Method for simultaneously reconstructing temperature and radiation characteristic parameters of particulate matters in complex combustion system

Abstract

The invention provides a method for simultaneously reconstructing temperature and radiation characteristic parameters of particulate matters in a complex combustion system. The method comprises the following steps: measuring combustion flames of the particulate matters to be detected by using a spectrograph to obtain detection spectral radiation intensity; based on a wavelength-temperature double-variable emissivity model, calculating flame theoretical spectral radiation intensity; initial parameters are set, and a residual function about unknown parameters is constructed; transforming the residual function into a quadratic model by using Taylor expansion; solving and updating the iteration step length by combining a Cholesky decomposition method and a Cauchy point method; updating unknown parameters by utilizing an iteration step length, calculating a residual sum of squares and performing convergence judgment; and when the convergence condition is met, unknown parameters required by the radiation characteristic parameter model and the real temperature of the particulate matter are updated. According to the method, high-precision particle temperature and radiation characteristic parameters are given based on flame spectrum radiation intensity, and the method has the advantages of high precision and rapid convergence in a gas-liquid-solid multi-phase coupled complex combustion environment.

Description

Technical Field

[0001] The present invention relates to the technical field of measuring particle temperature and radiation characteristic parameters in complex combustion systems, and in particular to a method for simultaneously reconstructing the temperature and radiation characteristic parameters of solid rocket engine plume particles. Background Art

[0002] Solid rocket engines have been widely used in military and civilian fields due to their advantages such as fast response, short preparation time, high thrust and low cost. The plume of solid rocket engines has the characteristics of high temperature, high speed and strong radiation. The plume temperature is an important parameter to characterize the working state and performance of the engine. Accurately measuring the plume temperature of solid rocket engines has important reference value for understanding the internal combustion of the engine and the overall performance of the engine. Solid rocket plumes are different from conventional liquid rocket plumes. They not only have the flow characteristics of high temperature, high speed and high pressure, but also are accompanied by phase-changing solid particles Al2O3 in their combustion gas. The gas-liquid-solid multiphase coupling leads to low measurement accuracy and poor global convergence of the traditional radiation temperature measurement method based on the "wavelength-emissivity" radiation characteristic parameter model. Therefore, it is of great significance to develop a method for simultaneously reconstructing the temperature and radiation characteristic parameters of particles in complex combustion systems based on the "wavelength-temperature-emissivity" radiation characteristic parameter model. Summary of the invention

[0003] The purpose of the present invention is to provide a method for solving the temperature and radiation characteristic parameters of particles in complex combustion systems, especially those related to solid rocket engine plume, based on flame spectral radiation intensity, while reconstructing the temperature and radiation characteristic parameters of particles in complex combustion systems.

[0004] The object of the present invention can be achieved by the following technical measures: a method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system, the method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system comprising:

[0005] Step 1, using a spectrometer to measure the combustion flame of the particles to be tested to obtain the detection spectrum radiation intensity;

[0006] Step 2, based on the wavelength-temperature dual variable emissivity model, calculate the flame theoretical spectral radiation intensity;

[0007] Step 3, setting initial parameters, and constructing a residual function about unknown parameters based on the flame detection spectrum radiation intensity and the theoretical spectrum radiation intensity;

[0008] Step 4, apply Taylor expansion to transform the residual function into a quadratic model;

[0009] Step 5, combining the Cholesky decomposition method and the Cauchy point method to solve the update iteration step size;

[0010] Step 6: Update the unknown parameters using the iteration step size, calculate the residual sum of squares and make a convergence judgment;

[0011] Step 7: When the convergence condition is met, the unknown parameters required for the radiation characteristic parameter model and the actual temperature of the particles are updated.

[0012] The purpose of the present invention can also be achieved by the following technical measures:

[0013] In step 1, based on the wavelength-temperature bivariate emissivity model, the flame theoretical spectral radiation intensity is expressed as the following formula 1:

[0014]

[0015] Where I is the theoretical spectral radiation intensity of the flame, in W / m 3 / sr; a0 and a1 are the unknown coefficients of the radiation characteristic parameter model; T is the particle temperature, in K; λ is the waveband used in the calculation, in m; c1 and c2 are Planck constants.

