CN113295043B - Tank vertical stabilizer neural self-adaptive control method based on BP neural network - Google Patents
Tank vertical stabilizer neural self-adaptive control method based on BP neural network Download PDFInfo
- Publication number
- CN113295043B CN113295043B CN202110558564.7A CN202110558564A CN113295043B CN 113295043 B CN113295043 B CN 113295043B CN 202110558564 A CN202110558564 A CN 202110558564A CN 113295043 B CN113295043 B CN 113295043B
- Authority
- CN
- China
- Prior art keywords
- tank
- neural network
- vertical stabilizer
- neural
- uncertainty
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F41—WEAPONS
- F41A—FUNCTIONAL FEATURES OR DETAILS COMMON TO BOTH SMALLARMS AND ORDNANCE, e.g. CANNONS; MOUNTINGS FOR SMALLARMS OR ORDNANCE
- F41A27/00—Gun mountings permitting traversing or elevating movement, e.g. gun carriages
- F41A27/06—Mechanical systems
- F41A27/18—Mechanical systems for gun turrets
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F41—WEAPONS
- F41A—FUNCTIONAL FEATURES OR DETAILS COMMON TO BOTH SMALLARMS AND ORDNANCE, e.g. CANNONS; MOUNTINGS FOR SMALLARMS OR ORDNANCE
- F41A27/00—Gun mountings permitting traversing or elevating movement, e.g. gun carriages
- F41A27/06—Mechanical systems
- F41A27/18—Mechanical systems for gun turrets
- F41A27/20—Drives for turret movements
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F41—WEAPONS
- F41H—ARMOUR; ARMOURED TURRETS; ARMOURED OR ARMED VEHICLES; MEANS OF ATTACK OR DEFENCE, e.g. CAMOUFLAGE, IN GENERAL
- F41H7/00—Armoured or armed vehicles
- F41H7/02—Land vehicles with enclosing armour, e.g. tanks
Abstract
The invention discloses a neural self-adaptive control method for a vertical tank stabilizer based on a BP (Back propagation) neural network, which comprises the following steps of: establishing a tank vertical stabilizer system nonlinear mathematical model with lumped uncertainty; predicting and compensating lumped uncertainty of the tank vertical stabilizer system through a designed BP neural network; and designing a neural self-adaptive control process based on the BP neural network based on the established mathematical model, and realizing the neural self-adaptive control on the tank vertical stabilizer through the designed neural self-adaptive control process. The invention designs the neural adaptive strategy of uncertainty factors in the tank vertical stabilizer by utilizing the special advantage of the BP neural network that any nonlinear function is approximated with any precision, so that the system can still keep better control performance under the conditions of basic vibration of a vehicle body, unmodeled dynamics and the like, and the pointing control precision of the tank vertical stabilizer on a tank gun can be effectively improved.
Description
Technical Field
The invention relates to the technical field of tank artillery control, in particular to a neural self-adaptive control method for a tank vertical stabilizer based on a BP neural network.
Background
In order to reduce adverse effects of the vehicle body fluctuating movement along with the terrain on the shooting precision of the tank gun in the tank advancing process, the tank stabilizer drives the gun barrel and the gun turret to rotate around the rotary shafts of the trunnion and the gun turret through a set of electro-hydraulic servo system and a motor servo system, and the directional control of the tank gun is realized. Traditional tank stabilizer controller generally designs according to linear steady system, and the influence of various nonlinear factors in the unable effective compensating system, along with the continuous improvement of performance requirements such as tank mobility, shooting precision, it has been difficult to satisfy the directional control demand of tank under the high maneuvering condition gradually, needs to seek the new tank stabilizer control method that can nonlinear factor influences among the compensating system urgently.
Disclosure of Invention
The invention aims to: in order to overcome the defects in the prior art, the neural self-adaptive control method of the vertical stabilizer of the tank based on the BP neural network is provided, the problem that the existing gun stabilizer controller designed according to a linear steady system cannot compensate uncertain factors in the system is solved, the control precision of the vertical stabilizer of the tank to the direction of a gun under different running working conditions can be effectively improved, and the shooting precision of the tank during running can be improved.
