Composite solid propellant burning rate control method and system

Abstract

The application discloses a composite solid propellant burning rate regulation method, relates to the technical field of composite solid propellant, and comprises the following steps: collecting the burning rate and the mixer torque value of the composite solid propellant under different conditions; constructing a correlation model of the burning rate and the mixer torque value; inputting the mixer torque value measured in the production process into the correlation model for calculation to obtain the propellant burning rate prediction value under different oxidant granularity and oxidant content conditions; and adjusting the oxidant grading according to the conformity and dispersion of the burning rate prediction value and design indexes. The application further discloses a composite solid propellant burning rate regulation system, which comprises a data acquisition module, a data analysis module, a burning rate prediction module and an oxidant adjustment module. The application proposes to combine multivariate regression analysis, real-time data prediction and fine oxidant grading adjustment, so that high-precision burning rate control can be realized, and the adaptability and optimization space are strong.

Description

Technical Field

[0001] This invention relates to the field of composite solid propellant technology, specifically to a method and system for controlling the burning rate of composite solid propellants. Background Technology

[0002] Composite solid propellants are a widely used high-energy propulsion material, extensively used in the propulsion systems of rockets, missiles, and other spacecraft. Their performance directly affects the effectiveness of the propulsion system. Burning rate is one of the important parameters of propellant performance. The burning rate of the propellant not only affects the thrust of the thruster, but is also closely related to factors such as the ratio of oxidizer and fuel, particle size, and curing conditions. Therefore, precise control of the burning rate is crucial for the design and production of propellants.

[0003] Traditional burning rate control methods rely on experience and experimental data. However, with the increase in production scale and the complexity of production processes, relying solely on experiments often fails to achieve the desired control effect. Achieving accurate prediction and real-time control of the burning rate has become a significant challenge. Existing technologies mainly rely on simple physical modeling of the burning rate of solid propellants, lacking in-depth analysis of the complex relationships between multiple factors. Linear regression analysis and mathematical modeling methods have been gradually applied in this field, but there is still considerable room for improvement in multivariate control and high-precision regulation. Therefore, developing a multivariate regression analysis-based method for controlling the burning rate of composite solid propellants, combined with actual production data and predictive models, to optimize propellant proportions and performance, has become a key technology for improving production efficiency and product quality. Summary of the Invention

[0004] To address the aforementioned technical problems, a method and system for controlling the burning rate of composite solid propellants are provided. This technical solution solves the problems described above.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method and system for controlling the burning rate of a composite solid propellant, comprising:

[0007] The burning rate and mixer torque values ​​of the composite solid propellant under different oxidant particle sizes, oxidant contents, and curing conditions were collected and preprocessed to obtain a sample set;

[0008] Mathematical regression was performed on the burning rate and mixer torque values ​​in the sample set to obtain a linear regression equation that describes the relationship between propellant burning rate and mixer torque. Based on the obtained regression results, a correlation model between burning rate and mixer torque values ​​was constructed.

[0009] The torque value of the mixer measured during the production process is input into the associated model for calculation to obtain the predicted value of the propellant burning rate under different oxidant particle size and oxidant content conditions;

[0010] Adjust the oxidant gradation based on the conformity and dispersion of the predicted burn rate with the design specifications.

[0011] Preferably, the sample set obtained by preprocessing the collected composite solid propellant combustion rate and mixer torque values ​​under different oxidizer particle sizes, oxidizer contents, and curing conditions specifically includes:

[0012] Check the collected data for missing values ​​and handle the missing values ​​using interpolation.

[0013] The interpolation formula is as follows:

[0014]

[0015] In the formula, x1x2 are the x-coordinates of two known data points before and after the point where the missing value is located, y1y2 are the corresponding y-values, and x... i It is the i-th position that needs interpolation, y i The i-th estimated missing value;

[0016] The moving average method is used to remove noise caused by sensor fluctuations and external interference to ensure data quality. The standard deviation statistical method is used to identify outliers in the data. For each set of data for combustion rate and mixer torque, the deviation from the overall data distribution is calculated. If a data point deviates from the overall distribution by more than a set threshold, it is considered abnormal data. Combining the physical laws of the propellant production process and the working parameters of the equipment, data points that cannot meet the physical constraints are eliminated according to the preset physical constraints.

