Two-stage linear radiation surface space atmospheric density sensor heating cover

Abstract

The application provides a two-section linear radiation surface space atmospheric density sensor heating cover, the top middle part of which is provided with a circular opening for the entry and exit of space atmospheric molecules; the inner radiation surface is divided into two sections, the generatrix of each section of the radiation surface is a straight line with a certain slope, and the slope of the generatrix of the first section of the radiation surface is smaller than that of the second section; the bottom of the heating cover is provided with a mounting seat in the circumferential direction for being fixed with a spacecraft body. The generatrix of each section of the radiation surface of the heating cover is designed according to a self-guided online machine learning method, the horizontal span and height of each section of the generatrix are selected as factor variables, and the heating time is selected as a target variable, and a heuristic algorithm is combined with a deep neural network to obtain the optimal variable design value. The two-section linear radiation surface space atmospheric density sensor heating cover provided by the application can enhance the radiation heat transfer characteristics between the sensor and the front quartz crystal vibration balance, reduce the heating time required for the desorption of space atmospheric molecules, and thus reduce the power consumption of the spacecraft in orbit flight.

Description

Technical Field

[0001] This invention belongs to the field of space atmospheric density measurement technology, specifically relating to a heating cover for a two-section linear radiating surface space atmospheric density sensor. Background Technology

[0002] Space atmospheric density sensors are primarily used to measure and assess the surface density of rarefied gases in space, aiding in the analysis and calculation of drag experienced by spacecraft during orbital flight. The sensor heats a heating shroud via electrical heating, and the radiating surface of the shroud heats a quartz crystal microbalance (QCM) through radiative heat transfer, facilitating the thermal desorption of atmospheric molecules. The surface density of the space atmosphere is then calculated by measuring the frequency drift of the crystal oscillator. The radiative heat transfer characteristics between the radiating surface and the QCM are directly related to the configuration of the radiating surface. By optimizing the structure of the radiating surface, the radiative heat transfer per unit time can be effectively enhanced, allowing the QCM surface temperature to reach the target value more quickly. This effectively reduces the electrical heating required for the space atmospheric density sensor, thus meeting the low power consumption requirements of spacecraft during orbital operation. Currently, most space atmospheric density sensor heating shrouds use a single-segment linear radiating surface, which cannot maximize the radiative heat transfer per unit time on the QCM surface, thus preventing further reduction in heating time and lowering the power consumption of the spacecraft during orbital operation.

[0003] Current front-end components of space atmospheric density sensors include a thin-film heating element, a heating shroud, a QCM crystal coating, and a quartz plate. The generatrix of the original heating shroud's radiating surface is a straight line with a certain slope. The heating shroud of the space atmospheric density sensor, using a single-segment linear radiating surface, failed to maximize the radiative heat transfer per unit time to the QCM surface, thus preventing further reduction of heating time and lowering of spacecraft on-orbit power consumption. Summary of the Invention

[0004] The purpose of this invention is to provide a two-section linear radiating surface space atmospheric density sensor heating cover that can effectively enhance the radiative heat transfer capability between the heating cover and the QCM, reduce the heating time required for the desorption of atmospheric molecules from the crystal oscillator surface, and reduce the power consumption of the spacecraft during on-orbit operation.

[0005] The present invention discloses a heating cover for a two-section linear radiating surface space atmospheric density sensor, comprising a housing, an atmospheric molecule inlet at the top of the housing, and a mounting base at the bottom of the housing for mounting and fixing the housing. The outer surface of the housing is in contact with an electric heating element. The atmospheric molecule inlet is the first radiating surface, and the lower end of the atmospheric molecule inlet to the bottom of the housing is the second radiating surface. The generatrices of the first and second radiating surfaces are both straight lines with slopes, and the slope of the generatrices of the second radiating surface is greater than that of the first radiating surface.

[0006] The slope of the generatrix of the first radiating surface is 1.67 to 1.81, and the slope of the generatrix of the second radiating surface is 17.2 to 18.6.

[0007] The method for designing the slope values ​​of the generatrix of the first and second radiating surfaces includes the following steps:

[0008] Step 1: Define factor variables and constraints, and randomly generate the initial sample space using the central composite design method;

[0009] Step 2: Establish a numerical model of heat transfer from a space atmospheric density sensor based on the sample space, and calculate the corresponding actual heating time; then, based on a self-guided machine learning optimization method, obtain an implicit function according to the variables and heating time in the sample space; establish a deep neural network model (DNN) based on the variable values ​​and heating time, and then train it based on the implicit function to obtain an objective function based on the minimum heating time; and then search for new variables and predict the corresponding heating time based on the objective function.