The technical scheme is as follows: in order to achieve the aim, the invention provides a neural self-adaptive control method of a tank vertical stabilizer based on a BP neural network, which comprises the following steps:
s1: establishing a nonlinear mathematical model of the tank vertical stabilizer system with lumped uncertainty;
s2: predicting and compensating lumped uncertainty of the tank vertical stabilizer system through a designed BP neural network;
s3: and (2) designing a neural self-adaptive control process based on the BP neural network based on the mathematical model established in the step (S1), and realizing the neural self-adaptive control on the tank vertical stabilizer through the designed neural self-adaptive control process.
Further, the method for establishing the nonlinear mathematical model of the tank vertical stabilizer system in the step S1 comprises the following steps:
a1: according to the mounting structure of the hydraulic cylinder in the tank vertical stabilizer electro-hydraulic servo system, the output displacement of the hydraulic rod is determined as follows:
in the formula (1), θ s 、θ g 、θ h Respectively is a tank aiming angleThe rotating angle of the artillery around the mass center in the high-low direction and the rotating angle of the tank body around the mass center in the high-low direction; l and Δ l are the initial length and the telescopic length of the hydraulic cylinder, respectively.
A2: according to the composition of the tank vertical stabilizer electro-hydraulic servo system, the mathematical model can be expressed as follows:
in the formula (2), m is the total mass equivalent to the piston and the load; p is a radical of 1 、p 2 The hydraulic flow and pressure of two cavities of the hydraulic cylinder are respectively subjected to oil source pressure p s Oil return pressure p r Controlling the influence of the input signal u; a. The 1 、A 2 The effective piston area of the two cavities of the hydraulic cylinder; f. of t The external load of the hydraulic cylinder; b is the effective viscous damping coefficient; d n Is unmodeled dynamics; beta is a beta e The elastic modulus of the hydraulic oil is shown; v 01 、V 02 The initial volume of two cavities of the hydraulic cylinder; c t Is the actuator leakage coefficient; g is the total gain in flow relative to the control input; s (u) is a sign function;
a3: unknown parameters existing in the vertical stabilizer system caused by the tank body base vibration and model uncertainty are defined as:
in the formula (3), the reaction mixture is,
and
respectively representing nominal values and uncertainties of the unknown parameters;
a4: and integrating the uncertainty of each unknown parameter and deducing a lumped uncertainty characterization equation of the tank vertical stabilizer system.
Further, the derivation process of the lumped uncertainty characterization equation of the tank vertical stabilizer system in step A4 is as follows:
state variables defining the system:
in the formula (4), x 3 ＝[p 1 ,p 2 ] T Then the state equation of the system is:
in formula (5):
then delta is 2 、Δ 3 I.e., a lumped uncertainty characterization equation derived for the tank vertical stabilizer system, where Δ f 3 、Δg 3 Are the components of the lumped uncertainty.
Further, the step S2 specifically includes the following steps:
b1: lumped uncertainty Δ for tank vertical stabilizer system 2 、Δ 3 Defining a prediction model as follows:
F k (X)＝O k (X)+ε k (X) (10)
in the formula (10), F k (X) is the actual value of the lumped uncertainty, ε k (X) is the prediction error of the BP neural network, O k (X) is the predicted value of the BP neural network, expressed as:
in the formula (11), X = (X) 1 ,x 2 ...x n-1 ,x n ) Is an input vector; w is a ji 、w kj Weights between the input layer and the hidden layer and between the hidden layer and the output layer are respectively; b i 、ω i Threshold values for hidden layer neurons and output layer neurons, respectively; m, n and h are the node numbers of the output layer, the input layer and the hidden layer respectively; f. of 1 (x)、f 2 (x) The activation functions of the hidden layer and the output layer respectively;
b2: for lumped uncertainty Δ 2 Defining the neural network input vector as:
X 1 ＝[x 1 ,y d ,x 2 ,χ 3 ,α 1 ] T (14)
meanwhile, the output expected value of the neural network is defined as:
training the neural network according to the updating rule of the weight and the threshold in the BP neural network based on the input and output sample data to obtain the lumped uncertainty delta 2 Predicted value of (A) O 1 (X 1 )；
B3: for lumped uncertainty Δ 3 Defining the neural network input vector as:
X 2 ＝[x 1 ,y d ,x 2 ,χ 3 ,α 1 ,α 2 ] T (16)
meanwhile, the output expected value of the neural network is defined as:
in the formula (17), α 2 Virtual control in designing a control method;
training the neural network according to the updating rule of the weight and the threshold in the BP neural network based on the input and output sample data to obtain the lumped uncertainty delta 3 Predicted value of (A) O 2 (X 2 )；
B4: in the design of the neural self-adaptive control method of the tank vertical stabilizer, lumped uncertainty delta can be utilized 2 、Δ 3 Predicted value of (A) O 1 (X 1 )、O 2 (X 2 ) The uncertainty of the system is compensated, and the control performance is improved.
