Method for controlling preset time of self-adaptive neural network of electro-hydraulic system

Abstract

The invention discloses an electro-hydraulic system adaptive neural network preset time control method, which integrates a neural network real-time learning technology, innovatively introduces a preset time performance function, and designs an adaptive neural network preset time performance controller. For the electro-hydraulic system position tracking control problem, it can be ensured that system output transient performance and steady-state performance converge to a specified performance index range within a preset time, safe and reliable operation of the system is ensured, the unknown dynamic state of the system can be learned in real time through a neural network, high-precision motion control performance is achieved, and the control precision is improved. And the problem of differential explosion in traditional backstepping control of the electro-hydraulic system can be avoided, and the influence of measurement noise on the control precision is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of electromechanical servo control, and in particular to an adaptive neural network preset time control method (ANNPTC) for an electro-hydraulic system. Background Art

[0002] Electro-hydraulic systems, with their high power density, high force / torque output, and fast dynamic response, play a crucial role in robotics, heavy machinery, high-performance load testing equipment, and other fields. Electro-hydraulic systems are typically nonlinear systems, characterized by numerous nonlinear characteristics and modeling uncertainties. These nonlinear characteristics include input nonlinearities such as hysteresis and saturation, proportional servo valve flow and pressure nonlinearities, and friction nonlinearities. Modeling uncertainties include parameter uncertainty and uncertainty nonlinearities. Parameter uncertainty primarily relates to load mass, actuator viscous friction coefficient, leakage coefficient, proportional servo valve flow gain, and hydraulic fluid elastic modulus. Uncertain nonlinearities primarily include unmodeled friction dynamics, high-order system dynamics, external disturbances, and unmodeled leakage. As electro-hydraulic systems develop towards higher precision and higher frequency response, the impact of these nonlinear characteristics on system performance becomes increasingly significant. Furthermore, modeling uncertainty can lead to instability or degradation of controllers designed based on the nominal system model. Therefore, the nonlinear characteristics and modeling uncertainty of electro-hydraulic systems are significant factors limiting system performance. With the continuous advancement of technology in the industrial and defense sectors, controllers designed based on traditional linear theory are no longer able to meet the high-performance requirements of these systems. Therefore, it is necessary to develop more advanced nonlinear control strategies that address the nonlinear characteristics of electro-hydraulic systems.

[0003] Many methods have been proposed to address the nonlinear control problems of electro-hydraulic systems. Adaptive control methods are very effective for dealing with parameter uncertainty and can achieve asymptotic tracking steady-state performance. However, they are inadequate for uncertain nonlinearities such as external load disturbances. Excessively large uncertain nonlinearities can cause system instability. Actual electro-hydraulic systems all have uncertain nonlinearities, so adaptive control methods cannot achieve high-precision control performance in practical applications. As a robust control method, classical sliding mode control can effectively handle any bounded modeling uncertainty and achieve asymptotic tracking steady-state performance. However, the discontinuous controller designed for classical sliding mode control is prone to chattering of the sliding mode surface, which degrades the tracking performance of the system. To simultaneously address both parameter uncertainty and uncertain nonlinearity, adaptive robust control methods have been proposed. These control methods can achieve deterministic transient and steady-state performance in the presence of both modeling uncertainties. To achieve high-precision tracking performance, the feedback gain must be increased to reduce the tracking error. However, due to measurement noise, excessively large feedback gain often results in chattering of the control input, which degrades control performance and may even cause system instability. Summary of the Invention

[0004] The purpose of the present invention is to provide an electro-hydraulic system adaptive neural network preset time control method, which can not only ensure that the system output transient performance and steady-state performance converge to a specified performance index range within a preset time, ensuring the safe and reliable operation of the system, but also utilize the neural network to learn the unknown dynamics of the system in real time to achieve high-precision motion control performance, and avoid the differential explosion problem in traditional backstepping control of electro-hydraulic systems, reducing the impact of measurement noise on control accuracy.

[0005] The technical solution to achieve the purpose of the present invention is: a method for controlling preset time of an electro-hydraulic system using an adaptive neural network, comprising the following steps:

[0006] Step 1: Establish a mathematical model of the electro-hydraulic system and proceed to step 2.

[0007] Step 2: Based on the mathematical model of the electro-hydraulic system, design an adaptive neural network preset time performance controller and proceed to step 3.

[0008] Step 3: Use Lyapunov stability theory to prove the stability of the adaptive neural network preset time performance controller, and obtain the result that the system tracking error is asymptotically stable.

