Cooking-off characteristic prediction and analysis method for charge fuze

Abstract

The invention discloses a cook-off characteristic prediction and analysis method for a charge fuze. The method comprises the following steps: step 1, establishing an equal-proportion calculation model of the charge fuze used in a cook-off test; 2, carrying out numerical calculation on the established model to obtain characteristic parameters; 3, a neural network model is established, the characteristic parameters serve as training samples, the charge length, the charge diameter, the heating rate and the thickness of the explosive-proof plate and the booster tube shell serve as input parameters, the cook-off response time, the hot spot coordinates and the hot spot temperature serve as output, and training is carried out; 4, calling the trained neural network model, inputting grain length, grain diameter, shell thickness and cook-off heating rate, and predicting cook-off response time, hot spot coordinates and hot spot temperature. When the size of the grain and the thickness of the end cover are changed, a calculation model does not need to be reconstructed, numerical simulation calculation does not need to be conducted, and a rapid and continuous cook-off response result can be obtained through the neural network model.

Description

Technical Field

[0001] The invention belongs to the technical field of finite element numerical calculation, and in particular relates to a method for predicting and analyzing the baking characteristics of a charge fuze. Background Art

[0002] With the continuous improvement of the destructive performance of modern ammunition and weapons, it is urgent to improve the safety performance of weapons in production, transportation, storage and battlefield use. For ammunition and weapons, thermal safety performance is of paramount importance. In order to ensure that ammunition and weapons will not easily respond to the external thermal environment during the whole process, it is of great significance to study their safety under thermal stimulation.

[0003] Cook-off test and numerical calculation method are important means to test and evaluate the thermal safety of ammunition. Due to the high risk and high cost of ammunition, the cook-off test also faces these problems. In addition, due to the short time and high power of the ammunition explosion, it is also very difficult to obtain the internal data of the ammunition. The numerical calculation method can change the parameters such as the cook-off heating rate, the shell thickness of the ammunition, and the aspect ratio of the ammunition according to the actual situation of the ammunition and the thermal environment in which it is located, and obtain the temperature distribution cloud map inside the ammunition during the cook-off process and the data such as the cook-off response position. However, when using the numerical simulation method for cook-off calculation, every time these parameters change, the calculation model needs to be remodeled, and because the physical time of the cook-off test itself is long, the time consumed when performing numerical calculations will also be very long. In view of this, the current research on the cook-off characteristics of the charge usually uses a fixed calculation model, selects several specific heating rates for simulation, and cannot obtain the relationship between the continuous cook-off heating rate and the cook-off characteristics. Summary of the invention

[0004] The present invention provides a method for predicting and analyzing the bake-off characteristics of a charge fuze, which can predict the bake-off characteristics of the fuze. When the size of the charge column and the thickness of the end cover are changed, there is no need to rebuild the calculation model and perform numerical simulation calculations. Through the neural network model, a fast and continuous bake-off response result can be obtained.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for predicting and analyzing the cooking characteristics of a charge fuze comprises the following steps:

[0007] Step 1: Establish a proportional calculation model of the charge fuze used in the baking test, the calculation model includes a hood, air, potting material, shell, insurance mechanism, detonating charge, explosion-proof plate, explosion-transmitting tube shell, and explosion-transmitting charge. The inside of the hood is air, and the electronic components are equivalent to the potting material. The shell is connected to the hood and the potting material by threads, the insurance mechanism is fixed to the shell by threaded connection with the explosion-proof plate and the explosion-transmitting tube shell, the detonating charge is filled between the insurance mechanism and the explosion-proof plate, and the explosion-transmitting charge is filled between the explosion-transmitting tube shell and the explosion-proof plate; the insurance mechanism puts the fuze in a safe state, the explosion-proof plate isolates the explosive elements of each level in the explosion sequence, the detonating charge and the explosion-transmitting charge form a charge, the charge is composed of RDX-based charge and adhesive, and the length and diameter of the charge and the thickness of the explosion-proof plate and the explosion-transmitting tube shell are specifically set according to actual needs;

[0008] Step 2: numerically calculate the established model; the self-heating reaction source term of the charge is set using the multi-grid method, and the thermal conductivity and specific heat capacity are obtained according to the superposition principle. The specific calculation formula is:

