A fuzzy adaptive underwater glider fault diagnosis method

Abstract

The present application relates to a kind of fuzzy adaptive underwater glider fault diagnosis method based on, it is related to underwater glider fault detection field.The present application includes: the nonlinear system of establishing the pitch angle of centroid adjustment mechanism control;According to the nonlinear system of centroid adjustment mechanism control pitch angle, design fault detection observer, the difference of state variable and the estimated value of state variable in fault detection observer is as state estimation error, whether there is fault is judged by the threshold of state estimation error;Design fault diagnosis estimator, the size of fault is identified.Through the design of fault detection observer and fault diagnosis estimator, the online diagnosis and estimation of actuator failure are realized, the precision of residual signal is improved, the phase lag is reduced, the disturbance signal can be observed, the occurrence of fault is detected in time and measures are taken.

Description

Technical Field

[0001] The present invention relates to the field of underwater glider fault detection, and more particularly to a fuzzy self-adaptive underwater glider fault diagnosis method. Background Art

[0002] Underwater gliders, as one of many high-tech marine products, are favored by experts in related fields for their powerful ocean exploration capabilities. Their role in the development of maritime military programs in various countries is becoming increasingly prominent. As underwater vehicles operating in complex ocean environments for extended periods, even minor faults can affect their navigation performance. In severe cases, they can even cause loss of control, resulting in significant economic losses. Research on underwater glider fault diagnosis can prevent and resolve faults online, improving navigation reliability and effectively reducing potential economic losses.

[0003] Scholars have conducted extensive research in the field of fault diagnosis and proposed numerous methods for fault detection and identification. Hardware redundancy, analytical redundancy, and signal-based methods are widely used. However, fault redundancy methods take up more space and increase weight on equipment, making them unsuitable for critical equipment. Analytical redundancy methods primarily determine whether a fault has occurred by calculating whether the residual between the actual state output and the estimated state output exceeds a set threshold. However, in actual industrial environments, the residual signal can be affected by interference signals, leading to inaccurate fault identification. Signal-based methods bypass the system's input-output model and use actual measured signals for fault diagnosis. However, these methods rely too heavily on expert prior knowledge, resulting in a lack of objective judgment. Summary of the Invention

[0004] The technical problems to be solved by the present invention are:

[0005] In order to avoid the deficiencies of the prior art, the present invention provides a method for underwater glider fault diagnosis based on fuzzy self-adaptation.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] A fuzzy adaptive underwater glider fault diagnosis method is characterized by comprising:

[0008] Establish the longitudinal state space equation of underwater glider;

[0009] A mathematical model of the fault is established. Based on the mathematical model, the actual control force and command control force models of the underwater glider during motion are established. Substituting the model into the longitudinal state space equation of the underwater glider, a nonlinear system for the center of mass adjustment mechanism to control the pitch angle is obtained.

[0010] A fault detection observer is designed based on the nonlinear system of the center of mass adjustment mechanism controlling the pitch angle. The difference between the state variable and the estimated value of the state variable in the fault detection observer is used as the state estimation error. The threshold of the state estimation error is used to determine whether a fault exists.

[0011] Design a fault diagnosis estimator to identify the magnitude of the fault.

[0012] A further technical solution of the present invention is as follows: The actual control force and command control force model of the underwater glider during movement is:

[0013] u=(1-l)u c +Δu

[0014] Among them, u c is the actual control force, u is the command control force, Δu is the additive fault force, and l∈[0,1] represents the effectiveness loss of the actuator.

[0015] A further technical solution of the present invention is that the nonlinear system of the center of mass adjustment mechanism for controlling the pitch angle is:

[0016]

[0017] x=ω z

[0018]

[0019]

[0020]

[0021]

[0022]

[0023] Where f = -Blu c +BΔu is defined as the combined effect of the fault; d is the unknown disturbance; m is the total weight of the glider, λ 26 ,λ 66 is the additional mass; ω z is the pitch angular velocity; ρ is the density of water; v is the speed of the underwater glider; S is the maximum cross-sectional area; L is the length of the underwater glider; is the rotation coefficient of the dimensionless angular velocity; J zz is the moment of inertia; m p is the mass of the slider; G = mg is the gravity of the glider; x p is the displacement of the slider; y c is the coordinate of the center of gravity in the carrier coordinate system; v yis the vertical velocity in the longitudinal plane; v x is the horizontal velocity in the longitudinal plane.

