An intelligent solution method for high-speed aircraft fire control model

Abstract

The present invention discloses an intelligent solution method for a high-speed aircraft fire control model, comprising the following steps: step 1, establishing a proxy model of a high-speed aircraft carrier delivery platform considering flight-thrust coupling according to existing data; step 2, constructing a fire control model of a high-speed aircraft platform, including a target motion prediction model, a missile outer ballistic model and a fire control hit problem solution model; step 3, solving the attack area according to the flight characteristics of the high-speed aircraft in combination with the Archimedean optimization algorithm, and inversely solving the initial command signal of the carrier. The present invention has few control parameters, good robustness, and can solve the optimization problem by generating an objective function value with the minimum error.

Description

Technical Field

[0001] The invention belongs to the technical field of high-speed aircraft control and relates to an intelligent solution method for a high-speed aircraft fire control model. Background Art

[0002] In recent years, all major countries have strengthened their research on high-speed aircraft and fire control methods. Moreover, unlike traditional aircraft, reusable high-speed aircraft have lower mission costs. Moreover, due to its extremely fast flight speed and super maneuverability, the flight mode of the aircraft will also be more flexible. The aviation fire control system is an important part of high-speed aircraft; the current development of artificial intelligence technology has brought new directions to the future fire control system. On the basis of improving the level of informatization, realizing the intelligence of the aviation fire control system is not only grasping the technical direction of the development of the aviation fire control system, but also an important task for the development of the aviation fire control system. The intelligent fire control system has a very broad application scenario. However, compared with traditional aircraft, the flight speed of high-speed aircraft has been significantly improved, and its flight environment is more complex and the frequency of disturbance is high. Therefore, it has the characteristics of short system response time and high real-time performance. Therefore, more new requirements are put forward for the fire control system, and the requirements for the accuracy of the algorithm solution are also improved. It is necessary to explore new intelligent algorithms to optimize the traditional fire control solution method. The existing fire control methods are mostly applied to subsonic / supersonic manned aircraft; however, there is a lack of research on fire control methods for high-speed aircraft. Therefore, in order to meet the needs of future high-speed aircraft fire control, it is necessary to design an intelligent fire control solution strategy suitable for high-speed aircraft. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent solution method for a high-speed aircraft fire control model, which can solve the optimization problem by generating an objective function value with the minimum error.

[0004] In order to solve the above technical problems, the present invention provides an intelligent solution method for a high-speed aircraft fire control model, comprising the following steps:

[0005] Step 1: Based on the existing data, establish a proxy model of the high-speed aircraft carrier delivery platform considering the flight-thrust coupling;

[0006] Step 2: Construct a fire control model for the high-speed aircraft platform, including a target motion prediction model, a missile exterior ballistics model, and a fire control hit problem solving model;

[0007] Step 3: According to the flight characteristics of the high-speed aircraft, the attack area is solved in combination with the Archimedes optimization algorithm, and the initial command signal of the carrier aircraft is reversed.

[0008] Preferably, in step 1, establishing a proxy model of a high-speed aircraft carrier delivery platform considering flight-propulsion coupling based on existing data specifically includes the following steps:

[0009] Step 11, assuming that the aircraft is flying on a rotating spherical earth, derive a nonlinear mathematical model of a high-speed aircraft on a rotating circular earth;

[0010] Step 12: Data sampling is performed based on the aircraft aerodynamic data, geometric parameters and propulsion coefficients in the existing database, and polynomial fitting is performed on the sampled hypersonic segment data. After fitting, the fitted polynomial proxy model is evaluated using the goodness of fit test method.

[0011] Preferably, in step 2, establishing the target motion prediction model specifically comprises: using an interactive multi-model algorithm to predict the target, and decomposing the target's motion mode into a combination of the following multiple motion state models:

[0012] (1)CV model;

[0013] When the target is non-maneuvering, i.e., performing uniform linear motion or uniformly accelerated linear motion, the following second-order constant speed CV model is used respectively;

[0014] Assume that the target moves in a uniform straight line, the target displacement is recorded as x(t), and the speed is recorded as Under the condition that the target speed has random disturbance, it is assumed that the random disturbance of speed has a mean of zero and a variance of δ 2 Gaussian white noise a(t);

[0015] Get system state variables According to Newton's laws of motion,

[0016]

[0017] Written in matrix form:

[0018]

[0019] That is, the CV model is

[0020]

[0021] Among them, A(t) is the system matrix of the model and B(t) is the input matrix.