[0009] λ Ct =αλ1+βλ2

[0010] C Ct =αC1+βC2

[0011] Where λ is thermal conductivity, C is specific heat capacity, 1 is the index of RDX-based charge, 2 is the index of binder, and Ct is the index of mixed charge; in the charge, the volume ratio of RDX-based charge is α, and the volume ratio of binder is β. The multi-step decomposition process of RDX-based charge under heating follows the Arrhenius equation, and its thermal decomposition process is expressed as:

[0012]

[0013]

[0014]

[0015] Where: r is the charge reaction rate, unit mol·L -1 ·s -1 , subscripts 1, 2, 3 are reaction numbers; reaction 1 is a primary chemical reaction of charge endothermic reaction, reaction 2 is a primary chemical reaction of solid intermediate product into gaseous intermediate product, reaction 3 is a secondary chemical reaction of gaseous intermediate product further decomposition into final gaseous product; Z is pre-exponential factor, unit s -1 ; E is the charge activation energy, unit J·mol -1 ; R is the universal gas constant, which is 8.314 J·mol -1 ·K -1; A is RDX-based charge; B is solid intermediate product; C is gaseous intermediate product; D is gaseous final product; ρ A , B , C , D are the mass concentrations of A, B, C, and D, in kg·m -3 , T is the reaction temperature, unit K;

[0016] where ρ A , B , C , D The representation method is as follows:

[0017]

[0018] ρ i is the mass concentration of the ith product, M i is the mass of the i-th product, M RDX is the mass of RDX-based charge, V is the volume of RDX-based charge, α i is the proportion of the i-th product to the total mass of the RDX-based charge, ρ RDX is the density of RDX-based charge; at the initial moment ρ A =ρ RDX , ρ B =ρ C =ρ D =0;

[0019] According to the principle of continuous reaction, the reaction kinetics equation can be expressed as:

[0020]

[0021]

[0022]

[0023]

[0024] t is time; Substitute formulas (1), (2), and (3) into formulas (4) to (7), and use the Runge-Kutta method to solve formulas (4) to (7) to obtain r1, r2, and r3;

[0025] During the thermal decomposition of RDX-based charge, the total heat generated by decomposition per unit time is the sum of the heat generated by each step of the reaction:

[0026] S1=Q1r1+Q2r2+Q3r3 (8)

[0027] Q is the heat of reaction;

[0028] The binder absorbs heat when it decomposes. The decomposition process is represented by a single-step reaction model:

[0029]

[0030] The heat generated per unit time during the thermal decomposition process of the binder is:

[0031] S2=ρ Kel-F Q4r4 (10)

[0032] where ρ Kel-F is the density of the binder;

[0033] The mixed thermal decomposition unit method is used for the computational grid. The heat released by decomposition is the sum of the heat absorbed or released by each step of the reaction between the RDX-based charge and the binder. Therefore, the heat generated per unit time during the decomposition of the charge is:

[0034] S ct =αS1+βS2 (11)

[0035] The calculation process of formulas (1)-(11) is written into a UDF in C language, and the UDF is loaded into the calculation model. The temperature rise rate is applied to the calculation model as a boundary condition to perform calculations to obtain the overall cooking response characteristics of the fuze; the local cooking response characteristics of the charge inside the fuze are extracted to obtain the temperature distribution cloud map inside the ammunition during the cooking process, as well as the cooking hotspot coordinates, hotspot temperature, and response time characteristic parameters;

[0036] Step 3: Establish a neural network model. Use the characteristic parameters obtained in step 2 as training samples. Take charge length, charge diameter, heating rate, explosion-proof plate and explosion tube shell thickness as input parameters. Use cooking response time, hotspot coordinates and hotspot temperature as outputs for training. According to the training results, add training samples through numerical calculation until the accuracy requirement is met.

[0037] Step 4: Call the trained neural network model, input the length of the charge, the diameter of the charge, the thickness of the shell, the rate of heating during cooking, predict the cooking response time, hotspot coordinates, and hotspot temperature, draw the two-dimensional plane of the detonating charge and the transfer charge, mark the response points, and mark their temperatures and response times.

[0038] Furthermore, the adhesive is fluororubber Kel-F.

[0039] Furthermore, the heating rates include 3.3 K / h, 1 K / min, 2 K / min, 4 K / min, and 8 K / min.