[0017] The mixer torque and combustion rate data are standardized and converted to the 0-1 range.

[0018] Preferably, the method for identifying outliers in the data based on standard deviation statistical methods specifically includes:

[0019] The formula for the statistical method of standard deviation is as follows:

[0020]

[0021] In the formula, x i Let Z be the i-th data point, μ be the mean of the dataset, σ be the standard deviation of the dataset, and Z be the mean of the dataset. i The standard deviation of the i-th data point;

[0022] Based on the standard deviation of data points, outlier data is identified. If the absolute value of the standard deviation of a data point exceeds twice the set threshold, the data point is considered an outlier.

[0023] Preferably, the step of performing mathematical regression on the burning rate and mixer torque values ​​in the sample set to obtain a linear regression equation describing the relationship between propellant burning rate and mixer torque, and constructing a correlation model between burning rate and mixer torque values ​​based on the obtained regression results, specifically includes:

[0024] Obtain a sample set containing combustion speed and hybrid torque values, and construct a linear regression model. The formula for the linear regression model is:

[0025] Y = a + bX

[0026] In the formula, Y is the propellant burning rate, X is the mixer torque value, a is the constant term of the regression equation, and b is the regression coefficient;

[0027] The linear regression model is used to find the best line for the data, minimizing the sum of squared errors between the actual observed values ​​and the predicted values.

[0028] The regression coefficients are obtained by taking the derivatives of a and b and setting the derivatives to zero.

[0029] Using the obtained regression equation, combustion rate is predicted for different mixer torque values.

[0030] Preferably, the step of finding the best-fit line for the data based on a linear regression model and minimizing the sum of squares of the errors between the actual observed values ​​and the predicted values ​​specifically includes:

[0031] The formula for the sum of squares of the errors is:

[0032]

[0033] In the formula, E is the sum of squared errors, i is the index, n is the total number of data points, and y i The i-th observation, where a is the intercept of the regression equation, b is the slope of the regression equation, and x... i The value of the i-th independent variable;

[0034] The value of E is minimized by taking the derivatives of a and b.

[0035] Preferably, the step of taking the derivatives of a and b and setting the derivatives to zero to obtain the regression coefficients specifically includes:

[0036] Take the partial derivatives with respect to a and b respectively, and set them to zero. The formula for calculating the partial derivatives is:

[0037]

[0038] In the formula, y represents the partial derivative of the sum of squared errors E with respect to the regression coefficient a. i Let x be the true dependent variable value corresponding to the i-th observation, a be the intercept in the regression equation, b be the slope in the regression equation, and x be the slope. i Let i be the value of the i-th independent variable. This represents the partial derivative of the sum of squared errors E with respect to the regression coefficient a;

[0039] By solving these two equations, we obtain the regression coefficients a and b, and thus the linear regression equation.

[0040] Preferably, the step of inputting the mixer torque value measured during the production process into the associated model for calculation to obtain the propellant burning rate prediction value under different oxidant particle size and oxidant content conditions specifically includes:

[0041] The torque value of the mixer measured in real time during the production process is input into the trained correlation model. The model predicts the burning rate based on the input torque value, oxidant particle size and oxidant content. The predicted value is the propellant burning rate under different oxidant particle size and oxidant content conditions.

[0042] The predicted combustion rate is compared with the design target value to check whether it meets the requirements.

[0043] Preferably, adjusting the oxidant gradation based on the conformity and dispersion of the predicted burning rate with the design specifications specifically includes:

[0044] Calculate the current burning rate error and dispersion, compare the prediction results of the regression model with the design index, calculate the error between the actual burning rate and the design index, and calculate the dispersion of the burning rate.

[0045] The formula for calculating the dispersion of the burning rate is as follows:

[0046]

[0047] In the formula, n is the total number of sample data. For the predicted burning rate of the i-th observation, J is the average of all observations, and J is the dispersion of the burning rate.

[0048] Based on experimental and production data, we analyzed the effects of different oxidant gradations, including the mass ratio of oxidant to fuel, on combustion rate and dispersibility.

[0049] Based on the error between the predicted burning rate and the design target, the ratio of oxidizer is adjusted to make the burning rate closer to the design target.

[0050] By gradually changing the oxidant gradation ratio and making predictions, the difference between the predicted burning rate and the design index, as well as the change in dispersibility, are observed. The oxidant gradation is continuously adjusted until the burning rate meets the design requirements and the dispersibility meets the requirements.