Further, the update rule of the weight and the threshold in the BP neural network in step S2 is:
in equation (12), η is a correction law of constant gain.
Further, the method for designing the neural adaptive control process based on the BP neural network in step S3 specifically includes:
c1: firstly, a gyroscope is arranged on a vehicle body to measure the pitching motion condition of the vehicle body in real time, and the rotation angle theta of the vehicle body around the tank mass center in the high and low directions is determined h Based on desired gun sighting angle theta s And the angle theta of the artillery around the center of mass in the high-low direction g Calculating the desired output displacement y of the hydraulic cylinder according to equation (1) d ；
C2: the state error of the tank vertical stabilizer system is defined as follows:
c3: in the lumped uncertainty delta 2 、Δ 3 On the basis of prediction, a neural self-adaptive control program of the vertical tank stabilizer is designed and obtained.
Further, the neural adaptive control process of the tank vertical stabilizer in the step C3 specifically comprises the following steps:
in formula (19):
the invention applies the BP neural network to the neural self-adaptive control of the tank vertical stabilizer, the BP neural network has the special advantage of approaching any nonlinear function with any precision, the neural self-adaptive control method of the tank vertical stabilizer is designed based on the BP neural network, and the method has important significance for improving the pointing control precision of the tank under maneuvering conditions and effectively improving the shooting precision of the tank during traveling.
Has the beneficial effects that: compared with the prior art, the invention designs the neural adaptive strategy of the uncertainty factor in the tank vertical stabilizer by utilizing the special advantage of the BP neural network that the BP neural network approaches to any nonlinear function with any precision, solves the problem that the existing artillery stabilizer controller designed according to a linear steady system can not compensate the uncertainty factor in the system, ensures that the system can still keep better control performance under the conditions of basic vibration of a vehicle body, unmodeled dynamic state and the like, and can effectively improve the pointing control precision of the tank vertical stabilizer to the tank artillery.
Drawings
FIG. 1 is a connecting structure of a hydraulic cylinder and a tank gun;
FIG. 2 is a control schematic diagram of an electro-hydraulic servo system of a tank vertical stabilizer;
FIG. 3 is a schematic diagram of a neural adaptive control method of a tank vertical stabilizer based on a BP neural network;
FIG. 4 is a schematic diagram of an electro-hydraulic coupling kinetic model of a tank;
FIG. 5 shows a predicted value O of a first BP neural network under the action of a neural adaptive controller of a tank vertical stabilizer designed by the invention 1 (X 1 ) And the lumped uncertainty actual value delta 2 A comparative graph of (a);
FIG. 6 shows a predicted value O of a second BP neural network under the action of a neural adaptive controller of a vertical stabilizer of a tank designed by the invention 2 (X 2 ) And the lumped uncertainty actual value delta 3 A comparative graph of (a);
FIG. 7 shows the actual output displacement y and the expected output displacement y of the hydraulic rod under the action of the neural self-adaptive controller of the tank vertical stabilizer designed by the invention d A comparative graph of (a);
FIG. 8 is a diagram showing a comparison of vertical angular displacements of a tank cradle under the action of a neural adaptive controller of a tank vertical stabilizer, a backstepping controller without a BP neural network and a PID controller designed by the invention.
Detailed Description
The present invention is further illustrated by the following figures and specific examples, which are to be understood as illustrative only and not as limiting the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.