[0009] Compared with the prior art, the present invention has the following significant advantages: (1) ensuring that the transient performance and steady-state performance of the system output converge to the specified performance index range within the preset time, ensuring the safe and reliable operation of the system; (2) using the neural network to learn the unknown dynamics of the system in real time to achieve high-precision motion control performance; (3) avoiding the differential explosion problem in the traditional backstepping control of the electro-hydraulic system, reducing the influence of measurement noise on the control accuracy, and the simulation results verify its effectiveness; (4) realizing the electro-hydraulic system adaptive neural network preset time control with system output preset time performance regulation, strong anti-interference ability and high tracking performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a schematic diagram of the principle of the electro-hydraulic system adaptive neural network preset time control method of the present invention.

[0011] Figure 2 It is a schematic diagram of the electro-hydraulic system principle of the present invention.

[0012] Figure 3 It is a curve diagram of the tracking process of the system output to the expected instruction under the action of the ANNPTC controller designed by the present invention.

[0013] Figure 4 It is a graph showing the tracking error of the system changing with time under the action of the ANNPTC controller designed by the present invention.

[0014] Figure 5 This is a comparison curve of the tracking errors of the system under the action of the ANNPTC controller designed by the present invention and the traditional PID controller.

[0015] Figure 6 It is a control input curve diagram of the system under the action of the ANNPTC controller designed by the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] Combine Figure 1 and Figure 2 The present invention provides an electro-hydraulic system adaptive neural network preset time control method, comprising the following steps:

[0018] Step 1: Establish a mathematical model of the electro-hydraulic system.

[0019] Step 1-1: The electro-hydraulic system is applied to the linear motion of large-scale industrial heavy-load mechanical equipment, wherein the load is fixedly connected to the piston rod on the hydraulic cylinder, and the electro-hydraulic proportional servo valve controls the movement of the piston rod on the hydraulic cylinder, thereby driving the load to move;

[0020] According to Newton's second law, the force balance equation of the electro-hydraulic system is:

[0021]

[0022] In formula (1), m represents the mass of the load, y represents the displacement of the hydraulic cylinder piston rod, Indicates the speed of the hydraulic cylinder piston rod, It represents the acceleration of the hydraulic cylinder piston rod, A represents the effective working area of ​​the hydraulic cylinder piston, and the oil pressure difference P between the inlet and outlet oil chambers on both sides of the cylinder L =P1-P2, P1 represents the oil pressure in the hydraulic cylinder inlet chamber, P2 represents the oil pressure in the hydraulic cylinder outlet chamber, represents the friction force on the load, d1(t) represents the unmodeled mechanical disturbance of the system, and t represents time.

[0023] Then formula (1) can be rewritten as:

[0024]

[0025] In the electro-hydraulic system, ignoring the leakage of oil from the cylinder, the dynamic pressure equation is:

[0026]

[0027] Formula (3), β e Indicates the effective elastic modulus of the oil, C t Indicates the leakage coefficient of the hydraulic cylinder, the control volume of the oil inlet chamber V1=V 01 +Ay, the control volume of the oil outlet chamber V2=V 02 -Ay, V 01 Indicates the initial volume of the oil inlet chamber, V 02 represents the initial volume of the oil outlet cavity, Q1 represents the flow rate of the oil inlet cavity, Q2 represents the flow rate of the oil outlet cavity, q1 represents the unmodeled interference of Q1, q2 represents the unmodeled interference of Q2, Indicates P L The first derivative of .

[0028] Q1 and Q2 are respectively related to the displacement of the spool of the electro-hydraulic proportional servo valve x v There are the following relationships:

[0029]

[0030] Among them, the electro-hydraulic proportional servo valve coefficient C d represents the flow coefficient of the electro-hydraulic proportional servo valve, w0 represents the valve core area gradient of the electro-hydraulic proportional servo valve, ρ represents the oil density, P s Indicates the oil supply pressure, P r represents the return oil pressure, s(·) represents the function of the intermediate variable ·, and is defined as:

[0031]

[0032] Ignoring the dynamics of the electro-hydraulic proportional servo valve spool, assume that the control input u acting on the spool and the spool displacement x v Proportional relationship, that is, satisfying x v =k i u, where k i represents the voltage-spool displacement gain coefficient, so Equation (4) is rewritten as:

[0033]

[0034] Formula (6), intermediate variable k u =k q k i , intermediate variables Intermediate variables

[0035] Step 1-2, define state variables: Among them, the intermediate variable x1=y, the intermediate variable Intermediate variable x3 = AP L / m, then transform equation (2) into the state equation:

[0036]

[0037] Formula (7), represents the first-order derivative of x1, represents the first-order derivative of x2, represents the first-order derivative of x3, the unknown dynamics of the system Δ1=d1(t) / m, and the intermediate variable F(x2)=F f (x2) / m, intermediate variable Intermediate variables Intermediate variables System unknown dynamics T stands for transpose.