[0016] In step 3, the initial parameters a0, a1, T and the initial trust region radius Δ0 are set, and the residual function is constructed as shown in the following formula 2:

[0017]

[0018] In the formula, is the residual, and parameter a=[a0, a1, T] is the required parameter.

[0019] In step 4, the residual function f(a) is expanded using Taylor near different bands, and the quadratic approximation model of the residual function is expressed as the following formula 3:

[0020]

[0021] In the formula, p is the iteration step length, is the transpose of the gradient, H is the Hessian matrix of the residual function, p T is the transpose of the iteration step.

[0022] In step 4, constructing a quadratic approximation model of the residual function includes:

[0023] 4a. Calculate the transpose of the gradient according to the following formula 7

[0024]

[0025] In the formula, and They are the partial derivatives of the residual sum of squares f(a) with respect to the unknown coefficients of the radiation characteristic parameter model and the particle temperature;

[0026] 4b. Calculate the second-order partial derivative of each unknown parameter based on the residual function f(a) and construct the Hessian matrix H according to the following formula 8:

[0027]

[0028] In step 5, based on the Cholesky decomposition method and the Cauchy point method, the trust region radius Δ k The inner solution iteration step length makes the quadratic approximation model m(p k ) to obtain the minimum value, and the iterative step length solution is expressed as the following formula 4:

[0029]

[0030] Where a0 and a1 are the unknown coefficients of the radiation characteristic parameter model; T is the particle temperature in K; k is the number of calculations, and p is the iteration step.

[0031] In step 5, the flame temperature is solved based on the Cholesky decomposition method and the Cauchy point method, including:

[0032] 5a. The iteration step size p can be set by setting the gradient of the quadratic approximation model m(p) According to formulas 9 and 10, we can get:

[0033]

[0034] 5b. When the matrix H is positive definite, the Cholesky decomposition method is used to decompose H according to the following formula 11:

[0035] H=LL T (11)

[0036] Where L is a lower triangular matrix;

[0037] 5c. Obtain the iteration step length p according to Formula 7-Formula 11;

[0038] 5d. When the matrix H is not positive definite or cannot be decomposed, the Cauchy point method is used to solve the iteration step p within the initial trust region radius Δ0. The iteration step direction of the Cauchy point is set to the negative gradient direction. The calculation of the iteration step o is expressed as the following formula 12:

[0039]

[0040] Where α is the iteration step factor, and the iteration step p is a vector consisting of the changes in the unknown parameters.

[0041] In step 6, using the iteration step size p k The updated solution parameters are expressed as the following formula 5:

[0042]

[0043] The Cholesky decomposition method and the Cauchy point method are combined to solve the update iteration step. After using the iteration step to update the unknown parameters, the residual sum of squares is calculated again and convergence is judged. If convergence occurs, the result is the unknown parameters of the radiation characteristic parameter model and the true temperature of the particle. If the convergence condition is not met, the trust region radius is corrected based on the ratio of the actual residual function reduction to the predicted residual function reduction, and the unknown parameters of the radiation characteristic parameter model and the particle temperature are updated.

[0044] In step 6, if the convergence condition is not met, the amount Δf is reduced according to the actual residual function actual And the prediction residual function reduction Δf pred , calculate the ratio ρ according to the following formula 6 to determine the validity of the current iteration step:

[0045]

[0046] If ρ>0.75, the trust region radius needs to be expanded. If ρ<0.25, the trust region radius needs to be reduced. Other situations remain unchanged.

[0047] In step 7, the adjusted trust region radius Δ is used k+1 , substitute the updated parameters into step 3, and repeat steps 4 to 6 until the convergence condition is met; at this time, the calculated and T k+1 They are the parameters required for the radiation characteristic parameter model and the true temperature of the particles respectively.