[0010] Step 3: Repeat step 2. When the relative error between the predicted heating time and the actual heating time is less than 5%, the convergence requirement is met. Then stop the calculation and extract the minimum heating time and corresponding variable value predicted by the neural network model. Calculate the actual heating time and the values ​​of the busbar and busbar slope.

[0011] The factor variables in step 1 are H1, H2, L1, and L2. H1 is the vertical height of the generatrix of the first radiating surface, L1 is the horizontal span of the generatrix of the first radiating surface, H2 is the vertical height of the generatrix of the second radiating surface, and L2 is the horizontal span of the generatrix of the second radiating surface.

[0012] The factor variables H1, H2, L1, L2 in step 1 follow the following constraints:

[0013] H1 + H2 = 19.5 (1)

[0014] L1 + L2 = 3.4 (2)

[0015] The implicit function mentioned in step 2 is τ = F(H1,H2,L1,L2), where τ is the heating time; the objective function is τ = argminf(H1',H2',L1',L2'), where H1',H2',L1',L2' are new variable values ​​obtained by training the neural network.

[0016] The slope values ​​of the first radiation surface (6) generatrix (3) and the second radiation surface (7) generatrix (5) in step 3 are calculated as follows:

[0017]

[0018]

[0019] Wherein, i1 is the slope value of the first segment of the radiating surface generatrix, and i2 is the slope value of the second segment of the radiating surface generatrix.

[0020] The present invention provides a two-section linear radiating surface space atmospheric density sensor heating hood, which has at least the following advantages: by changing the geometric configuration of the radiating surface of the heating hood, the product of the inner radiating surface area and the radiative heat transfer angle coefficient between the inner radiating surface and the QCM is increased, thereby enhancing the radiative heat transfer to the QCM surface per unit time. Compared with the original atmospheric density sensor heating hood, the heating time is reduced by approximately 15%. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall structure of the present invention.

[0022] Figure 2 This is a design flowchart as described in the example of the present invention.

[0023] Figure 3 This is the analysis flowchart described in the example of the present invention. Detailed Implementation

[0024] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0025] See Figure 1 The present invention includes a housing 4 and an atmospheric molecule inlet 2 opened at the top 1 of the housing 4, and a mounting base 8 located at the bottom of the housing 4 for mounting and fixing the housing 4; the outer surface of the housing 4 is in contact with the electric heating element, the atmospheric molecule inlet 2 is a first radiating surface 6, and the lower end of the atmospheric molecule inlet 2 to the bottom of the housing is a second radiating surface 7, the generatrix 3 of the first radiating surface 6 and the generatrix 5 of the second radiating surface 7 are both straight lines with slope, and the slope of the generatrix 5 of the second radiating surface 7 is greater than the slope of the generatrix 3 of the first radiating surface 6.

[0026] The design range for the slope of the busbar 3 of the first radiating surface 6 is 1.67 to 1.81, and the design range for the slope of the busbar 5 of the second radiating surface 7 is 17.2 to 18.6.

[0027] This embodiment uses the example of a slope of 1.71 for the busbar 3 of the first radiating surface 6 and a slope of 17.44 for the busbar 5 of the second radiating surface 7 for illustrative purposes.

[0028] like Figure 2 As shown, the design of the optimal slope value of the generatrix 3 of the first radiating surface 6 and the generatrix 5 of the second radiating surface 7 of a two-section linear radiating surface space atmospheric density sensor heating cover is carried out according to the following steps:

[0029] Step 1: Define factor variables and constraints, and randomly generate the initial sample space using the central composite design method;

[0030] Step 2: Establish a numerical model of heat transfer from a space atmospheric density sensor based on the sample space, and calculate the corresponding actual heating time; then, based on a self-guided machine learning optimization method, obtain an implicit function according to the variables and heating time in the sample space; establish a deep neural network (DNN) model based on the variable values ​​and heating time, and then train it based on the implicit function to obtain an objective function based on the minimum heating time; and then search for new variables and predict the corresponding heating time based on the objective function.

[0031] Step 3: Repeat Step 2. When the relative error between the predicted heating time and the actual heating time is less than 5%, the convergence requirement is met. Then stop the calculation and extract the minimum heating time predicted by the neural network model and the corresponding variable value; calculate the actual heating time.

[0032] Further, please refer to Figure 3 :

[0033] Step 1: Randomly generate m0 sets of H1, L1, H2, L2 values ​​as the sample space using the central design method. In this invention, m0 is 10. The constraints are H1+H2=19.5 and L1+L2=3.4.

