deviation exceeds a preset threshold, or when there is a deviation between the theoretical differential thermogravimetric curve and the actual differential thermogravimetric curve of the mixed coal, and the deviation exceeds a preset threshold, the kinetic parameters are corrected.

[0010] In one embodiment of the present invention, S32 further includes: S321, the morphological similarity between the theoretical and actual thermogravimetric curves of the blended coal is evaluated using the Pearson correlation coefficient. The calculation formula is as follows:

[0011] in, The thermogravimetric curve of the mixed coal in the first... i The values ​​of each temperature sampling point This represents the average response level of the entire mixed coal thermogravimetric curve. The thermogravimetric curve of the reference coal in the 1st... i The values ​​of each temperature sampling point This represents the average response level of the entire reference coal thermogravimetric curve. S322, calculate the deviations in ignition temperature, maximum weight loss rate peak, and burnout temperature between the theoretical and actual thermogravimetric curves. The calculation formula is as follows: ; ; ; in, This is the actual ignition temperature. The theoretical ignition temperature, This represents the temperature corresponding to the actual maximum rate of weight loss. This represents the temperature corresponding to the theoretical maximum rate of weight loss. This is the actual burnout temperature. The theoretical burnout temperature; S323, using conversion rate deviation and weight loss rate deviation, calculates the difference in reaction stages of blended coal with temperature. The formula for calculating the difference in reaction stages of blended coal with temperature is:

[0012] in, This represents the actual conversion rate. Theoretical conversion rate; S324, based on the number, intensity, width and interpeak distance of the curves, analyzes the similarity between the theoretical differential thermogravimetric curve of mixed coal and the actual differential thermogravimetric curve of mixed coal.

[0013] In one embodiment of the present invention, S33 further includes: S331, the loss function formula is defined by correcting for dynamic parameters as follows: ; in, , , These are the corresponding weighting coefficients, used to adjust the degree of influence of different physical constraints during the parameter correction process. This is used to characterize the degree of importance attached to the temperature deviation of the maximum weight loss rate, in order to constrain the consistency between the main combustion reaction zone of the blended coal and the reference coal. Used to characterize the constraint strength on burnout temperature deviation, in order to reflect the degree of matching between the burnout stage and tail reaction characteristics; This is used to adjust the weight of the overall thermogravimetric curve morphology similarity in the calibration process, so as to ensure that the simulated curve maintains a consistent trend with the experimental curve throughout the entire heating range. This represents the peak deviation of the maximum weightlessness rate. To correct the burnout temperature deviation, an optimization algorithm is used to correct the kinetic parameters based on the loss function; S332: Based on the corrected kinetic parameters, obtain the corrected theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal, and compare the similarity between the corrected theoretical thermogravimetric curve and the actual thermogravimetric curve. If there is a deviation between the corrected theoretical thermogravimetric curve and the actual thermogravimetric curve of the mixed coal and the deviation exceeds a preset threshold, or if there is a deviation between the corrected theoretical differential thermogravimetric curve and the actual differential thermogravimetric curve of the mixed coal and the deviation exceeds a preset threshold, repeat step S331.

[0014] In one embodiment of the present invention, the reference thermogravimetric curve database in step S1 includes actual thermogravimetric curves, actual differential thermogravimetric curves, and characteristic parameters of water precipitation peaks and combustion peaks in the actual differential thermogravimetric curves obtained under standardized thermogravimetric experimental conditions. The characteristic parameters of water precipitation peaks and combustion peaks in the actual differential thermogravimetric curves include ignition temperature, temperature corresponding to the maximum weight loss rate, burnout temperature, characteristic rate, and weight loss stage ratio, which serve as the core characteristic data of the database.

[0015] In one embodiment of the present invention, the theoretical coal quality parameters in step S1 include the ash content, volatile matter, fixed carbon, lower heating value and metal oxide content of each blended coal type, and the blended coal combustion kinetic input parameters are generated according to the blending ratio of each blended coal type.