[0040] Compared with the prior art, the present invention has the following significant advantages:

[0041] (1) The present invention can perform a bake-off simulation on a charge fuze having different ratios of RDX-based charge and binder, and obtain a relatively accurate numerical simulation result, so as to obtain the bake-off characteristics of the charge fuze at a target heating rate;

[0042] (2) The present invention predicts the cooking characteristics of the charge fuze when the size of the charge column, the thickness of the explosion-proof plate shell, the thickness of the booster shell, and the heating rate are changed through a neural network, without the need to remodel the calculation model and waste a large amount of calculation resources in the numerical simulation process;

[0043] (3) The neural network can be used to predict the response characteristic parameters under a large number of heating rates, and the continuity relationship between the heating rate and the response characteristic parameters can be obtained, which can provide corresponding guidance for the subsequent thermal desensitization design of the fuze after burning. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is the structural diagram of the overall fuze physical model.

[0045] Figure 2 It is the temperature cloud diagram of the overall fuze at the response moment under a certain heating rate.

[0046] Figure 3 A neural network diagram with two hidden layers.

[0047] Among them, 1- wind cap, 2- air, 3- potting material, 4- shell, 5- insurance mechanism, 6- detonating charge, 7- explosion-proof plate, 8- detonating tube shell, 9- detonating charge. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0049] The present invention provides a method for predicting and analyzing the cooking characteristics of a charge fuze, and the specific steps are as follows:

[0050] Step 1: Establish a proportional calculation model of the charge fuze used in the cook-off test, as shown in the attached figure. Figure 1As shown, the calculation model includes: a hood 1, air 2, a potting material 3, a shell 4, a safety mechanism 5, a detonating charge 6, a flameproof plate 7, a detonating tube shell 8, and a detonating charge 9. The inside of the hood 1 is air 2, and electronic components and other components not involving safety are equivalent to the potting material 3, and only their heat transfer performance is considered. The shell 4 is connected to the hood 1 and the potting material 3 through threads, and the safety mechanism 5 is fixed to the shell 4 through threaded connection with the flameproof plate 7 and the detonating tube shell 8. The detonating charge 6 is filled between the safety mechanism 5 and the flameproof plate 7, and the detonating charge 9 is filled between the detonating tube shell 8 and the flameproof plate 7; the safety mechanism 5 puts the fuze in a safe state, and the flameproof plate 7 isolates the explosive components of each level in the explosion sequence, so the primer and detonator of the fuze will not cause the primer 6 and the detonator 9 to ignite during the baking process, so the primer and the detonator are ignored in the calculation model; the parts connected by threads in the fuze are simplified to a cylindrical fuze shell.

[0051] The length and diameter of the detonator 6 and the booster 9 in the calculation model, as well as the thickness of the explosion-proof plate 7 and the booster shell 8 can be specifically set according to actual needs;

[0052] Step 2: numerically calculate the established model; the self-heating reaction source term of the charge (priming charge 6 and booster charge 9) is set using the multi-grid method. The charge is composed of RDX-based charge and binder. The thermal conductivity and specific heat capacity are obtained according to the superposition principle. The specific calculation formula is:

[0053] λ Ct =αλ1+βλ2

[0054] C Ct =αC1+βC2

[0055] Where λ is thermal conductivity and C is specific heat capacity. 1 is the index of RDX-based charge, 2 is the index of binder, and Ct is the index of mixed charge; the volume ratio of RDX-based charge in the charge is α, and the volume ratio of binder is β. The multi-step decomposition process of RDX-based charge under heating follows the Arrhenius equation, and its thermal decomposition process is expressed as:

[0056]

[0057]

[0058]

[0059] Where: r is the charge reaction rate, unit mol·L -1 ·s -1 , subscripts 1, 2, 3 are reaction numbers; Z is the pre-exponential factor, unit s -1 ; E is the charge activation energy, unit J·mol -1; R is the universal gas constant, which is 8.314 J·mol -1 ·K -1 ; A is RDX-based charge; B is solid intermediate product; C is gaseous intermediate product; D is gaseous final product; ρ A , B , C , D are the mass concentrations of A, B, C, and D, in kg·m -3 , T is the reaction temperature, unit K; Reaction 1 is a primary chemical reaction of charge endothermic reaction, Reaction 2 is a primary chemical reaction of solid intermediate product into gaseous intermediate product, Reaction 3 is a secondary chemical reaction of gaseous intermediate product further decomposing into final gaseous product;

[0060] where ρ A , B , C , D The representation method is as follows:

[0061]

