Throttle valve pressure sensor fault-tolerant control method based on self-adaptive observer

Abstract

The invention discloses a throttle valve pressure sensor fault-tolerant control method based on a self-adaptive observer, and aims to solve the problems of stability and control precision of a throttle valve pressure sensor under the conditions of faults and external disturbance. The method comprises the following steps: firstly, establishing a throttle valve pressure control system model containing external disturbance, aiming at a pressure sensor fault, estimating real information of a pressure sensor by designing a self-adaptive observer, generating residual information of the system and the observer, and designing a self-adaptive threshold fault detection algorithm based on the information; on the basis of fault detection, parameters of the observer model are adjusted in a self-adaptive mode, and the true value of the pressure sensor is reconstructed. Meanwhile, an undisturbed unit is designed to enhance the robustness of the system under an external disturbance condition. By means of the method, the system stability of the throttle valve pressure sensor under the conditions of faults and external disturbance can be ensured, and the method has high control precision, fault-tolerant capability and disturbance resistance.

Description

Technical Field

[0001] The invention belongs to the technical field of throttle pressure sensor control, and in particular relates to a throttle valve pressure sensor fault-tolerant control method based on an adaptive observer. Background Art

[0002] As oil drilling technology continues to develop towards more complex formations, the problem of narrow density safe drilling has become increasingly prominent and has become a bottleneck problem restricting oil exploration and development technology. In this context, the robustness and fault tolerance of the automatic throttling control system are increasingly required. In the automatic throttling control system, pressure sensor fault detection is crucial. If the sensor is damaged, the data collected by the PLC will be unreliable, which may cause the PLC to output an erroneous control signal, resulting in malfunction of the throttle valve. This may cause overflow, well kick, or even blowout and other serious accidents. Therefore, it is very necessary to realize fault-tolerant control of the pressure control loop.

[0003] For automatic throttling control, it is necessary to make adjustments based on the feedback information transmitted by the bottom hole pressure. Generally, an observer is used to estimate the bottom hole pressure information during design. The commonly used observer is the Romberg observer. For the Romberg observer, the structure is relatively simple and fixed, and the convergence speed is not fast enough. For the adaptive observer, by reasonably designing the adaptive law, when the model is mismatched, the observer can track the sensor signal without error.

[0004] For fault-tolerant control, the main method has a reduced fault-tolerant performance when the object does not match the observer model, cannot adapt to the disturbance in the well, and does not consider that when the sensor fails, the direct use of the observer output will produce a sudden change and cause disturbance to the system. Therefore, the external disturbance is considered when designing the adaptive observer, and the non-disturbance unit is added to ensure that when the sensor fails, the adaptive observer reconstructed value has no static error with the true value. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that the throttle valve pressure sensor in the automatic throttling system fails, is subject to external interference, and the output signal of the observer changes suddenly after the failure occurs.

[0006] To achieve the above object, the present invention provides a fault-tolerant control method for a throttle valve pressure sensor based on an adaptive observer, and the specific steps are as follows:

[0007] S1. Considering the system failure and external disturbance, the closed-loop transfer function is converted into an augmented state space equation to construct a throttle valve pressure control model;

[0008] S2. Based on the throttle valve pressure control model in step S1, an adaptive observer is designed, and a gain matrix of the adaptive observer and an estimation matrix of disturbance and fault are adjusted in real time according to an adaptive law;

[0009] S3, designing a fault detection unit based on an adaptive observer and residual for fault detection;

[0010] S4. Design disturbance-free units to ensure system stability and build a fault-tolerant control system.