[0038] To facilitate controller design, the following assumptions are made:

[0039] Assumption 1: The system is expected to track the position command x d It is second-order continuous, and the system expects that the position command, velocity command and acceleration command are all bounded.

[0040] Assumption 2: The unknown dynamics Δ1 and Δ2 of the system satisfy:

[0041]

[0042] In formula (8), δ1 and δ2 are both unknown positive constants.

[0043] Go to step 2.

[0044] Step 2: Based on the mathematical model of the electro-hydraulic system, design an adaptive neural network preset time performance controller. The specific steps are as follows:

[0045] Step 2-1: To facilitate controller design, define the error ζ1 = z1 / Φ(t), where the system tracking error z1 = x1-x d , x d is the position command that the system expects to track, Φ(t) represents the preset time performance function, in order to facilitate the system state x1 to track the expected position command x as accurately as possible under the designed controller drive d , and let z1 always satisfy |z1|<Φ(t), and ensure that the tracking error ζ1 tends to 0, as follows:

[0046] The tracking error z1 satisfies the preset range:

[0047] -Φ(t)<z1<Φ(t) (9),

[0048] Formula (9) presets the time performance function Φ(t), which is in the form of

[0049]

[0050] Among them, Φ0 represents a constant positive, Φ ∞ represents a constant positive, c represents a constant positive, T f Represents a constant that is always positive and satisfies the condition Φ0>Φ ∞ , tan represents the tangent function, and π represents the pi constant.

[0051] Design the following nonlinear filter:

[0052]

[0053] Formula (11), filter gain τ1>0, v1 represents the virtual control of x2, and the filter error ω1 of v1=v 1f -v1,v 1f represents the filtered signal of v1, σ(t) represents a function that is always positive and satisfies Where ν represents the integration variable, represents a constant that is always positive, represents the first-order derivative of v1, Indicates v 1f The first derivative of The upper bound l1 of represents a positive constant.

[0054] Taking the derivative of the tracking error ζ1, we get:

[0055]

[0056] in, represents the first-order derivative of ζ1, v 1f The error with x2 is ζ2 = x2 - v 1f ; Represents x d The first derivative of ; Represents the first-order derivative of the preset time performance function Φ(t).

[0057] Select Lyapunov function We can get:

[0058]

[0059] in, represents the first-order derivative of L1.

[0060] Design virtual control v1 as:

[0061]

[0062] Formula (14), gain k1>0, then

[0063]

[0064] Step 2-2, design the following nonlinear filter:

[0065]

[0066] Formula (16), filter gain τ2>0, v2 represents the virtual control of x3, and the filter error ω2 of v2=v 2f -v2,v 2f represents the filtered signal of v2, represents the first-order derivative of v2, Indicates v 2f The first derivative of The upper bound l2 represents a positive constant.

[0067] Derivative of the error ζ2 yields:

[0068]

[0069] in, represents the first-order derivative of ζ2, v 2f The error with x3 is ζ3 = x3 - v 2f .

[0070] Select Lyapunov function We can get:

[0071]

[0072] in, represents the first-order derivative of L2.

[0073] Design virtual control v2 as:

[0074]

[0075] Formula (19), gain k2>0, v s represents an intermediate variable, γ1 represents a positive constant, represents the estimated value of F(x2), The specific form is:

[0076]

[0077] Formula (20), W a The estimated value of W a represents the weights of the neural network, represents the activation function of the neural network, X a Represents the input of the neural network.

[0078] The weight update law of the neural network is designed as:

[0079]

[0080] Formula (21), express The first derivative of , Γ a Represents the weight gain matrix of the neural network, and Proj represents the discontinuous mapping function.

[0081] Substituting formula (19) into formula (18), we get:

[0082]

[0083] Formula (22), the neural network weight W a The estimated error ω a Represents the approximation error of the neural network.

[0084] Step 2-3, take the derivative of ζ3 and get:

[0085]

[0086] in, represents the first-order derivative of ζ3.

[0087] Select Lyapunov function have to:

[0088]

[0089] in, represents the first-order derivative of L3.