[0051] Preferably, the step of gradually changing the oxidant gradation ratio and making predictions, observing the difference between the predicted burning rate and the design index, as well as the change in dispersibility, and continuously adjusting the oxidant gradation until the burning rate meets the design requirements and the dispersibility meets the requirements specifically includes:

[0052] By adjusting the oxidizer gradation, the error between the predicted burning rate and the design target is reduced, so that the predicted burning rate is as close as possible to the target burning rate. A tolerance range is set to ensure that the burning rate is within this range. If the burning rate exceeds this range, the oxidizer ratio is adjusted.

[0053] Through actual production verification, it was found that the propellant burning rate under the adjusted oxidizer gradation met the design target and had a small degree of dispersion. If it still did not meet the requirements, the adjustment was continued. The adjusted burning rate met the target, but there was still some dispersion. The dispersion was reduced by further refining the adjustment of the oxidizer gradation.

[0054] After multiple rounds of optimization, the optimal oxidant gradation ratio was obtained.

[0055] A composite solid propellant burn rate control system, comprising:

[0056] Data acquisition module: The data acquisition module is used to collect the burning rate and mixer torque values ​​of composite solid propellant under different oxidant particle sizes, oxidant contents and curing conditions, and obtain a sample set after preprocessing;

[0057] Data Analysis Module: The data analysis module is electrically connected to the data acquisition module. The data analysis module is used to perform mathematical regression on the burning rate and mixer torque values ​​in the sample set to obtain a linear regression equation that describes the relationship between propellant burning rate and mixer torque. Based on the obtained regression results, a correlation model between burning rate and mixer torque values ​​is constructed.

[0058] Burn rate prediction module: The burn rate prediction module is electrically connected to the data analysis module. The burn rate prediction module is used to input the mixer torque value measured during the production process into the associated model for calculation to obtain the propellant burn rate prediction value under different oxidant particle size and oxidant content conditions.

[0059] Oxidant Adjustment Module: The oxidant adjustment module is electrically connected to the combustion rate prediction module. The oxidant adjustment module is used to adjust the oxidant gradation based on the conformity and dispersion of the combustion rate prediction value with the design index.

[0060] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0061] This invention proposes a method that combines multivariate regression analysis, real-time data prediction, and refined oxidizer gradation adjustment to achieve not only high-precision burning rate control but also strong adaptability and optimization potential, significantly improving propellant production efficiency and product quality. Attached Figure Description

[0062] Figure 1 This is a flowchart outlining the steps of the present invention.

[0063] Figure 2 This is a system framework diagram of the present invention. Detailed Implementation

[0064] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0065] Reference Figure 1 As shown, a method and system for controlling the burning rate of a composite solid propellant includes:

[0066] A method and system for controlling the burning rate of a composite solid propellant, comprising:

[0067] Step 1:

[0068] Check the collected data for missing values ​​and handle the missing values ​​using interpolation.

[0069] The interpolation formula is as follows:

[0070]

[0071] In the formula, x1x2 are the x-coordinates of two known data points before and after the point where the missing value is located, y1y2 are the corresponding y-values, and x... i It is the i-th position that needs interpolation, y i The i-th estimated missing value;

[0072] The moving average method is used to remove noise caused by sensor fluctuations and external interference to ensure data quality. The standard deviation statistical method is used to identify outliers in the data. For each set of data for combustion rate and mixer torque, the deviation from the overall data distribution is calculated. If a data point deviates from the overall distribution by more than a set threshold, it is considered abnormal data. Combining the physical laws of the propellant production process and the working parameters of the equipment, data points that cannot meet the physical constraints are eliminated according to the preset physical constraints.

[0073] The formula for the statistical method of standard deviation is as follows:

[0074]

[0075] In the formula, x i Let Z be the i-th data point, μ be the mean of the dataset, σ be the standard deviation of the dataset, and Z be the mean of the dataset. i The standard deviation of the i-th data point;

[0076] Based on the standard deviation of data points, outlier data is identified. If the absolute value of the standard deviation of a data point exceeds twice the set threshold, the data point is considered an outlier.

[0077] The torque and combustion rate data of the mixer are standardized and converted to the range of 0-1.