The invention provides a neural self-adaptive control method of a tank vertical stabilizer based on a BP neural network, which comprises the following steps of:
step 1: establishing a nonlinear mathematical model of the tank vertical stabilizer system with lumped uncertainty, which comprises the following steps:
referring to fig. 1, according to the mounting structure of the hydraulic cylinder in the tank vertical stabilizer electro-hydraulic servo system, the output displacement of the hydraulic rod is determined as follows:
in the formula (1), θ s 、θ g 、θ h The tank sighting angle, the rotation angle of the gun around the mass center in the high-low direction and the rotation angle of the vehicle body around the tank mass center in the high-low direction are respectively; l, delta l initial length and telescopic length of the hydraulic cylinder.
Referring to fig. 2, the mathematical model of the tank vertical stabilizer electro-hydraulic servo system can be expressed as:
in the formula (2), m is the total mass equivalent to the piston and the load; p is a radical of 1 、p 2 The hydraulic flow and pressure of two cavities of the hydraulic cylinder are respectively subjected to the pressure p of the oil source s Oil return pressure p r Controlling the influence of the input signal u; a. The 1 、A 2 The effective piston area of the two cavities of the hydraulic cylinder; f. of t An external load for the hydraulic cylinder; b is the effective viscous damping coefficient; d n Is unmodeled dynamics; beta is a beta e The elastic modulus of the hydraulic oil is measured; v 01 、V 02 The initial volume of the two cavities of the hydraulic cylinder; c t Is the actuator leakage coefficient; g is the total gain in flow relative to the control input; s (u) is a sign function.
Unknown parameters existing in the vertical stabilizer system caused by tank body base vibration and model uncertainty are defined as:
in the formula (3), the reaction mixture is,
and
the nominal values and uncertainties of these unknown parameters are represented separately.
And integrating the uncertainty of each unknown parameter and deducing a lumped uncertainty characterization equation of the tank vertical stabilizer system. State variables defining the system:
in the formula (4), x 3 ＝[p 1 ,p 2 ] T Then the state equation of the system is:
in formula (5):
then a 2 、Δ 3 I.e. the derived lumped uncertainty characterization equation of the tank vertical stabilizer system. Wherein, Δ f 3 、Δg 3 Are the components of the lumped uncertainty.
And 2, step: a series of BP neural networks are designed to predict and compensate the lumped uncertainty of the tank vertical stabilizer system, and the method specifically comprises the following steps:
lumped uncertainty Δ for tank vertical stabilizer system 2 、Δ 3 Defining a prediction model as follows:
F k (X)＝O k (X)+ε k (X) (10)
in the formula (10), F k (X) is the actual value of the lumped uncertainty, ∈ k (X) is the prediction error of the BP neural network, O k (X) is the predictor of BP neural network, expressed as
In formula (11), X = (X) 1 ,x 2 ...x n-1 ,x n ) Is an input vector; w is a ji 、w kj Weights between the input layer and the hidden layer and between the hidden layer and the output layer are respectively; b i 、ω i Threshold values for hidden layer neurons and output layer neurons, respectively; m, n and h are the node numbers of the output layer, the input layer and the hidden layer respectively; f. of 1 (x)、f 2 (x) The activation functions of the hidden layer and the output layer, respectively.
The update rule of the weight and the threshold in the BP neural network in this embodiment is as follows:
η in the equation (12) is a correction law of constant gain.
For lumped uncertainty Δ 2 The number of input layer nodes n =5, and the number of output layer nodes m =1.
Defining the neural network input vector as:
X 1 ＝[x 1 ,y d ,x 2 ,χ 3 ,α 1 ] T (14)
meanwhile, the output expected value of the neural network is defined as:
based on input and output sample data, training the neural network according to the equations (12) and (13) to obtain the lumped uncertainty delta 2 Predicted value of (A) O 1 (X 1 )。
For lumped uncertainty Δ 3 The number of input layer nodes n =6, and the number of output layer nodes m =1.
Defining the neural network input vector as:
X 2 ＝[x 1 ,y d ,x 2 ,χ 3 ,α 1 ,α 2 ] T (16)
meanwhile, the output expected value of the neural network is defined as:
in the formula (17), α 2 Virtual control in the design of a control method.
