Aircraft infrared detectability calculation method

Abstract

The invention belongs to the technical field of aircraft infrared stealth and infrared detectability analysis, and particularly relates to an aircraft infrared detectability calculation method. An iterative calculation method is replaced by a sectioning method with a higher convergence order to realize acceleration of an iterative process, and further, a Bayesian optimization process is introduced into a detection distance iterative calculation logic, so that the problem that the detection distance is more accurate under the condition that the target intrinsic radiation intensity is lower is effectively solved. And the problem that the detection distance cannot be solved after background radiation and path radiation correction is introduced is solved. A target infrared detection distance calculation program developed by the method can call the MODTRAN in real time in an iteration process to calculate atmospheric transmissivity, background radiation and path radiation in different path directions and under different observation distances, and compared with a mode of establishing a corresponding atmospheric transmissivity, background radiation and path radiation database according to the MODTRAN in advance, the method has the advantages that the calculation efficiency is improved, and the calculation cost is reduced. The method has higher flexibility and wider applicability while ensuring the calculation precision, and the consumed time of iterative calculation is within an acceptable range.

Description

Technical Field

[0001] The invention belongs to the technical field of aircraft infrared stealth and infrared detectability analysis, and specifically is a method for calculating the infrared detectability of an aircraft. Background Art

[0002] Reliable assessment of aircraft infrared detectability is of great significance for accurately grasping the infrared radiation characteristic level of aircraft and evaluating the effectiveness of infrared characteristic suppression measures. Usually, the infrared range of point source targets is an indicator used to characterize the performance of infrared detection systems, but the detection envelope composed of ranges in different detection directions is also often used to evaluate the infrared detectability and sensitivity of flying targets.

[0003] The infrared action distance of a point source target is related to many factors such as target radiation characteristics, atmospheric transmittance, detector performance parameters, background and path radiation. The currently commonly used point source target action distance model is a correction model that introduces the influence of atmospheric transmittance, background and path radiation on the basis of the RD Little Hudson model. The atmospheric transmittance, background radiation and path radiation at different distances are calculated by calling the atmospheric radiation transmission professional software MODTRAN developed by the US Air Force Geophysical Laboratory. Since the infrared action distance model of a point source target is an implicit function with distance as the independent variable, it is impossible to obtain an analytical solution and can only be solved by iteration. Patent CN116127252A discloses a "method and system for calculating infrared action distance based on dichotomy", constructs a detection distance iteration format under a square format, and uses dichotomy to iteratively calculate the final action distance R. Although the patent significantly shortens the iterative calculation time by using the dichotomy method, the detection distance model it uses does not take into account the influence of background and path radiation, and the method of simplifying the detection distance calculation process by fitting the relationship between atmospheric transmittance and observation distance will also introduce additional errors. Patent CN101320475A discloses a "method for estimating the range of infrared imaging systems under complex background conditions". By constructing a new equation, the range equation curve, the relationship curve between atmospheric transmittance and observation distance are used to obtain the range through a graphical method. However, this method is only applicable to solving simpler range models, and the patent does not consider the impact of background and path radiation on the detection distance. Patent CN107368617A discloses a "method for calculating the range of ground-to-air detection infrared imaging systems based on Lowtran7 atmospheric software", which considers the impact of background and path radiation on the detection distance solution, and then ensures that all solutions can be calculated by traversing all paths within the range of 0 to 100km. However, this calculation method is inefficient. Although the time consumption caused by real-time calling Lowtran calculations is avoided and the calculation efficiency is improved by constructing the atmospheric transmittance, background radiation and path radiation databases in advance, it also limits its scope of application and lacks flexibility when facing new environmental conditions. The article "The Influence of Atmospheric Background on Target Detection", published in "Infrared and Laser Engineering, 2014, 43(2): 438-441", points out that as the observation distance increases, the difference between the path radiation and the background radiation decreases, which has a non-negligible impact on the solution of the detection distance. Specifically, when the target intrinsic radiation characteristics are weak and the observation distance is close, the difference between the path radiation and the background radiation is large, which may cause the sum of the difference between the two and the target intrinsic radiation intensity to be negative, further leading to errors in the solution based on the Little Hudson and corresponding modified models. However, there is currently no public literature that discusses this problem in detail and proposes a solution.

[0004] The existing solutions to the infrared range model of point source targets mostly use the dichotomy method, curve intersection method or variable step-size approximation method. Among them, the dichotomy method and variable step-size step-by-step approximation method require a large number of iterations to obtain the calculation results. In addition, the solution objects of the above methods are mostly the simpler Little Hudson model or its deformation form without considering the background and path radiation effects. For the solution of the range model that introduces the background and path radiation effects, in order to reduce the iteration time, the atmospheric transmittance, background radiation and path radiation under different detection directions and distances are usually calculated in advance and a database is established. In the actual calculation, the values ​​in the database are interpolated to obtain the required results, but this method will lead to an increase in the calculation error. In addition, when the intrinsic radiation characteristics of the target are weak and the observation distance is close, the difference between the path radiation and the background radiation is large, which may cause the detection distance to be unable to be solved, which is not solved in the existing methods. Summary of the invention

[0005] The purpose of the present invention is to avoid the shortcomings of the prior art and provide a method for calculating the infrared detectability of an aircraft, which solves the point source target detection distance model proposed by Little Hudson and various improved models proposed based on this by coupling the chord cut method and the Bayesian optimization algorithm, and develops a corresponding target infrared detection distance calculation method based on the coupled solution method.