[0078] Step Two:

[0079] Obtain a sample set containing combustion speed and hybrid torque values, and construct a linear regression model. The formula for the linear regression model is:

[0080] Y = a + bX

[0081] In the formula, Y is the propellant burning rate, X is the mixer torque value, a is the constant term of the regression equation, and b is the regression coefficient;

[0082] The linear regression model is used to find the best line for the data, minimizing the sum of squared errors between the actual observed values ​​and the predicted values.

[0083] The formula for the sum of squares of the errors is:

[0084]

[0085] In the formula, E is the sum of squared errors, i is the index, n is the total number of data points, and y i The i-th observation, where a is the intercept of the regression equation, b is the slope of the regression equation, and x... i The value of the i-th independent variable;

[0086] By taking the derivatives of a and b, the value of E is minimized;

[0087] Take the partial derivatives with respect to a and b respectively, and set them to zero. The formula for calculating the partial derivatives is:

[0088]

[0089] In the formula, y represents the partial derivative of the sum of squared errors E with respect to the regression coefficient a. i Let x be the true dependent variable value corresponding to the i-th observation, a be the intercept in the regression equation, b be the slope in the regression equation, and x be the slope. i Let i be the value of the i-th independent variable. This represents the partial derivative of the sum of squared errors E with respect to the regression coefficient a;

[0090] By solving these two equations, we obtain the regression coefficients a and b, and thus the linear regression equation.

[0091] Using the obtained regression equation, combustion rate is predicted for different mixer torque values.

[0092] Step 3:

[0093] The torque value of the mixer measured in real time during the production process is input into the trained correlation model. The model predicts the burning rate based on the input torque value, oxidant particle size and oxidant content. The predicted value is the propellant burning rate under different oxidant particle size and oxidant content conditions.

[0094] The predicted combustion rate is compared with the design target value to check whether it meets the requirements.

[0095] Step Four:

[0096] Calculate the current burning rate error and dispersion, compare the prediction results of the regression model with the design index, calculate the error between the actual burning rate and the design index, and calculate the dispersion of the burning rate.

[0097] The formula for calculating the dispersion of the burning rate is as follows:

[0098]

[0099] In the formula, n is the total number of sample data. For the predicted burning rate of the i-th observation, J is the average of all observations, and J is the dispersion of the burning rate.

[0100] Based on experimental and production data, we analyzed the effects of different oxidant gradations, including the mass ratio of oxidant to fuel, on combustion rate and dispersibility.

[0101] Based on the error between the predicted burning rate and the design target, the ratio of oxidizer is adjusted to make the burning rate closer to the design target.

[0102] By gradually changing the gradation ratio of the oxidant and making predictions, the difference between the predicted burning rate and the design index, as well as the change in dispersibility, are observed. The gradation of the oxidant is continuously adjusted until the burning rate meets the design requirements and the dispersibility meets the requirements.

[0103] By adjusting the oxidizer gradation, the error between the predicted burning rate and the design target is reduced, so that the predicted burning rate is as close as possible to the target burning rate. A tolerance range is set to ensure that the burning rate is within this range. If the burning rate exceeds this range, the oxidizer ratio is adjusted.

[0104] Through actual production verification, it was found that the propellant burning rate under the adjusted oxidizer gradation met the design target and had a small degree of dispersion. If it still did not meet the requirements, the adjustment was continued. The adjusted burning rate met the target, but there was still some dispersion. The dispersion was reduced by further refining the adjustment of the oxidizer gradation.

[0105] After multiple rounds of optimization, the optimal oxidant gradation ratio was obtained.

[0106] A composite solid propellant burn rate control system, comprising:

[0107] Data acquisition module: The data acquisition module is used to collect the burning rate and mixer torque values ​​of composite solid propellant under different oxidant particle sizes, oxidant contents and curing conditions, and obtain a sample set after preprocessing;

[0108] Data Analysis Module: The data analysis module is electrically connected to the data acquisition module. The data analysis module is used to perform mathematical regression on the burning rate and mixer torque values ​​in the sample set to obtain a linear regression equation that describes the relationship between propellant burning rate and mixer torque. Based on the obtained regression results, a correlation model between burning rate and mixer torque values ​​is constructed.

[0109] Burn rate prediction module: The burn rate prediction module is electrically connected to the data analysis module. The burn rate prediction module is used to input the mixer torque value measured during the production process into the associated model for calculation to obtain the propellant burn rate prediction value under different oxidant particle size and oxidant content conditions.