[0048] The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particles in a complex combustion system in the present invention relates to a method for reconstructing the temperature and radiation characteristic parameters of particles in a complex combustion system based on the flame spectrum radiation intensity, and in particular to the temperature and radiation characteristic parameters of particles in a solid rocket engine plume. During the measurement of the present invention, a spectrometer is first used to collect the spectral radiation intensity of the combustion flame of the particles to be measured, and the band range used for calculation, the unknown parameters of the "wavelength-temperature-emissivity" radiation characteristic parameter model, the particle temperature and the trust region radius are initialized, and then the residual square is calculated band by band and the residual function is constructed, and the residual function is deformed into a quadratic model by applying Taylor expansion, and the Cho1esky decomposition method and the Cauchy point method are combined to solve the update iteration step length, and after the unknown parameters are updated using the iteration step length, the residual square sum is calculated again and convergence is judged, and if converged, the result at this time is the unknown parameters of the radiation characteristic parameter model and the true temperature of the particles, and if the convergence condition is not met, the trust region radius is corrected based on the ratio of the actual residual function reduction amount and the predicted residual function reduction amount, and the unknown parameters of the radiation characteristic parameter model and the particle temperature are updated. The unknown parameters of the "wavelength-temperature-emissivity" radiation characteristic parameter model and the particle temperature calculation method provided by the present invention have the advantages of high precision and fast convergence in a complex combustion environment of gas-liquid-solid multiphase coupling, and can provide high-precision particle temperature and radiation characteristic parameters based on the flame spectral radiation intensity. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flowchart of a specific embodiment of the method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to the present invention;

[0050] Figure 2 A graph showing the response value of a spectrometer and the radiation intensity of a detected spectrum in a specific embodiment of the present invention;

[0051] Figure 3 It is a schematic diagram of the result of the fitting radiation intensity changing with the number of iterations before the residual square sum converges in a specific embodiment of the present invention;

[0052] Figure 4 It is a schematic diagram of the results of the particle temperature and the residual sum of squares based on the spectral radiation intensity of the solid rocket engine plume changing with the number of iterations in a specific embodiment of the present invention;

[0053] Figure 5 It is a schematic diagram of the result of the emissivity changing with the number of iterations based on the “wavelength-temperature-emissivity” radiation characteristic parameter model in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0054] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0055] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations and / or combinations thereof.

[0056] like Figure 1 As shown, Figure 1 The flowchart of the method for simultaneously reconstructing the temperature and radiation characteristic parameters of the particulate matter in a complex combustion system of the present invention. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of the particulate matter in a complex combustion system comprises the following steps:

[0057] 1) Using a spectrometer to measure the flame to be tested to obtain the detection spectrum radiation intensity;

[0058] 2) Based on the wavelength-temperature dual variable emissivity model, the flame theoretical spectral radiation intensity is expressed as the following formula 1:

[0059]

[0060] Where I is the flame spectral radiation intensity, unit is W / m 3 / sr; a0 and a1 are the unknown coefficients of the radiation characteristic parameter model; T is the particle temperature, in K; λ is the waveband used in the calculation, in m; c1 and c2 are Planck constants.

[0061] 3) Set the initial parameters a0, a1, T, the initial trust region radius Δ0, and construct the residual function expressed as the following formula 2:

[0062]

[0063] In the formula, is the residual, and parameter a=[a0, a1, T] is the required parameter.

[0064] The initial parameters a0, a1, T, and the initial trust region radius Δ0 can be set to 0.01, 0.00 1, 1000, and 1, respectively.

[0065] 4) The residual function f(a) is expanded using Taylor near different bands, and the quadratic approximation model of the residual function is expressed as the following formula 3:

[0066]

[0067] In the formula, p is the iteration step length, is the transpose of the gradient, p T is the transpose of the iteration step, and H is the Hessian matrix of the residual function.

[0068] Constructing the quadratic approximation model of the residual function includes the following sub-steps:

[0069] 4a. Calculate the transpose of the gradient according to the following formula 7

[0070]

[0071] In the formula, and are the partial derivatives of the residual sum of squares f(a) with respect to the unknown coefficients of the radiation characteristic parameter model and the particle temperature, respectively.

[0072] 4b. Calculate the second-order partial derivative of each unknown parameter based on the residual function f(a) and construct the Hessian matrix H according to the following formula 8:

[0073]

[0074] 5) Based on the Cho1esky decomposition method and Cauchy point method, in the trust region radius Δ k The inner solution iteration step length makes the quadratic approximation model m(o k ) to obtain the minimum value, and the iterative step length solution is expressed as the following formula 4:

[0075]

[0076] In the formula, k is the number of calculations.

[0077] Solving the flame temperature based on the Cholesky decomposition method and the Cauchy point method includes the following sub-steps:

[0078] 5a. The iteration step size p can be set by setting the gradient of the quadratic approximation model m(p) According to formulas 9 and 10, we can get:

[0079]

[0080] 5b. When the matrix H is positive definite, the Cholesky decomposition method is used to decompose H according to the following formula 11:

[0081] H=LL T (11)

[0082] Where L is a lower triangular matrix.