[0006] To achieve the above object, the technical solution adopted by the present invention is a method for calculating the infrared detectability of an aircraft, comprising the following steps:

[0007] Step 1: Obtain the detection direction to be solved, the detection band range to be solved, the target projection area under each detection direction and the integrated infrared radiation intensity of the target within the detection band range to be solved;

[0008] Among them, the detection direction to be solved is calculated based on the target and the target tail direction is 0°, which is written by the user into a text file. When the program is running, it reads and obtains the detection direction setting from the text file; the detection band range to be solved means that the user needs to select the band range to be calculated from the commonly used infrared detection band range, or input a user-defined band that does not exceed the above band range; the integrated infrared radiation intensity of the target in the detection band set by the user and the projection area of ​​the set detection direction to be solved are obtained by the calculation program according to the target integrated radiation brightness distribution text file provided by the user;

[0009] Step 2: Obtain detection system parameters, including:

[0010] Optical system parameters: F number of optical components of the detection system, average spectral transmittance τ opt , entrance pupil diameter D opt and the target type used in calculating sampling efficiency when taking into account the effects of diffraction;

[0011] Among them, the F number of the optical components of the detection system, the average spectral transmittance τ opt and entrance pupil diameter D opt The user writes a text file according to his / her needs, and the program reads and obtains the above parameters from the text file when it is running; the user specifies in the text file whether to consider the influence of the diffraction effect. If the diffraction effect is considered, the user needs to choose to regard the target as an ideal point source target or a finite size target, and choose the target image center at the pixel center, corner point or average position, all of which are written by the user in the text file, and the program reads and obtains the specific settings of the above parameters;

[0012] Detector parameters: false alarm rate FAR, system bandwidth f0, detection probability P d , signal-to-noise ratio SNR and how to obtain it, pixel size d, total field of view Ω, frame rate f p , scanning efficiency η sc , the number of pixels N pixel , noise equivalent bandwidth Δf and its acquisition method, spectral responsivity D * (λ) or average spectral responsivity

[0013] The above detection system parameters (including optical system parameters and detector parameters) are input into the program in the form of text files. The user can choose to directly give the signal-to-noise ratio SNR, thereby avoiding the input of the false alarm rate FAR, system bandwidth f0 and detection probability P. d ; In the case of uncertain signal-to-noise ratio, the user can also choose to give the false alarm rate FAR, system bandwidth f0 and detection probability P d , the program reads the false alarm rate FAR, system bandwidth f0 and detection probability P d After that, the signal-to-noise ratio SNR can be calculated by the user. Similarly, the user can choose to directly give the noise equivalent bandwidth Δf, thereby avoiding the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel ; In the case of uncertain noise equivalent bandwidth, the user can also choose to give the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel , the program reads the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel Then you can find the noise equivalent bandwidth Δf by yourself;

[0014] Step 3: Obtain parameters related to atmospheric transmittance, background radiation and path radiation calculation, including calculation band, atmospheric model, aerosol type, path direction, path starting point and end point elevation, observation distance, sun and surface albedo model;

[0015] Among the above parameters, the atmospheric model, aerosol type, calculation band, solar and surface albedo model need to be set by the user and used as program input in the form of a text file; the path direction, path start and end elevations, and observation distance are solved by the program itself during operation; after the program obtains the above parameters, it will call MODTRAN to calculate the atmospheric transmittance, background radiation, and path radiation under different observation directions and observation distances;

[0016] Step 4: Based on the parameters obtained from the first to third steps, construct an iterative format and iteratively obtain the maximum infrared detectable distance value R based on the chord cut method and Bayesian optimization algorithm max ; Based on the Little Hudson range model, the sampling efficiency, background radiation and path radiation correction are introduced to obtain the improved point source target range equation shown in formula (1); then the formula (1) is sorted and transformed to obtain the iterative formats shown in formulas (2) and (3) respectively:

[0017]

[0018] Among them, R is the maximum infrared detectable distance of the target obtained in the end; ξ is the sampling efficiency; D opt is the entrance pupil diameter of the optical element of the detection system; λ1 is the lower limit of the band interval; λ2 is the lower limit of the band interval; τ opt is the spectral average transmittance of the optical element; J T (λ) is the target spectral radiation intensity; τ a (λ,R) is the atmospheric transmittance; L P (λ) is the path spectral radiance; L B (λ) is the background spectral radiance; A T D is the projection area of ​​the target in the object space in the detection direction; * (λ) is the spectral responsivity of the detector; F # is the F number of the optical element; Ω is the total field of view of the detector focal plane; Δf is the equivalent noise bandwidth of the detection system; SNR is the signal-to-noise ratio of the detector; NEI is the equivalent noise illumination of the detection system; f(R) and g(R) are the functions corresponding to the different constructed iterative formats respectively;