[0090] According to formula (24), the control input of the valve core, that is, the adaptive neural network preset time performance controller u, is:

[0091]

[0092] Formula (25), gain k3>0, u s represents an intermediate variable, and γ2 represents a positive constant.

[0093] Substituting formula (25) into formula (24) yields:

[0094]

[0095] Go to step 3.

[0096] Step 3: Use Lyapunov stability theory to prove the stability of the adaptive neural network preset time performance controller, and obtain the result that the system tracking error is asymptotically stable, as follows:

[0097] The Lyapunov function L is defined as follows:

[0098]

[0099] Derivative (27) and substitute (11), (16), (21) and (26) into it to obtain:

[0100]

[0101] in, represents the first derivative of L.

[0102] Taking into account |Δ1|+ω a ≤γ1 and |Δ2|≤γ2, we can get the expression:

[0103]

[0104] Notice

[0105]

[0106] Substituting formula (30) into formula (29), we can get

[0107]

[0108] Formula (31), intermediate variable Integrating both sides of equation (31) we can get:

[0109]

[0110] From Equation (32), we can see that L is bounded and Ψ is integrally bounded. It can be concluded that all signals in the system are bounded. Therefore, Ψ is uniformly continuous. According to Barbalat’s lemma, when time tends to positive infinity, the tracking error z1 tends to 0.

[0111] Therefore, it is concluded that by adjusting the gains k1, k2, k3 and the filter gains τ1, τ2, the adaptive neural network preset time performance controller designed for the electro-hydraulic system can enable the system to innovatively obtain the result that the tracking error converges to 0 asymptotically. The principle diagram of the electro-hydraulic system adaptive neural network preset time performance controller is shown in the figure. Figure 1 shown.

[0112] Example

[0113] In order to evaluate the performance of the designed controller, the physical parameters of the electro-hydraulic system in the simulation are shown in Table 1:

[0114] Table 1 System physical parameters

[0115] Physical parameters Numerical Physical parameters Numerical <![CDATA[A(m 2 )]]> <![CDATA[2×10 -4 ]]> <![CDATA[β e (Well)]]> <![CDATA[2×10 8 ]]> m(kg) 40 B (N·s / m) 80 <![CDATA[C t (m 5 / (N·s))]]> <![CDATA[7×10 -12 ]]> <![CDATA[k u (m / V)]]> <![CDATA[4×10 -8 ]]> <![CDATA[V 01 (m 3 )]]> <![CDATA[1×10 -3 ]]> <![CDATA[V 02 (m 3 )]]> <![CDATA[1×10 -3 ]]> <![CDATA[P s (MPa)]]> 7 <![CDATA[P r (MPa)]]> 0

[0116] Given a system with the expected instruction x d =0.02sin(πt)×(1-e -t )m.

[0117] The following controllers are used for comparison in the simulation:

[0118] Adaptive neural network preset time control method (ANNPTC) for electro-hydraulic system: take gains k1=150, k2=60, k3=10, Γ a =diag{100,100,100,100,100},τ1=2000,τ2=2000,Φ0=1,Φ ∞ =0.02, c=0.5, T f =10.

[0119] PID controller: The steps for selecting PID controller parameters are: first, ignoring the nonlinear dynamics of the electro-hydraulic system, obtain a set of controller parameters through the PID parameter self-tuning function in Matlab, and then fine-tune the obtained self-tuning parameters after adding the nonlinear dynamics of the system to achieve the best tracking performance. The selected controller parameters are k P =1000,k I =500,k D =1.

[0120] The expected command of the system, the tracking error of the ANNPTC controller, and the tracking error comparison between the ANNPTC controller and the PID controller are as follows: Figure 3 、 Figure 4 and Figure 5 As shown. Figure 4 It can be seen that under the action of the ANNPTC controller, the position output of the electro-hydraulic system has a high tracking accuracy for the command, and the amplitude of the steady-state tracking error is about 4×10 -5 m. From Figure 5 The comparison of the tracking errors of the two controllers shows that the tracking error of the ANNPTC controller proposed in the present invention is much smaller than that of the PID controller, and the tracking performance is more superior.

[0121] Figure 6 This is a graph showing the change of the control input of the electro-hydraulic system over time under the action of the ANNPTC controller. It can be seen from the figure that the obtained control input is a smooth and continuous signal, which is more conducive to execution in practical applications.