[0024] A further technical solution of the present invention is: the fault detection observer is:

[0025]

[0026] in, is the estimated value of the state variable x in the fault detection observer, L d ∈R 1×1 is a positive gain matrix.

[0027] A further technical solution of the present invention is characterized in that the threshold is:

[0028]

[0029] Among them, λ min [·] represents the minimum eigenvalue of the matrix; L d is the observer gain matrix; l g is the Lipschitz constant; Disturbance after the system is discretized.

[0030] A further technical solution of the present invention is characterized in that the fault diagnosis estimator is:

[0031]

[0032] in, is the error of the dynamic estimate of f, represents the derivative of the fault effect, is the estimated value error of x, matrix K, L e are the estimator parameters that need to be designed.

[0033] A computer system, characterized in that it includes: one or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned method.

[0034] A computer-readable storage medium is characterized by storing computer-executable instructions, which are used to implement the above method when executed.

[0035] The beneficial effects of the present invention are:

[0036] The present invention provides a fuzzy adaptive underwater glider fault diagnosis method. By designing a fault detection observer and a fault diagnosis estimator, this method enables online diagnosis and estimation of actuator faults. This improves the accuracy of residual signals, reduces phase lag, and can observe disturbance signals, allowing for timely detection of faults and the implementation of appropriate measures. Compared with existing technologies, this method offers the following technical advantages:

[0037] 1. It does not require additional space for the equipment and is very suitable for equipment with important requirements;

[0038] 2. The fault observer proposed in the present invention takes the difference between the state variable x and the estimated value of the state variable in the fault detection observer as the state estimation error, and determines whether a fault exists through the threshold of the state estimation error, thereby reducing the degree to which the residual signal in the existing method is affected by the interference signal.

[0039] 3. The method proposed in the present invention uses the state variables in the system as the basis for judgment, does not rely on the prior knowledge of experts, and can obtain objective judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like parts throughout the drawings.

[0041] Figure 1 The overall framework of the present invention.

[0042] Figure 2 Pitch angle control diagram.

[0043] Figure 3 Slider position diagram.

[0044] Figure 4 Vertical motion trajectory.

[0045] Figure 5 Pitch rate estimation error and fault detection threshold.

[0046] Figure 6 Fault estimate and actual fault value.

[0047] Figure 7 Fault estimation error. DETAILED DESCRIPTION

[0048] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0049] This paper proposes a fuzzy adaptive underwater glider fault diagnosis method to address actuator failures in underwater gliders, enabling fault diagnosis and identification. First, a fault detection observer is designed to detect faults and a threshold is given to determine whether a fault has occurred. Next, a fault estimator is designed to estimate state variables and fault impacts. This method addresses the fault diagnosis problem for underwater glider systems with Lipschitz nonlinearities. Finally, simulation experiments using the proposed fault detection observer and fault estimator verify the feasibility of this method.

[0050] like Figure 1 As shown, the following steps are included:

[0051] Step 1: Establish a simplified model of the underwater glider's longitudinal plane;

[0052] Step 2: Define the nonlinear system and the mathematical models under three failure modes;

[0053] Step 3: Based on the underwater glider model established in step 1, model the actual control force and the command control force during the movement of the underwater glider;

[0054] Step 4: Design a fault detection observer. Analyze and judge the state estimation error and provide a fault detection strategy. When the state estimation error exceeds the threshold, it can be determined that the actuator has failed.

[0055] Step 5: Design a fault estimator. After completing the fault detection, it is necessary to identify the size and time characteristics of the fault.