[0110] Oxidant Adjustment Module: The oxidant adjustment module is electrically connected to the combustion rate prediction module. The oxidant adjustment module is used to adjust the oxidant gradation based on the conformity and dispersion of the combustion rate prediction value with the design index.

[0111] The process of using this invention is as follows:

[0112] Step 1: Collect propellant burning rate and mixer torque values ​​under different oxidant particle sizes, oxidant contents, and curing conditions;

[0113] Step 2: Conduct a preliminary check on the collected sample data to identify any missing values;

[0114] Step 3: Use interpolation to fill in the missing data;

[0115] Step 4: Use the moving average method to remove noise caused by sensor fluctuations and external interference;

[0116] Step 5: Identify outlier data points using standard deviation statistical methods;

[0117] Step Six: Based on preset physical constraints, eliminate abnormal data that does not meet production conditions;

[0118] Step 7: Standardize the mixer torque value and combustion rate data, converting them into a range of 0-1;

[0119] Step 8: Based on the processed data, construct a linear regression model between combustion speed and the mixer torque value;

[0120] Step 9: Establish the mathematical relationship between combustion speed and mixer torque using a linear regression model formula;

[0121] Step 10: Optimize the regression coefficients in the regression equation using the method of minimizing the sum of squared errors;

[0122] Step 11: Take the partial derivatives of the constant term and regression coefficients of the regression equation, set them to zero, and solve for the optimal regression coefficients;

[0123] Step 12: Predict the burning rate using the optimized regression equation;

[0124] Step 13: Input the real-time measured mixer torque value during the production process into the trained regression model;

[0125] Step Fourteen: Predict the propellant burning rate based on the input mixer torque value, oxidizer particle size, and oxidizer content;

[0126] Step 15: Compare the predicted burning rate with the design target burning rate to check for compliance;

[0127] Step 16: Calculate the error between the actual burning rate and the design target, and calculate the dispersion of the burning rate;

[0128] Step 17: Based on experimental data, analyze the effect of different oxidant and fuel mass ratios on burning rate and dispersibility;

[0129] Step 18: Adjust the oxidizer ratio based on the error between the predicted burn rate and the design target;

[0130] Step 19: By gradually adjusting the oxidant gradation ratio, observe the difference between the burning rate and the design target, and make adjustments to reduce errors and dispersion;

[0131] Step 20: After multiple rounds of optimization, verify whether the burning rate meets the design specifications and ensure the optimization of the oxidant gradation until the dispersion meets the predetermined requirements.

[0132] In summary, the advantages of this invention are:

[0133] By collecting data on burning rate and mixer torque under different oxidant particle sizes, oxidant contents, and curing conditions, and combining this with mathematical regression analysis, a precise correlation model between propellant burning rate and mixer torque can be established. This model can not only predict burning rate during the experimental stage, but also adjust the oxidant ratio in real time during the production process to ensure that the burning rate reaches the predetermined design target.

[0134] By preprocessing the collected data, including missing value interpolation, noise removal, and outlier removal, the high quality of the model input data is ensured, thereby improving the prediction accuracy and reliability of the regression model. This process guarantees the accuracy of the data, effectively avoids errors caused by data fluctuations or interference, and enhances the stability of the model.

[0135] By using the standard deviation statistical method to identify and eliminate outliers, data anomalies caused by sensor errors or external interference are reduced, and the burning rate control in the production process is optimized. This helps to ensure the consistency and stability of propellant performance and reduces uncertainties in production.

[0136] By predicting the error and dispersion between the burning rate and the design target through the model, the oxidant gradation can be adjusted in real time, so that the burning rate gradually approaches the target value. Through multiple rounds of optimization, the ratio of oxidant can be continuously adjusted until the design requirements are met and the dispersion is minimized. This process makes the production process more flexible and precise, and can quickly respond to deviations that occur during the production process.

[0137] This method can predict and adjust the burning rate of the propellant in real time, which greatly shortens the production cycle and reduces the workload of experimental verification. Since the relationship between the burning rate and the oxidizer gradation is accurately modeled, frequent experimental debugging and excessive manual intervention are no longer required in the production process, which improves the automation and efficiency of production.