[0062] ρ i is the mass concentration of the ith product, M i is the mass of the i-th product, M RDX is the mass of RDX-based charge, V is the volume of RDX-based charge, α i is the proportion of the i-th product to the total mass of the RDX-based charge, ρ RDX is the density of RDX-based charge; at the initial moment ρ A =ρ RDX , ρ B =ρ C =ρ D =0;

[0063] According to the principle of continuous reaction, the reaction kinetics equation can be expressed as:

[0064]

[0065]

[0066]

[0067]

[0068] t is time; Substitute formula (1), (2), and (3) into formula (4)-(7), use the Runge-Kutta method to solve formula (4) to formula (7), and finally obtain r1, r2, and r3;

[0069] Therefore, during the thermal decomposition of RDX-based charge, the total heat generated per unit time is the sum of the heat generated by each step of the reaction:

[0070] S1=Q1r1+Q2r2+Q3r3 (8)

[0071] Q is the heat of reaction;

[0072] The binder is fluororubber Kel-F, which absorbs heat when decomposing. The decomposition process can be represented by a single-step reaction model:

[0073]

[0074] The heat generated per unit time during the thermal decomposition process of the binder is:

[0075] S2=ρ Kel-F Q4r4 (10)

[0076] where ρ Kel-F is the density of the binder;

[0077] The mixed thermal decomposition unit method is used for the computational grid. The heat released by decomposition is the sum of the heat absorbed or released by each step of the reaction between the RDX-based charge and the binder. Therefore, the heat generated per unit time during the decomposition of the charge is:

[0078] S ct =αS1+βS2 (11)

[0079] The calculation process of formulas (1)-(11) is written into UDF in C language, and the UDF is loaded into the calculation model. The calculation is performed after applying typical heating rates of 3.3K / h, 1K / min, 2K / min, 4K / min, 8K / min, etc. as boundary conditions on the calculation model. Figure 2 As shown, the overall cooking response characteristics of the fuze are obtained; the local cooking response characteristics of the charge inside the fuze are extracted to obtain the temperature distribution cloud map inside the ammunition during the cooking process and the cooking hotspot coordinates, hotspot temperature, and response time characteristic parameters (for example, the VariableMaximum function in the CFD-POST software can be used to create and export the characteristic parameters);

[0080] Step 3, as attached Figure 3As shown, a neural network model is established (the input layer consists of four nodes, namely, charge length, charge diameter, heating rate, and shell thickness; the output layer contains three nodes, namely, response time, hot spot coordinates, and hot spot temperature). The characteristic parameters obtained by numerical calculation are used as training samples, and the charge length, charge diameter, heating rate, explosion-proof plate and explosion tube shell thickness are used as input parameters, and the baking response time, hot spot coordinates, and hot spot temperature are used as outputs. After training, according to the training results, training samples are added through numerical calculation until the accuracy requirements are met (for example, sample data can be imported into Matlab software and trained to obtain a trained neural network model);

[0081] Step 4: Call the trained neural network model, input the length of the charge column, the diameter of the charge column, the thickness of the shell, and the rate of temperature rise during cooking, and then predict the cooking response time, hotspot coordinates, and hotspot temperature. Draw the two-dimensional plane of the detonating charge column and the transmission charge column, mark the response points, and mark their temperatures and response times.