[0022] (2) CT model;

[0023] The CT model is a constant turning model. When the target makes a constant turning motion with a constant speed but a constantly changing direction, it is described by the CT model. The discrete time domain form of the CT model is expressed as:

[0024]

[0025] In the formula, x k and k is the position component of the target state along the x-axis and y-axis in the Cartesian coordinate system at time k, and is the velocity component corresponding to time k, and The corresponding mean is zero and the variance is q 2 Gaussian process white noise, w is the constant turning rate, which can reflect the target maneuvering situation, p is the radar sampling interval, and the state transfer equation describes the recursive relationship of the target state from time k to time k+1;

[0026] (3) Singer model;

[0027] The Singer model is a first-order time-dependent model with zero acceleration mean. Assuming that the time-dependent function of the target maneuver acceleration is in the form of exponential decay, the time-dependent function Ra(τ) of the Singer model is:

[0028]

[0029] Where: a is the unknown parameter that determines the target maneuvering characteristics in the interval (t, t+τ). b(t) is the maneuvering acceleration, is the maneuver acceleration variance.

[0030] Preferably, in step 2, the missile outer ballistic model is established as follows: when a high-speed aircraft is used as a carrier delivery platform, the model establishment process is different from that of a traditional aircraft, and the flight characteristics of the aircraft cause the initial delivery speed and altitude to change; the delivery altitude is closer to near space than that of a traditional aircraft, and the initial delivery speed is also faster. Therefore, the centrifugal acceleration caused by the curvature of the earth and the Coriolis acceleration caused by rotation need to be considered in the process of establishing the model.

[0031] Preferably, in step 2, in the air-to-air mission of the high-speed aircraft, since the carrier platform is a reusable fly-thrust coupled high-speed aircraft with excellent maneuverability and a large flight envelope, the comprehensive performance of the platform is greatly improved compared to traditional aircraft. The high performance of the platform can reduce the requirements for the delivery objects carried by the platform compared to traditional aircraft; for example, during the delivery process, the initial velocity of the delivered object is very large, so the delivery object can be unpowered, and the form of unpowered guidance can be considered to reduce the cost of delivery. The establishment of a fire control hit problem solving model is specifically: the rapid simulation method is used for simulation in the fire control launch, which specifically includes the following steps:

[0032] (1) The mathematical models of missile and target maneuvers are simplified as necessary. The missile, aircraft and target are considered as moving particles located in the same plane. The forces acting on the missile are simplified to thrust, drag and lift acting on the missile particles. The missile adopts proportional guidance law and the role of the control system is not considered.

[0033] (2) Solve the missile's motion differential equations using a fast variable-step integration method;

[0034] (3) Optimize program design and use structural modularization and other means to improve computing speed;

[0035] Establish a mathematical model for the high-speed aircraft carrier delivery platform to perform air-to-air fire control tasks, and the line-of-sight angular velocity between the missile and the target for

[0036]

[0037] Distance change rate for

[0038]

[0039] Where, V1 is the speed of the aircraft, V M —target speed, r—target distance, θ—velocity deviation angle of the carrier, q—line of sight angle, θ M —Target velocity inclination, η M —target heading angle, η—aircraft heading angle, q0 is the initial value of the line of sight angle between the missile and the target, r0 is the initial value of the distance between the missile and the target, and Δt is the target flight time;

[0040] The target's motion characteristic equation is described by the following equation:

[0041]

[0042] Where T is the target’s flight time at the moment of missile launch, θ M0 —Initial value of target velocity inclination, η M0 —Initial value of target heading angle, G M is the target maneuvering overload, is the target speed angle change rate, when G M = 0, the target moves in a straight line at a constant speed. M When ≠0, the target moves in a constant circular motion;

[0043] After integrating the above formula, we get:

[0044]

[0045] Among them, θ M0 —Initial value of target velocity inclination, η M0—Initial value of target heading angle;

[0046] After integrating the acceleration of the missile, the velocity of the missile at any moment is

[0047]

[0048] Where V0 is the initial velocity of the missile, is the missile speed change rate.