[0056] The above steps are as follows:

[0057] Step 1: First, establish a simplified model of the underwater glider in the longitudinal plane, make the lateral force, yaw moment and roll moment zero, and make the motion of the underwater glider a small maneuver, and the center of gravity position (x c ,y c ,z c ) is a first-order small quantity. In this case, the buoyancy center is selected as the origin of the carrier coordinate system, and the simplified longitudinal motion equations of the underwater glider can be obtained:

[0058]

[0059] Where m is the total weight of the glider, λ 11 ,λ 22 ,λ 26 ,λ 66 is the additional mass; x c ,y c are the coordinates of the center of gravity in the carrier coordinate system; C xis the longitudinal force coefficient; and is the position force coefficient; and is the rotation coefficient of the dimensionless angular velocity, and the corresponding rotation force is generated when the underwater glider performs a stable rotation motion; and is the horizontal rudder force coefficient, which corresponds to the rudder force generated when the underwater glider operates the horizontal rudder; ΔG = GB, where G = mg is the weight of the glider, B = ρgV is the buoyancy of the glider, V is the total volume of the glider, and ρ is the density of water; S is the maximum cross-sectional area; L / v is the dimensionless coefficient, where L is the length of the underwater glider; v is the speed of the underwater glider; J zz is the moment of inertia; v x is the horizontal velocity in the longitudinal plane; v y is the vertical velocity in the longitudinal plane; ω z is the pitch angular velocity; x c =6m p x p / m, where m p is the mass of the slider, x p and δ are the slider displacement and the horizontal rudder size respectively, which serve as the two control inputs of the system, and θ is the heading angle.

[0060] The longitudinal state space equation of the glider can be organized into:

[0061]

[0062] Where:

[0063]

[0064] The present invention mainly considers the fault diagnosis of the execution mechanism of the center of mass adjustment system.

[0065] Let δ = 0 and consider the unknown interference d, equation (2) can be further simplified to:

[0066]

[0067] Let x=ω in the above formula z , A=n,B=-c,u=x p ,g(x)=g(ω z ), formula (4) can be organized as

[0068]

[0069] Step 2: Define the original nonlinear system as:

[0070]

[0071] The three failure modes are paranoid failure, failure failure, and stuck failure. Their mathematical models can be expressed as follows.

[0072] (1) Paranoid failure:

[0073]

[0074] Where ΔU is the influence of the bias fault constant.

[0075] (2) Failure:

[0076]

[0077] Where Γ=diag[λ1,λ2,λ3,…] is the failure factor, where λ i ∈[0,1].λ i =1 means no fault occurs, λ i =0 indicates a complete failure.

[0078] (3) Stuck fault:

[0079]

[0080] Where Γ=diag[λ1,λ2,λ3,…] is the fault factor and ΔU is the bias fault constant. i =0, ΔU F When ≠0, it can be considered that a stuck fault has occurred.

[0081] Step 3: Based on the fault mathematical model (failure fault) defined in step 2, given equation (8), let Γ = I1-l, where l∈R and l∈[0,1] represents the effectiveness loss of the actuator, and model the actual control force and the command control force during the underwater glider's motion as follows:

[0082] u=(1-l)u c +Δu (10)

[0083] Where u c is the actual control force; u is the command control force; Δu is the additive fault force.

[0084] Substituting formula (10) into formula (5) yields:

[0085]

[0086] Where, f = -Blu c +BΔu is defined as the combined effect of the fault.

[0087] Step 4: Based on the nonlinear system of the center of mass adjustment mechanism controlling the pitch angle in step 3, the fault detection observer is designed as:

[0088]

[0089] Where, is the estimated value of the state variable x in the fault detection observer, L d ∈R 1×1 is a positive gain matrix.

[0090] make is the state estimation error, and by discretizing formula (12), we can obtain:

[0091]

[0092] The state error is analyzed and judged, and a fault detection strategy is given. When the state estimation error in formula (13) exceeds the threshold, it can be judged that the actuator has failed. The threshold ξ is:

[0093]

[0094] Where λ min [·] represents the minimum eigenvalue of the matrix; L d is the observer gain matrix; l g is the Lipschitz constant; Disturbance after the system is discretized.

[0095] Step 5: After completing the fault detection, it is necessary to identify the magnitude of the fault. In step 3, f=-Blu c +BΔu is used as the total fault impact, and an observer is designed to estimate the fault.

[0096] The fault estimator designed by the present invention is:

[0097]

[0098]

[0099] Where, are the estimated values ​​of x and f respectively; matrices K and L e are the estimator parameters that need to be designed.