[0011] Furthermore, it is characterized in that the specific implementation of step S1 is as follows:

[0012] S101. The calculation formula for the pressure difference between the front and rear ends of the throttle valve is:

[0013]

[0014] Where, Δp is the throttling pressure difference, MPa; Q is the drilling fluid flow rate, L / S; ρ is the throttling valve fluid density, kg / m 3 ; x is the valve core displacement, m;

[0015] S102, the electro-hydraulic actuator is composed of a proportional amplifier, an electro-hydraulic proportional valve and a hydraulic cylinder;

[0016] The transfer function of the proportional amplifier is:

[0017]

[0018] Where, I(s) is the output current, A; U(s) is the input voltage, A; K a is the proportional amplifier gain;

[0019] The transfer function of the electro-hydraulic proportional valve is:

[0020]

[0021] Where, Q1(s) is the no-load flow rate of the electro-hydraulic proportional directional valve, m 3 / s, Q1(s)=K p X v (s);X v (s) is the valve core displacement of the electro-hydraulic proportional reversing valve, m; ω p is the bandwidth of the valve rad / s, K p is the proportionality coefficient;

[0022] Transfer function of hydraulic cylinder:

[0023]

[0024] in:

[0025]

[0026] Among them, β e is the effective volume elastic modulus of hydraulic oil, MPa; ω h is the natural frequency of the hydraulic cylinder, rad / s; b is the hydraulic damping ratio; K ce is the pressure flow coefficient;

[0027] S103, ratio link of pressure sensor:

[0028]

[0029] Where: U f is the feedback voltage, V; P(s) is the pressure at the detection point, MPa; K f is the pressure sensor gain, V / MPa;

[0030] S104. The closed-loop transfer function of the system is obtained through the throttle valve, the electro-hydraulic actuator and the pressure sensor:

[0031]

[0032] Where K1 = K a K p K f / A p ;

[0033] Considering the system failure and external disturbance, the closed-loop transfer function is converted into an augmented state space equation:

[0034]

[0035] Where A is the system state matrix, B is the input matrix, C is the output matrix, D is the disturbance impact matrix, and F is the fault impact matrix; x is the system state variable, and y is the sensor measurement value.

[0036] Furthermore, it is characterized in that the adaptive observer in step S2 is:

[0037]

[0038] in is the state variable of the observer, is the observed value of y, L is the observer gain matrix, Γ is the disturbance estimation matrix, and Ψ is the fault estimation matrix.

[0039] Furthermore, it is characterized in that the specific implementation method of adjusting the gain matrix of the adaptive observer and the estimation matrix of disturbance and fault according to the adaptive law in step S2 is as follows:

[0040] Assume that the system state estimation error is:

[0041] The error dynamic equation is: Where e is the state estimation error; the disturbance and fault impact of the system are determined according to the error equation, and the observer gain and disturbance and fault estimation matrices are adjusted according to the adaptive law;

[0042] The adaptive law of gain matrix L is: where γ L is a positive regulating factor; is the gain matrix adjustment rate;

[0043] Perturbation estimation matrix adaptation law: where γ Γ is the adjustment factor for disturbance estimation;

[0044] Adjustment law of fault estimation matrix Ψ: in is the fault estimation matrix adjustment rate, γ ψ is the adjustment factor for fault estimation;

[0045] The stability analysis of the system is carried out according to the Lyapunov function:

[0046]

[0047] Its time derivative is:

[0048]

[0049] By selecting the gain matrix L and the adaptive law to adjust the parameter γ L , γ Γ , γ Ψ ensure This proves the global stability of the system under the influence of disturbances and faults.

[0050] Furthermore, it is characterized in that step S3 specifically comprises:

[0051] Define the residual: The threshold value is dynamically adjusted according to the historical information of the residual. The expectation of r(t) at time t is:

[0052]

[0053] The standard deviation is:

[0054]

[0055] Where M is the data window size; ΔT is the control system sampling time; the threshold adaptive law is Φ(t) = |μ(t)| + εσ(t), ε is a coefficient related to the confidence level, when the sensor fails, r(t) < Φ(t), and the fault detection unit output is 1.

[0056] Furthermore, it is characterized in that the design of the non-disturbance unit in step S4 is specifically as follows:

[0057] In fault-tolerant control, the observer output Reconstructing sensor output Assuming that a fault occurs at time t0, a disturbance-free unit is constructed, and the observer output and sensor measurement values ​​are processed as follows:

[0058]

[0059] Ensure that the residual does not mutate before and after the fault. When r(t) does not mutate, y out There will be no sudden changes before and after the failure.