[0049] The missile's proportional guidance equation is:

[0050]

[0051] Where K is the proportional guidance coefficient, —rate of change of velocity angle;

[0052] The movement of the target will not only affect the movement characteristics of the target itself, but also affect the guidance law of the missile itself:

[0053]

[0054] In the formula, is the rate of change of sight angle.

[0055] The fire control model of the high-speed aircraft carrier delivery platform is:

[0056]

[0057] Based on the fast simulation calculation equations in the established fire control model, after determining the initial conditions, the maximum dynamic range of the missile is used as the initial value of the distance to calculate the parameter changes of each point of the missile during the attack process, and then the hit limit conditions stipulated by the missile characteristics such as the hit and miss amount are used to judge whether the missile hits the target. If it does not hit, the distance minus the miss amount is used as the initial value for the next calculation, and the calculation is repeated in this way to finally obtain the maximum launch distance of the missile; similarly, the minimum launch distance of the missile can be obtained, and then the initial value q0 is continuously changed for calculation, and finally the missile attack area can be obtained;

[0058] The program uses variable step sizes for integration: a larger step size is used at the beginning of the calculation, and a smaller step size is used when the missile approaches or hits the target. This greatly improves the calculation speed while meeting certain accuracy conditions.

[0059] Preferably, in step 3, based on the fast simulation method, combined with the Archimedes intelligent optimization method, different initial missile launch sight angles are switched respectively, the initial attitude command of the missile is iteratively calculated from 0° to 180° for the maximum attack distance and the minimum attack distance of the launch at this time, and the optimal solution is obtained through the Archimedes intelligent optimization algorithm, and finally the initial attitude command that reaches the maximum attack distance under different launch conditions is obtained. According to the flight characteristics of the high-speed aircraft, the attack area is solved in combination with the Archimedes optimization algorithm, and the initial command signal of the carrier aircraft is reversed, which specifically includes the following steps:

[0060] Step 31: Establishing Archimedes intelligent optimization algorithm model;

[0061] (2) Initialize the positions of all objects

[0062] O i =lb i +rand×(ub i -lb i );i=1,2,...,N (14)

[0063] In the formula, O i is the i-th object in a set of N objects, lb i andub i are the lower and upper bounds of the search space respectively;

[0064] Use formula (15) to initialize the volume vol of the i-th object i and density den i :

[0065]

[0066] Where rand is a D-dimensional vector that randomly generates numbers between [0,1]. Finally, initialize the acceleration acc of the i-th object i ;

[0067] acc i =lb i +rand×(ub i -lb i ); i=1,2,...,N (16)

[0068] (2) Update density and volume;

[0069] The update formula for the density and volume of the i-th object in the t+1th iteration is:

[0070]

[0071] Among them, vol best and den bestThe volume and density of the best objects ever discovered;

[0072] (3) Define the transfer operator and density operator;

[0073] At the beginning, the objects collide with each other. After a period of time, the objects try to reach a state of equilibrium. In the Archimedean algorithm, this is achieved through the transfer operator TF. The transfer operator TF converts the algorithm flow from exploration mode to development mode. The definition of the transfer operator TF is:

[0074]

[0075] The transfer operator TF gradually increases with time until it reaches 1, where t and t max are the current number of iterations and the maximum number of iterations, respectively. Similarly, the density factor d helps the algorithm to search from global to local, and d decreases as the iteration time increases;

[0076]

[0077] (4) Selection of exploration mode;

[0078] If the transfer operator TF≤0.5, the objects collide, randomly select an object and update the acceleration of the object in the t+1th iteration using formula (20):

[0079]

[0080] where den i ,vol i , and acc i is the density, volume and acceleration of object i, and acc mr ,den mr and vol mr are the acceleration, density and volume of the random material;

[0081] (5) Selection of development model;

[0082] If the transfer operator TF>0.5 and there is no collision between the objects, the t+1 iteration acceleration of the object is updated using formula (21);

[0083]

[0084] Among them, acc best is the optimal acceleration of the object;

[0085] (6) Normalized acceleration;

[0086] The acceleration of the object is normalized using equation (22) to calculate the percentage change.