Training the neural network according to the equations (12) and (13) based on input and output sample data to obtain lumped uncertainty delta 3 Predicted value of (A) O 2 (X 2 )。
Neural self-adaptive control of tank vertical stabilizerIn the design of the method, lumped uncertainty delta can be used 2 、Δ 3 Predicted value of (A) O 1 (X 1 )、O 2 (X 2 ) The uncertainty of the system is compensated, and the control performance is improved.
And step 3: a tank vertical stabilizer neural self-adaptive control method based on a BP neural network is designed based on a mathematical model, and comprises the following steps:
firstly, a gyroscope is arranged on a vehicle body to measure the pitching motion condition of the vehicle body in real time, and the rotation angle theta of the vehicle body around the tank mass center in the high and low directions is determined h Based on the desired gun aiming angle theta s And the angle theta of the artillery around the center of mass in the high-low direction g Calculating the desired output displacement y of the hydraulic cylinder according to equation (1) d 。
The state error of the tank vertical stabilizer system is defined as follows:
in the pair of lumped uncertainty Δ 2 、Δ 3 On the basis of prediction, the designed neural self-adaptive control method of the vertical stabilizer of the tank comprises the following steps:
in formula (19):
the embodiment also provides a neural self-adaptive controller of the tank vertical stabilizer based on the BP neural network, which comprises a network interface, a memory and a processor; the network interface is used for receiving and sending signals in the process of receiving and sending information with other external network elements; a memory for storing computer program instructions executable on the processor; a processor for, when executing the computer program instructions, performing the steps of the consensus method described above.
The present embodiment also provides a computer storage medium storing a computer program that when executed by a processor can implement the method described above. The computer-readable medium may be considered tangible and non-transitory. Non-limiting examples of a non-transitory tangible computer-readable medium include a non-volatile memory circuit (e.g., a flash memory circuit, an erasable programmable read-only memory circuit, or a mask read-only memory circuit), a volatile memory circuit (e.g., a static random access memory circuit or a dynamic random access memory circuit), a magnetic storage medium (e.g., an analog or digital tape or hard drive), and an optical storage medium (e.g., a CD, DVD, or blu-ray disc), among others. The computer program includes processor-executable instructions stored on at least one non-transitory tangible computer-readable medium. The computer program may also comprise or rely on stored data. The computer programs may include a basic input/output system (BIOS) that interacts with the hardware of the special purpose computer, a device driver that interacts with specific devices of the special purpose computer, one or more operating systems, user applications, background services, background applications, and the like.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Based on the scheme, in order to verify the effect of the scheme, the embodiment applies the neural adaptive controller of the vertical stabilizer of the tank in an example, as shown in fig. 4, the neural adaptive controller is compiled into an S function form in Matlab/Simulink, and the tank electro-hydraulic coupling dynamic model is established based on 20km/h and D levelThe control performance of the control method is verified under the working condition of 0-degree aiming angle of the road surface. Wherein the control parameter is set as k 1 ＝500,k 2 ＝400,k 3 =250. Setting eta for the first neural network 1 =0.1, setting η for the second neural network 2 =0.01, and furthermore the initial weights w of the two neural networks ji 、w kj And a threshold value b j 、ω k Are all set to zero, the number of hidden layer neurons is all set to 5, and the activation function selects f (x) = tansig (x). In addition, a PID controller (k) is selected p ＝300，k i ＝60，k d = 1), a back-stepping controller without a BP neural network (parameter settings are the same as those of the neural adaptive controller designed according to the present invention) as a comparison.
FIG. 5 shows a first BP neural network O under the action of a neural adaptive controller of a tank vertical stabilizer designed by the invention 1 (X 1 ) To the lumped uncertainty actual value delta 2 The predicted situation of (2); FIG. 6 shows a second BP neural network O 2 (X 2 ) To the lumped uncertainty actual value delta 3 The predicted situation of (2); FIG. 7 shows the actual output displacement y of the hydraulic rod of the vertical stabilizer of the tank versus the expected output displacement y d The tracking situation of (1). The figure shows that the neural self-adaptive controller of the tank vertical stabilizer, which is designed by the invention, can effectively predict and supplement the influence of uncertainty factors in a system, so that the actual output displacement of a hydraulic rod can accurately track the expected output displacement.