[0060] Furthermore, it is characterized in that the specific calculation method of the transfer function of the hydraulic cylinder in step S102 is:

[0061] Based on the working characteristics of the hydraulic cylinder and combined with the continuity equation, the hydraulic cylinder oil inlet chamber equation, oil return chamber equation and the change in the hydraulic cylinder volume are obtained:

[0062]

[0063] Among them, Q1 is the flow rate at the inlet of the electro-hydraulic proportional valve, Q2 is the flow rate at the outlet of the electro-hydraulic proportional valve, C1 is the internal leakage coefficient of the hydraulic cylinder; C2 is the external leakage coefficient of the hydraulic cylinder; p1 is the pressure of the hydraulic cylinder without rod chamber, p2 is the pressure of the hydraulic cylinder with rod chamber, V1 is the volume of the hydraulic cylinder oil inlet chamber, m 3 ; V2 is the volume of the hydraulic cylinder oil return chamber, m 3 β e is the effective volume elastic modulus of the hydraulic oil, MPa; V is the total volume of the hydraulic cylinder, m 3 ; V0 is the initial volume of the hydraulic cylinder, y is the displacement of the hydraulic cylinder, A is the effective area of ​​the hydraulic oil pre-action piston, and t is the time;

[0064] Assume that the initial volumes of the two chambers of the hydraulic cylinder are The flow rate is 1 / 2 of the sum of the inlet and outlet flows; therefore, the flow equation and load flow equation of the hydraulic cylinder when the throttle valve moves in the positive and negative directions, as well as the force balance equation of the throttle valve are obtained:

[0065]

[0066] Where Q mis the load flow of the electro-hydraulic proportional valve, A P is the effective working area of ​​the hydraulic cylinder, m 2 ;p L is the load pressure, m is the total mass of the actuator, kg; ζ is the viscous damping coefficient; k is the spring stiffness, N / m; F2 is the hydraulic cylinder load force, N; F1 is the hydraulic oil force, N;

[0067] The flow equation of the hydraulic cylinder when the throttle valve moves in the positive and negative directions, the load flow equation, and the force balance equation of the throttle valve are Laplace transformed to obtain the following equation:

[0068] Q m (s) = K q x1(s)-K c P L (s)

[0069]

[0070] p L (s)=[(ms 2 +ζs+K)Y(s)+F d ] / A P

[0071] Where s is the instantaneous speed of the hydraulic cylinder piston, K q K is the flow-displacement gain of the electro-hydraulic proportional valve when it is working stably. c is the pressure-flow coefficient when the electro-hydraulic proportional valve is working stably, Y(s) is the displacement of the hydraulic cylinder piston, F d It is the force of drilling fluid on the throttle valve core;

[0072] According to the equation after Laplace transformation, by the displacement superposition principle, and because the actuator is inelastic load and the damping coefficient generated by leakage ζ is the conventional damping coefficient. After simplifying the equation after Laplace transformation, the transfer function of the hydraulic cylinder can be obtained.

[0073] Beneficial effects:

[0074] 1. The present invention designs the observer gain matrix and its adaptive law according to the system, and introduces the designed adaptive law and adjustment strategy in the case of external interference and fault, so as to maintain accurate estimation of the system in the case of fault and external interference. The design of adaptive threshold using standard deviation for fault detection makes it more sensitive to faults, improves the response speed of detection, and can detect potential faults and abnormalities earlier.

[0075] 2. The present invention provides a fault-tolerant control method for a throttle valve pressure sensor based on an adaptive observer. The method is a control method that has a fault-tolerant function and can overcome external interference. An adaptive observer is designed to estimate the pressure sensor information, thereby solving the problem of mismatch between the observer and the model during a fault and the problem of interference in the measurement caused by external disturbances of the system. A non-disturbance unit is introduced to solve the problem of sudden change in the observer output when the pressure sensor fails, thereby ensuring the fault tolerance and robustness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 is a flow chart of the method provided by the present invention;

[0077] Figure 2 is a structural diagram of a throttle valve pressure control model provided by the present invention;

[0078] Figure 3 It is a block diagram of the throttle valve fault-tolerant control principle of the adaptive observer provided by the present invention;

[0079] Figure 4 It is a mathematical model block diagram of the throttle valve pressure control system provided by the present invention;