[0087]

[0088] In the formula, u and l are the normalized ranges, and the values ​​of u and l are set to 0.9 and 0.1 respectively. It defines the percentage of the step size that each individual will change. If the object is far from the global optimum, the acceleration value will be high, which means that the target will be in exploration mode; otherwise, in the exploitation phase;

[0089] (7) Update location;

[0090] (a) If TF≤0.5, that is, the exploration phase, the position of the i-th object at iteration t+1 is

[0091]

[0092] In the formula, x rand is the individual position randomly generated by the algorithm. C1 is a constant, which is assigned a value of 2 according to specific requirements;

[0093] (b) If TF>0.5, that is, the development stage, the position of the i-th object at iteration t+1 is

[0094]

[0095] C2 is a constant of 6, T increases with time, proportional to the transfer operator, T = C3 × TF, T increases with iteration in the range [C3 × 0.3, 1], and initially takes a certain percentage from the optimal position of the initial object;

[0096] In formula (24), F is the sign of the object's motion direction;

[0097]

[0098] Where, P = 2*rand-C4;

[0099] (8) Evaluation;

[0100] Evaluate each individual using the objective function f, note the best solution found, and assign the individual with the best fitness x best , optimal density den best , optimal volume vol best and the optimal acceleration acc best ;

[0101] Step 32, based on the fast simulation method of fire control model solution, combined with Archimedes intelligent optimization algorithm, solve the fire control command;

[0102] The fitness function of the Archimedean algorithm is:

[0103]

[0104] Where, V1 is the speed of the aircraft, V M —Target speed, θ—Aircraft speed deviation angle, q—Line of sight angle, θ M —Target speed inclination;

[0105] During the calculation process, for each attack distance iterated using the fast simulation method, the flight status instruction of the carrier is used to calculate the fitness value of each individual in the population according to the fitness function, and return to the Archimedean algorithm to iteratively update the position, volume and acceleration information of each individual until the function reaches the allowable error or the algorithm reaches the maximum number of iterations, and then output the solved attack distance and the aircraft status instruction at the attack distance.

[0106] The beneficial effects of the present invention are as follows: the present invention establishes fire control models of hypersonic platforms according to the target mission requirements of future high-speed aircraft fire control, including a target motion prediction model, a missile outer ballistic model and a fire control hit problem solving model; then, according to the flight characteristics of the high-speed aircraft, the attack area is solved in combination with the Archimedes optimization algorithm, and the initial command signal of the carrier is reversed; this fire control scheme selects a high-speed aircraft as a delivery carrier, adopts a new intelligent optimization algorithm combined with a traditional numerical integration method, and intelligently solves the fire control instructions of the carrier to realize intelligent decision-making of the fire control model, which can not only give play to the performance advantages of the carrier's high speed and high maneuverability, greatly improve the mission execution efficiency, but also meet the future needs of air fire control and the development trend of the intelligent fire control system; the present invention has few control parameters and good robustness, and can solve the optimization problem by generating an objective function value with the minimum error. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0108] Figure 2 It is a schematic diagram of the high-speed aircraft carrier delivery platform of the present invention performing an air-to-air mission.

[0109] Figure 3 The present invention provides a flow chart for solving the high-speed aircraft fire control model equation group.

[0110] Figure 4 This is a flow chart of the problem solving method of the Archimedean algorithm in the present invention.

[0111] Figure 5 Schematic diagram of the entire process of solving fire control instructions using the Archimedes intelligent algorithm based on fast simulation.