[0016] To achieve the above objectives, a second aspect of the present invention provides a device for predicting the combustion performance of blended coal based on thermogravimetric analysis, comprising: The reference thermogravimetric curve database establishment module is used to establish a reference thermogravimetric curve database containing the design coal type and the optimal adaptability coal type, and to calculate the theoretical coal quality parameters of the blended coal based on the blended coal quality parameter calculation module. The coal quality parameter calculation module for blended coal uses a weighted algorithm to simulate the mass loss behavior of the blended coal during the heating process, based on the theoretical coal quality parameters of the blended coal and the kinetic characteristics of each individual coal contained in the blended coal, in order to obtain the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the blended coal. The multi-dimensional similarity analysis and parameter correction module is used to perform multi-dimensional similarity analysis between the theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal and the actual thermogravimetric curve and differential thermogravimetric curve of the mixed coal. Based on the differences in curve shape, characteristic temperature deviation and weight loss rate, the parameter correction mechanism is activated. The kinetic parameters in the kinetic characteristics are adjusted iteratively through the optimization algorithm to improve the prediction accuracy.

[0017] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0018] The method, apparatus, and storage medium of this invention can accurately predict the nonlinear superposition effect of the coal combustion dynamics process, significantly improve the quantitative evaluation accuracy of boiler combustion performance and adaptability, and realize the optimization of coal blending scheme and dynamic matching of combustion state.

[0019] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a method for predicting the combustion performance of blended coal based on thermogravimetric curve analysis according to an embodiment of the present invention; Figure 2 This is a structural diagram of a coal combustion performance prediction device based on thermogravimetric analysis according to an embodiment of the present invention. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0023] The following description, with reference to the accompanying drawings, describes a method and apparatus for predicting the combustion performance of mixed coal based on thermogravimetric analysis, according to an embodiment of the present invention.

[0024] Example 1 Figure 1 This is a flowchart of a method for predicting the combustion performance of mixed coal based on thermogravimetric analysis, according to an embodiment of the present invention.

[0025] like Figure 1 As shown, the method for predicting the combustion performance of blended coal based on thermogravimetric analysis includes the following steps: S1. Establish a database of reference thermogravimetric curves containing the design coal type and the optimal adaptability coal type, and calculate the theoretical coal quality parameters of the blended coal based on the blended coal quality parameter calculation module. Specifically, the step of "establishing a baseline thermogravimetric curve database for the design coal type and the optimal adaptable coal type" is the foundational step for quantitative evaluation and optimization decision-making of blended coal combustion performance in this invention. Its technical implementation principle is based on standardized thermogravimetric analysis (TGA) and curve characteristic parameter extraction, aiming to construct a representative database of coal pyrolysis and combustion behavior, providing a reliable reference standard for subsequent blended coal simulation and similarity analysis.

[0026] In some implementations, this step first selects the coal type used in the power plant design and a well-adapted coal type that has performed well in historical operation as benchmark samples. As one implementation method, in this invention, the coal sample must meet standard experimental conditions, including a temperature rise rate set to... The coal powder particle size was controlled at 40 mesh, and the coal sample mass was kept constant (usually 10-20 mg). The experimental atmosphere was air, with an air flow rate of approximately 280 L / h. Under these conditions, the mass change of the coal sample during the heating process was recorded using a thermogravimetric analyzer to obtain TG (thermogravimetric curve) and DTG (differential thermogravimetric curve). Data preprocessing was performed, such as normalization, smoothing, and noise removal, to improve data quality and comparability.

[0027] Subsequently, key characteristic parameters, including ignition temperature, were extracted from the DTG curve. Temperature corresponding to the maximum rate of weight loss Burnout temperature Characteristic rates (such as the maximum rate of weightlessness) The parameters include the ignition characteristics, combustion activity, and burnout capacity of the coal, as well as the mass loss ratio at each stage of weight loss (such as moisture evaporation, volatile matter release, and coke oxidation). These parameters reflect the ignition characteristics, combustion activity, and burnout capacity of the coal, and are core indicators for evaluating the combustion adaptability of a boiler.

[0028] Furthermore, the extracted TG / DTG curves and their characteristic parameters will be stored in a baseline thermogravimetric curve database, and coupled with parameters such as ash content, volatile matter, fixed carbon, and lower heating value from the coal quality characteristic database to construct a complete data system of coal thermochemical properties. This database not only provides input for coal blending simulation but also provides a benchmark for subsequent similarity analysis and combustion state mapping.

[0029] The technical value of this step lies in providing a high-precision reference model for simulating and predicting the combustion behavior of blended coal through standardized experiments and feature parameter extraction, thereby improving the scientific nature of coal blending schemes and the stability of boiler operation. It plays a crucial supporting role in blended coal combustion dynamics modeling, adaptive model correction, and the generation of combustion optimization suggestions.

