Iterative optimization method for power transmission performance of radome
By iteratively optimizing the radome wall thickness using a radial basis function neural network surrogate model, the problem of high computation time in existing technologies is solved, and the globally optimal design of radome power transmission performance is achieved, meeting the performance requirements in complex electromagnetic environments.
Patent Information
- Application Number
- CN202410981892.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-26
AI Technical Summary
Existing technologies struggle to efficiently optimize the power transmission performance of radomes, especially in terms of computationally expensive global optimal designs, which fail to meet design requirements in complex electromagnetic environments.
Iterative optimization is performed using a radial basis function neural network surrogate model (RBFNN). By simulating the mapping relationship between the input wall thickness and the power transmission performance of the radome, the neural network parameters are updated using Latin hypercube sampling and electromagnetic full-wave algorithm to gradually approach the optimal wall thickness design.
It achieves optimal radome power transmission performance under optimal wall thickness, reduces computation time cost, and improves the global optimization efficiency of the design.
Smart Images

Figure CN119026455A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of design optimization methods for antenna cover power transmission performance, and in particular to an iterative optimization method for antenna cover power transmission performance. Background Art
[0002] The radome is an important device for protecting the radar antenna. It is located at the front end of the system and needs to meet not only the appearance requirements but also the power transmission performance requirements of the radar system. As the electromagnetic environment becomes increasingly complex, the radar system places higher design requirements on the power transmission performance of the radome.
[0003] The design of the power transmission performance of the traditional radome relies on a large number of manual repetitive simulations and rule induction, and the direction of further optimization is determined based on experience, but it is difficult to achieve the global optimum. In the prior art, the advantage of using genetic algorithms to design the power transmission performance of the radome is that it can achieve the global optimum, but the disadvantage is that the objective function needs to be calculated multiple times, and the calculation time cost of the power transmission performance of the radome is very high, which is unacceptable for the optimization problem of the power transmission performance of the radome.
[0004] Therefore, as the requirements for the power transmission performance of the radome become higher and higher, higher requirements are also put forward for the design optimization of the radome. The present invention aims to provide an iterative optimization method for the power transmission performance of the radome, thereby obtaining the optimal power transmission performance data of the radome under the optimal wall thickness condition.
[0005] It should be understood that the above statements merely provide background technology related to the present invention and do not necessarily constitute prior art. Summary of the invention
[0006] The object of the present invention is to provide an iterative optimization method for the power transmission performance of a radome, thereby obtaining optimal power transmission performance data of the radome under the condition of optimal wall thickness.
[0007] In order to achieve the above object, the present invention provides a method for iteratively optimizing the power transmission performance of a radome, which specifically comprises the following steps:
[0008] S1. Set initial parameters, including outline, operating frequency, dielectric constant, sample set S, and radome wall thickness x k The initial value and range of the radial basis function neural network proxy model RBFNN(x k ,θ k ), where θ k are the radial basis function neural network proxy model parameters, and the subscript k represents the number of iterations.
[0009] S2. Sample the initial wall thickness data x within the range of the radome wall thickness.k , and calculate the wall thickness x k Corresponding radome power transmission performance data y k .
[0010] S3, {x k ,y k} is added to the sample set S.
[0011] S4, according to the sample set S, update the radial basis function neural network proxy model RBFNN (x k ,θ k ) parameters, and get θ k+1 .
[0012] S5, using the constrained optimization method, in x k Within the range of values, let the parameter θ of the radial basis function neural network proxy model RBFNN be k+1 Unchanged, search for the current radial basis function neural network proxy model RBFNN(x k ,θ k+1 ) Potential optimal solution of wall thickness x k+1 .
[0013] S6. Calculation and wall thickness x k+1 Corresponding radome power transmission performance data y k+1 .
[0014] S7, calculate the radial basis function neural network proxy model RBFNN (x k+1 ,θ k+1 )’s output value y*.
[0015] S8. Calculate y* and y k+1 The difference is taken as the error value Err of the radial basis function neural network proxy model RBFNN.
[0016] S9, determine whether the error value Err is less than the target error value eps; if the error value Err ≤ the target error value eps, then output x k+1 As the optimal wall thickness design value, the operation ends; if the error value Err> target error value eps, the data is updated so that x k =x k+1 ,y k =y k+1 ,θ k =θ k+1 , k=k+1, and go to S3 to recalculate.