[0112] Figure 6 The figure is a schematic diagram of a missile attack area obtained in the specific implementation of the present invention. DETAILED DESCRIPTION

[0113] like Figure 1 As shown, an intelligent solution method for a high-speed aircraft fire control model includes the following steps:

[0114] Step 1: According to the flight characteristics of the high-speed aircraft platform, the motion equation of the high-speed aircraft on the rotating circular ground is derived. According to the model data of the existing aircraft, the polynomial proxy model structure is used to perform polynomial fitting of the data. After fitting, the fitted polynomial proxy model is evaluated by the goodness of fit test method.

[0115]

[0116] Among them, SSR is the regression sum of squares of the data, SST is the introduced total sum of squares, SSE is the error sum of squares of the data, and SST = SSE + SSR.

[0117] Step 2: Construct a fire control model for the high-speed aircraft platform, including target motion prediction, missile exterior ballistic model, and fire control hit problem solving model.

[0118] The target motion prediction model is established using the interactive multi-model algorithm, which decomposes the target motion state into a combination of multiple motion states.

[0119] Next, the missile's external ballistic model is established. Similarly, since the high-speed aircraft is used as a carrier delivery platform, it is different from the traditional aircraft model establishment, and the flight characteristics of the aircraft make the initial delivery speed and altitude change. Therefore, in the establishment of the model, it is necessary to consider the centrifugal acceleration caused by the curvature of the earth and the Coriolis acceleration caused by the rotation. Therefore, it is necessary to establish a missile motion model under the rotating circular earth. Figure 2 As shown, the missile's particle motion equation is as follows:

[0120]

[0121] In the formula, r is the distance from the center of mass of the missile aircraft to the center of the earth, θ is the longitude, φ is the latitude, V is the speed of the missile aircraft, γ is the ballistic inclination, and ψ is the ballistic deviation. X, D, L are the thrust, drag and lift of the aircraft, g is the local gravity acceleration at the location of the aircraft, α is the angle of attack of the missile aircraft, σ is the roll angle of the missile aircraft, ω e is the angular velocity of the Earth's rotation.

[0122] Finally, a fire control hit problem solution model is established. In the air-to-air mission performed by high-speed aircraft, since the delivery carrier platform is a reusable high-speed aircraft with fly-thrust coupling, it has excellent maneuverability and a large flight envelope, and the overall performance of the platform is greatly improved compared with traditional aircraft. The high performance of the platform can reduce the requirements for the delivery object of the platform compared with the delivery object of traditional aircraft; for example, during the delivery process, the initial velocity of the projectile is very large, so the delivery object can be unpowered, and the form of unpowered guidance can be considered to reduce the delivery cost. Based on the problem, the fire control hit problem model is established as follows:

[0123]

[0124] Due to the hypersonic flight characteristics of the carrier aircraft platform, high requirements are placed on the speed and accuracy of delivery. Therefore, the rapid simulation method is used for simulation in fire control launch.

[0125] Based on the fast simulation calculation equations in the established fire control model, under the determined initial conditions, the maximum dynamic range of the missile can be used as the initial value of the distance to calculate the parameter changes of each point in the missile attack process, and then the hit limit conditions specified by the missile characteristics such as the hit and miss amount can be used to determine whether the missile hits the target. If it does not hit, the distance minus the miss amount is used as the initial value for the next calculation, and the calculation is repeated in this way to finally obtain the maximum launch distance of the missile; similarly, the minimum launch distance of the missile can be obtained. By continuously changing the initial value for calculation, the missile attack area can be finally obtained.

[0126] In addition, the use of variable step lengths in the process of calculating the missile attack zone with a smaller step length is an important measure to increase the speed of calculation and ensure the completion of the calculation. If the step size is too large during the calculation, the calculation error will increase or even diverge, especially when the missile approaches the target, and the distance within a step length Δt time is greater than twice the hit and miss amount, it may lead to wrong judgment, and the hit will be judged as a miss, so that the calculated attack zone boundary is also incorrect; conversely, if the step size is too small, it will affect the speed of calculation. Due to the above reasons, variable step sizes are used for integration in the program: at the beginning of the calculation, a larger step size can be selected, and when the missile approaches or hits the target, a smaller step size is taken. In this way, the entire calculation can meet a certain accuracy condition and greatly improve the calculation speed. The specific flow chart is as follows Figure 3 shown.