[0080] Figure 5 It is a constant deviation fault and a constant gain fault diagram measured by the adaptive observer provided by the present invention;

[0081] Figure 6 is the fault detection result provided by the present invention;

[0082] Figure 7 This is the constant gain fault tolerance effect provided by the present invention;

[0083] Figure 8 This is the constant deviation fault tolerance effect provided by the present invention. DETAILED DESCRIPTION

[0084] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0085] The application principle of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0086] Embodiment 1:

[0087] like Figure 1As shown in the figure, a fault-tolerant control method for a throttle valve pressure sensor based on an adaptive observer of the present invention is as follows: first, a throttle valve pressure control model is established considering sensor failure and external disturbance; then, an adaptive observer is designed based on the throttle valve pressure control model to estimate the pressure information of the sensor; then, a fault-tolerant control method is designed, and a non-disturbance unit is introduced in consideration of the signal mutation problem of the observer when the sensor fails. The principle block diagram of the whole system is as follows Figure 2 The specific implementation steps are as follows:

[0088] S1. Establish throttle valve pressure control model

[0089] The throttle valve pressure control model consists of three parts: throttle valve, electro-hydraulic actuator and pressure sensor. Figure 3 As shown, the throttle valve pressure control system is mainly controlled by the PLC controller sending out control signals. After receiving the signals, the proportional amplifier amplifies the signals and transmits the processed signals to the electro-hydraulic proportional valve. The electro-hydraulic proportional valve pumps the hydraulic oil into the hydraulic cylinder according to the direction and size of the received electrical signals, thereby pushing the valve stem, driving the valve core to move, changing the flow area of ​​the throttle valve, and thus changing the size of the back pressure. Specifically:

[0090] S101, throttle valve

[0091] The magnitude of the throttling pressure is mainly related to the flow rate of the fluid flowing through the throttle valve, the fluid density, etc. The drilling fluid returned from the wellhead flows through the throttle valve, and the pressure difference between the two ends of the throttle valve is calculated as follows:

[0092]

[0093] Where, Δp is the throttling pressure difference, MPa; Q is the drilling fluid flow rate, L / S; ρ is the throttling valve fluid density, kg / m 3 ; x is the valve core displacement, m.

[0094] S102, electro-hydraulic actuator modeling

[0095] The electro-hydraulic actuator consists of three parts: proportional amplifier, electro-hydraulic proportional valve and hydraulic cylinder.

[0096] The time constant of the proportional amplifier is very small, and the time delay is generally not considered. It is only regarded as a proportional amplification link. The transfer function of the proportional amplifier is:

[0097]

[0098] Where, I(s) is the output current, A; U(s) is the input voltage, A; K a is the proportional amplifier gain.

[0099] Without considering the dead zone characteristics, the electro-hydraulic proportional valve can be represented by a first-order inertia link to express its mathematical model. The transfer function of the electro-hydraulic proportional valve is:

[0100]

[0101] Where, Q1(s) is the no-load flow rate of the electro-hydraulic proportional directional valve, m 3 / s, Q1(s)=K p X v (s), X v (s) is the valve core displacement of the electro-hydraulic proportional reversing valve, m; ω p is the bandwidth of the valve rad / s, K p is the proportional coefficient, and s is the instantaneous speed of the hydraulic cylinder piston.

[0102] Based on the working characteristics of the hydraulic cylinder (the working characteristics of the hydraulic cylinder describe the performance and response ability of the hydraulic cylinder under different working conditions, usually including characteristics such as displacement, speed, pressure, thrust, etc.), and combined with the continuity equation, the hydraulic cylinder oil inlet chamber equation shown in formula (1-4), the oil return chamber equation shown in formula (1-5), and the change in the hydraulic cylinder volume shown in formula (1-6) can be obtained.