FIG. 8 is a diagram showing a comparison of vertical angular displacements of a tank cradle under the action of a neural adaptive controller of a tank vertical stabilizer, a backstepping controller without a BP neural network and a PID controller designed by the invention. Compared with a backstepping controller and a PID (proportion integration differentiation) controller which do not contain a BP (back propagation) neural network, the neural self-adaptive controller of the vertical stabilizer of the tank, which is designed by the invention, can enable the cradle to be stabilized near an expected 0-degree aiming angle, and improves the control precision of the vertical stabilizer of the tank on the direction of a gun in the vertical direction.
Claims (2)
1. A neural self-adaptive control method of a tank vertical stabilizer based on a BP neural network is characterized by comprising the following steps:
s1: establishing a nonlinear mathematical model of the tank vertical stabilizer system with lumped uncertainty;
s2: predicting and compensating lumped uncertainty of the tank vertical stabilizer system through a designed BP neural network;
s3: designing a neural self-adaptive control program based on the BP neural network based on the mathematical model established in the step S1, and realizing the neural self-adaptive control on the vertical stabilizer of the tank through the designed neural self-adaptive control program;
the method for establishing the nonlinear mathematical model of the tank vertical stabilizer system in the step S1 comprises the following steps:
a1: according to the mounting structure of the hydraulic cylinder in the tank vertical stabilizer electro-hydraulic servo system, the output displacement of the hydraulic rod is determined as follows:
in the formula (1), θ s 、θ g 、θ h The tank sighting angle, the rotation angle of the gun around the mass center in the high-low direction and the rotation angle of the vehicle body around the mass center of the tank in the high-low direction are respectively; l and delta l are respectively the initial length and the telescopic length of the hydraulic cylinder; a is the distance between the center point of the trunnion and the installation position of the hydraulic cylinder on the turret; l. the d The distance between the center point of the trunnion and the installation position of the hydraulic cylinder on the artillery is taken as the distance; alpha is the angle of the diagonal of the side where the hydraulic cylinder is located in a triangle with the mounting central points of the two ends of the hydraulic cylinder and the central point of the trunnion as the vertexes;
a2: according to the composition of the tank vertical stabilizer electro-hydraulic servo system, the mathematical model can be expressed as follows:
in the formula (2), m is the total mass equivalent to the piston and the load; p is a radical of formula 1 、p 2 Are respectively hydraulicPressure in two chambers of cylinder, subject to source pressure p s Oil return pressure p r Controlling the influence of the input signal u; a. The 1 、A 2 The effective piston area of the two cavities of the hydraulic cylinder; f. of t The external load of the hydraulic cylinder; b is the effective viscous damping coefficient; d n Is unmodeled dynamics; beta is a e The elastic modulus of the hydraulic oil is measured; v 01 、V 02 The initial volume of two cavities of the hydraulic cylinder; c t Is the actuator leakage coefficient; g is the total gain in flow relative to the control input; s (u) is a sign function;
a3: unknown parameters existing in the vertical stabilizer system caused by the tank body base vibration and model uncertainty are defined as:
in the formula (3), the reaction mixture is,
and
respectively representing nominal values and uncertainties of the unknown parameters;
a4: integrating the uncertainty of each unknown parameter, and deducing a lumped uncertainty characterization equation of the tank vertical stabilizer system;
the derivation process of the lumped uncertainty characterization equation of the tank vertical stabilizer system in the step A4 is as follows:
state variables defining the system:
in the formula (4), x 3 ＝[x 3 ,x 4 ] T ＝[p 1 ,p 2 ] T Then the state equation of the system is:
in formula (5):
then a 2 、Δ 3 I.e., a lumped uncertainty characterization equation derived for the tank vertical stabilizer system, where Δ f 3 、Δg 3 As a component of the lumped uncertainty;
the step S2 specifically includes the steps of:
b1: lumped uncertainty Δ for tank vertical stabilizer system 2 、Δ 3 Defining the following prediction model:
F k (X I )＝O k (X I )+ε k (X I ) (10)
in the formula (10), F k (X I ) For the actual value of the lumped uncertainty, ε k (X I ) Prediction error for BP neural network, O k (X I ) Is a predicted value of the BP neural network, and is expressed as:
in formula (11), x Ii ＝(x I1 ,x I2 ...x I(n-1) ,x In ) Is an input vector; w is a ji 、w kj Weights between the input layer and the hidden layer and between the hidden layer and the output layer are respectively; b j 、ω k Threshold values for hidden layer neurons and output layer neurons, respectively; m, n and h are the node numbers of the output layer, the input layer and the hidden layer respectively; f. of 1 (·)、f 2 () activation functions for the hidden layer and the output layer, respectively;
b2: for lumped uncertainty Δ 2 Defining the neural network input vector as:
X 1 ＝[x 1 ,y d ,x 2 ,χ 3 ,α 1 ] T (14)
wherein, y d A desired output displacement for the hydraulic cylinder;
meanwhile, the output expected value of the neural network is defined as:
training the neural network according to the updating rule of the weight and the threshold in the BP neural network based on the input and output sample data to obtain the lumped uncertainty delta 2 Predicted value of (A) O 1 (X 1 )；
B3: for lumped uncertainty Δ 3 Defining the neural network input vector as:
X 2 ＝[x 1 ,y d ,x 2 ,χ 3 ,α 1 ,α 2 ] T (16)
meanwhile, the output expected value of the neural network is defined as:
in the formula (17), α 2 Designed for control methodVirtual control;
training the neural network according to the updating rule of the weight and the threshold in the BP neural network based on the input and output sample data to obtain the lumped uncertainty delta 3 Predicted value of (A) O 2 (X 2 )；
B4: in the neural self-adaptive control method design of the tank vertical stabilizer, lumped uncertainty delta can be utilized 2 、Δ 3 Predicted value of (A) O 1 (X 1 )、O 2 (X 2 ) Compensating for system uncertainty;
the method for designing the neural adaptive control process based on the BP neural network in the step S3 specifically comprises the following steps:
c1: firstly, a gyroscope is arranged on a vehicle body to measure the pitching motion condition of the vehicle body in real time, and the rotation angle theta of the vehicle body around the tank mass center in the high and low directions is determined h Based on the desired gun aiming angle theta s And the angle theta of the artillery around the center of mass in the high-low direction g Calculating the desired output displacement y of the hydraulic cylinder according to equation (1) d ；
C2: the state error of the tank vertical stabilizer system is defined as follows:
c3: in the pair of lumped uncertainty Δ 2 、Δ 3 On the basis of prediction, a neural self-adaptive control program of the vertical stabilizer of the tank is designed and obtained;
the neural self-adaptive control process of the tank vertical stabilizer in the step C3 specifically comprises the following steps:
wherein k is 1 、k 2 、k 3 Designing parameters for the controller;
in formula (19):
2. the neural adaptive control method for the vertical stabilizer of the tank based on the BP neural network as claimed in claim 1, wherein the update rule of the weight and the threshold in the BP neural network in the step S2 is as follows:
in the equation (12), η is a correction law of constant gain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110558564.7A CN113295043B (en) | 2021-05-21 | 2021-05-21 | Tank vertical stabilizer neural self-adaptive control method based on BP neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110558564.7A CN113295043B (en) | 2021-05-21 | 2021-05-21 | Tank vertical stabilizer neural self-adaptive control method based on BP neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113295043A CN113295043A (en) | 2021-08-24 |
| CN113295043B true CN113295043B (en) | 2022-10-14 |
Family
ID=77323692
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110558564.7A Active CN113295043B (en) | 2021-05-21 | 2021-05-21 | Tank vertical stabilizer neural self-adaptive control method based on BP neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113295043B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07210207A (en) * | 1994-01-17 | 1995-08-11 | Hitachi Ltd | Adaptive controller |