[0083] 5c. According to formula 7-formula 11, the iteration step size is obtained.

[0084] 5d. When the matrix H is not positive definite or cannot be decomposed, the Cauchy point method is used to solve the iteration step p within the initial trust region radius Δ0. The iteration step direction of the Cauchy point is set to the negative gradient direction. The calculation of the iteration step p is expressed as the following formula 12:

[0085]

[0086] Where θ is the iteration step factor, and the iteration step p is a vector consisting of the changes in the unknown parameters.

[0087] 6) Use iteration step size p k The updated solution parameters are expressed as the following formula 5:

[0088]

[0089] Calculate the residual according to the updated solution parameters to determine whether the convergence condition is met. If so, then T k is the final particle temperature.

[0090] 7) If the convergence condition is not met, reduce the amount Δf according to the actual residual function actual And the prediction residual function reduction Δf pred , calculate the ratio ρ according to the following formula 6 to determine the validity of the current iteration step:

[0091]

[0092] If ρ>0.75, the trust region radius needs to be expanded. If ρ<0.25, the trust region radius needs to be reduced. Other situations remain unchanged.

[0093] Judging the validity of the selected iteration step and modifying the trust region radius includes the following sub-steps:

[0094] 7a. Actual residual function reduction Δf actual And the prediction residual function reduction Δf pred They are respectively expressed as follows:

[0095] Δf actual =f(a)-f(a+p) (13)

[0096] Δf pred =m(0)-m(p) (14)

[0097] 7b. Reduce the amount Δf according to the actual residual function actual And the prediction residual function reduction Δf pred , calculate the ratio ρ according to the following formula 6 to determine the validity of the current iteration step:

[0098]

[0099] If ρ>0.75, the trust region radius Δ can be expanded k+1 =2Δ k ; If ρ<0.25, the trust region radius Δ needs to be reduced k+1 =0.5Δ k ; If 0.25<ρ<0.75, the trust region radius Δ needs to be maintained k+1 =Δ k constant.

[0100] 8) Use the adjusted trust region radius Δ k+1 , substitute the updated parameters into step 3), and repeat steps 4) to 7) until the convergence condition is met. and T k+1 They are the parameters required for the final “wavelength-temperature-emissivity” radiation characteristic parameter model and the actual temperature of the particles.

[0101] The present invention uses a spectrometer to measure the spectral radiation intensity curve of a solid rocket engine plume, limits the wavelength range of the spectral radiation curve, uses the spectral radiation intensity within the range as input data, and reconstructs the radiation characteristic parameters and particle temperature based on the "wavelength-temperature-emissivity" radiation characteristic parameter model, the Cholesky decomposition method, and the Cauchy point method. The residual sum of squares of the detected spectral radiation intensity and the fitted spectral radiation intensity is used as the convergence basis, and the unknown parameters after iterative correction are input to repeat the next iteration. The above steps are repeated until the iteration converges, which are the parameters required by the final "wavelength-temperature-emissivity" radiation characteristic parameter model and the particle temperature.

[0102] In a specific embodiment of the present invention, the method of simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system includes:

[0103] 101. Use a spectrometer to measure the flame spectral radiation intensity, output a continuous band spectral radiation intensity curve, set the continuous band λ range to 640nm-760nm, parameters a0=0.01, a1=0.001, T=1200K, and the initial trust region radius Δ0=1.

[0104] 102. Based on the wavelength-temperature bivariate emissivity model, the flame theoretical spectral radiation intensity is expressed as the following formula 1:

[0105]

[0106] Where I is the theoretical spectral radiation intensity of the flame, in W / m 3 / sr; a0 and a1 are the unknown coefficients of the emissivity model; T is the particle temperature, in K; λ is the waveband used in the calculation, in m; c1 and c2 are Planck constants.

[0107] 103. Set the initial parameters a0, a1, T, the initial trust region radius Δ0, and construct the residual function expressed as the following formula 2:

[0108]

[0109] In the formula, is the flame spectrum radiation intensity measured by the spectrometer, is the flame spectrum radiation intensity obtained based on the emissivity model, and the parameter a=[a0, a1, T] is the required parameter.