[0103]

[0104] Among them, Q1 is the flow rate at the inlet of the electro-hydraulic proportional valve, Q2 is the flow rate at the outlet of the electro-hydraulic proportional valve, C1 is the internal leakage coefficient of the hydraulic cylinder; C2 is the external leakage coefficient of the hydraulic cylinder; p1 is the pressure of the hydraulic cylinder without rod chamber, p2 is the pressure of the hydraulic cylinder with rod chamber, V1 is the volume of the hydraulic cylinder oil inlet chamber, m 3 ; V2 is the volume of the hydraulic cylinder oil return chamber, m 3 β e is the effective volume elastic modulus of the hydraulic oil, MPa; V is the total volume of the hydraulic cylinder, m 3 ; V0 is the initial volume of the hydraulic cylinder, y is the displacement of the hydraulic cylinder, A is the effective area of ​​the hydraulic oil pre-action piston, and t is the time.

[0105] Assume that the initial volumes of the two chambers of the hydraulic cylinder are And the piston moves left and right. So the flow rate is 1 / 2 of the sum of the inlet and outlet flows. Therefore, we can get the flow equation of the hydraulic cylinder when the throttle valve moves in the positive and negative directions and the load flow equation (1-7) and (1-8) as well as the throttle valve force balance equation (1-9).

[0106]

[0107] Where Q m is the load flow of the electro-hydraulic proportional valve, A Pis the effective working area of ​​the hydraulic cylinder, m 2 ;p L is the load pressure, m is the total mass of the actuator, kg; ζ is the viscous damping coefficient; K is the spring stiffness, N / m; F2 is the hydraulic cylinder load force, N; F1 is the hydraulic oil force, N;

[0108] By performing Laplace transform on the above equations (1-7) and (1-9), we can obtain equations as shown in equations (1-10) to (1-12):

[0109] Q m (s) = K q x1(s)-K c P L (s) (1-10)

[0110]

[0111] p L (s)=[(ms 2 +ζs+K)Y(s)+F d ] / A P (1-12)

[0112] Where s is the instantaneous speed of the hydraulic cylinder piston, K q K is the flow-displacement gain of the electro-hydraulic proportional valve when it is working stably. c is the pressure-flow coefficient when the electro-hydraulic proportional valve is working stably, Y(s) is the displacement of the hydraulic cylinder, F d It is the force exerted by drilling fluid on the throttle valve core.

[0113] According to (1-10) to (1-12), and by the displacement superposition principle, since the actuator (the actuator is the piston-piston rod) is non-elastic load and the damping coefficient A generated by leakage is P 2 / (C1+C2+δ Q ) is much larger than the conventional damping coefficient ζ, that is, Therefore, after simplifying (1-10) to (1-12), we can get the transfer function of the hydraulic cylinder:

[0114]

[0115] in:

[0116]

[0117]

[0118] Among them, β e is the effective volume elastic modulus of hydraulic oil, MPa; ω his the natural frequency of the hydraulic cylinder, rad / s; b is the hydraulic damping ratio; K ce is the pressure flow coefficient.

[0119] S103, Pressure Sensor

[0120] Since the time constant of the pressure sensor is very small, it acts as a proportional link:

[0121]

[0122] Where: U f is the feedback voltage, V; P(s) is the pressure at the detection point, MPa; K f is the pressure sensor gain, V / MPa.

[0123] S104. Based on the analysis of the throttle valve, electro-hydraulic actuator and pressure sensor, the mathematical model block diagram of the controlled object can be obtained as follows: Figure 4 As shown, the closed-loop transfer function of the system is obtained:

[0124]

[0125] Where K1 = K a K p K f / A p .

[0126] Considering the system failure and external disturbance, the closed-loop transfer function (i.e. the mathematical model of the throttle valve pressure control system) is converted into the augmented state space equation:

[0127]

[0128] Where A is the system state matrix, B is the input matrix, C is the output matrix, D is the disturbance impact matrix, and F is the fault impact matrix. x is the system state variable, and y is the sensor measurement value.

[0129] After the system is abstracted into a transfer function, it is converted into a state-space equation to facilitate the construction of an adaptive observer (the construction of an adaptive observer requires the state-space equation of the system). Compared with the traditional state-space equation, the augmented state-space equation takes into account the impact of external disturbances and faults.

[0130] S2. According to the throttle valve pressure control model, the adaptive observer is designed as:

[0131]

[0132] in is the state variable of the observer, is the observed value of y, L is the observer gain matrix, Γ is the disturbance estimation matrix, and Ψ is the fault estimation matrix.