[0017] Preferably, in S1, the number of iterations k is initially set to 0.
[0018] Preferably, in S2, the method for sampling the initial wall thickness data is the Latin hypercube method.
[0019] Preferably, in S2 and S6, the method for calculating the antenna cover power transmission performance data corresponding to the wall thickness is an electromagnetic full-wave algorithm.
[0020] Preferably, in S4, the algorithm for updating the parameters of the radial basis function neural network proxy model is a regression algorithm.
[0021] Preferably, in S9, the error value eps is in the range of 0.1 to 1.
[0022] In summary, compared with the prior art, the present invention provides an iterative optimization method for the power transmission performance of a radome, which utilizes the approximation ability of a radial basis function neural network to complex nonlinear functions, and replaces the full-wave calculation method by simulating the mapping relationship between the input wall thickness and the power transmission performance of the radome, and continuously updates the radial basis function neural network parameters during the iterative optimization process, thereby obtaining the optimal radome power transmission performance data under the optimal wall thickness condition. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 is a flow chart of the method of the present invention;
[0024] Figure 2 is an outline diagram of an exemplary radome of the present invention;
[0025] FIG3(a) is a graph showing an example radome in the present invention, where the power transmission performance of the radome varies with the beam angle in azimuth after the radome is optimized and designed by the method of the present invention; FIG3(b) is a graph showing an example radome in the present invention, where the power transmission performance of the radome varies with the beam angle in elevation after the radome is optimized and designed by the method of the present invention;
[0026] Figure 4 The following is an example of the optimal wall thickness design result of the radome after the radome is optimized by the method described in the present invention. DETAILED DESCRIPTION
[0027] The following is combined with Figures 1 to 4 , the present invention is further explained by describing a preferred specific embodiment in detail.
[0028] It should be noted that the drawings are in a very simplified form and use non-precise proportions. They are only used to conveniently and clearly assist in explaining the embodiments of the present invention, and are not used to limit the conditions for the implementation of the present invention. Therefore, they have no substantive technical significance. Any structural modification, change in proportional relationship or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical content disclosed by the present invention.
[0029] It should be noted that, in the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only the elements explicitly listed, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0030] like Figure 1 As shown, the present invention provides an iterative optimization method for the power transmission performance of a radome. When the outer surface, material properties and operating frequency of the radome are known, a radial basis function neural network proxy model is used to design the wall thickness of the radome so that the radome has the best performance under the wall thickness. The method specifically comprises the following steps:
[0031] S1. Set initial parameters, including radome outline, operating frequency, dielectric constant, sample set S, radome wall thickness x k The initial value and range of the radial basis function neural network proxy model RBFNN(x k ,θ k ), where x k is the wall thickness of the radome, θ k is the radial basis function neural network proxy model parameter, the subscript k represents the number of iterations, and is initially set to 0;
[0032] S2, the wall thickness of the radome x k In the range of values, the Latin hypercube method is used to sample the initial wall thickness data x k and calculated by electromagnetic full-wave algorithm with wall thickness x k Corresponding radome power transmission performance data y k ;
[0033] S3, {x k ,y k}Add to sample set S;
[0034] S4, based on the sample set S, use the regression algorithm to update the radial basis function neural network proxy model RBFNN (x k ,θ k ) parameters, and get θ k+1 ;
[0035] S5, using the constrained optimization method, in x k Within the range of values, let the parameter θ of the radial basis function neural network proxy model RBFNN be k+1Unchanged, search for the current radial basis function neural network proxy model RBFNN(x k ,θ k+1 ) Potential optimal solution of wall thickness x k+1 ;
[0036] S6, using electromagnetic full wave algorithm to calculate the wall thickness x k+1 Corresponding radome power transmission performance data y k+1 ;
[0037] S7, calculate the radial basis function neural network proxy model RBFNN (x k+1 ,θ k+1 ) output value y*; where y* is the output value y* of RBFNN(x k+1 ,θ k+1 ) when the wall thickness is x k+1 , parameter θ k+1 The calculated value under
[0038] S8. Calculate y* and y k+1 The difference is used as the error value Err of the radial basis function neural network proxy model RBFNN;
[0039] S9, determine whether the error value Err is less than the target error value eps, where eps ranges from 0.1 to 1; if Err is less than or equal to eps, output x k+1 As the optimal wall thickness design value, the operation ends; if Err is greater than eps, the data is updated so that x k =x k+1 ,y k =y k+1 ,θ k =θ k+1 , k=k+1, and go to S3 to recalculate.