[0100] Defining Error for:

[0101]

[0102]

[0103] The dynamic characteristics of the estimation error can be obtained as

[0104]

[0105]

[0106] Where,

[0107] To verify the effectiveness of the method of the present invention, we conducted a hardware-in-the-loop simulation experiment. During the normal gliding process, a slider offset fault was imposed and the method was used to detect the fault and estimate the impact of the fault.

[0108] The mission type used is a direction finding mission, with the diving buoyancy set to -3.5L, the diving depth set to -800m, and the diving pitch angle set to -30°. The ascent buoyancy is set to 3.5L, the ascent depth is set to -5m, and the ascent pitch angle is set to 30°.

[0109] Through semi-physical simulation experiments, we verified the effectiveness of the present invention. Figure 3 Shows the movement of the slider. Figure 4 The vertical trajectory is shown. From the experimental results, it can be seen that when the simulation is carried out for 1500 seconds, the slider fails. Figure 2 The pitch angle shown in is out of the stable state at the current moment. Figure 6 The actual fault value in suddenly increases. After the fault occurs, Figure 5 The pitch angular velocity error in the yaw rate increases rapidly and exceeds the fault detection threshold, and the system detects the occurrence of a fault. Figure 6 The fault estimate in also increases rapidly, Figure 7 The fault estimation error in the CNN converges quickly to near zero, achieving accurate tracking of the actual fault value.

[0110] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present invention, and these modifications or replacements should all be included in the scope of protection of the present invention.

Claims

1. A method for underwater glider fault diagnosis based on fuzzy adaptive method, characterized in that: include: Establish the longitudinal state space equation of underwater glider; A fault mathematical model is established, and a model of the actual control force and the command control force during the motion of the underwater glider is established based on the fault mathematical model. The model is substituted into the longitudinal state space equation of the underwater glider to obtain a nonlinear system for the center of mass adjustment mechanism to control the pitch angle. The model of the actual control force and the command control force during the motion of the underwater glider is: u=(1-l)u c +Δu Among them, u c is the actual control force, u is the command control force, Δu is the additive fault force, and l∈[0,1] represents the effectiveness loss of the actuator; The nonlinear system of the center of mass adjustment mechanism controlling the pitch angle is: x=ω z Where f = -Blu c +BΔu is defined as the combined effect of the fault; d is the unknown disturbance; m is the total weight of the glider, λ 26 ,λ 66 is the additional mass; ω z is the pitch angular velocity; ρ is the density of water; v is the speed of the underwater glider; S is the maximum cross-sectional area; L is the length of the underwater glider; is the rotation coefficient of the dimensionless angular velocity; J zz is the moment of inertia; m p is the mass of the slider; G = mg is the gravity of the glider; x p is the displacement of the slider; y c is the coordinate of the center of gravity in the carrier coordinate system; v y is the vertical velocity in the longitudinal plane; v x is the horizontal velocity in the longitudinal plane; A fault detection observer is designed based on the nonlinear system of the center of mass adjustment mechanism controlling the pitch angle. The difference between the state variable and the estimated value of the state variable in the fault detection observer is used as the state estimation error. The threshold of the state estimation error is used to determine whether a fault exists. Design a fault diagnosis estimator to identify the magnitude of the fault.

2. The method for underwater glider fault diagnosis based on fuzzy adaptive method according to claim 1, characterized in that: The fault detection observer is: in, is the estimated value of the state variable x in the fault detection observer, L d ∈R 1×1 is a positive gain matrix.

3. The method for fault diagnosis and fault-tolerant control of underwater glider based on fuzzy adaptive control according to claim 2, characterized in that: The thresholds are: Among them, λ min [·] represents the minimum eigenvalue of the matrix; L d is the observer gain matrix; l g is the Lipschitz constant; Disturbance after the system is discretized.

4. The method for fault diagnosis and fault-tolerant control of underwater glider based on fuzzy adaptive control according to claim 1, characterized in that: The fault diagnosis estimator is: in, is the error of the dynamic estimate of f, represents the derivative of the fault effect, is the estimated value error of x, matrix K, L e are the estimator parameters that need to be designed.

5. A computer system, characterized in that include: One or more processors, a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that Computer-executable instructions are stored, and when the instructions are executed, they are used to implement the method according to any one of claims 1 to 4.

Images