[0034] Step 2: Establish a numerical model of heat transfer for the space atmospheric density sensor based on the generated variable sample space, calculate the corresponding actual heating time τ, and obtain the objective function τ0=F(H1,L1,H2,L2) through a heuristic algorithm;

[0035] Step 3: Based on the sample space in Step 1 and the target value in Step 2, establish a self-guided online machine learning DNN network model, and obtain the prediction objective function τ1=f(H1,L1,H2,L2) between the target and factor variables through training;

[0036] Step 4: Based on the function argmin f(H1,L1,H2,L2), solve for the new variable values ​​H1',L1',H2',L2', and calculate the actual heating time τ according to Step 2. If If the convergence requirement is met, output the corresponding H1, L1, H2, L2, and τ. If the convergence requirement is not met, compare whether the current H1', ​​L1', H2', L2' are equal to the existing variable values ​​in the sample space. If they are, add them along with the corresponding τ value as new samples to the DNN network model for continued training. If not, add perturbations at H1', ​​L1', H2', L2', calculate the τ value according to step 2, and add it to the DNN network model until the predicted value of τ1 meets the convergence requirement.

[0037] Step 5: Calculate the actual τ value according to Step 2 based on the final output H1', ​​L1', H2', L2', and calculate the slope value of each segment of the radiation surface generatrix accordingly.

[0038] In this embodiment, the heating time required for the optimized two-section linear atmospheric density sensor heating hood is reduced by approximately 15% compared to the original structure.

Claims

1. A heating cover for a two-section linear radiating surface space atmospheric density sensor, characterized in that, Includes a shell (4) and an atmospheric molecule inlet (2) opened at the top (1) of the shell (4) and a mounting base (8) located at the bottom of the shell (4) for mounting and fixing the shell (4); the outer surface of the shell (4) is in contact with the electric heating element, the atmospheric molecule inlet (2) is a first radiation surface (6), the lower end of the atmospheric molecule inlet (2) to the bottom of the shell is a second radiation surface (7), the generatrix (3) of the first radiation surface (6) and the generatrix (5) of the second radiation surface (7) are both straight lines with slope, and the slope of the generatrix (5) of the second radiation surface (7) is greater than the slope of the generatrix (3) of the first radiation surface (6); The slope design method for the generatrix (3) of the first radiating surface (6) and the generatrix (5) of the second radiating surface (7) includes the following steps: Step 1: Define factor variables and constraints, and randomly generate the initial sample space using the central composite design method; Step 2: Establish a numerical model of heat transfer from a space atmospheric density sensor based on the sample space, and calculate the corresponding actual heating time; then, based on a self-guided machine learning optimization method, obtain an implicit function according to the variables and heating time in the sample space; establish a deep neural network model (DNN) based on the variable values ​​and heating time, and then train it based on the implicit function to obtain an objective function based on the minimum heating time; and then search for new variables and predict the corresponding heating time based on the objective function. Step 3, repeat step 2. When the relative error between the predicted heating time and the actual heating time is less than 5%, the convergence requirement is met. Then stop the calculation and extract the minimum heating time and corresponding variable value predicted by the neural network model. Calculate the actual heating time and the slope values ​​of the generatrix (3) of the first radiation surface (6) and the generatrix (5) of the second radiation surface (7).

2. The heating shroud for the two-section linear radiating surface space atmospheric density sensor according to claim 1, characterized in that, The slope of the first radiation surface generatrix (3) is 1.67 to 1.81, and the slope of the second radiation surface generatrix (5) is 17.2 to 18.

6.

3. The heating shroud for the two-section linear radiating surface space atmospheric density sensor according to claim 1, characterized in that, The factor variables in step 1 are: H 1, H 2, L 1, L 2, the aforementioned H 1 represents the vertical height of the generatrix (3) of the first radiating surface (6). L 1 represents the horizontal span of the generatrix (3) of the first radiating surface (6). H 2 is the vertical height of the generatrix (5) of the second radiating surface (7), the aforementioned L 2 is the horizontal span of the generatrix (5) of the second radiation surface (7).

4. The heating shroud for the two-section linear radiating surface space atmospheric density sensor according to claim 1, characterized in that, Factor variables in step 1 H 1, H 2, L 1, L 2. The following constraints must be met: H 1+ H 2=19.5（1） L 1+ L 2=3.4（2）。 5. The heating shroud for the two-section linear radiating surface space atmospheric density sensor according to claim 1, characterized in that, The implicit function mentioned in step 2 is τ = F ( H 1, H 2, L 1, L 2), τ Heating time; objective function τ = argmin f ( H 1', H 2', L 1', L 2'), H 1', H 2', L 1', L 2' New variable values ​​obtained from neural network training.

6. The heating shroud for the two-section linear radiating surface space atmospheric density sensor according to claim 1, characterized in that, The slope values ​​of the first radiation surface (6) generatrix (3) and the second radiation surface (7) generatrix (5) in step 3 are calculated as follows: （3） （4） Among them, the i 1 represents the slope value of the generatrix (3) of the first segment of the radiating surface (6). i 2 is the slope value of the second radiation surface (7) generatrix (5).

Images