[0133] To ensure the convergence and accuracy of the adaptive observer in complex environments, the gain matrix L and the estimation matrices Γ and Ψ for disturbances and faults are adjusted in real time through the adaptive law.

[0134] Assume that the system state estimation error is:

[0135]

[0136] The error dynamic equation is:

[0137]

[0138] Where e is the state estimation error.

[0139] The error dynamic equation shows that when the estimation error is large, the disturbance and fault influence of the system are more significant. At this time, the model cannot accurately track the sensor signal under complex environments such as external interference and faults. Therefore, it is necessary to design a suitable adaptive law to adjust L, Γ and Ψ.

[0140] According to the Lyapunov stability theory, the following adaptive law is designed to adjust the observer gain L and the disturbance and fault estimation matrices Γ and Ψ:

[0141] The adaptive law of gain matrix L is:

[0142]

[0143] where γ L is a positive adjustment factor, used to control the gain adjustment speed; is the gain matrix adjustment law.

[0144] Perturbation estimation matrix adaptation law:

[0145]

[0146] where γ Γ is the adjustment factor for disturbance estimation.

[0147] Adjustment law of fault estimation matrix Ψ:

[0148]

[0149] in is the fault estimation matrix adjustment rate, γ ψ is the adjustment factor for fault estimation.

[0150] In order to ensure that the design of the observer has global asymptotic stability, the stability analysis of the system is carried out by constructing the following Lyapunov function:

[0151]

[0152] Its time derivative is:

[0153]

[0154] By selecting the gain matrix L and the adaptive law to adjust the parameter γ L , γ Γ , γ Ψ Ensure that the time derivative of the Lyapunov function is less than zero, that is, Thus, the global asymptotic stability of the system under the influence of disturbances and faults is proved.

[0155] S3. Design of Fault Detection Unit Based on Residual

[0156] To achieve fault detection, we first define the residual

[0157]

[0158] The threshold value is dynamically adjusted according to the historical information of the residual. The expectation of r(t) at time t is:

[0159]

[0160] The standard deviation is:

[0161]

[0162] Where M is the data window size; ΔT is the control system sampling time. The threshold adaptive law is Φ(t)=|μ(t)|+εσ(t), ε is the coefficient related to the confidence level. When the sensor fails, r(t)<Φ(t), and the fault detection unit output is 1.

[0163] S4. To solve the problem of sudden signal changes when the sensor fails, the non-disturbance unit is introduced as follows:

[0164] In fault-tolerant control, Reconstructing sensor output Assuming that a fault occurs at time t0, a disturbance-free unit is constructed, and the observer output and sensor measurement values ​​are processed as follows:

[0165]

[0166] Ensure that the residual does not mutate before and after the fault. When r(t) does not mutate, y out There will be no sudden changes before and after the failure:

[0167]

[0168] Through the above system joint simulation, when the sensor is normal, the adaptive observer unit can track the sensor measurement value without error through the adaptive law; when the sensor fails, the fault detection unit detects the fault, and then uses the adaptive observer to reconstruct the sensor measurement value and output it through the non-disturbance unit to replace the measurement value of the faulty sensor, so that the sensor failure will not cause disturbance to the original control system, and fault-tolerant control of the sensor failure is realized.

[0169] Implementation 2:

[0170] The present invention aims at the fault-tolerant control problem when a throttle pressure sensor fails. The observer detects the sensor fault, and then designs a switching observer output to compensate for the fault according to the fault detection result, thereby achieving system stability in the case of a throttle pressure sensor failure.

[0171] The effectiveness of the fault-tolerant control method of the present invention is verified by MATLAB / Simulink simulation experiments and the experimental results are given. Given system parameters as shown in Table 1:

[0172] Table 1 System parameters

[0173]

[0174] like Figure 5 The designed adaptive observer accurately observed that the output signal of the pressure sensor was added with a constant deviation fault of 1 and a constant gain fault of 0.6 at the 30th second. The fault detection results are shown in Figure 6 As shown in the figure, after the fault is detected, the sensor output signal is quickly reconstructed through the sensor, and the system is not affected by the fault and runs stably. Figure 7 is the constant gain fault tolerance effect, Figure 8 It is the constant deviation fault tolerance effect.