Claims

1. A method for controlling preset time of an electro-hydraulic system using an adaptive neural network, characterized in that: The following steps are involved: Step 1: Establish a mathematical model of the electro-hydraulic system and proceed to step 2; Step 2: Based on the mathematical model of the electro-hydraulic system, design an adaptive neural network preset time performance controller and proceed to step 3; Step 3: Use Lyapunov stability theory to prove the stability of the adaptive neural network preset time performance controller, and obtain the result that the system tracking error is asymptotically stable.

2. The electro-hydraulic system adaptive neural network preset time control method according to claim 1, characterized in that: In step 1, a mathematical model of the electro-hydraulic system is established as follows: Step 1-1: An electro-hydraulic system is used for linear motion of large industrial heavy-load mechanical equipment. The load is fixedly connected to the piston rod of a hydraulic cylinder. An electro-hydraulic proportional servo valve controls the movement of the piston rod of the hydraulic cylinder, thereby driving the load. Based on the dynamic characteristics of the load, hydraulic cylinder, and electro-hydraulic proportional servo valve, a mathematical model of the electro-hydraulic system is obtained. Step 1-2: To facilitate controller design, define state variables and convert the mathematical model of the electro-hydraulic system into state space equations.

3. The electro-hydraulic system adaptive neural network preset time control method according to claim 2, characterized in that: Step 1-1: The electro-hydraulic system is used for linear motion of large industrial heavy-load mechanical equipment. The load is fixedly connected to the piston rod on the hydraulic cylinder. The electro-hydraulic proportional servo valve controls the movement of the piston rod on the hydraulic cylinder, thereby driving the load movement. Based on the dynamic characteristics of the load, hydraulic cylinder, and electro-hydraulic proportional servo valve, the mathematical model of the electro-hydraulic system is obtained as follows: According to Newton's second law, the force balance equation of the electro-hydraulic system is: In formula (1), m represents the mass of the load, y represents the displacement of the hydraulic cylinder piston rod, Indicates the speed of the hydraulic cylinder piston rod, It represents the acceleration of the hydraulic cylinder piston rod, A represents the effective working area of ​​the hydraulic cylinder piston, and the oil pressure difference P between the inlet and outlet oil chambers on both sides of the cylinder L =P1-P2, P1 represents the oil pressure in the hydraulic cylinder inlet chamber, P2 represents the oil pressure in the hydraulic cylinder outlet chamber, represents the friction force on the load, d1(t) represents the unmodeled mechanical disturbance of the system, and t represents time; Then formula (1) can be rewritten as: In the electro-hydraulic system, ignoring the leakage of oil from the cylinder, the dynamic pressure equation is: Formula (3), β e Indicates the effective elastic modulus of the oil, C t Indicates the leakage coefficient of the hydraulic cylinder, the control volume of the oil inlet chamber V1=V 01 +Ay, the control volume of the oil outlet chamber V2=V 02 -Ay, V 01 Indicates the initial volume of the oil inlet chamber, V 02 represents the initial volume of the oil outlet cavity, Q1 represents the flow rate of the oil inlet cavity, Q2 represents the flow rate of the oil outlet cavity, q1 represents the unmodeled interference of Q1, q2 represents the unmodeled interference of Q2, Indicates P L The first derivative of ; Q1 and Q2 are respectively related to the displacement of the spool of the electro-hydraulic proportional servo valve x v There are the following relationships: Among them, the electro-hydraulic proportional servo valve coefficient C d Indicates the electro-hydraulic ratio The flow coefficient of the servo valve, w0 represents the valve core area gradient of the electro-hydraulic proportional servo valve, ρ represents the oil density, P s Indicates the oil supply pressure, P r represents the return oil pressure, s(·) represents the function of the intermediate variable ·, and is defined as: Ignoring the dynamics of the electro-hydraulic proportional servo valve spool, assume that the control input u acting on the spool and the spool displacement x v Proportional relationship, that is, x v =k i u ， Among them, k i represents the voltage-spool displacement gain coefficient, so Equation (4) is rewritten as: Formula (6), intermediate variable k u =k q k i , intermediate variables Intermediate variables 4. The electro-hydraulic system adaptive neural network preset time control method according to claim 3, characterized in that: Step 1-2: To facilitate controller design, define state variables and convert the obtained electro-hydraulic system mathematical model into a state space equation, as follows: Define state variables: Among them, the intermediate variable x1=y, the intermediate variable Intermediate variable x3 = AP L / m ， Then transform Equation (2) into the state space equation: Formula (7), represents the first-order derivative of x1, represents the first-order derivative of x2, represents the first-order derivative of x3, the unknown dynamics of the system Δ1=d1(t) / m, and the intermediate variable F(x2)=F f (x2) / m, intermediate variable Intermediate variables Intermediate variables System unknown dynamics T stands for transpose.