| CN102981407A (en) * | 2012-11-29 | 2013-03-20 | 北京理工大学 | Tank gun control servo control method based on auto-disturbance rejection control technology |
| CN103411479A (en) * | 2013-08-01 | 2013-11-27 | 北京理工大学 | Sliding mode and active disturbance rejection based composite control method of tank gun controlling system |
| CN104850889A (en) * | 2014-11-19 | 2015-08-19 | 北京航空航天大学 | Airplane rotation actuator drive unit adaptive fault detection, isolation and confidences assessment method |
| CN104914851A (en) * | 2015-05-21 | 2015-09-16 | 北京航空航天大学 | Adaptive fault detection method for airplane rotation actuator driving device based on deep learning |
| CN112818471A (en) * | 2021-02-18 | 2021-05-18 | 西北工业大学 | Uncertainty analysis method for transmission between tank travels |
-
2021
- 2021-05-21 CN CN202110558564.7A patent/CN113295043B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07210207A (en) * | 1994-01-17 | 1995-08-11 | Hitachi Ltd | Adaptive controller |
| CN102981407A (en) * | 2012-11-29 | 2013-03-20 | 北京理工大学 | Tank gun control servo control method based on auto-disturbance rejection control technology |
| CN103411479A (en) * | 2013-08-01 | 2013-11-27 | 北京理工大学 | Sliding mode and active disturbance rejection based composite control method of tank gun controlling system |
| CN104850889A (en) * | 2014-11-19 | 2015-08-19 | 北京航空航天大学 | Airplane rotation actuator drive unit adaptive fault detection, isolation and confidences assessment method |
| CN104914851A (en) * | 2015-05-21 | 2015-09-16 | 北京航空航天大学 | Adaptive fault detection method for airplane rotation actuator driving device based on deep learning |
| CN112818471A (en) * | 2021-02-18 | 2021-05-18 | 西北工业大学 | Uncertainty analysis method for transmission between tank travels |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113295043A (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108614426B (en) | Multi-mobile-robot formation robust control method based on disturbance observer | |
| CN107992681B (en) | Composite control method for active front wheel steering system of electric automobile | |
| CN107415939B (en) | Steering stability control method for distributed driving electric automobile | |
| CN108228975B (en) | Motor servo system parameter identification method and backlash elimination control method | |
| CN108107731B (en) | Automobile stability control method based on tire nonlinear characteristics | |
| CN110928182B (en) | Robust self-adaptive repetitive control method of hydraulic servo system based on state estimation | |
| CN109976161B (en) | Finite time optimization tracking control method of uncertain nonlinear system | |
| CN107831761A (en) | A kind of path tracking control method of intelligent vehicle | |
| CN104638999B (en) | Dual-servo-motor system control method based on segmentation neutral net friction model | |
| CN112558468B (en) | Launching platform adaptive robust output feedback control method based on double observers | |
| CN110362110B (en) | Fixed self-adaptive neural network unmanned aerial vehicle track angle control method | |
| CN110162073B (en) | Sectional sine attitude maneuver trajectory planning method adaptive to boundary constraint | |
| CN107165892B (en) | A kind of sliding-mode control of electrohydraulic servo system | |
| CN114047773A (en) | Underwater ore collection robot backstepping sliding mode self-adaptive attitude control method based on extended state observer | |
| CN113295043B (en) | Tank vertical stabilizer neural self-adaptive control method based on BP neural network | |
| WO2023010828A1 (en) | Lateral control of driverless device | |
| CN107765548A (en) | Flat pad high-precision motion control method based on double observers | |
| CN109995278B (en) | Motor servo system self-adjustment control method considering input limitation | |
| CN113741172A (en) | Slip rate active disturbance rejection control method and controller based on bandwidth parameter setting | |
| CN115182407A (en) | Method and device for controlling arm support, controller and engineering machinery | |
| CN113791539A (en) | Self-adaptive neural network control method based on nonlinear active suspension | |
| Lei et al. | Vehicle direct yaw moment control based on tire cornering stiffness estimation | |
| CN108726380B (en) | Finite time tracking controller design method based on drive lacking bridge crane | |
| CN117369287B (en) | Lu Bangchao spiral sliding mode control method for multi-axle vehicle | |
| CN115618586A (en) | Design method and device of controller and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |