Coal spontaneous combustion multi-step kinetics parameter calculation method based on directional adjustment and random optimization

Abstract

The application discloses a coal spontaneous combustion multi-step kinetics parameter calculation method based on directional adjustment and random optimization, and obtains an actual TG curve; a multi-step reaction process to be solved is drafted, and the TG curve is solved and calculated through kinetics parameters and the multi-step reaction process; the solid mass of each temperature in the multi-step reaction is calculated, error values of the TG curve and each temperature point thereof are obtained, a temperature point corresponding to a maximum error value is obtained, the adjustment direction of the kinetics parameters is determined according to the relative size of the kinetic characteristics and the actual mass and the calculated mass, and the kinetics parameters of each step are directionally adjusted; the random range is determined according to the adjusted kinetics parameters and a relaxation factor, and the kinetics parameter combination is generated in a random optimization mode according to the relaxation number; and the kinetics parameters of the multi-step reaction are finally obtained through continuous iteration. The method can effectively reduce the iteration number in the process of solving the kinetics parameters of the coal spontaneous combustion multi-step reaction, thereby shortening the solving time and effectively improving the calculation efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to a coal spontaneous combustion multi-step kinetics parameter calculation method based on directional adjustment and random optimization, belonging to the technical field of coal spontaneous combustion research. BACKGROUND

[0002] Coal is a complex organic mixture, and its combustion process cannot be described by a single-step chemical reaction. In recent years, the multi-step reaction process of coal combustion has been proposed, and numerous multi-step reaction kinetics processes for coal have been extensively studied. For example, the paper with DOI 10.1016 / j.combustflame.2013.12.013 "Smouldering combustion of peat in wildfires:Inverse modelling of the drying and the thermal and oxidative decomposition kinetics" proposed an inverse algorithm based on genetic algorithm to obtain the corresponding multi-step reaction kinetics parameters. The paper with DOI 10.1016 / j.energy.2021.121299 "Heat effects and kinetics of coal spontaneous combustion at various oxygen contents" uses Gaussian peak fitting to obtain the combustion heat of different reactions in the coal combustion heat release process. The patent with application publication number CN110728039A "Carbonized combustible single-peak pyrolysis kinetics parameter calculation method based on genetic algorithm" uses genetic algorithm to solve the single-peak pyrolysis kinetics parameters of carbonized combustible.

[0003] The above methods mainly optimize the coal combustion kinetics parameters through genetic algorithm. The input parameters of the genetic algorithm are set as the kinetics parameters, the TG curve (i.e. TG curve) is calculated through software programming, the error is set as the fitness function, and finally a set of kinetics parameters with the minimum error is obtained, which is the solved kinetics parameters. However, when using group optimization algorithm, each iteration needs to be calculated for the corresponding number of population, or each possible kinetics parameter needs to be calculated, which results in a long time for each iteration. Since the direction of optimization cannot be determined, more iterations will be further increased, which finally makes the current method of calculating kinetics parameters too time-consuming and inefficient. SUMMARY

[0004] In view of the problems in the prior art, the present application provides a coal spontaneous combustion multi-step kinetic parameter calculation method based on directional adjustment and random optimization, which can effectively reduce the iteration number in the process of solving the coal spontaneous combustion multi-step reaction kinetic parameters, thereby shortening the solving time and effectively improving the calculation efficiency by means of the combination of directional adjustment and random optimization.

[0005] In order to achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows: a coal spontaneous combustion multi-step kinetic parameter calculation method based on directional adjustment and random optimization, and the specific steps are as follows:

[0006] S1, obtaining the actual TG curve of the coal sample (i.e., the curve of the actual mass of the coal sample at different temperatures) according to the thermogravimetric experiment;

[0007] S2, preparing the multi-step reaction process to be solved, and calculating the TG curve (i.e., the curve of the calculated mass of the coal sample at different temperatures) by means of the kinetic parameters and the multi-step reaction process;

[0008] S3, obtaining the calculated mass of each solid at each temperature of the multi-step reaction, and then obtaining the error value of the calculated TG curve and the corresponding actual mass and calculated mass at each temperature point, and obtaining the temperature point corresponding to the maximum error value;

[0009] S4, determining the kinetic characteristics, and determining the adjustment direction of the kinetic parameters according to the relative size of the actual mass and the calculated mass, determining the adjustment weight of the content of each solid according to the size of the calculated mass of each solid, and then adjusting the kinetic parameters of each step in combination with the size of the adjustment coefficient;

[0010] S5, generating the kinetic parameter combination in the manner of random optimization according to the adjusted kinetic parameters and the relaxation factor, and determining the random range according to the relaxation number;

[0011] S6, calculating the error value of the kinetic parameter combination generated in step S5, and if the error value is lower than the set error threshold or the maximum iteration number is reached, the current kinetic parameters are obtained as the optimal kinetic parameters of the multi-step reaction; otherwise, the directional adjustment and random optimization of steps S4 and S5 are continuously repeated until the error value is lower than the set error threshold or the maximum iteration number is reached.

[0012] Further, the process of determining the kinetic characteristics in step S4 is as follows:

[0013] The kinetic characteristics in the directional adjustment refer to the relationship between the residual mass of the reactant at a certain temperature and the kinetic parameters, which is obtained by the partial derivative of the mass with respect to the kinetic parameters, and the obtaining method is as follows:

[0014] Each step of the multi-step reaction of coal spontaneous combustion is equivalent to the following formula, and the reaction rate follows the Arrhenius chemical reaction rate equation:

[0015] B1+C1→D1+F1 (1)

[0016] According to the Arrhenius chemical reaction rate equation, the reaction rate constant of a single-step reaction is:

[0017]

[0018] Wherein: k is the reaction rate constant; B and C are the concentrations of reactants B1 and C1, respectively; m and n are the reaction orders of B1 and C1, respectively; A is the pre-exponential factor; E is the reaction activation energy; R is the gas constant; T is the reaction temperature;

[0019] The reaction rate of B1 reactant is:

[0020]

[0021] When a fixed heating rate is used, T = βt + T0, then dt = dT / β, and the relationship between B1 reactant and temperature is expressed as:

[0022]

[0023] Let u = E / RT, then dT = -Edu / Ru 2 , then the above formula is transformed as:

[0024]

[0025] Integrating both sides gives:

[0026]

[0027] The Starink method is used to determine the temperature integral function by the following approximation:

[0028]

[0029] Wherein, b is a constant, s is a constant, and D is a constant.

[0030] Set the reaction order m to 1, then the result of the above integral is as follows:

[0031]

[0032] Since B is greater than zero, we have:

[0033]

[0034] From the above formula, we have:

[0035] B=f(A)=exp(-Ak1)

[0036] Where k1 is a constant greater than zero, the larger A is, the smaller B is;

[0037] For E.

[0038]

[0039] Among them, k2 is a constant greater than zero, s is about 1.8, then the smaller E is, the greater E s-1 The smaller, The larger the The smaller is , the smaller is B;

[0040] The kinetic characteristics are thus obtained: the residual mass of the reactants decreases with decreasing E and increasing A.

[0041] Furthermore, in step S4, the initial value of the first step reaction in the multi-step reaction of coal spontaneous combustion is calculated, specifically:

[0042] According to the expression of B, calculate the initial slope expression, set k = -Ae D C n E 1-s / (βR 1-s ),but:

[0043]

[0044] in The initial value of the TG curve is obtained, and the remaining parameters are obtained through experimental conditions and commonly used values. The relative relationship between A and E is obtained and the initial value is determined.

[0045] Furthermore, when the kinetic parameters of each step are adjusted in a directional manner in step S4, rapid adjustment is required in the early stage of the directional adjustment to find the range of the optimal solution. The adjustment amplitude is reduced in the later stage to obtain the accurate solution of the kinetic parameters and set the attenuation term. As the number of iterations increases, the adjustment amplitude of the kinetic parameters continues to decrease. Therefore, the adjustment of the kinetic parameters is determined according to the adjustment direction, adjustment coefficient, weight and attenuation term, as shown in the following formula:

[0046]

[0047] Among them, E t+1 is the adjusted activation energy at time t+1, E t is the activation energy at time t, (-1) ξFor the adjustment direction determined according to the maximum error, wherein the value of ξ is 1 or 2, the adjustment direction is selected according to the need, α is the adjustment coefficient, N is the maximum iteration number, h is the iteration step number at this time, θ is the convergence lower limit of the attenuation term, and the adjustment mode of the remaining kinetic parameters is performed according to the above adjustment mode of the activation energy.

[0048] Further, in the directional adjustment process of the step S4, the temperature points corresponding to the multiple larger error values are selected for adjustment according to the actual needs by selecting the temperature points corresponding to the maximum error values or by sorting the error values from large to small; wherein the specific process of adjusting the temperature points corresponding to the multiple larger error values is as follows: according to the influence of low-temperature adjustment on the high-temperature reaction, when the maximum error value is selected, the maximum error is further adjusted by the contraction coefficient, and after the maximum error value is selected, the error of the temperature point corresponding to the maximum error value is set to zero, then the temperature point corresponding to the next maximum error value is selected, and so on, the temperature points corresponding to the multiple larger error values are sequentially selected, and the kinetic parameters are directionally adjusted.

[0049] Further, the step S5 is specifically as follows:

[0050] The random range of the kinetic parameters is determined according to the kinetic parameters and the relaxation factor, which is [kinetic parameter*(1-relaxation factor), kinetic parameter*(1+relaxation factor)]; then a random function generating function is used to generate [relaxation number] kinetic parameters in the above random range, each kinetic parameter is combined in order to generate [relaxation number] kinetic parameter combinations, and the kinetic parameter combination after directional adjustment is added to form [relaxation number+1] kinetic parameter combinations, which includes the kinetic parameter combination after directional adjustment and the relaxation number of random kinetic parameter combinations in the range of the relaxation factor.

[0051] Compared with the prior art, the application combines Arrhenius chemical reaction rate equation and Starink method, uses differential transformation to obtain the relationship between the initial coal sample reaction content and the kinetic parameters, so as to determine the direction and initial value setting of active optimization, and realizes active optimization and random optimization of the kinetic parameters through software, including the following steps: obtaining the TG curve of the coal sample according to the thermogravimetric experiment; formulating the multi-step reaction process to be solved, and solving the TG curve through the kinetic parameters and the multi-step reaction process; obtaining the mass of each solid at each temperature of the multi-step reaction, obtaining the error value of the TG curve and each temperature point thereof, obtaining the temperature point corresponding to the maximum error value, determining the adjustment direction of the kinetic parameters according to the relative size of the kinetic characteristics and the actual mass and the calculated mass, determining the adjustment weight of each solid content according to the size of the amount of substance of each solid, and then combining the size of the adjustment coefficient to adjust the kinetic parameters of each step in a directional manner; determining the random range according to the adjusted kinetic parameters and the relaxation factor, generating the kinetic parameter combination in a random optimization manner according to the relaxation number; calculating the error value of the result, and continuing the directional adjustment and random optimization, and finally obtaining the kinetic parameters of the multi-step reaction. The method can effectively reduce the iteration number in the solving process of the multi-step reaction kinetic parameters of the coal spontaneous combustion, and further shorten the solving time, and effectively improve the calculation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 is a flowchart of the application;

[0053] Figure 2 is a curve graph of the change of the calculated weight and the actual weight at different temperatures in the test verification of the application;

[0054] Figure 3 is a mass percentage composition graph of each component in the test verification.

[0055] Figure 4 is a maximum error change graph of the calculation process of the application in the test verification;

[0056] Figure 5 is an average percentage error change graph of the calculation process of the application in the test verification. DETAILED DESCRIPTION

[0057] The application will be further described below.

[0058] As shown in the following specific steps of the application: Figure 1

[0059] S1, obtaining the actual TG curve of the coal sample (i.e. the curve formed by the actual mass of the coal sample at different temperatures) according to the thermogravimetric experiment;

[0060] ​S2, a multi-step reaction process to be solved is formulated, and a TG curve (i.e. a curve of the calculated mass of the coal sample formed at different temperatures) is calculated through the kinetic parameters and the multi-step reaction process;

[0061] S3, the calculated mass of each solid at each temperature of the multi-step reaction is obtained through calculation, and then the calculated TG curve and the error value of the actual mass and the calculated mass at each temperature point are obtained, and the temperature point corresponding to the maximum error value is obtained;

[0062] S4, the kinetic characteristics are determined first, and the adjustment direction of the kinetic parameters is determined according to the relative size of the actual mass and the calculated mass, the adjustment weight of the content of each solid is determined according to the size of the calculated mass of each solid, and then the kinetic parameters of each step are adjusted in a directional manner according to the size of the adjustment coefficient;

[0063] The process of determining the kinetic characteristics is as follows:

[0064] The kinetic characteristics in the directional adjustment refer to the relationship between the residual mass of the reactant at a certain temperature and the kinetic parameters, which is obtained through the partial derivative of the mass with respect to the kinetic parameters, and the obtaining method is as follows:

[0065] Each step of the multi-step reaction of coal spontaneous combustion is equivalent to the following formula, and the reaction rate follows the Arrhenius chemical reaction rate equation:

[0066] B1+C1→D1+F1 (1)

[0067] According to the Arrhenius chemical reaction rate equation, the reaction rate constant of a single-step reaction is:

[0068]

[0069] Wherein: k is the reaction rate constant; B and C are the concentrations of reactants B1 and C1 respectively; m and n are the reaction orders of B1 and C1 respectively; A is the pre-exponential factor; E is the reaction activation energy; R is the gas constant; T is the reaction temperature;

[0070] The reaction rate of B1 reactant is:

[0071]

[0072] When a fixed heating rate is used, T = βt + T0, then dt = dT / β, and the relationship between B1 reactant and temperature is represented as:

[0073]

[0074] Let u = E / RT, then dT = -Edu / Ru 2 , the above formula is transformed into:

[0075]

[0076] Integrating both sides gives:

[0077]

[0078] The Starink method is used to determine the temperature integral function by approximating it in the following way:

[0079]

[0080] where b is a constant, s is a constant, and D is a constant;

[0081] Let the reaction order m be 1, then the result of the above integral is as follows:

[0082]

[0083] Since B is greater than zero, then:

[0084]

[0085] From the above equation, for A:

[0086] B = f(A) = exp(-Ak1)

[0087] where k1 is a constant greater than zero, then the larger A is, the smaller B is;

[0088] For E:

[0089]

[0090] where k2 is a constant greater than zero, and s is approximately 1.8, then the smaller E is, the larger E s-1 is, the smaller B is;

[0091] Thus the kinetic characteristic is obtained: the remaining mass of the reactant decreases as E decreases and A increases.

[0092] Next, the initial value of the first step reaction in the multi-step reaction of coal spontaneous combustion is calculated, which is:

[0093] According to the expression of B, the initial slope expression is calculated, let k = -Ae D C n E 1-s / (βR 1-s ), then:

[0094]

[0095] ​in The initial value of the TG curve is obtained, and the remaining parameters are obtained through experimental conditions and commonly used values. The relative relationship between A and E is obtained and the initial value is determined.

[0096] When directional adjustment of the kinetic parameters of each step is performed, the directional adjustment needs to be adjusted quickly in the early stage to find the range of the optimal solution. The adjustment amplitude is reduced in the later stage to obtain the accurate solution of the kinetic parameters and set the attenuation term. As the number of iterations increases, the adjustment amplitude of the kinetic parameters continues to decrease. Therefore, the adjustment of the kinetic parameters is determined according to the adjustment direction, adjustment coefficient, weight and attenuation term, as shown in the following formula:

[0097]

[0098] Among them, E t+1 is the adjusted activation energy at time t+1, E t is the activation energy at time t, (-1) ξ is the adjustment direction determined by the maximum error, where the value of ξ is 1 or 2, which is selected according to the direction to be adjusted. α is the adjustment coefficient, N is the maximum number of iterations, h is the number of iteration steps at this time, and θ is the convergence lower limit of the attenuation term. The adjustment methods of the remaining kinetic parameters are all carried out in accordance with the above-mentioned activation energy adjustment method.

[0099] In addition, during the directional adjustment process, the temperature point corresponding to the maximum error value is selected according to actual needs, or the error values ​​are sorted from large to small, and multiple temperature points corresponding to larger error values ​​are selected for adjustment; the specific process of adjusting the temperature points corresponding to multiple larger error values ​​is: since the low-temperature adjustment will affect the high-temperature reaction, when the maximum error value is selected, the maximum error is further adjusted by the shrinkage coefficient, and after the maximum error value is selected, the error of the temperature point corresponding to the maximum error value is reset to zero, and then the temperature point corresponding to the next maximum error value is selected, and so on, multiple temperature points corresponding to larger error values ​​are selected in turn to perform directional adjustment of the dynamic parameters.

[0100] S5. Based on the adjusted kinetic parameters and setting the relaxation factor, a random range is determined, and a kinetic parameter combination is generated by random optimization based on the relaxation number, specifically:

[0101] According to the kinetic parameters and the relaxation factor, the random range of the kinetic parameters is determined as [kinetic parameter*(1-relaxation factor), kinetic parameter*(1+relaxation factor)]; then a random function generating function is used to generate [relaxation number] kinetic parameters in the above random range, each kinetic parameter is combined in order to generate [relaxation number] kinetic parameter combinations, and the kinetic parameter combination after directional adjustment is added to form [relaxation number+1] kinetic parameter combinations, which includes the kinetic parameter combination after directional adjustment and the relaxation number of random kinetic parameter combinations within the relaxation factor range.

[0102] S6, calculate the error value of the kinetic parameter combination generated in step S5, if it is lower than the set error threshold or reaches the maximum iteration number, the current kinetic parameter is obtained as the optimal kinetic parameter of the multi-step reaction; otherwise, continue to repeat steps S4 and S5 for directional adjustment and random optimization until the error threshold is lower than the set error threshold or the maximum iteration number is reached.

[0103] Test proves:

[0104] ① Obtain the thermogravimetric curve of the coal sample by the simultaneous thermal analyzer, and calculate the mass percentage at each temperature by the maximum mass;

[0105] ② Use python / matlab and other software to write a program [calculation function program] to realize the calculation of the actual mass and the calculated mass under different temperature points through the multi-step reaction process, kinetic parameters and different temperature points as shown in Figure 2 , and the mass between different components as shown in Figure 3 , and determine the maximum error point by the maximum error between the total mass and the actual curve

[0106] ③ According to the relative size between the calculated mass and the actual mass at the maximum error point, determine the direction of kinetic parameter adjustment, for example: if the calculated mass is less than the actual mass, adjust in the direction of increasing activation energy and decreasing pre-exponential factor, when adjusting the activation energy, ξ is 2, and when adjusting the pre-exponential factor, ξ is 1; calculate the weight of each component according to the mass of each component in Figure 3 ; calculate the decay factor according to the number of runs; combine the decay factor, adjustment weight, and adjustment direction to calculate the adjusted kinetic parameters according to formula (12).

[0107] ④According to the adjusted kinetic parameter value and the relaxation factor, the range of the randomly adjusted kinetic parameter is calculated: [kinetic parameter*(1-relaxation factor), kinetic parameter*(1+relaxation factor)], a random function generating function is used to generate [relaxation number] kinetic parameters in the above range, each kinetic parameter is combined in order to generate [relaxation number] kinetic parameter combinations, and the kinetic parameter combination after directional adjustment is added to form [relaxation number+1] kinetic parameter combinations, and the optimal kinetic parameter combination under each combination is obtained through [calculation function program].

[0108] ⑤According to the calculated optimal kinetic parameter combination, steps ③ and ④ are cycled to realize the cyclic directional adjustment and random adjustment of the kinetic parameters, and the average percentage error and the maximum error point at each iteration are recorded, and the Figure 4 and 5 It can be seen from the figure that when the method of the application is used, the kinetic parameter combination with an average error of 3.58% can be found in the fifth iteration, and only about 10 [calculation program functions] need to be calculated each time, which consumes less time. This is because the directional adjustment method can quickly adjust the kinetic parameters to a smaller result; according to Figure 4 and Figure 5 It can be seen that the final convergence result error is small, because the random optimization method is used, which avoids the situation that the directional adjustment jumps between two solutions and cannot continue to converge.

[0109] The above is only the preferred embodiment of the application, it should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A method for calculating multi-step kinetic parameters of coal spontaneous combustion based on directional adjustment and random optimization, characterized in that: The specific steps are: S1. Obtain the actual TG curve of the coal sample according to the thermogravimetric experiment; S2. Formulate a multi-step reaction process to be solved, and solve and calculate the TG curve through kinetic parameters and multi-step reaction process; S3. Obtaining the calculated mass of each solid at each temperature of the multi-step reaction by calculation, and then obtaining the calculated TG curve and the error value between the actual mass and the calculated mass at each temperature point, and obtaining the temperature point corresponding to the maximum error value; S4. First determine the kinetic characteristics. Specifically, the kinetic characteristics in the directional adjustment refer to the relationship between the residual mass of the reactants and the kinetic parameters at a certain temperature. The relationship is obtained by partial differential of the mass with respect to the kinetic parameters. The method for obtaining the relationship is as follows: Each step of the multi-step reaction of coal spontaneous combustion is equivalent to the following formula, and the reaction rate follows the Arrhenius chemical reaction rate equation: B1+C1→D1+F1 (1) According to the Arrhenius chemical reaction rate equation, the reaction rate constant of a single-step reaction is: Where: k is the reaction rate constant; B and C are the concentrations of reactants B1 and C1 respectively; m and n are the reaction orders of reactants B1 and C1 respectively; A is the pre-exponential factor; E is the reaction activation energy; R is the gas state constant; T is the reaction temperature; The reaction rate of reactant B1 is: When a fixed heating rate is used, T = βt + T0, then dt = dT / β, and the relationship between B1 reactant and temperature is expressed as: Let u=E / RT, then dT=-Edu / Ru 2 , then the above formula is transformed into: Integrating both sides yields: The temperature integral function determined by the Starink method is approximated in the following way: Among them, b is a constant, s is a constant, and D is a constant; If the reaction order m is set to 1, the result of the above integration is as follows: Since B is greater than zero, then: According to the above formula, for A: B=f(A)=exp(-Ak1) Where k1 is a constant greater than zero, the larger A is, the smaller B is; For E: Among them, k2 is a constant greater than zero, the smaller E is, the higher E s-1 The smaller, The larger the The smaller is , the smaller is B; The kinetic characteristics are thus obtained: the residual mass of the reactants decreases with the decrease of E and the increase of A; Then calculate the initial value of the first step reaction in the multi-step reaction of coal spontaneous combustion, specifically: According to the expression of B, calculate the initial slope expression, set k = -Ae D C n E 1-s / (βR 1-s ),but: in The initial value of the TG curve is obtained, and the remaining parameters are obtained through experimental conditions and common values. The relative relationship between A and E is obtained and the initial value is determined; Then, the adjustment direction of the kinetic parameters is determined based on the relative size of the actual mass and the calculated mass. The adjustment weight of each solid content is determined according to the calculated mass of each solid. Then, combined with the size of the adjustment coefficient, the kinetic parameters of each step are adjusted directionally. Specifically, the directional adjustment needs to be adjusted quickly in the early stage to find the range of the optimal solution. In the later stage, the adjustment amplitude is reduced to obtain the accurate solution of the kinetic parameters and set the attenuation term. As the number of iterations increases, the adjustment amplitude of the kinetic parameters continues to decrease. Therefore, the adjustment of the kinetic parameters is determined according to the adjustment direction, adjustment coefficient, weight and attenuation term, as shown in the following formula: Among them, E t+1 is the adjusted activation energy at time t+1, E t is the activation energy at time t, (-1) ξ is the adjustment direction determined by the maximum error, α is the adjustment coefficient, N is the maximum number of iterations, h is the number of iteration steps at this time, θ is the convergence lower limit of the attenuation term, and the adjustment methods of the remaining kinetic parameters are all carried out in the same way as the above activation energy adjustment; S5. According to the adjusted kinetic parameters and setting the relaxation factor, a random range is determined, and a kinetic parameter combination is generated by random optimization according to the relaxation number, specifically: according to the kinetic parameters and the relaxation factor, the random range of the kinetic parameters is determined to be [kinetic parameter*(1-relaxation factor), kinetic parameter*(1+relaxation factor)]; then a random function is used to generate a function to generate [relaxation number] kinetic parameters within the above random range, each kinetic parameter is combined in order to generate a [relaxation number] kinetic parameter combination, and the kinetic parameter combination after the direction adjustment is added to form a [relaxation number+1] kinetic parameter combination, which includes the kinetic parameter combination after the direction adjustment and a [relaxation number] random kinetic parameter combination within the relaxation factor range; S6. Calculate the error value of the kinetic parameter combination generated in step S5. If the error value is lower than the set error threshold or reaches the maximum number of iterations, the current kinetic parameters are obtained as the optimal kinetic parameters for the multi-step reaction; otherwise, continue to repeat steps S4 and S5 for directional adjustment and random optimization until the error value is lower than the set error threshold or the maximum number of iterations is reached.

2. The method for calculating multi-step kinetic parameters of coal spontaneous combustion based on directional adjustment and random optimization according to claim 1 is characterized in that: During the directional adjustment process of step S4, the temperature point corresponding to the maximum error value is selected according to actual needs, or the error values ​​are sorted from large to small, and multiple temperature points corresponding to larger error values ​​are selected for adjustment; the specific process of adjusting the temperature points corresponding to multiple larger error values ​​is: according to the low-temperature adjustment, the high-temperature reaction will be affected. When the maximum error value is selected, the maximum error is further adjusted by the contraction coefficient, and after the maximum error value is selected, the error of the temperature point corresponding to the maximum error value is reset to zero, and then the temperature point corresponding to the next maximum error value is selected, and so on, and multiple temperature points corresponding to larger error values ​​are selected in turn to perform directional adjustment of the dynamic parameters.

Images