[0030] S2, based on the theoretical coal quality parameters of the blended coal and the kinetic characteristics of each individual coal included in the blended coal, uses a weighted algorithm to simulate the mass loss behavior of the blended coal during the heating process, so as to obtain the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the blended coal.

[0031] Further, step S2 includes: S21. A single coal kinetic model was constructed using the Arrhenius kinetic model to describe the coal pyrolysis and combustion process of a single coal and to obtain the single coal kinetic parameters. S22, weights are set based on the proportion of various types of coal in the blend, and the overall reaction rate of the blend is calculated using a weighted superposition model; S23, based on the overall reaction rate of the mixed coal, the conversion rate of the mixed coal with temperature is calculated using the numerical integration method. Based on this conversion rate, the theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal are calculated. The calculation formula is as follows: ; ; in, The conversion rate of the mixed coal as a function of temperature. This refers to the initial mass of the mixed coal.

[0032] Specifically, in some implementations, the specific operation methods include: first, retrieving key parameters such as ash content, volatile matter, fixed carbon, lower heating value, alkali metal content, and sulfur content of each individual coal from the coal quality characteristic database; second, based on the coal type ratio input by the user, using a weighted calculation method, linearly superimposing or nonlinearly correcting the coal quality parameters of each component to generate the theoretical coal quality parameters of the blended coal.

[0033] The pyrolysis and combustion process of single coal is described using the Arrhenius kinetic model, the basic form of which is:

[0034] in, Indicates conversion rate. The reaction rate constant is temperature-dependent. The reaction order. The rate constant. Calculated using the following formula:

[0035] in, For frequency factors, As the apparent activation energy, It is the universal gas constant (8.314 J / mol·K). This refers to the current temperature. In practical applications, the temperature of each individual coal unit... , , The parameters can be obtained from experimental data or literature and stored in a baseline thermogravimetric curve database.

[0036] Furthermore, the overall reaction rate of the mixed coal was calculated using a weighted superposition algorithm, and its expression is as follows:

[0037] in, For the first The blending ratio of different types of coal, , , These are the kinetic parameters for this type of coal. This model considers the nonlinear superposition effect of different coal types during pyrolysis and combustion, thus more accurately reflecting the combustion behavior of blended coal.

[0038] To generate TG and DTG curves, the system uses numerical integration methods (such as Runge-Kutta, Adams-Bashforth, or Euler Method) to solve the above dynamic equations.

[0039] As one implementation method, the initial conditions are set as follows: , rate of temperature rise The temperature range is typically set to 25–1000°C, with a step size of 1–5°C to ensure the continuity and accuracy of the curve. The conversion rate is obtained through integration calculation. Then, generate the TG and DTG curves according to the following formula:

[0040]

[0041] in, The initial coal sample mass is typically set to 100 mg. The DTG curve can be obtained by numerically differencing the TG curve. The differencing method can be central difference or Savitzky-Golay smoothing difference to reduce the influence of noise.

[0042] The technical value of this step lies in its ability to predict the combustion characteristics of blended coal without conducting experiments every time, providing a theoretical baseline for subsequent experimental verification. Simultaneously, by comparing with experimental curves, a self-learning mechanism can be triggered to iteratively optimize the kinetic parameters, thereby improving the model's adaptability and prediction accuracy. This method is widely applicable to the blending optimization of various coal types in thermal power plants, demonstrating significant engineering practicality and promotional value.

[0043] S3. A multi-dimensional similarity analysis is performed between the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the mixed coal and the actual thermogravimetric curve and the differential thermogravimetric curve of the mixed coal. Based on the differences in curve shape, characteristic temperature deviation and weight loss rate, a parameter correction mechanism is initiated. The kinetic parameters in the kinetic characteristics are adjusted iteratively through the optimization algorithm to improve the prediction accuracy.

[0044] Further, step S3 includes: S31. The actual thermogravimetric curve and differential thermogravimetric curve of the mixed coal are obtained through experiments, and its experimental characteristic points are extracted. The experimental characteristic points include ignition temperature, burnout temperature, maximum weight loss rate and stage ratio. S32. Using machine learning, the method judges whether there is a significant deviation between the theoretical and actual thermogravimetric curves of the mixed coal based on the morphological similarity between the theoretical and actual thermogravimetric curves, the deviation of ignition temperature, the deviation of the peak of maximum weight loss rate, the deviation of burnout temperature, and the difference in reaction stages with temperature. It also judges whether there is a significant difference between the theoretical and actual differential thermogravimetric curves based on the structural similarity between the peak shape of the theoretical and actual differential thermogravimetric curves. S321, the morphological similarity between the theoretical and actual thermogravimetric curves of the blended coal is evaluated using the Pearson correlation coefficient. The calculation formula is as follows:

[0045] in, The thermogravimetric curve of the mixed coal in the first... i The values ​​of each temperature sampling point This represents the average response level of the entire mixed coal thermogravimetric curve. The thermogravimetric curve of the reference coal in the 1st... i The values ​​of each temperature sampling point This represents the average response level of the entire reference coal thermogravimetric curve. S322, calculate the deviations in ignition temperature, maximum weight loss rate peak, and burnout temperature between the theoretical and actual thermogravimetric curves. The calculation formula is as follows: ; ; ; in, This is the actual ignition temperature. The theoretical ignition temperature, This represents the temperature corresponding to the actual maximum rate of weight loss. This represents the temperature corresponding to the theoretical maximum rate of weight loss. This is the actual burnout temperature. The theoretical burnout temperature; S323, using conversion rate deviation and weight loss rate deviation, calculates the difference in reaction stages of blended coal with temperature. The formula for calculating the difference in reaction stages of blended coal with temperature is:

[0046] in, for, for; S324, Based on the number, intensity, width and interpeak distance of the curves, analyze the similarity between the theoretical differential thermogravimetric curve of mixed coal and the actual differential thermogravimetric curve of mixed coal. S33, When there is a deviation between the theoretical thermogravimetric curve and the actual thermogravimetric curve of the mixed coal, and the deviation exceeds a preset threshold, or when there is a deviation between the theoretical differential thermogravimetric curve and the actual differential thermogravimetric curve of the mixed coal, and the deviation exceeds a preset threshold, the kinetic parameters are corrected. S331, the loss function formula is defined by correcting for dynamic parameters as follows: ; in, , , These are the corresponding weighting coefficients, used to adjust the degree of influence of different physical constraints during the parameter correction process. This is used to characterize the degree of importance attached to the temperature deviation of the maximum weight loss rate, in order to constrain the consistency between the main combustion reaction zone of the blended coal and the reference coal. Used to characterize the constraint strength on burnout temperature deviation, in order to reflect the degree of matching between the burnout stage and tail reaction characteristics; This is used to adjust the weight of the overall thermogravimetric curve morphology similarity in the calibration process, so as to ensure that the simulated curve maintains a consistent trend with the experimental curve throughout the entire heating range. This represents the peak deviation of the maximum weightlessness rate. To correct the burnout temperature deviation, an optimization algorithm is used to correct the kinetic parameters based on the loss function; S332: Based on the corrected kinetic parameters, obtain the corrected theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal, and compare the similarity between the corrected theoretical thermogravimetric curve and the actual thermogravimetric curve. If there is a deviation between the corrected theoretical thermogravimetric curve and the actual thermogravimetric curve of the mixed coal and the deviation exceeds a preset threshold, or if there is a deviation between the corrected theoretical differential thermogravimetric curve and the actual differential thermogravimetric curve of the mixed coal and the deviation exceeds a preset threshold, repeat step S331.

[0047] Specifically, in the learning and evaluation module (50) of the similarity analysis of the thermogravimetric curves of mixed coal, the multi-index similarity analysis of the simulated thermogravimetric curves and the actual experimental thermogravimetric curves is a key step in realizing the quantitative evaluation of the combustion characteristics of mixed coal and the adaptability of the boiler. This step constructs an adaptive correction mechanism by quantifying the differences in curve shape, characteristic temperature deviation and weight loss rate, thereby dynamically optimizing the weight allocation of kinetic parameters and coal quality parameters, and improving the model prediction accuracy and engineering applicability.