[0019] Then, the critical zenith angle is calculated to determine whether the zenith angle corresponding to the current detection direction exceeds the critical zenith angle; the lower bound R1 and upper bound R2 of the initial interval of the detection distance iteration are assigned, and the lower bound and upper bound of the initial interval of the iteration are respectively substituted into formula (2) to obtain f(R1) and f(R2), and then the chord cut method is used to iteratively solve formula (2) to obtain the maximum infrared detectable distance R max; Alternatively, substitute the lower and upper bounds of the initial interval into formula (3) to obtain g(R1) and g(R2), determine whether there is a valid solution, and use the chord cut method to iteratively solve formula (3) to obtain the maximum infrared detectable distance R max ;

[0020] Step 5: Determine whether the calculation result of the previous step meets the point source target assumption, and give the final calculation result of the maximum detectable distance in the current detection direction;

[0021] First, the distance R0 required to meet the point source hypothesis under the current detection direction and target projection area needs to be calculated according to formula (4), where A T is the projected area of ​​the target in the object space, A p is the area of ​​the pixel. max <R0 is true, if true then output -R max , indicating that the current calculation result does not meet the point source assumption and should be converted to the extended source detection distance model for calculation; if not, output R max , as the calculation result of the maximum infrared detectable distance of the target in the current detection direction;

[0022]

[0023] Step 6: Determine whether the detection distance calculation in all predetermined detection directions is completed. If not, repeat steps 4 and 5 to perform the calculation in the next detection direction until the maximum detectable distance calculation in all detection directions is completed. Save and output all calculation results in the form of a text file.

[0024] Furthermore, in the fourth step, determining whether the zenith angle corresponding to the current detection direction exceeds the critical zenith angle specifically includes:

[0025] (1) If the zenith angle of the path direction does not exceed the critical zenith angle, determine whether there is an effective solution for the maximum detectable distance, and iterate to find the effective solution. Substitute the lower bound and upper bound of the initial interval of iteration into equation (3) to obtain g(R1) and g(R2), and determine whether g(R1)>0 and g(R2)>0 are true;

[0026] If it is established, the existence judgment process and iterative solution process are carried out. First, the embedded Bayesian optimization algorithm is used to determine whether there is a valid solution. After confirming that there is a valid solution, the maximum infrared detectable distance R is obtained through an iterative process. max ;

[0027] -g(R) is used as the objective function of waiting for optimization, and the Bayesian algorithm is used to find its maximum value. The specific steps are as follows: uniformly randomly sample 5 samples in the initial interval of detection distance iteration, calculate the -g(R) corresponding to each sample in turn, and judge whether -g(R)>0 is established. If it is established, it indicates that there is a valid solution and exits the optimization judgment process; if it is not established, 10 optimization calculations are performed after completing 5 prior calculations; the detection distance value that makes -g(R)>0 is used as the lower bound of the new iteration interval, and after updating the iteration interval, the chord cut method and formula (3) are used to start iteration. The solution that finally makes g(R)=0 is the maximum infrared detectable distance R required in the end. max , output, store R max The value of and exit the iterative calculation process; if there is no detection distance that makes -g(R)>0 after 10 optimization calculations, it means that there is no valid solution under the current calculation condition settings, and -1 is returned, and then the iterative calculation process is exited and directly enter the sixth step;

[0028] If it is not true, directly enter the iterative solution process of the chord cutting method to obtain the maximum infrared detectable distance R max ; Use formula (2) as the iteration format, substitute the lower bound R1 and upper bound R2 of the initial iteration interval of the detection distance into formula (2) to obtain f(R1) and f(R2); According to R3=R2-(R2-R1)*f(R2) / [f(R2)-f(R1)], obtain the intermediate distance R3, substitute R3 into formula (2) to obtain f(R3), and judge whether f(R3)<e holds, where e is the convergence residual; If f(R3)<e holds, the final required maximum infrared detectable distance R is obtained. max , return R max The value of and exit the iterative calculation process, otherwise repeat the process;

[0029] (2) If the zenith angle of the path direction exceeds the critical zenith angle, calculate the restricted distance R in the current path direction. lim , determine whether the current detection is limited by the curvature of the earth and solve the maximum infrared detectable distance R max Specifically, R lim Substituting into formula (2) we can obtain f(R lim ), judge f(R lim )>0 (i.e. limited by the curvature of the earth)?

[0030] If true, directly return the maximum infrared detectable distance R max =R lim , and exit the iterative calculation process;

[0031] If it is not true, enter the iterative solution process and finally output the maximum infrared detectable distance R max; Assign values ​​to the lower bound R1 and upper bound R2 of the initial interval of the detection distance iteration, and substitute R1 and R2 into formula (3) to obtain g(R1) and g(R2), respectively, and determine whether g(R1)>0 and g(R2)>0. The subsequent process is the same as in step (1), that is, if g(R1)>0 and g(R2)>0 holds, enter the solution existence judgment process and iterative solution process; if g(R1)>0 and g(R2)>0 does not hold, enter the chord cut method iterative solution process.