[0127] Step 3: Establish a fast simulation method combined with the Archimedes intelligent optimization method, switch different initial missile launch sight angles, iterate the missile's initial posture command from 0° to 180° to calculate the maximum attack distance and minimum attack distance at this time, and obtain the optimal solution through the Archimedes intelligent optimization algorithm, and finally obtain the initial posture command that achieves the maximum attack distance under different launch conditions, such as Figure 4 and Figure 5 shown.

[0128] In the Archimedes intelligent optimization algorithm, each iteration updates the volume and density of the object:

[0129]

[0130] Among them, vol best and den best are the volume and density of the best object found so far and rand is a uniformly distributed random number.

[0131] When the transfer factor TF in the algorithm is ≤ 0.5, the objects collide and the algorithm enters the exploration mode. At this time, the acceleration update formula of the objects is:

[0132]

[0133] where den i ,vol i , and acc i is the density, volume and acceleration of object i, and acc mr ,den mr and vol mr are the acceleration, density and volume of the random material.

[0134] At this time, the position of the i-th object at iteration t+1 is:

[0135]

[0136] If the transfer operator TF>0.5, there is no collision between objects, and the algorithm enters the development mode. The acceleration update formula of the object in this mode is:

[0137]

[0138] Then the position update formula of the i-th object is:

[0139]

[0140] The main advantage of a hypersonic aircraft carrier platform is that it can perform fire control tasks on targets at a long distance. Therefore, in order to give full play to the performance characteristics of the aircraft carrier platform, the Archimedean algorithm fitness function of the present invention is:

[0141]

[0142] The initial simulation conditions set by the present invention are that the initial flight altitude of the carrier aircraft is 23000m, the speed is 1180m / s (Ma=4), and the average speed of the target within 10 seconds is 450m / s. The angular velocity limit of the missile tracking the target is 20° / s, and a proportional guidance method is adopted, K=4, and the missile is tracked by its own radar. The maximum detection angle of the missile radar beacon is 2rad (114°), so the absolute value of the line of sight angle is limited to less than 114°, and the relative speed required to destroy the target is 5m / s. In the Archimedean algorithm, the population size is set to 30, the maximum number of iterations is set to 1000, C1=2, C2=6, C3=2, and C4=0.5. After preliminary simulation, considering the track deviation angle that reaches the maximum distance under different line of sight angles, the above assumption is also adopted, and the target line of sight angle is limited to less than 360°, then the following attack area centered on the target can be obtained as follows Figure 6 shown.

[0143] From the above figure, we can find that the maximum attack distance appears at a line of sight angle of about 180°, which shows that when performing air-to-air missions, high-speed aircraft can often drop objects in front of or to the side of the target at a long distance, using an over-the-shoulder launch mode, and then quickly disengage with its own high speed and high maneuverability after the drop. This fire control method can better exert the extremely strong flight performance of high-speed aircraft and greatly improve the survivability of the carrier aircraft.