[0048] At the technical implementation level, this step first uses the Pearson correlation coefficient. The overall morphological similarity between simulated and experimental curves is evaluated and defined as follows:

[0049] in, The thermogravimetric curve of the mixed coal in the first... i The values ​​of each temperature sampling point This represents the average response level of the entire mixed coal thermogravimetric curve. The thermogravimetric curve of the reference coal in the 1st... i The values ​​of each temperature sampling point This represents the average response level of the entire reference coal thermogravimetric curve. The closer the value is to 1, the more similar the curve shapes are. By averaging the thermogravimetric response values, the overall offset caused by differences in initial mass, ash content, and residual mass of different coal samples can be eliminated, allowing the similarity analysis to focus on reflecting the dynamic trends and morphological consistency of the pyrolysis and combustion processes.

[0050] In some implementations, the mean and It can also be replaced with the median or weighted average to enhance robustness to local outliers.

[0051] In addition, characteristic temperature deviations, including ignition temperature, need to be calculated. Temperature of maximum weight loss rate Burnout temperature Its expression is:

[0052]

[0053]

[0054] in, This is the actual ignition temperature. The theoretical ignition temperature, This represents the temperature corresponding to the actual maximum rate of weight loss. This represents the temperature corresponding to the theoretical maximum rate of weight loss. This is the actual burnout temperature. Theoretical burnout temperature At the same time, by calculating the conversion rate deviation Assess the differences in reaction stages at different temperature points:

[0055] in, This represents the actual conversion rate. This represents the theoretical conversion rate.

[0056] Furthermore, based on the peak structure similarity of the DTG curves, the system analyzes characteristics such as the number of peaks, intensity, width, enclosed area, and inter-peak distance to determine whether the simulated curves reasonably reflect the combustion stage distribution of the blended coal. When the comprehensive evaluation result of the above indicators deviates from the set threshold, the system will automatically activate the adaptive correction mechanism.

[0057] Regarding parameter settings, the loss function The overall deviation between the simulation and experimental curves is used to quantify the difference, and it is defined as:

[0058] in, , , These are the corresponding weighting coefficients, used to adjust the degree of influence of different physical constraints during the parameter correction process. This is used to characterize the degree of importance attached to the temperature deviation of the maximum weight loss rate, in order to constrain the consistency between the main combustion reaction zone of the blended coal and the reference coal. Used to characterize the constraint strength on burnout temperature deviation, in order to reflect the degree of matching between the burnout stage and tail reaction characteristics; The weighting of the overall thermogravimetric curve morphology similarity in the calibration process is used to ensure that the simulated curve maintains a consistent trend with the experimental curve throughout the entire heating range. Using an optimization algorithm, the kinetic parameters can be corrected based on the loss function. As one implementation method, the optimization algorithm can employ genetic algorithm (GA), particle swarm optimization (PSO), or gradient descent to adjust the frequency factors in the Arrhenius model. Apparent activation energy Reaction order Perform iterative corrections.

[0059] In application scenarios, this step is widely used in coal blending optimization in thermal power plants, boiler combustion stability assessment, and combustion parameter adjustment decisions. Through closed-loop feedback with experimental data, the system can continuously learn and adapt the model to the combustion behavior of blended coal, thereby improving prediction accuracy and engineering adaptability.

[0060] The technical value of this step lies in its ability to effectively solve the prediction bias problem caused by the nonlinear superposition of mixed coal combustion dynamics through multi-dimensional similarity analysis and adaptive correction mechanism, providing a scientific basis for the quantitative evaluation and optimization of boiler combustion performance.

[0061] The nonlinear superposition effect prediction method for mixed coal combustion dynamics based on thermogravimetric curve analysis in this invention can effectively predict the nonlinear superposition effect in the mixed coal combustion process, improve the quantitative evaluation accuracy of mixed coal combustion performance and boiler adaptability, and realize the optimization of coal blending scheme and dynamic matching of combustion state.

[0062] Example 2 Figure 2 This is a structural diagram of a coal combustion performance prediction device based on thermogravimetric curve analysis according to an embodiment of the present invention.

[0063] like Figure 2As shown, a device for predicting the combustion performance of blended coal based on thermogravimetric analysis includes: The reference thermogravimetric curve database establishment module is used to establish a reference thermogravimetric curve database containing the design coal type and the optimal adaptability coal type, and to calculate the theoretical coal quality parameters of the blended coal based on the blended coal quality parameter calculation module. The coal quality parameter calculation module for blended coal uses a weighted algorithm to simulate the mass loss behavior of the blended coal during the heating process, based on the theoretical coal quality parameters of the blended coal and the kinetic characteristics of each individual coal contained in the blended coal, in order to obtain the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the blended coal. The multi-dimensional similarity analysis and parameter correction module is used to perform multi-dimensional similarity analysis between the theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal and the actual thermogravimetric curve and differential thermogravimetric curve of the mixed coal. Based on the differences in curve shape, characteristic temperature deviation and weight loss rate, the parameter correction mechanism is activated. The kinetic parameters in the kinetic characteristics are adjusted iteratively through the optimization algorithm to improve the prediction accuracy.