[0040] It can be understood that the target error value eps is set according to actual conditions.
[0041] Further, such as Figure 2 As shown in the figure, the example radome shape diagram of the iterative optimization method of the present invention, in this example, the radome shape line equation is (r+B) 2 +p 2 =R 2 ,in L0 is the length of the radome, 500 mm, and D0 is the bottom diameter of the radome, 300 mm.
[0042] Further, as shown in Figures 3(a) and 3(b), the power transmission performance of the radome in azimuth and elevation varies with the beam angle after the optimization design by the method of the present invention. It can be seen from the figure that the minimum value of the power transmission performance of the radome in azimuth is 92.3%, and the minimum value in elevation is 92.0%.
[0043] Further, such as Figure 4 As shown in FIG. 1 , the optimal wall thickness design result of the radome after the optimization design by the method of the present invention is shown. As can be seen from the figure, the wall thickness x of the radome changes with the p coordinate value, with a maximum value of 5.84 mm and a minimum value of 5.06 mm.
[0044] In summary, the present invention provides an iterative optimization method for the power transmission performance of a radome. The method utilizes the approximation capability of a radial basis function neural network to complex nonlinear functions, and replaces the full-wave calculation method by simulating the mapping relationship between the input wall thickness and the power transmission performance of the radome. The radial basis function neural network parameters are continuously updated during the iterative optimization process, thereby obtaining the optimal power transmission performance data of the radome under the optimal wall thickness condition.
[0045] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be appreciated that the above description should not be considered as a limitation of the present invention. After reading the above content, it will be apparent to those skilled in the art that various modifications and substitutions of the present invention will occur. Therefore, the protection scope of the present invention should be limited by the appended claims.
Claims
1. A method for iterative optimization of radome power transmission performance, characterized in that: The specific steps include: S1. Set initial parameters, including outline, operating frequency, dielectric constant, sample set S, and radome wall thickness x k The initial value and range of the radial basis function neural network proxy model RBFNN(x k ,θ k ), where θ k is the radial basis function neural network proxy model parameter, and the subscript k represents the number of iterations; S2. Sample the initial wall thickness data x within the range of the radome wall thickness. k , and calculate the wall thickness x k Corresponding radome power transmission performance data y k ; S3, {x k ,y k }Add to sample set S; S4, according to the sample set S, update the radial basis function neural network proxy model RBFNN (x k ,θ k ) parameters, and get θ k+1 ; S5, using the constrained optimization method, in x k Within the range of values, let the parameter θ of the radial basis function neural network proxy model RBFNN be k+1 Unchanged, search for the current radial basis function neural network proxy model RBFNN(x k ,θ k+1 ) Potential optimal solution of wall thickness x k+1 ; S6. Calculation and wall thickness x k+1 Corresponding radome power transmission performance data y k+1 ; S7, calculate the radial basis function neural network proxy model RBFNN (x k+1 ,θ k+1 )’s output value y*; S8. Calculate y* and y k+1 The difference is used as the error value Err of the radial basis function neural network proxy model RBFNN; S9, determine whether the error value Err is less than the target error value eps; if the error value Err ≤ the target error value eps, then output x k+1 As the optimal wall thickness design value, the operation ends; if the error value Err> target error value eps, the data is updated so that x k =x k+1 ,y k =y k+1 ,θ k =θ k+1 , k=k+1, and go to S3 to recalculate.
2. The method for iterative optimization of radome power transmission performance according to claim 1, characterized in that: In S1, the number of iterations k is initially set to 0.
3. The iterative optimization method for the power transmission performance of a radome according to claim 1, characterized in that: In the above-mentioned S2, the method for sampling the initial wall thickness data is the Latin hypercube method.
4. The iterative optimization method for the power transmission performance of a radome according to claim 1, characterized in that: In the above-mentioned S2 and S6, the method for calculating the power transmission performance data of the radome corresponding to the wall thickness is an electromagnetic full-wave algorithm.
5. The method for iterative optimization of radome power transmission performance according to claim 1, characterized in that: In the above S4, the algorithm for updating the parameters of the radial basis function neural network proxy model is a regression algorithm.
6. The method for iterative optimization of radome power transmission performance according to claim 1, characterized in that: In the above S9, the error value eps ranges from 0.1 to 1.