[0082] The above shows and describes 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 above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for predicting and analyzing the cooking characteristics of a charge fuze, characterized in that: The following steps are involved: Step 1: Establish a proportional calculation model of the charge fuze used in the cooking test, the calculation model comprising a hood (1), air (2), a potting material (3), a shell (4), a safety mechanism (5), a detonating charge (6), a flameproof plate (7), a detonating tube shell (8), and a flameproof charge (9). The hood (1) contains air (2), and the electronic components are equivalent to the potting material (3). The shell (4) is connected to the hood (1) and the potting material (3) through threads, and the safety mechanism (5) is connected to the flameproof plate (7) and the detonating tube shell (8) through threads. The explosive charge (6) is connected and fixed on the housing (4), the explosive charge (6) is filled between the safety mechanism (5) and the explosion-proof plate (7), and the booster charge (9) is filled between the booster tube shell (8) and the explosion-proof plate (7); the safety mechanism (5) puts the fuze in a safe state, the explosion-proof plate (7) isolates the explosive elements of each level in the explosion sequence, the explosive charge (6) and the booster charge (9) form a charge, the charge is composed of an RDX-based charge and a binder, and the length and diameter of the charge and the thickness of the explosion-proof plate (7) and the booster tube shell (8) are specifically set according to actual needs; Step 2: numerically calculate the established model; the self-heating reaction source term of the charge is set using the multi-grid method, and the thermal conductivity and specific heat capacity are obtained according to the superposition principle. The specific calculation formula is: l Ct =al1+bl2 C Ct =αC1+βC2 Where λ is thermal conductivity, C is specific heat capacity, 1 is the index of RDX-based charge, 2 is the index of binder, and Ct is the index of mixed charge; in the charge, the volume ratio of RDX-based charge is α, and the volume ratio of binder is β. The multi-step decomposition process of RDX-based charge under heating follows the Arrhenius equation, and its thermal decomposition process is expressed as: Where: r is the charge reaction rate, unit mol·L -1 ·s -1 , subscripts 1, 2, 3 are reaction numbers; reaction 1 is a primary chemical reaction of charge endothermic reaction, reaction 2 is a primary chemical reaction of solid intermediate product into gaseous intermediate product, reaction 3 is a secondary chemical reaction of gaseous intermediate product further decomposition into final gaseous product; Z is pre-exponential factor, unit s -1 ; E is the charge activation energy, unit J·mol -1 ; R is the universal gas constant, which is 8.314 J·mol -1 ·K -1 ; A is RDX-based charge; B is solid intermediate product; C is gaseous intermediate product; D is gaseous final product; ρ A , B , C , D are the mass concentrations of A, B, C, and D, in kg·m -3 , T is the reaction temperature, unit K; where ρ A , B , C , D The representation method is as follows: ρ i is the mass concentration of the ith product, M i is the mass of the i-th product, M RDX is the mass of RDX-based charge, V is the volume of RDX-based charge, α i is the proportion of the i-th product to the total mass of the RDX-based charge, ρ RDX is the density of RDX-based charge; at the initial moment ρ A =ρ RDX , ρ B =ρ C =ρ D =0; According to the principle of continuous reaction, the reaction kinetics equation can be expressed as: t is time; Substitute formulas (1), (2), and (3) into formulas (4) to (7), and use the Runge-Kutta method to solve formulas (4) to (7) to obtain r1, r2, and r3; During the thermal decomposition of RDX-based charge, the total heat generated by decomposition per unit time is the sum of the heat generated by each step of the reaction: S1=Q1r1+Q2r2+Q3r3 (8) Q is the heat of reaction; The binder absorbs heat when it decomposes. The decomposition process is represented by a single-step reaction model: The heat generated per unit time during the thermal decomposition process of the binder is: S2=ρ Kel-F Q4r4 (10) where ρ Kel-F is the density of the binder; The mixed thermal decomposition unit method is used for the computational grid. The heat released by decomposition is the sum of the heat absorbed or released by each step of the reaction between the RDX-based charge and the binder. Therefore, the heat generated per unit time during the decomposition of the charge is: S ct =αS1+βS2 (11) The calculation process of formulas (1)-(11) is written into a UDF in C language, and the UDF is loaded into the calculation model. The temperature rise rate is applied to the calculation model as a boundary condition to perform calculations to obtain the overall cooking response characteristics of the fuze; the local cooking response characteristics of the charge inside the fuze are extracted to obtain the temperature distribution cloud map inside the ammunition during the cooking process, as well as the cooking hotspot coordinates, hotspot temperature, and response time characteristic parameters; Step 3: Establish a neural network model, use the characteristic parameters obtained in step 2 as training samples, use charge length, charge diameter, heating rate, explosion-proof plate and explosion tube shell thickness as input parameters, and use cooking response time, hotspot coordinates, and hotspot temperature as outputs for training. According to the training results, add training samples through numerical calculation until the accuracy requirements are met; Step 4: Call the trained neural network model, input the length of the charge, the diameter of the charge, the thickness of the shell, the rate of heating during cooking, predict the cooking response time, hotspot coordinates, and hotspot temperature, draw the two-dimensional plane of the detonating charge and the transfer charge, mark the response points, and mark their temperatures and response times.

2. The method for predicting and analyzing the cooking characteristics of a charge fuze according to claim 1, characterized in that: The adhesive is fluororubber Kel-F.

3. The method for predicting and analyzing the cooking characteristics of a charge fuze according to claim 1, characterized in that: The heating rates include 3.3K / h, 1K / min, 2K / min, 4K / min, and 8K / min.

Images