[0064] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a method for predicting the combustion performance of mixed coal based on thermogravimetric analysis.

[0065] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0066] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for predicting the combustion performance of blended coal based on thermogravimetric analysis, characterized in that, include: S1. Establish a database of reference thermogravimetric curves containing the design coal type and the optimal adaptability coal type, and calculate the theoretical coal quality parameters of the blended coal based on the blended coal quality parameter calculation module. S2, based on the theoretical coal quality parameters of the blended coal and the kinetic characteristics of each individual coal included in the blended coal, a weighted algorithm is used to simulate the mass loss behavior of the blended coal during the heating process, so as to obtain the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the blended coal. S3. A multi-dimensional similarity analysis is performed between the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the mixed coal and the actual thermogravimetric curve and the differential thermogravimetric curve of the mixed coal. Based on the differences in curve shape, characteristic temperature deviation and weight loss rate, a parameter correction mechanism is initiated. The kinetic parameters in the kinetic characteristics are adjusted iteratively through the optimization algorithm to improve the prediction accuracy.

2. The method as described in claim 1, characterized in that, S2 further includes: S21. A single coal kinetic model was constructed using the Arrhenius kinetic model to describe the coal pyrolysis and combustion process of a single coal and to obtain the single coal kinetic parameters. S22, weights are set based on the proportion of various types of coal in the blend, and the overall reaction rate of the blend is calculated using a weighted superposition model; S23, based on the overall reaction rate of the mixed coal, the conversion rate of the mixed coal with temperature is calculated using the numerical integration method. Based on this conversion rate, the theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal are calculated. The calculation formula is as follows: ； ； in, The conversion rate of the mixed coal as a function of temperature. This refers to the initial mass of the mixed coal.

3. The method as described in claim 1, characterized in that, The formula for calculating the overall reaction rate of the mixed coal in step S22 is as follows: ； in, For the first i The blending ratio of different types of coal, For the first i Frequency factor of coal cultivation For the first i The apparent activation energy of coal For the first i The reaction order of the coal. R This is the universal gas constant. T This is the current temperature.

4. The method as described in claim 1, characterized in that, S3 further includes: S31. The actual thermogravimetric curve and differential thermogravimetric curve of the mixed coal are obtained through experiments, and its experimental characteristic points are extracted. The experimental characteristic points include ignition temperature, burnout temperature, maximum weight loss rate and stage ratio. S32. Using machine learning, the method judges whether there is a significant deviation between the theoretical and actual thermogravimetric curves of the mixed coal based on the morphological similarity between the theoretical and actual thermogravimetric curves, the deviation of ignition temperature, the deviation of the peak of maximum weight loss rate, the deviation of burnout temperature, and the difference in reaction stages with temperature. It also judges whether there is a significant difference between the theoretical and actual differential thermogravimetric curves based on the structural similarity between the peak shape of the theoretical and actual differential thermogravimetric curves. S33. When there is a deviation between the theoretical thermogravimetric curve and the actual thermogravimetric curve of the mixed coal, and the deviation exceeds a preset threshold, or when there is a deviation between the theoretical differential thermogravimetric curve and the actual differential thermogravimetric curve of the mixed coal, and the deviation exceeds a preset threshold, the kinetic parameters are corrected.