[0138] By continuously optimizing the oxidizer gradation, we can not only ensure that the propellant burning rate meets the design requirements, but also effectively control the dispersion and ensure that the propellant performance remains consistent throughout the production process. This is crucial for high-precision aerospace and missile applications.

[0139] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for controlling the burning rate of a composite solid propellant, characterized in that, include: The burning rate and mixer torque values ​​of the composite solid propellant under different oxidant particle sizes, oxidant contents, and curing conditions were collected and preprocessed to obtain a sample set; Mathematical regression was performed on the burning rate and mixer torque values ​​in the sample set to obtain a linear regression equation that describes the relationship between propellant burning rate and mixer torque. Based on the obtained regression results, a correlation model between burning rate and mixer torque values ​​was constructed. The torque value of the mixer measured during the production process is input into the associated model for calculation to obtain the predicted value of the propellant burning rate under different oxidant particle size and oxidant content conditions; Adjust the oxidant gradation based on the conformity and dispersion of the predicted burn rate with the design specifications.

2. The method for controlling the burning rate of a composite solid propellant according to claim 1, characterized in that, The collected composite solid propellant combustion rate and mixer torque values ​​under different oxidizer particle sizes, oxidizer contents, and curing conditions were preprocessed to obtain a sample set, specifically including: Check the collected data for missing values ​​and handle the missing values ​​using interpolation. The interpolation formula is as follows: In the formula, x1x2 are the x-coordinates of two known data points before and after the point where the missing value is located, y1y2 are the corresponding y-values, and x... i It is the i-th position that needs interpolation, y i The i-th estimated missing value; The moving average method is used to remove noise caused by sensor fluctuations and external interference to ensure data quality. The standard deviation statistical method is used to identify outliers in the data. For each set of data for combustion rate and mixer torque, the deviation from the overall data distribution is calculated. If a data point deviates from the overall distribution by more than a set threshold, it is considered abnormal data. Combining the physical laws of the propellant production process and the working parameters of the equipment, data points that cannot meet the physical constraints are eliminated according to the preset physical constraints. The mixer torque and combustion rate data are standardized and converted to the 0-1 range.

3. The method for controlling the burning rate of a composite solid propellant according to claim 2, characterized in that, The method for identifying outliers in data based on standard deviation statistical methods is specifically... include: The formula for the statistical method of standard deviation is as follows: In the formula, x i Let Z be the i-th data point, μ be the mean of the dataset, σ be the standard deviation of the dataset, and Z be the mean of the dataset. i The standard deviation of the i-th data point; Based on the standard deviation of data points, outlier data is identified. If the absolute value of the standard deviation of a data point exceeds twice the set threshold, the data point is considered an outlier.

4. The method for controlling the burning rate of a composite solid propellant according to claim 3, characterized in that, The mathematical regression of the burning rate and mixer torque values ​​in the sample set yields a linear regression equation describing the relationship between propellant burning rate and mixer torque. Based on the regression results, the specific steps for constructing a correlation model between burning rate and mixer torque values ​​include: Obtain a sample set containing combustion speed and hybrid torque values, and construct a linear regression model. The formula for the linear regression model is: Y = a + bX In the formula, Y is the propellant burning rate, X is the mixer torque value, a is the constant term of the regression equation, and b is the regression coefficient; The linear regression model is used to find the best line for the data, minimizing the sum of squared errors between the actual observed values ​​and the predicted values. The regression coefficients are obtained by taking the derivatives of a and b and setting the derivatives to zero. Using the obtained regression equation, combustion rate is predicted for different mixer torque values.

5. The method for controlling the burning rate of a composite solid propellant according to claim 4, characterized in that, The method, based on a linear regression model, finds the straight line that best suits the data, minimizing the sum of squared errors between the actual observed and predicted values. include: The formula for the sum of squares of the errors is: In the formula, E is the sum of squared errors, i is the index, n is the total number of data points, and y i The i-th observation, where a is the intercept of the regression equation, b is the slope of the regression equation, and x... i The value of the i-th independent variable; The value of E is minimized by taking the derivatives of a and b.