[0175] The simulation experiment results confirm that the throttle valve pressure sensor fault-tolerant control method based on the adaptive observer proposed by the method of the present invention can realize the fault diagnosis and fault-tolerant control of the throttle valve pressure sensor.

[0176] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A fault-tolerant control method for a throttle valve pressure sensor based on an adaptive observer, characterized in that: The specific steps are as follows: S1. Considering the system failure and external disturbance, the closed-loop transfer function is converted into an augmented state space equation to construct a throttle valve pressure control model; S2. Based on the throttle valve pressure control model in step S1, an adaptive observer is designed, and a gain matrix of the adaptive observer and an estimation matrix of disturbance and fault are adjusted in real time according to an adaptive law; S3, designing a fault detection unit based on an adaptive observer and residual for fault detection; S4. Design disturbance-free units to ensure system stability and build a fault-tolerant control system.

2. The throttle valve pressure sensor fault-tolerant control method based on adaptive observer according to claim 1 is characterized in that: The specific implementation of step S1 is as follows: S101. The calculation formula for the pressure difference between the front and rear ends of the throttle valve is: Where, Δp is the throttling pressure difference, MPa; Q is the drilling fluid flow rate, L / S; ρ is the throttling valve fluid density, kg / m 3 ; x is the valve core displacement, m; S102, the electro-hydraulic actuator is composed of a proportional amplifier, an electro-hydraulic proportional valve and a hydraulic cylinder; The transfer function of the proportional amplifier is: Where, I(s) is the output current, A; U(s) is the input voltage, A; K a is the proportional amplifier gain; The transfer function of the electro-hydraulic proportional valve is: Where, Q1(s) is the no-load flow rate of the electro-hydraulic proportional directional valve, m 3 / s, Q1(s)=K p X v (s);X v (s) is the valve core displacement of the electro-hydraulic proportional reversing valve, m; ω p is the bandwidth of the valve rad / s, K p is the proportionality coefficient; Transfer function of hydraulic cylinder: in: Among them, β e is the effective volume elastic modulus of hydraulic oil, MPa; ω h is the natural frequency of the hydraulic cylinder, rad / s; b is the hydraulic damping ratio; K ce is the pressure flow coefficient; S103, ratio link of pressure sensor: Where: U f is the feedback voltage, V; P(s) is the pressure at the detection point, MPa; K f is the pressure sensor gain, V / MPa; S104. The closed-loop transfer function of the system is obtained through the throttle valve, the electro-hydraulic actuator and the pressure sensor: Where K1 = K a K p K f / A p ; Considering the system failure and external disturbance, the closed-loop transfer function is converted into an augmented state space equation: Where A is the system state matrix, B is the input matrix, C is the output matrix, D is the disturbance impact matrix, and F is the fault impact matrix; x is the system state variable, and y is the sensor measurement value.

3. The throttle valve pressure sensor fault-tolerant control method based on adaptive observer according to claim 1, characterized in that: The adaptive observer in step S2 is: in is the state variable of the observer, is the observed value of y, L is the observer gain matrix, Γ is the disturbance estimation matrix, and Ψ is the fault estimation matrix.

4. The throttle valve pressure sensor fault-tolerant control method based on adaptive observer according to claim 1, characterized in that: The specific implementation of adjusting the gain matrix of the adaptive observer and the estimation matrix of disturbance and fault according to the adaptive law in step S2 is as follows: Assume that the system state estimation error is: The error dynamic equation is: Where e is the state estimation error; the disturbance and fault impact of the system are determined according to the error equation, and the observer gain and disturbance and fault estimation matrices are adjusted according to the adaptive law; The adaptive law of gain matrix L is: where γ L is a positive regulating factor; is the gain matrix adjustment rate; Perturbation estimation matrix adaptation law: where γ Γ is the adjustment factor for disturbance estimation; Adjustment law of fault estimation matrix Ψ: in is the fault estimation matrix adjustment rate, γ ψ is the adjustment factor for fault estimation; The stability analysis of the system is carried out according to the Lyapunov function: Its time derivative is: By selecting the gain matrix L and the adaptive law to adjust the parameter γ L , γ Γ , γ Ψ ensure This proves the global stability of the system under the influence of disturbances and faults.