5. The method as described in claim 1, characterized in that, S32 further includes: S321, the morphological similarity between the theoretical and actual thermogravimetric curves of the blended coal is evaluated using the Pearson correlation coefficient. The calculation formula is as follows: in, The thermogravimetric curve of the mixed coal in the first... i The values ​​of each temperature sampling point This represents the average response level of the entire mixed coal thermogravimetric curve. The thermogravimetric curve of the reference coal in the 1st... i The values ​​of each temperature sampling point This represents the average response level of the entire reference coal thermogravimetric curve. S322, calculate the deviations in ignition temperature, maximum weight loss rate peak, and burnout temperature between the theoretical and actual thermogravimetric curves. The calculation formula is as follows: ； ； ； in, This is the actual ignition temperature. The theoretical ignition temperature, This represents the temperature corresponding to the actual maximum rate of weight loss. This represents the temperature corresponding to the theoretical maximum rate of weight loss. This is the actual burnout temperature. The theoretical burnout temperature; S323, using conversion rate deviation and weight loss rate deviation, calculates the difference in reaction stages of blended coal with temperature. The formula for calculating the difference in reaction stages of blended coal with temperature is: in, This represents the actual conversion rate. Theoretical conversion rate; S324, based on the number, intensity, width and interpeak distance of the curves, analyzes the similarity between the theoretical differential thermogravimetric curve of mixed coal and the actual differential thermogravimetric curve of mixed coal.

6. The method as described in claim 5, characterized in that, S33 further includes: S331, the loss function formula is defined by correcting for dynamic parameters as follows: ； in, , , These are the corresponding weighting coefficients, used to adjust the degree of influence of different physical constraints during the parameter correction process. This is used to characterize the degree of importance attached to the temperature deviation of the maximum weight loss rate, in order to constrain the consistency between the main combustion reaction zone of the blended coal and the reference coal. Used to characterize the constraint strength on burnout temperature deviation, in order to reflect the degree of matching between the burnout stage and tail reaction characteristics; This is used to adjust the weight of the overall thermogravimetric curve morphology similarity in the calibration process, so as to ensure that the simulated curve maintains a consistent trend with the experimental curve throughout the entire heating range. This represents the peak deviation of the maximum weightlessness rate. To address the burnout temperature deviation, an optimization algorithm is used to correct the kinetic parameters based on a loss function. S332: Based on the corrected kinetic parameters, obtain the corrected theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal, and compare the similarity between the corrected theoretical thermogravimetric curve and the actual thermogravimetric curve. If there is a deviation between the corrected theoretical thermogravimetric curve and the actual thermogravimetric curve of the mixed coal and the deviation exceeds a preset threshold, or if there is a deviation between the corrected theoretical differential thermogravimetric curve and the actual differential thermogravimetric curve of the mixed coal and the deviation exceeds a preset threshold, repeat step S331.

7. The method as described in claim 1, characterized in that, The reference thermogravimetric curve database mentioned in step S1 includes actual thermogravimetric curves, actual differential thermogravimetric curves, and characteristic parameters of water precipitation peaks and combustion peaks in the actual differential thermogravimetric curves obtained under standardized thermogravimetric experimental conditions. The characteristic parameters of water precipitation peaks and combustion peaks in the actual differential thermogravimetric curves include ignition temperature, temperature corresponding to the maximum weight loss rate, burnout temperature, characteristic rate, and weight loss stage ratio, which serve as the core characteristic data of the database.

8. The method as described in claim 1, characterized in that, The theoretical coal quality parameters mentioned in step S1 include the ash content, volatile matter, fixed carbon, lower heating value and metal oxide content of each blended coal type. The input parameters for the combustion kinetics of the blended coal are generated according to the blending ratio of each blended coal type.

9. A device for predicting the combustion performance of blended coal based on thermogravimetric analysis, characterized in that, include: The reference thermogravimetric curve database establishment module is used to establish a reference thermogravimetric curve database containing the design coal type and the optimal adaptability coal type, and to calculate the theoretical coal quality parameters of the blended coal based on the blended coal quality parameter calculation module. The coal quality parameter calculation module for blended coal uses a weighted algorithm to simulate the mass loss behavior of the blended coal during the heating process, based on the theoretical coal quality parameters of the blended coal and the kinetic characteristics of each individual coal contained in the blended coal, in order to obtain the theoretical thermogravimetric curve and the theoretical differential thermogravimetric curve of the blended coal. The multi-dimensional similarity analysis and parameter correction module is used to perform multi-dimensional similarity analysis between the theoretical thermogravimetric curve and theoretical differential thermogravimetric curve of the mixed coal and the actual thermogravimetric curve and differential thermogravimetric curve of the mixed coal. Based on the differences in curve shape, characteristic temperature deviation and weight loss rate, the parameter correction mechanism is activated. The kinetic parameters in the kinetic characteristics are adjusted iteratively through the optimization algorithm to improve the prediction accuracy.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as claimed in any one of claims 1-8.

Images