Claims

1. An intelligent solution method for a high-speed aircraft fire control model, characterized in that: The steps include: Step 1: Based on the existing data, establish a proxy model of the high-speed aircraft carrier delivery platform considering the flight-thrust coupling; Step 2: Construct a fire control model for a high-speed aircraft platform, including a target motion prediction model, a missile exterior trajectory model, and a fire control hit problem solving model; specifically, the following steps are included: (1) The mathematical model of missile and target maneuvers is simplified. The missile, aircraft and target are regarded as moving particles located in the same plane. The forces acting on the missile are simplified to thrust, drag and lift acting on the missile particles. The missile adopts proportional guidance law and the role of the control system is not considered. (2) Solve the missile's motion differential equations using a fast variable-step integration method; (3) Optimize program design and use modular structure to improve computing speed; Establish a mathematical model for air-to-air guidance of a high-speed aircraft carrier delivery platform, and the line-of-sight angular velocity between the missile and the target for Distance change rate for Where, V1 is the speed of the aircraft, V M —target speed, r—target distance, θ—velocity deviation angle of the carrier, q—line of sight angle, θ M —target velocity inclination, q0 is the initial value of the line of sight angle between the missile and the target, r0 is the initial value of the distance between the missile and the target, and Δt is the target flight time; The target's motion characteristic equation is described by the following equation: Where T is the target’s flight time at the moment of missile launch, θ M0 —Initial value of target velocity inclination, η M0 —Initial value of target heading angle, G M is the target maneuvering overload, is the target speed angle change rate, when G M = 0, the target moves in a straight line at a constant speed. M When ≠0, the target moves in a constant circular motion; After integrating the above formula, we get: Among them, θ M0 —Initial value of target velocity inclination, η M0 —Initial value of target heading angle; After integrating the acceleration of the missile, the velocity of the missile at any moment is Where V0 is the initial velocity of the missile, is the missile speed change rate; The missile's proportional guidance equation is: Where K is the proportional guidance coefficient, —rate of change of velocity angle; The movement of the target will not only affect the movement characteristics of the target itself, but also affect the guidance law of the missile itself: In the formula, is the rate of change of sight angle; The fire control model of the high-speed aircraft carrier delivery platform is: Based on the fast simulation calculation equations in the established fire control model, under the determined initial conditions, the maximum dynamic range of the missile is used as the initial value of the distance to calculate the parameter changes of each point of the missile during the attack process, and then the hit and miss amount is determined by the hit restriction conditions specified by the missile characteristics to determine whether the missile hits the target. If it does not hit, the distance minus the miss amount is used as the initial value for the next calculation, and the calculation is repeated in this way to finally obtain the maximum launch distance of the missile; similarly, the minimum launch distance of the missile can be obtained, and then the initial value q0 is continuously changed for calculation, and finally the missile attack area can be obtained; The program uses variable step size for integration: at the beginning of the calculation, a larger step size is selected, and when the missile approaches or hits the target, a smaller step size is used, which greatly improves the calculation speed while meeting certain accuracy conditions; Step 3: According to the flight characteristics of the high-speed aircraft, the attack area is solved by combining the Archimedes optimization algorithm, and the initial command signal of the carrier aircraft is reversed; specifically, the steps include: Step 31: Establishing Archimedes intelligent optimization algorithm model; (1) Initialize the positions of all objects O i =lb i +rand×(ub i -lb i );i=1,2,...,N (14) In the formula, O i is the i-th object in a set of N objects, lb i andub i are the lower and upper bounds of the search space respectively; Use formula (15) to initialize the volume vol of the i-th object i and density den i : Where rand is a D-dimensional vector that randomly generates a number between [0,1]. Finally, the acceleration acc of the i-th object is initialized. i ; <h2 style=";text-align:left;direction:ltr">acc<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> =lb<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> +rand×(ub<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> -lb<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> );i=1,2,...,N (16) (2) Update density and volume; The update formula for the density and volume of the i-th object in the t+1th iteration is: Among them, vol best and den best are the volume and density of the best object found so far, and rand is a uniformly distributed random number; (3) Define the transfer operator and density operator; At the beginning, the objects collide with each other. After a period of time, the objects try to reach a state of equilibrium. In the Archimedean algorithm, this is achieved through the transfer operator TF. The transfer operator TF converts the algorithm flow from exploration mode to development mode. The definition of the transfer operator TF is: The transfer operator TF gradually increases with time until it reaches 1, where t and t max are the current number of iterations and the maximum number of iterations, respectively. Similarly, the density factor d helps the algorithm to search from global to local, and d decreases as the iteration time increases; (4) Selection of exploration mode; If the transfer operator TF≤0.5, the objects collide, randomly select an object and update the acceleration of the object in the t+1th iteration using formula (20): where den i 、vol i ,acc i is the density, volume and acceleration of the ith object, and acc mr ,den mr and vol mr are the acceleration, density and volume of the random material; (5) Selection of development model; If the transfer operator TF>0.5 and there is no collision between the objects, the t+1 iteration acceleration of the object is updated using formula (21); Among them, acc best is the optimal acceleration of the object; (6) Normalized acceleration; The acceleration of the object is normalized using formula (22) to calculate the percentage change; In the formula, u and l are the normalized ranges, and the values ​​of u and l are set to 0.9 and 0.1 respectively. It defines the percentage of the step size that each individual will change. If the object is far from the global optimum, the acceleration value will be high, which means that the target will be in exploration mode; otherwise, in the exploitation phase; (7) Update location; (a) If TF≤0.5, that is, the exploration phase, the position of the i-th object at iteration t+1 is In the formula, x rand is the individual position randomly generated by the algorithm, C1 is a constant, which is assigned a value of 2 according to specific needs; (b) If TF>0.5, that is, the development stage, the position of the i-th object at iteration t+1 is C2 is a constant of 6, T increases with time, proportional to the transfer operator, T = C3 × TF, T increases with iteration in the range [C3 × 0.3, 1], and initially takes a certain percentage from the optimal position of the initial object; In formula (24), F is the sign of the object's motion direction; Where, P = 2*rand-C4; (8) Evaluation; Evaluate each individual using the objective function f, note the best solution found, and assign the individual with the best fitness x best , optimal density den best , optimal volume vol best and the optimal acceleration acc best ; Step 32, based on the fast simulation method of fire control model solution, combined with Archimedes intelligent optimization algorithm, solve the fire control command; The fitness function of the Archimedean algorithm is: During the calculation process, for each attack distance iterated using the fast simulation method, the flight status instruction of the carrier is used to calculate the fitness value of each individual in the population according to the fitness function, and then return to the algorithm to iteratively update the position, volume and acceleration information of each individual until the function reaches the allowable error or the algorithm reaches the maximum number of iterations, and then output the solved attack distance and the aircraft status instruction at the attack distance.