[0110] 104. The residual function f(a) is expanded using Taylor near different bands, and the quadratic approximation model of the residual function is expressed as the following formula 3:

[0111]

[0112] In the formula, p is the iteration step length, is the transpose of the gradient, p T is the transpose of the iteration step, and H is the Hessian matrix of the residual function.

[0113] The transpose of the gradient is calculated according to the following formula 15

[0114]

[0115] In the formula, and are the partial derivatives of the residual sum of squares f(a) with respect to the unknown coefficients of the radiation characteristic parameter model and the particle temperature, respectively.

[0116] The second-order partial derivative of each unknown parameter is calculated based on the residual function f(a), and the Hessian matrix H is constructed according to the following formula 8:

[0117]

[0118] 105. Solve the iterative step increment p within the trust region radius Δ0 so that the quadratic approximation model m(p) achieves the minimum value. The iterative step solution is expressed as the following formula 16:

[0119]

[0120] 106. Using the iteration step size p to update the solution parameters is expressed as the following formula 17:

[0121]

[0122] T 1 =T+p T (17)

[0123] Calculate the residual according to the updated solution parameters to determine whether the convergence conditions are met. If so, output the result.

[0124] 107. If the convergence condition is not met, reduce the amount Δf according to the actual residual function actual And the prediction residual function reduction Δf pred , calculate the ratio ρ according to the following formula 6 to determine the validity of the current iteration step:

[0125]

[0126] If ρ>0.75, the trust region radius Δ2=2Δ1 can be expanded; if ρ<0.25, the trust region radius Δ2=0.5Δ1 needs to be reduced; if 0.25<ρ<0.75, the trust region radius Δ2=Δ1 needs to be kept unchanged.

[0127] 108. Use the adjusted trust region radius Δ2, substitute the updated parameters into Formula 2, and repeat steps 103 to 107 until the convergence condition is met. and T k+1 They are the parameters required for the final “wavelength-temperature-emissivity” radiation characteristic parameter model and the particle temperature.

[0128] A spectrometer was used to measure the plume temperature of a solid rocket engine under the condition of an incoming flow Mach number of 2.4. The measuring point was located on the engine centerline, 120 mm away from the engine nozzle. The measured solid rocket engine plume spectral radiation intensity correlation data was used as a calculation example.

[0129] Figure 2 The spectrometer response value and spectral radiation intensity curve of the solid rocket engine plume measured by the spectrometer. The spectral radiation intensity line 690nm-760nm band is selected to solve the unknown parameters. The analysis results are shown in Figure 3 , Figure 4 and Figure 5 . Figure 3 This is the result of the fitted radiation intensity changing with the number of iterations before the residual sum of squares converges. Figure 4 The particle temperature reconstructed based on the solid rocket motor plume flame spectral radiation intensity and the residual sum of squares change with the number of iterations. Figure 5 This is the result of the spectral emissivity changing with the number of iterations based on the “wavelength-temperature-emissivity” radiation characteristic parameter model.

[0130] For the above-mentioned embodiments of the present invention, common knowledge such as the well-known specific structures and characteristics in the scheme is not described in detail; the embodiments are described in a step-by-step manner, and the technical features involved in each embodiment can be combined with each other under the premise that there is no conflict between each other, and the same and similar parts between the embodiments can be referred to each other.

[0131] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as falling within the scope of protection of the present invention.

Claims

1. A method for simultaneously reconstructing the temperature and radiation characteristic parameters of particles in a complex combustion system, characterized in that: The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particles in a complex combustion system includes: Step 1, using a spectrometer to measure the combustion flame of the particles to be tested to obtain the detection spectrum radiation intensity; Step 2, based on the wavelength-temperature dual variable emissivity model, calculate the flame theoretical spectral radiation intensity; Step 3, setting initial parameters, and constructing a residual function about unknown parameters based on the flame detection spectrum radiation intensity and the theoretical spectrum radiation intensity; Step 4, apply Taylor expansion to transform the residual function into a quadratic model; Step 5, combine the Cho1esky decomposition method and the Cauchy point method to solve the update iteration step size; Step 6: Update the unknown parameters using the iteration step size, calculate the residual sum of squares and make a convergence judgment; Step 7: When the convergence condition is met, the unknown parameters required for the radiation characteristic parameter model and the actual temperature of the particles are updated.

2. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to claim 1, characterized in that: In step 1, based on the wavelength-temperature bivariate emissivity model, the flame theoretical spectral radiation intensity is expressed as the following formula 1: Where I is the theoretical spectral radiation intensity of the flame, in W / m 3 / sr; a0 and a1 are unknown coefficients of the radiation characteristic parameter model; T is the particle temperature, unit is K; λ is the waveband used for calculation, in m; c1 and c2 are Planck constants.

3. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to claim 2, characterized in that: In step 3, the initial parameters a0, a1, T and the initial trust region radius △0 are set, and the residual function is constructed as shown in the following formula 2: In the formula, is the residual, and parameter a=[a0, a1, T] is the required parameter.

4. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particles in a complex combustion system according to claim 3, characterized in that: In step 4, the residual function f(a) is expanded using Taylor near different bands, and the quadratic approximation model of the residual function is expressed as the following formula 3: In the formula, p is the iteration step length, is the transpose of the gradient, p T is the transpose of the iteration step, and H is the Hessian matrix of the residual function.

5. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to claim 4, characterized in that: In step 4, constructing a quadratic approximation model of the residual function includes: 4a. Calculate the transpose of the gradient according to the following formula 7 In the formula, and They are the partial derivatives of the residual sum of squares f(a) with respect to the unknown coefficients of the radiation characteristic parameter model and the particle temperature; 4b. Calculate the second-order partial derivative of each unknown parameter based on the residual function f(a) and construct the Hessian matrix H according to the following formula 8:

6. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to claim 5, characterized in that: In step 5, based on the Cho1esky decomposition method and the Cauchy point method, the trust region radius △ k The inner solution iteration step length makes the quadratic approximation model m(p k ) to obtain the minimum value, and the iterative step length solution is expressed as the following formula 4: Where a0 and a1 are the unknown coefficients of the radiation characteristic parameter model; T is the particle temperature in K; k is the number of calculations, and p is the iteration step.

7. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to claim 6, characterized in that: In step 5, the flame temperature is solved based on the Cholesky decomposition method and the Cauchy point method, including: 5a. The iteration step size p can be set by setting the gradient of the quadratic approximation model m(p) According to formulas 9 and 10, we can get: 5b. When the matrix H is positive definite, the Cholesky decomposition method is used to decompose H according to the following formula 11: H=LL T (11) Where L is a lower triangular matrix; 5c. Obtain the iteration step length p according to Formula 7-Formula 11; 5d. When the matrix H is not positive definite or cannot be decomposed, the Cauchy point method is used to solve the iteration step p within the initial trust region radius △0. The iteration step direction of the Cauchy point is set to the negative gradient direction. The calculation of the iteration step p is expressed as the following formula 12: Where α is the iteration step factor, and the iteration step p is a vector consisting of the changes in the unknown parameters.

8. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particles in a complex combustion system according to claim 7, characterized in that: In step 6, using the iteration step size p k The updated solution parameters are expressed as the following formula 5: The Cholesky decomposition method and the Cauchy point method are combined to solve the update iteration step. After using the iteration step to update the unknown parameters, the residual sum of squares is calculated again and convergence is judged. If convergence occurs, the result is the unknown parameters of the radiation characteristic parameter model and the true temperature of the particle. If the convergence condition is not met, the trust region radius is corrected based on the ratio of the actual residual function reduction to the predicted residual function reduction, and the unknown parameters of the radiation characteristic parameter model and the particle temperature are updated.

9. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to claim 8, characterized in that: In step 6, if the convergence condition is not met, the amount Δf is reduced according to the actual residual function actual And the prediction residual function reduction Δf pred , calculate the ratio ρ according to the following formula 6 to determine the validity of the current iteration step: If ρ>0.75, the trust region radius needs to be expanded. If ρ<0.25, the trust region radius needs to be reduced. Other situations remain unchanged.

10. The method for simultaneously reconstructing the temperature and radiation characteristic parameters of particulate matter in a complex combustion system according to claim 9, characterized in that: In step 7, use the adjusted trust region radius △ k+1 , substitute the updated parameters into step 3, and repeat steps 4 to 6 until the convergence condition is met; at this time, the calculated and T k+1 They are the parameters required for the radiation characteristic parameter model and the true temperature of the particles respectively.

Images