[0032] The beneficial effects of the present invention are as follows: the present invention provides a method for calculating infrared detectability of an aircraft, and the iterative process is accelerated by replacing the iterative calculation method with a chord cut method with a higher convergence order. Compared with the binary method, the time consumed in the iterative solution process is significantly shortened. In addition, the present invention effectively solves the problem that the detection distance cannot be solved after the background radiation and path radiation correction are introduced when the target intrinsic radiation intensity is low by introducing the Bayesian optimization process into the iterative calculation logic of the detection distance. Based on the aircraft infrared detectability calculation method based on the chord cut method and Bayesian optimization proposed by the present invention, the target infrared detection distance calculation program developed can call MODTRAN in real time during the iteration process to calculate the atmospheric transmittance, background radiation and path radiation under different path directions and different observation distances. Compared with the method of establishing the corresponding atmospheric transmittance, background radiation and path radiation database according to MODTRAN in advance, it can have stronger flexibility and wider applicability while ensuring the calculation accuracy, and the time consumption of the iterative calculation is also within an acceptable range. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a solution logic schematic diagram of the present invention;

[0034] Figure 2 is a schematic diagram of the spectral response range and spectral response curve of the detector used in the embodiment of the present invention;

[0035] Figure 3 is a schematic diagram of a pitch detection surface and a yaw detection surface when a target is used as a reference in an embodiment of the present invention;

[0036] Figure 4 It is an iterative logic diagram of the calculation method proposed by the present invention when the fourth step of determining that the path direction exceeds the critical zenith angle in the solution logic diagram of the present invention;

[0037] Figure 5 It is an iterative logic diagram of the calculation method proposed by the present invention when the fourth step of determining that the path direction does not exceed the critical zenith angle in the solution logic diagram of the present invention;

[0038] Figure 6It is an operation logic diagram for judging the existence of a solution based on Bayesian optimization in the calculation method proposed by the present invention;

[0039] Figure 7 The prediction result of the function curve obtained when the existence judgment of the solution is performed based on Bayesian optimization in the calculation method proposed by the present invention when the integrated radiation brightness distribution in the 3μm to 5μm band of the aircraft in the same flight state and different detection directions is used as input;

[0040] Figure 8 It is the maximum detectable envelope distribution in the pitch plane and yaw plane of the aircraft obtained according to the adopted parameters in the embodiment of the present invention.

[0041] In the figure, 1. Pitch detection surface; 2. Yaw detection surface. DETAILED DESCRIPTION

[0042] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0043] In order to achieve the above-mentioned purpose, the present invention provides the following specific implementation methods: the infrared detectability calculation method of an aircraft described above, based on the detection direction file to be solved, the detection band range information to be solved, the target integrated infrared radiation brightness file, the infrared detection system parameters and the atmospheric parameters provided by the user, starts from the first detection direction to be calculated, initializes some variables required before the iterative calculation starts, and determines whether the current detection direction exceeds the critical direction while considering the influence of the earth's curvature. If it does not exceed the critical direction, first set the initial iteration interval and substitute it into formula (3) to obtain the function values ​​at the upper and lower boundaries of the interval, and then analyze whether it is necessary to perform a solution existence judgment within the initial iteration interval. When it is necessary to perform a solution existence judgment, enter the Bayesian optimization process. When it is not necessary to perform a solution existence judgment, directly enter the chord cut method iteration process, and perform step-by-step iteration according to formula (2) to finally obtain the target infrared maximum detectable distance R max When the critical direction is exceeded, the restricted distance is first substituted into (3) and the infrared detection distance of the target at this distance is judged to be less than the restricted distance R according to whether the function value exceeds 0. lim . At the maximum infrared detectable distance R of the target max Less than the restricted distance R lim In the case of , reset the initial iteration interval and substitute it into formula (3). According to the function values ​​at the upper and lower bounds of the iteration interval, determine whether to enter the Bayesian optimization process or directly enter the chord cut method iteration process. Finally, the maximum detectable distance R of the target infrared can be obtained. maxThe target infrared detection distance calculation program developed by the method proposed in the present invention will automatically call MODTRAN to calculate the atmospheric transmittance, background radiation and path radiation during operation without human intervention, which significantly shortens the time consumed by repeatedly calling MODTRAN while having high accuracy.

[0044] like Figure 1 As shown, a method for calculating infrared detectability of an aircraft based on chord cut method and Bayesian optimization adopted in an embodiment of the present invention includes the following steps:

[0045] Step 1: Get the detection direction to be solved, the detection band range to be solved, the projected area of ​​the target under each detection direction, and the integrated infrared radiation intensity of the target within the detection band range to be solved. Among them, the detection direction to be solved is calculated based on the target as the reference and the target tail direction as 0°. The user writes it into a text file, and the program reads and obtains the detection direction setting from the text file when it runs; the detection band range to be solved means that the user needs to select the band range to be calculated from the commonly used infrared detection band range (i.e. 1~3μm, 3~5μm and 8~14μm), or enter a user-defined band that does not exceed the above band range. For example, if you want to calculate the detection distance of the target in the 3-5μm band, the user needs to set the calculation band start value to 3μm and the band end value to 5μm in the text file; if the user wants to calculate the detection distance of the target in the 3.5μm-4.8μm band, the user needs to set the calculation band start value to 3.5μm and the band end value to 4.8μm in the text file; the integrated infrared radiation intensity of the target in the detection band set by the user and the projection area of ​​the set detection direction to be solved are obtained by the calculation program according to the target integrated radiation brightness distribution text file provided by the user;