6. The method for controlling the burning rate of a composite solid propellant according to claim 5, characterized in that, The process of taking the derivatives of a and b and setting the derivatives to zero to obtain the regression coefficients specifically includes: Take the partial derivatives with respect to a and b respectively, and set them to zero. The formula for calculating the partial derivatives is: In the formula, y represents the partial derivative of the sum of squared errors E with respect to the regression coefficient a. i Let x be the true dependent variable value corresponding to the i-th observation, a be the intercept in the regression equation, b be the slope in the regression equation, and x be the slope. i Let i be the value of the i-th independent variable. This represents the partial derivative of the sum of squared errors E with respect to the regression coefficient a; By solving these two equations, we obtain the regression coefficients a and b, and thus the linear regression equation.

7. The method for controlling the burning rate of a composite solid propellant according to claim 6, characterized in that, The step of inputting the mixer torque value measured during the production process into the associated model for calculation to obtain the propellant burning rate prediction value under different oxidant particle size and oxidant content conditions specifically includes: The torque value of the mixer measured in real time during the production process is input into the trained correlation model. The model predicts the burning rate based on the input torque value, oxidant particle size and oxidant content. The predicted value is the propellant burning rate under different oxidant particle size and oxidant content conditions. The predicted combustion rate is compared with the design target value to check whether it meets the requirements.

8. The method and system for controlling the burning rate of a composite solid propellant according to claim 7, characterized in that, The adjustment of oxidant gradation based on the conformity and dispersion of the predicted combustion rate with the design specifications specifically includes: Calculate the current burning rate error and dispersion, compare the prediction results of the regression model with the design index, calculate the error between the actual burning rate and the design index, and calculate the dispersion of the burning rate. The formula for calculating the dispersion of the burning rate is as follows: In the formula, n is the total number of sample data. For the predicted burning rate of the i-th observation, J is the average of all observations, and J is the dispersion of the burning rate. Based on experimental and production data, we analyzed the effects of different oxidant gradations, including the mass ratio of oxidant to fuel, on combustion rate and dispersibility. Based on the error between the predicted burning rate and the design target, the ratio of oxidizer is adjusted to make the burning rate closer to the design target. By gradually changing the oxidant gradation ratio and making predictions, the difference between the predicted burning rate and the design index, as well as the change in dispersibility, are observed. The oxidant gradation is continuously adjusted until the burning rate meets the design requirements and the dispersibility meets the requirements.

9. The method and system for controlling the burning rate of a composite solid propellant according to claim 8, characterized in that, The process of gradually changing the oxidant gradation ratio and making predictions, observing the difference between the predicted burning rate and the design parameters, as well as changes in dispersibility, and continuously adjusting the oxidant gradation until the burning rate meets the design requirements and the dispersibility meets the requirements, specifically includes: By adjusting the oxidizer gradation, the error between the predicted burning rate and the design target is reduced, so that the predicted burning rate is as close as possible to the target burning rate. A tolerance range is set to ensure that the burning rate is within this range. If the burning rate exceeds this range, the oxidizer ratio is adjusted. Through actual production verification, it was found that the propellant burning rate under the adjusted oxidizer gradation met the design target and had a small degree of dispersion. If it still did not meet the requirements, the adjustment was continued. The adjusted burning rate met the target, but there was still some dispersion. The dispersion was reduced by further refining the adjustment of the oxidizer gradation. After multiple rounds of optimization, the optimal oxidant gradation ratio was obtained.

10. A composite solid propellant burn rate control system, characterized in that, include: Data acquisition module: The data acquisition module is used to collect the burning rate and mixer torque values ​​of composite solid propellant under different oxidant particle sizes, oxidant contents and curing conditions, and obtain a sample set after preprocessing; Data Analysis Module: The data analysis module is electrically connected to the data acquisition module. The data analysis module is used to perform mathematical regression on the burning rate and mixer torque values ​​in the sample set to obtain a linear regression equation that describes the relationship between propellant burning rate and mixer torque. Based on the obtained regression results, a correlation model between burning rate and mixer torque values ​​is constructed. Burn rate prediction module: The burn rate prediction module is electrically connected to the data analysis module. The burn rate prediction module is used to input the mixer torque value measured during the production process into the associated model for calculation to obtain the propellant burn rate prediction value under different oxidant particle size and oxidant content conditions. Oxidant Adjustment Module: The oxidant adjustment module is electrically connected to the combustion rate prediction module. The oxidant adjustment module is used to adjust the oxidant gradation based on the conformity and dispersion of the combustion rate prediction value with the design index.

Images