2. The intelligent solution method of the high-speed aircraft fire control model as claimed in claim 1, characterized in that: In step 1, based on the existing data, the proxy model of the high-speed aircraft carrier delivery platform considering the flight-thrust coupling is established, which specifically includes the following steps: Step 11, assuming that the aircraft is flying on a rotating spherical earth, derive a nonlinear mathematical model of a high-speed aircraft on a rotating circular earth; Step 12: Data sampling is performed based on the aircraft aerodynamic data, geometric parameters and propulsion coefficients in the existing database, and polynomial fitting is performed on the sampled hypersonic segment data. After fitting, the fitted polynomial proxy model is evaluated using the goodness of fit test method.

3. The intelligent solution method of the high-speed aircraft fire control model as claimed in claim 1, characterized in that: In step 2, the target motion prediction model is established by using an interactive multi-model algorithm to predict the target and decomposing the target's motion mode into a combination of the following multiple motion state models: (1)CV model; When the target is non-maneuvering, i.e., performing uniform linear motion or uniformly accelerated linear motion, the following second-order constant speed CV model is used respectively; Assume that the target moves in a uniform straight line, the target displacement is recorded as x(t), and the speed is recorded as Under the condition that the target speed has random disturbance, it is assumed that the random disturbance of speed has a mean of zero and a variance of δ 2 Gaussian white noise a(t); Get system state variables According to Newton's laws of motion, Written in matrix form: That is, the CV model is Among them, A(t) is the system matrix of the model, and B(t) is the input matrix; (2) CT model; The CT model is a constant turning model. When the target makes a constant turning motion with a constant speed but a constantly changing direction, it is described by the CT model. The discrete time domain form of the CT model is expressed as: In the formula, x k and k is the position component of the target state along the x-axis and y-axis in the Cartesian coordinate system at time k, and is the velocity component corresponding to time k, and The corresponding mean is zero and the variance is q 2 Gaussian process white noise, w is the constant turning rate, which can reflect the target maneuvering situation, p is the radar sampling interval, and the state transfer equation describes the recursive relationship of the target state from time k to time k+1; (3) Singer model; The Singer model is a first-order time-dependent model with zero acceleration mean. Assuming that the time-dependent function of the target maneuver acceleration is in the form of exponential decay, the time-dependent function Ra(τ) of the Singer model is: Where: a is the unknown parameter that determines the target maneuvering characteristics in the interval (t, t+τ), b(t) is the maneuvering acceleration, is the maneuver acceleration variance.

Images