[0046] Specifically, in this embodiment, the detection points corresponding to the directions to be solved are 73 points in the range of 0° to 360° in the pitch plane and 73 points in the range of 0° to 360° in the yaw plane, and the detection band range to be solved is 3μm to 5μm. The projection area of ​​the target in different detection directions in the pitch plane and the yaw plane and the infrared integrated radiation intensity of the target in the 3μm to 5μm band are shown in Table 1:

[0047] Table 1 Target projection area and radiation intensity in the pitch and yaw detection plane in the 3μm to 5μm band when Ma=0.9

[0048]

[0049]

[0050]

[0051] Step 2: Obtain the detection system parameters, including optical system parameters: F number of the detection system optical components, average spectral transmittance τ opt , entrance pupil diameter D opt and the target type used in calculating sampling efficiency when taking into account the effects of diffraction;

[0052] Among them, the F number of the optical components of the detection system, the average spectral transmittance τ opt and entrance pupil diameter D opt The user writes a text file according to his / her needs, and the program reads and obtains the above parameters from the text file when it is running; the user specifies in the text file whether to consider the influence of the diffraction effect. If the diffraction effect is considered, the user needs to choose to regard the target as an ideal point source target or a finite size target, and choose the target image center at the pixel center, corner point or average position, all of which are written by the user in the text file, and the program reads and obtains the specific settings of the above parameters;

[0053] Detector parameters: false alarm rate FAR, system bandwidth f0, detection probability P d , signal-to-noise ratio SNR and how to obtain it, pixel size d, total field of view Ω, frame rate f p , scanning efficiency η sc , the number of pixels N pixel , noise equivalent bandwidth Δf and its acquisition method, spectral responsivity D * (λ) or average spectral responsivity

[0054] The above detection system parameters (including optical system parameters and detector parameters) are input into the program in the form of text files. The user can choose to directly give the signal-to-noise ratio SNR, thereby avoiding the input of the false alarm rate FAR, system bandwidth f0 and detection probability P. d ; In the case of uncertain signal-to-noise ratio, the user can also choose to give the false alarm rate FAR, system bandwidth f0 and detection probability P d , the program reads the false alarm rate FAR, system bandwidth f0 and detection probability P d After that, the signal-to-noise ratio SNR can be calculated by the user. Similarly, the user can choose to directly give the noise equivalent bandwidth Δf, thereby avoiding the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel ; In the case of uncertain noise equivalent bandwidth, the user can also choose to give the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel , the program reads the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel Then you can find the noise equivalent bandwidth Δf by yourself;

[0055] Specifically, in this embodiment, F # =1.3,τ opt =0.8, D opt =15.2mm, the target image is regarded as an ideal point target and the average position is used when calculating the sampling efficiency; FAR = 0.08, f0 = 10000Hz and P d =0.9, calculated SNR = 6.01; d = 20 μm, Ω = 0.3357, f p =60Hz,η sc =1 and N pixel =327680, calculated Δf = 30Hz; spectral responsivity D * (λ) Figure 2 The peak responsivity is about 1.2037×10 11 cm·Hz 12 ·W -1 .

[0056] Step 3: Obtain parameters related to atmospheric transmittance, background radiation and path radiation calculation. In this embodiment, the calculation of atmospheric transmittance, background radiation and path radiation corresponding to the path direction and path length is realized by calling MODTRAN. Specifically, the atmospheric model set in the MODTRAN calculation is the mid-latitude summer atmospheric model and the forest surface reflectivity model. Solar radiation is considered in the calculation of spectral radiation in the form of multiple scattering. The CO2 mixing ratio is 380ppmv, the rural aerosol model is used in the boundary layer area, the spring-summer seasonal aerosol profile is used in the troposphere, and the background stratospheric profile and extinction model are selected in the stratosphere. The time and solar altitude angle are summer, June 21, 12:28, and 89.899°.

[0057] Step 4: Calculate the azimuth and critical zenith angle in the current detection direction, and determine whether the path direction exceeds the critical zenith angle. Figure 3 The distribution of the 107 detection directions calculated in this embodiment in the pitch plane and the yaw plane is shown. In the figure, 1 is the pitch detection plane, and 2 is the yaw detection plane; in order to mainly illustrate the implementation process of the calculation method proposed in the present invention, no specific detection direction is selected here to illustrate the subsequent steps.

[0058] Furthermore, when it is determined in the fourth step that the path direction does not exceed the critical zenith angle, the program enters Figure 4The operation logic is shown. The lower limit of the initial interval of the detection distance iteration is set to 1km, and the upper limit is calculated using the form of formula (1) without considering the influence of atmospheric attenuation, background radiation and path radiation. Usually, the distance value is hundreds of kilometers, and can even reach thousands of kilometers due to different detector performance. Substituting the lower limit of 1km and the upper limit of hundreds of kilometers into formula (3), we can get the solutions g(R1) and g(R2), and determine whether g(R1)>0 and g(R2)>0.