5. The throttle valve pressure sensor fault-tolerant control method based on adaptive observer according to claim 1, characterized in that: Step S3 is specifically as follows: Define the residual: The threshold value is dynamically adjusted according to the historical information of the residual. The expectation of r(t) at time t is: The standard deviation is: Where M is the data window size; ΔT is the control system sampling time; the threshold adaptive law is Φ(t) = |μ(t)| + εσ(t), ε is a coefficient related to the confidence level, when the sensor fails, r(t) < Φ(t), and the fault detection unit output is 1.

6. The throttle valve pressure sensor fault-tolerant control method based on adaptive observer according to claim 1, characterized in that: The design of the non-disturbance unit in step S4 is as follows: In fault-tolerant control, the observer output Reconstructing sensor output Assuming that a fault occurs at time t0, a disturbance-free unit is constructed, and the observer output and sensor measurement values ​​are processed as follows: Ensure that the residual does not mutate before and after the fault. When r(t) does not mutate, y out There will be no sudden changes before and after the failure.

7. The throttle valve pressure sensor fault-tolerant control method based on adaptive observer according to claim 2, characterized in that: The specific calculation method of the transfer function of the hydraulic cylinder in step S102 is: Based on the working characteristics of the hydraulic cylinder and combined with the continuity equation, the hydraulic cylinder oil inlet chamber equation, oil return chamber equation and the change in the hydraulic cylinder volume are obtained: Among them, Q1 is the flow rate at the inlet of the electro-hydraulic proportional valve, Q2 is the flow rate at the outlet of the electro-hydraulic proportional valve, C1 is the internal leakage coefficient of the hydraulic cylinder; C2 is the external leakage coefficient of the hydraulic cylinder; p1 is the pressure of the hydraulic cylinder without rod chamber, p2 is the pressure of the hydraulic cylinder with rod chamber, V1 is the volume of the hydraulic cylinder oil inlet chamber, m 3 ; V2 is the volume of the hydraulic cylinder oil return chamber, m 3 β e is the effective volume elastic modulus of the hydraulic oil, MPa; V is the total volume of the hydraulic cylinder, m 3 ; V0 is the initial volume of the hydraulic cylinder, y is the displacement of the hydraulic cylinder, A is the effective area of ​​the hydraulic oil pre-action piston, and t is the time; Assume that the initial volumes of the two chambers of the hydraulic cylinder are The flow rate is 1 / 2 of the sum of the inlet and outlet flows; therefore, the flow equation and load flow equation of the hydraulic cylinder when the throttle valve moves in the positive and negative directions, as well as the force balance equation of the throttle valve are obtained: Where Q m is the load flow of the electro-hydraulic proportional valve, A P is the effective working area of ​​the hydraulic cylinder, m 2 ;p L is the load pressure, m is the total mass of the actuator, kg; ζ is the viscous damping coefficient; k is the spring stiffness, N / m; F2 is the hydraulic cylinder load force, N; F1 is the hydraulic oil force, N; The flow equation of the hydraulic cylinder when the throttle valve moves in the positive and negative directions, the load flow equation, and the force balance equation of the throttle valve are Laplace transformed to obtain the following equation: Q m (s)=K q x1(s)-K c P L (s) p L (s)=[(ms 2 +ζs+K)Y(s)+F d ] / A P Where s is the instantaneous speed of the hydraulic cylinder piston, K q K is the flow-displacement gain of the electro-hydraulic proportional valve when it is working stably. c is the pressure-flow coefficient when the electro-hydraulic proportional valve is working stably, Y(s) is the displacement of the hydraulic cylinder piston, F d It is the force of drilling fluid on the throttle valve core; According to the equation after Laplace transformation, by the displacement superposition principle, and because the actuator is inelastic load and the damping coefficient generated by leakage ζ is the conventional damping coefficient. After simplifying the equation after Laplace transformation, the transfer function of the hydraulic cylinder can be obtained.

Images