5. The electro-hydraulic system adaptive neural network preset time control method according to claim 4, characterized in that: In step 1, to facilitate controller design, the following assumptions are made: Assumption 1: The system is expected to track the position command x d It is second-order continuous, and the system expects position command, velocity command and acceleration command to be bounded; Assumption 2: The unknown dynamics Δ1 and Δ2 of the system satisfy: In formula (8), δ1 and δ2 are both unknown positive constants; Go to step 2.

6. The electro-hydraulic system adaptive neural network preset time control method according to claim 5, characterized in that: In step 2, based on the mathematical model of the electro-hydraulic system, an adaptive neural network preset time performance controller is designed as follows: Step 2-1: To facilitate controller design, define the error ζ1 = z1 / Φ(t) ， Among them, the system tracking error z1=x1-x d , x d is the position command that the system expects to track, Φ(t) represents the preset time performance function, in order to facilitate the system state x1 to track the expected position command x as accurately as possible under the designed controller drive d , and let z1 always satisfy |z1|<Φ(t), and ensure that the tracking error ζ1 tends to 0, as follows: The tracking error z1 meets the preset range: -Φ(t)<z1<Φ(t) (9), Formula (9), preset time performance function Φ(t) ， The specific form is Among them, Φ0 represents a constant positive, Φ ∞ represents a constant positive, c represents a constant positive, T f Represents a constant that is always positive and satisfies the condition Φ0>Φ ∞ , tan represents the tangent function; Design the following nonlinear filter: Formula (11), filter gain τ1>0 ， v1 represents the virtual control of x2, and the filtering error of v1 is ω1=v 1f -v1, v 1f represents the filtered signal of v1, σ(t) represents a function that is always positive and satisfies Where ν represents the integration variable, represents a constant that is always positive, represents the first-order derivative of v1, Indicates v 1f The first derivative of The upper bound l1 represents a positive constant; Taking the derivative of the tracking error ζ1, we get: in, represents the first-order derivative of ζ1, v 1f The error with x2 is ζ2 = x2 - v 1f ; Represents x d The first derivative of ; represents the first derivative of the preset time performance function Φ(t); Select Lyapunov function We can get: in, represents the first-order derivative of L1; Design virtual control v1 as: Formula (14), gain k1>0, then Step 2-2, design the following nonlinear filter: Formula (16), filter gain τ2>0, v2 represents the virtual control of x3, and the filter error ω2 of v2=v 2f -v2, v 2f represents the filtered signal of v2, represents the first-order derivative of v2, Indicates v 2f The first derivative of The upper bound l2 represents a positive constant; Derivative of the error ζ2 yields: in, represents the first-order derivative of ζ2, v 2f The error with x3 is ζ3 = x3 - v 2f ; Select Lyapunov function We can get: in, represents the first-order derivative of L2; Design virtual control v2 as: Formula (19), gain k2>0, v s represents an intermediate variable, γ1 represents a positive constant, represents the estimated value of F(x2), The specific form is: Formula (20), W a The estimated value of W a represents the weights of the neural network, represents the activation function of the neural network, X a Represents the input of the neural network; The weight update law of the neural network is designed as: Formula (21), express The first derivative of , Γ a Represents the weight gain matrix of the neural network, Proj represents the discontinuous mapping function; Substituting formula (19) into formula (18), we get: Formula (22), the neural network weight W a The estimated error ω a represents the approximation error of the neural network; Step 2-3, take the derivative of ζ3 and get: in, represents the first-order derivative of ζ3; Select Lyapunov function have to: in, represents the first-order derivative of L3; According to formula (24), the control input of the valve core, that is, the adaptive neural network preset time performance controller u, is: Formula (25), gain k3>0, u s represents an intermediate variable, and γ2 represents a positive constant; Substituting formula (25) into formula (24) yields: Go to step 3.

7. The electro-hydraulic system adaptive neural network preset time control method according to claim 6, characterized in that: The stability of the adaptive neural network preset time performance controller is proved by using Lyapunov stability theory in step 3, and the result that the system tracking error is asymptotically stable is obtained, as follows: The Lyapunov function L is defined as follows: The stability is proved by using Lyapunov stability theory, and the result that the system tracking error is asymptotically stable is obtained.

Images