[0059] Furthermore, when it is determined that g(R1)>0 and g(R2)>0 exists, it means that the value of the target intrinsic radiation after atmospheric attenuation at the lower boundary of the iteration interval exceeds the difference between the path radiation and the background radiation, that is, J T (λ)τ a (λ,R)>(L P (λ)-L B (λ))A T , and at the upper limit of the iteration interval, the value of the target intrinsic radiation after atmospheric attenuation still exceeds the difference between the path radiation and the background radiation, that is, J T (λ)τ a (λ,R)≤(L P (λ)-L B (λ))A T , which means that there may be a situation between the lower bound and the upper bound where the value of the target intrinsic radiation after atmospheric attenuation is equal to the difference between the path radiation and the background radiation. Therefore, we enter the process of determining the existence of the solution using Bayesian optimization, that is, Figure 6 The operation logic diagram shown. After entering, take -g(R) as the objective function waiting for optimization. First, uniformly randomly sample 5 points in the initial iteration interval, calculate the atmospheric transmittance, background radiation and path radiation under the corresponding path length in turn, and then substitute each parameter into formula (3) to calculate g(R) and obtain the corresponding -g(R), and judge whether -g(R)>0 is established. If it is established, it means that there is indeed a valid solution in the initial iteration interval, and the Bayesian optimization process is exited; if it is not established, 10 optimization calculations are performed after completing 5 prior calculations. The detection distance that makes -g(R)>0 is used as the lower bound of the new iteration interval. After updating the iteration interval, the chord cut method and formula (3) are used to start iteration. The solution that finally makes g(R)=0 is the maximum infrared detectable distance R required. max , return, output and store R max If there is no detection distance that makes -g(R)>0 after 10 optimization calculations, it means that there is no valid solution under the current calculation condition settings, and -1 is returned, and then the iterative calculation process is exited. Figure 7It shows two sets of -g(R) function curves predicted by the Bayesian optimization process. The left picture shows the case where -g(R)=0 has a valid solution, and the right picture shows the case where -g(R)=0 does not have a valid solution.

[0060] Further, when it is judged that g(R1)>0 and g(R2)>0 do not exist, it means that either the target intrinsic radiation at the lower boundary of the initial interval exceeds the difference between the path radiation and the background radiation after atmospheric attenuation and the target intrinsic radiation intensity at the upper boundary is less than the difference between the path radiation and the background radiation after atmospheric attenuation, or the target intrinsic radiation intensity at the lower boundary is less than the difference between the path radiation and the background radiation after atmospheric attenuation and the target intrinsic radiation intensity at the upper boundary exceeds the difference between the path radiation and the background radiation, and the function g(R)=0 has a solution within the given iteration interval, so the chord cut method iterative solution process is entered. Using formula (2) as the iteration format, the lower bound R1 and upper bound R2 of the initial iteration interval of the detection distance are respectively substituted into formula (2) to obtain f(R1) and f(R2). According to R3=R2-(R2-R1)*f(R2) / [f(R2)-f(R1)], the intermediate distance R3 is obtained. R3 is substituted into formula (2) to obtain f(R3). It is determined whether f(R3)<e is established, where e is the convergence residual. In this embodiment, e=0.0001 is set. If f(R3)<e is established, the final required maximum infrared detectable distance R is obtained. max , return R max The value of and exit the iterative calculation process, otherwise repeat the process.

[0061] Furthermore, in the fourth step, when it is determined that the path direction exceeds the critical zenith angle, the program's operating logic is as follows: Figure 5 As shown. First, temporarily set the lower limit of the initial iteration interval of the detection distance to 1km, and the upper limit is set according to the distance value calculated based on the target intrinsic radiation intensity without considering the influence of atmospheric attenuation, background and path radiation. Then, calculate the distance between the target and the ground in the current direction, that is, the limited distance R lim Then R lim Substituting into equation (2) we can obtain f(R lim ), and determine whether there is f(R lim )>0.

[0062] Furthermore, in judging whether there is f(R lim )>0, then within the limited distance R lim There is J T (λ)τ a (λ,R)>(L P (λ)-L B (λ))A T , the true solution exceeds R limHowever, due to the limitation of the earth's curvature, the maximum infrared detectable distance in this path direction can only be R lim , that is, R max =R lim , return R max value and exit the iterative calculation process.

[0063] Furthermore, in judging whether there is f(R lim )≤0, this means J T (λ)τ a (λ,R)≤(L P (λ)-L B (λ))A T , the real solution is between 0 and R lim The detection of the target is not limited by the curvature of the earth. Change the upper limit of the initial iteration interval of the detection distance to R lim , and start the chord-cut method iteration process in the new detection distance iteration interval, substitute the upper and lower bounds of the detection distance iteration interval into formula (3) to obtain g(R1) and g(R2) respectively, and judge whether g(R1)>0 and g(R2)>0. Then, according to the judgment result, choose to enter the Bayesian solution existence judgment process or directly enter the chord-cut method iteration solution process.

[0064] Step 5: Calculate the distance R0 required to satisfy the point source assumption corresponding to the target projection area in the current detection direction according to formula (4). max If it is greater than or equal to R0, it means that the current detector responds normally and the point source target assumption is established, and the final calculation result R is output. max If it is less than R0, it means that the calculated distance value does not meet the conditions for the point source target assumption to be established, and the extended source action distance model should be used for solution. If the target is a large temperature difference target, the action distance model based on the small temperature difference assumption currently established for the extended source target cannot be used, then it is determined that the detector cannot respond normally to the target in the current direction, and outputs -R max .

[0065] Step 6: Determine whether the current detection point is the last detection point to be calculated. If not, repeat steps 4 and 5 to calculate the next detection direction until the maximum detectable distance calculation in all detection directions is completed, and the calculation results are saved and output. Figure 8 The present invention shows, for a flying target at a flight Mach number of 0.9, the maximum detectable envelope of the target calculated in the pitch plane and the yaw plane, the responsive area and the non-responsive area of ​​the air-based platform and the ground-based platform to the target infrared signal, the distance R0 required to meet the point source hypothesis in different detection directions, and other information.

[0066] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for calculating infrared detectability of an aircraft, characterized in that: The following steps are involved: Step 1: Obtain the detection direction to be solved, the detection band range to be solved, the target projection area under each detection direction, and the integrated infrared radiation intensity of the target within the detection band range to be solved; Step 2: Obtain detection system parameters, including: Optical system parameters: F number of optical components of the detection system, average spectral transmittance τ opt , entrance pupil diameter D opt and the target type used in calculating sampling efficiency when taking into account the effects of diffraction; Detector parameters: false alarm rate FAR, system bandwidth f0, detection probability P d , signal-to-noise ratio SNR and how to obtain it, pixel size d, total field of view Ω, frame rate f p , scanning efficiency η sc , the number of pixels N pixel , noise equivalent bandwidth Δf and its acquisition method, spectral responsivity D * (λ) or average spectral responsivity Step 3: Obtain parameters related to atmospheric transmittance, background radiation and path radiation calculation, including calculation band, atmospheric model, aerosol type, path direction, path starting point and end point elevation, observation distance, sun and surface albedo model; Step 4: Based on the parameters obtained from the first to third steps, construct an iterative format and iteratively obtain the maximum infrared detectable distance value R based on the chord cut method and Bayesian optimization algorithm max ; Based on the Little Hudson range model, the sampling efficiency, background radiation and path radiation correction are introduced to obtain the improved point source target range equation shown in formula (1); then the formula (1) is sorted and transformed to obtain the iterative formats shown in formulas (2) and (3) respectively: Among them, R is the maximum infrared detectable distance of the target obtained in the end; ξ is the sampling efficiency; D opt is the entrance pupil diameter of the optical element of the detection system; λ1 is the lower limit of the band interval; λ2 is the lower limit of the band interval; τ opt is the spectral average transmittance of the optical element; J T (λ) is the target spectral radiation intensity; τ a (λ,R) is the atmospheric transmittance; L P (λ) is the path spectral radiance; L B (λ) is the background spectral radiance; A T D is the projection area of ​​the target in the object space in the detection direction; * (λ) is the spectral responsivity of the detector; F # is the F number of the optical element; Ω is the total field of view of the detector focal plane; Δf is the equivalent noise bandwidth of the detection system; SNR is the signal-to-noise ratio of the detector; NEI is the equivalent noise illumination of the detection system; f(R) and g(R) are the functions corresponding to the different constructed iterative formats respectively; Then, the critical zenith angle is calculated to determine whether the zenith angle corresponding to the current detection direction exceeds the critical zenith angle; the lower bound R1 and upper bound R2 of the initial interval of the detection distance iteration are assigned, and the lower bound and upper bound of the initial interval of the iteration are respectively substituted into formula (2) to obtain f(R1) and f(R2), and then the chord cut method is used to iteratively solve formula (2) to obtain the maximum infrared detectable distance R max ; Alternatively, substitute the lower and upper bounds of the initial interval into formula (3) to obtain g(R1) and g(R2), determine whether there is a valid solution, and use the chord cut method to iteratively solve formula (3) to obtain the maximum infrared detectable distance R max ; Step 5: Determine whether the calculation result of the previous step meets the point source target assumption, and give the final calculation result of the maximum detectable distance in the current detection direction; First, the distance R0 required to meet the point source hypothesis under the current detection direction and target projection area needs to be calculated according to formula (4), where A T is the projected area of ​​the target in the object space, A p is the area of ​​the pixel; determine R max <R0 is true, if true then output -R max , indicating that the current calculation result does not meet the point source assumption and should be converted to the extended source detection distance model for calculation; if not, output R max , as the calculation result of the maximum infrared detectable distance of the target in the current detection direction; Step 6: Determine whether the detection distance calculation in all predetermined detection directions is completed. If not, repeat steps 4 and 5 to perform the calculation in the next detection direction until the maximum detectable distance calculation in all detection directions is completed. Save and output all calculation results in the form of a text file.

2. The method for calculating infrared detectability of an aircraft according to claim 1, characterized in that: In the first step, the detection direction to be solved is calculated based on the target and the target tail direction is 0°, which is written by the user into a text file. When the program is running, it reads and obtains the detection direction setting from the text file; the detection band range to be solved refers to the band range selected by the user to be calculated or the user-defined band entered; the integrated infrared radiation intensity of the target in the detection band set by the user and the projection area of ​​the set detection direction to be solved are obtained by the calculation program according to the target integrated radiation brightness distribution text file provided by the user.

3. The method for calculating infrared detectability of an aircraft according to claim 1, characterized in that: Among the optical system parameters of the second step, the F number and average spectral transmittance τ of the optical elements of the detection system are opt and entrance pupil diameter D opt The user writes a text file according to his / her own needs, and the program reads and obtains the above optical system parameters from the text file when it is running; the user specifies in the text file whether to consider the influence of the diffraction effect. If the diffraction effect is considered, the user needs to choose to regard the target as an ideal point source target or a finite size target, and choose the target image center at the pixel center, corner point or average position, all of which are written by the user in the text file, and the program reads and obtains the above specific parameter settings for the diffraction effect; In the detector parameters, the user chooses to directly give the signal-to-noise ratio SNR, thereby avoiding the input of the false alarm rate FAR, the system bandwidth f0 and the detection probability P d ; In the case of uncertain signal-to-noise ratio, the user chooses to give the false alarm rate FAR, system bandwidth f0 and detection probability P d , the program reads the false alarm rate FAR, system bandwidth f0 and detection probability P d The signal-to-noise ratio SNR can be calculated by the user; similarly, the user chooses to directly give the noise equivalent bandwidth Δf, thereby avoiding the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel ; In the case of uncertain noise equivalent bandwidth, the user chooses to give the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel , the program reads the input frame rate f p , scanning efficiency η sc and the number of pixels N pixel Then you can calculate the noise equivalent bandwidth Δf by yourself.

4. The method for calculating infrared detectability of an aircraft according to claim 1, characterized in that: The atmospheric model, aerosol type, calculation band, solar and surface albedo model in the third step need to be set by the user and used as program input in the form of a text file; the path direction, path starting point and end point elevations, and observation distance are solved by the program during operation; after the program obtains the above parameters, it will call MODTRAN to calculate the atmospheric transmittance, background radiation and path radiation under different observation directions and observation distances.

5. The method for calculating infrared detectability of an aircraft according to claim 1, characterized in that: In the fourth step, determining whether the zenith angle corresponding to the current detection direction exceeds the critical zenith angle specifically includes: (1) If the zenith angle of the path direction does not exceed the critical zenith angle, determine whether there is an effective solution for the maximum detectable distance, and iterate to find the effective solution; substitute the lower bound and upper bound of the initial interval of iteration into equation (3) to obtain g(R1) and g(R2), and determine whether g(R1)>0 and g(R2)>0 hold; If it is established, the existence judgment process and iterative solution process are carried out. First, the embedded Bayesian optimization algorithm is used to determine whether there is a valid solution. After confirming that there is a valid solution, the maximum infrared detectable distance R is obtained through an iterative process. max ; -g(R) is used as the objective function of waiting for optimization, and the Bayesian algorithm is used to find its maximum value. The specific steps are as follows: uniformly randomly sample 5 samples in the initial interval of detection distance iteration, calculate the -g(R) corresponding to each sample in turn, and judge whether -g(R)>0 is established. If it is established, it indicates that there is a valid solution and exits the optimization judgment process; if it is not established, 10 optimization calculations are performed after completing 5 prior calculations; the detection distance value that makes -g(R)>0 is used as the lower bound of the new iteration interval, and after updating the iteration interval, the chord cut method and formula (3) are used to start iteration. The solution that finally makes g(R)=0 is the maximum infrared detectable distance R required. max , output, store R max The value of and exit the iterative calculation process; if there is no detection distance that makes -g(R)>0 after 10 optimization calculations, it means that there is no valid solution under the current calculation condition settings, and -1 is returned, and then the iterative calculation process is exited and directly enter the sixth step; If it is not true, directly enter the iterative solution process of the chord cutting method to obtain the maximum infrared detectable distance R max ; Use formula (2) as the iteration format, substitute the lower bound R1 and upper bound R2 of the initial iteration interval of the detection distance into formula (2) to obtain f(R1) and f(R2); According to R3=R2-(R2-R1)*f(R2) / [f(R2)-f(R1)], obtain the intermediate distance R3, substitute R3 into formula (2) to obtain f(R3), and judge whether f(R3)<e holds, where e is the convergence residual; If f(R3)<e holds, the final required maximum infrared detectable distance R is obtained. max , return R max The value of and exit the iterative calculation process, otherwise repeat the process; (2) If the zenith angle of the path direction exceeds the critical zenith angle, calculate the restricted distance R in the current path direction. lim , determine whether the current detection is limited by the curvature of the earth and solve the maximum infrared detectable distance R max Specifically, R lim Substituting into formula (2) we can obtain f(R lim ), judge f(R lim )>0 (i.e. limited by the curvature of the earth)? If true, directly return the maximum infrared detectable distance R max =R lim , and exit the iterative calculation process; if it is not true, enter the iterative solution process and finally output the maximum infrared detectable distance R max ; Assign values ​​to the lower bound R1 and upper bound R2 of the initial interval of the detection distance iteration, and substitute R1 and R2 into formula (3) to obtain g(R1) and g(R2), respectively, and determine whether g(R1)>0 and g(R2)>0. If g(R1)>0 and g(R2)>0 hold, perform the existence judgment process and iterative solution process according to (1); if g(R1)>0 and g(R2)>0 do not hold, enter the chord-cut iterative solution process according to (1).

Images