An inertial platform gyro soft fault injection method

Abstract

This invention provides a soft fault injection method for inertial platform gyroscopes. It employs the transfer function method to establish a gyroscope measurement model, an inertial platform DC torque motor model, and an inertial platform load rotation integral model. Then, it establishes a soft fault differential simulation model for five types of gyroscope faults: high-frequency noise amplification fault, gyroscope power supply failure fault, gyroscope power supply instantaneous abnormality fault, gyroscope power supply random contact failure, and gyroscope motor and inner loop jamming fault, to obtain the gyroscope pitch and rotation speed signal under fault injection. Based on this, it uses a fault injection function to superimpose the fault signal to obtain the gyroscope measurement signal under fault injection, which is then compared with the inertial platform input signal to obtain the error signal. Finally, it establishes a proportional-integral-lead-derivative control model for the gyroscope platform, realizing dynamic calculation of the entire gyroscope platform under soft fault injection and achieving closed-loop calculation of fault injection and simulation.

Description

Technical Field

[0001] This invention relates to the field of aircraft inertial navigation and platform inertial navigation test equivalence and fault injection, and more specifically, to a method for soft fault injection of inertial platform gyroscopes. Background Technology

[0002] Gyroscopes are widely used in aircraft as angle measurement instruments, and in complex inertial navigation systems, inertial platforms also extensively employ gyroscopes. To ensure safe flight, testing of the inertial platform is a necessary pre-flight procedure. As a high-precision, high-reliability instrument, the gyroscope plays a crucial role, making its testing an indispensable sub-subject in inertial platform testing. However, precisely because of its high reliability and precision, direct testing of gyroscopes often fails to expose them to malfunctions. This is inconvenient for training system operators, as they lack exposure to gyroscope failure scenarios. Furthermore, repeated power-on testing of expensive real equipment inevitably damages gyroscope reliability, making it economically unprofitable. On the other hand, using a faulty inertial platform or faulty gyroscope can only simulate one type of fault, and even then, the number of faulty gyroscopes is often insufficient to meet the needs of a large number of trainees operating the testing system. Based on these background reasons, research on soft fault injection for inertial platforms has high engineering practical value. Currently, fault injection for gyroscopes generally only focuses on power supply continuity issues. Research on actual faults such as random power supply contact problems and gyroscope motor / inner loop jamming is insufficient, or there are no suitable publicly available fault injection methods for reference. Based on the current research status, this invention establishes a soft fault differential simulation method for five types of gyroscope faults: high-frequency noise amplification faults in gyroscope measurements, power supply failure faults, instantaneous power supply anomalies, random power supply contact problems, and gyroscope motor / inner loop jamming. This method has high theoretical, economic, and engineering practical value.

[0003] It should be noted that the information in the background section above is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to provide a soft fault injection method for inertial platform gyroscopes, thereby overcoming the problem that the simulation of gyroscope faults is not comprehensive and detailed due to the limitations and defects of related technologies.

[0005] According to one aspect of the present invention, a method for injecting soft faults into an inertial platform gyroscope is provided, comprising the following steps:

[0006] Step S10: Set the initial value of the input signal of the platform pitch gyroscope to 0. Establish the measurement dynamic model of the pitch gyroscope using a difference equation. Compare the input signal of the pitch gyroscope with the measurement signal of the pitch gyroscope to obtain the measurement error signal of the pitch gyroscope. Then, establish a measurement high-frequency noise amplification fault model, design a high-frequency noise amplification fault injection function, and obtain the measurement high-frequency noise amplification fault signal. Then, superimpose the measurement high-frequency noise amplification fault signal on the measurement error signal. Solve the calculation to obtain the measurement error signal of the gyroscope under fault injection. Perform low-pass operation and input it into the measurement dynamic model of the gyroscope to obtain the pitch angular acceleration signal of the gyroscope. Then, perform integration to obtain the pitch rotation angular velocity signal of the gyroscope.

[0007] Step S20: Based on the pitch rotation angular velocity signal of the gyroscope, establish a gyroscope power supply fault model and a gyroscope power outage fault model. First, capture the instantaneous angular velocity signal of the gyroscope power outage fault to obtain the initial angular velocity signal of the gyroscope power outage fault. Then, use a transfer function to establish a gyroscope speed decay model caused by power outage to obtain the decay speed signal of the gyroscope power outage fault. Next, establish a gyroscope power outage fault injection function and superimpose the pitch rotation angular velocity signal of the gyroscope to obtain the pitch rotation angular velocity signal under the injection of the gyroscope power outage fault.

[0008] Step S30: Establish a differential model for the instantaneous fault of the gyroscope power supply. First, capture the gyroscope angular velocity when the instantaneous fault of the gyroscope power supply occurs. Then, calculate the angular acceleration signal of the instantaneous fault of the gyroscope power supply based on the differential model of the instantaneous fault of the gyroscope power supply, and then integrate it to obtain the angular velocity signal of the instantaneous fault of the gyroscope power supply. Next, design the injection function for the instantaneous fault of the gyroscope power supply. Then, based on the pitch rotation angular velocity signal under the injection of the gyroscope power supply power failure fault, superimpose the angular velocity signal of the instantaneous fault of the gyroscope power supply to obtain the pitch rotation angular velocity signal of the gyroscope under the injection of the instantaneous fault of the gyroscope power supply.

[0009] Step S40: Establish a differential model for the random contact failure of the gyroscope power supply. First, capture the gyroscope rotation angular velocity signal when the random contact failure of the gyroscope power supply occurs. Then, calculate the equivalent negative angular acceleration signal of the gyroscope under the random contact failure fault according to the differential model of the random contact failure of the gyroscope power supply, and then integrate it to obtain the equivalent negative angular velocity signal of the gyroscope. Then, establish an injection function for the random contact failure of the gyroscope power supply. Based on the pitch rotation angular velocity signal of the gyroscope under the instantaneous abnormal fault injection of the gyroscope power supply, superimpose the equivalent negative angular velocity signal of the gyroscope to obtain the pitch rotation angular velocity signal of the gyroscope under the random contact failure of the gyroscope power supply.

[0010] Step S50: Establish a gyroscope motor and inner ring jamming fault model. At this time, the gyroscope experiences jamming and stalling jittering. First, use a sine function to calculate the gyroscope jamming jittering angular acceleration signal; then integrate it to obtain the gyroscope jamming jittering angular velocity signal; then establish a gyroscope motor and inner ring jamming fault injection function; then, based on the gyroscope pitch rotation angular velocity signal under the random poor contact fault of the gyroscope power supply, superimpose the gyroscope jamming jittering angular velocity signal to obtain the pitch rotation angular velocity signal under the gyroscope motor and inner ring jamming fault, and then integrate it to obtain the measurement signal of the pitch gyroscope.

[0011] Step S60: Based on the measurement signal from the pitch gyroscope, the platform input signal is set to white noise and compared with the measurement signal from the pitch gyroscope to obtain the platform pitch error signal. Based on the platform pitch error signal, error proportional signal, error integral signal, and error lead differential signal are constructed. Then, the platform pitch comprehensive control signal is superimposed. Next, a DC torque motor model is constructed to obtain the platform motor output torque signal. Then, a second-order transfer function is used to construct an inertial platform load rotation integral model, which is input to the platform motor output torque signal. The platform output angle signal is obtained and simultaneously used as the input signal for the platform pitch gyroscope, thus completing the closed-loop calculation of the entire system and realizing the simulation of gyroscope soft fault injection and dynamic output results for the inertial platform.

[0012] In one example embodiment of the present invention, the initial value of the input signal of the platform pitch gyroscope is set to 0. A measurement dynamic model of the pitch gyroscope is established using a difference equation. The measurement error signal of the pitch gyroscope is obtained by comparing the input signal of the pitch gyroscope with the measurement signal of the pitch gyroscope. Then, a measurement high-frequency noise amplification fault model is established, and a high-frequency noise amplification fault injection function is designed to obtain a measurement high-frequency noise amplification fault signal. The measurement error signal is then superimposed on the measurement high-frequency noise amplification fault signal. The measurement error signal of the gyroscope under fault injection is obtained by solving the problem. After low-pass processing, the signal is input into the measurement dynamic model of the gyroscope to obtain the pitch angular acceleration signal of the gyroscope. Finally, the pitch rotation angular velocity signal of the gyroscope is obtained by integration, including:

[0013] r(1) = 0;

[0014] e(n) = r(n) - y(n);

[0015] e wi (n)=a i sin(iω w t);

[0016]

[0017]

[0018] ebd (n)=(e a (n)-e b (n)) / T1;

[0019] e b (n+1)=e b (n)+e bd (n)T;

[0020]

[0021] ω w (n)=ω w (n+1)+Tω dd (n);

[0022] Where r(n) is the input signal of the pitch gyroscope; r(1) is the initial value of the input signal of the pitch gyroscope; e(n) is the measurement error signal of the pitch gyroscope; y(n) is the measurement signal of the pitch gyroscope; where ω wg Here, is a constant parameter, and is the fundamental frequency of the high-frequency noise; a i g is the amplitude of the i-th high-frequency noise amplification; i The fault injection function for the i-th high-frequency noise amplification; t is The start time for injecting faults that amplify high-frequency noise; t ie The fault injection end time is the i-th high-frequency noise amplification fault; j is the total number of high-frequency noise amplification faults; e wi (n) represents the measurement of high-frequency noise amplification fault signal, e a (n) represents the measurement error signal of the gyroscope under fault injection; where T1 is the low-pass time constant of the gyroscope; e bd (n) represents the pitch error low-pass rate signal; e b (n) represents the pitch error low-pass signal, initially set to 0; T is a constant integration parameter; ω n ε is the natural frequency parameter of the gyroscope, which is a constant; a ω is the damping ratio parameter of the gyroscope, which is a constant; dd (n) represents the pitch acceleration signal of the gyroscope; ω w (n) represents the pitch rotation angular velocity signal of the gyroscope;

[0023] In one exemplary embodiment of the present invention, the angular velocity signal at the instant of the gyroscope power failure is captured to obtain the initial angular velocity signal of the gyroscope power failure. Then, a gyroscope speed decay model caused by the power failure is established using a transfer function to obtain the decay speed signal of the gyroscope power failure. Next, a gyroscope power failure injection function is established, and the pitch rotation angular velocity signal of the gyroscope is superimposed to obtain the pitch rotation angular velocity signal under the gyroscope power failure injection, including:

[0024]

[0025]

[0026]

[0027] ω w2 (n)=ω w (n)-h1ω w1 (n);

[0028] Where t c1 ω represents the start time of the gyroscope power supply failure. wa T represents the initial angular velocity at which the gyroscope power supply fails; where T a ω is the time constant for the gyroscope's rotational speed to dissipate after a power outage; s is the differential operator of the transfer function; w1 (n) represents the fading rotational speed signal after a power outage of the gyroscope power supply; where t h1 t represents the start time of the gyroscope power supply failure. h2 h1 is the end time of the gyroscope power failure fault; h1 is the gyroscope power failure fault injection function; ω w2 (n) is the pitch rotation angular velocity signal injected under the power failure of the gyroscope power supply;

[0029] In one exemplary embodiment of the present invention, the gyroscope angular velocity at the moment of instantaneous power supply failure is captured; then, the angular acceleration signal of the instantaneous power supply failure is calculated based on the differential model of the instantaneous power supply failure model, and then integrated to obtain the angular velocity signal of the instantaneous power supply failure; then, a gyroscope power supply failure injection function is designed; then, based on the pitch rotation angular velocity signal under the injection of the gyroscope power supply failure, the instantaneous power supply failure angular velocity signal is superimposed to obtain the gyroscope pitch rotation angular velocity signal under the injection of the instantaneous power supply failure, including:

[0030]

[0031] ω dw3 (n)=c d (ω wb (n)-ω w3 (n)) / T b ;

[0032] ω w3 (n+1)=ω w3 (n)+Tω dw3 (n+1);

[0033]

[0034] ω w4(n)=ω w2 (n)+h2ω w3 (n);

[0035] Where ω wb t represents the gyroscope angular velocity at the moment of instantaneous abnormal failure of the gyroscope power supply. c2 ω represents the start time of the instantaneous abnormal fault in the gyroscope power supply. dw3 This refers to the instantaneous abnormal angular acceleration signal of the gyroscope power supply; c d T represents the intensity parameter of the instantaneous anomaly in the gyroscope power supply; it is a constant parameter. b ω is the time constant for instantaneous anomalies in the gyroscope power supply; it is a constant parameter. w3 h1 represents the instantaneous angular velocity signal of the gyroscope power supply; h2 represents the instantaneous fault injection function of the gyroscope power supply; ω w4 The pitch and rotation angular velocity signal of the gyroscope is injected under instantaneous abnormal fault of the gyroscope power supply;

[0036] In one exemplary embodiment of the present invention, the gyroscope rotation angular velocity signal is captured when a random contact failure of the gyroscope power supply occurs; then, the equivalent negative angular acceleration signal of the gyroscope under the random contact failure fault is calculated according to the differential model of the random contact failure of the gyroscope power supply, and then integrated to obtain the equivalent negative angular velocity signal of the gyroscope; then, an injection function for the random contact failure of the gyroscope power supply is established, and the equivalent negative angular velocity signal of the gyroscope is superimposed on the pitch rotation angular velocity signal of the gyroscope under the instantaneous abnormal fault injection of the gyroscope power supply to obtain the pitch rotation angular velocity signal of the gyroscope under the random contact failure of the gyroscope power supply, including:

[0037]

[0038] ω dw5 (n)=(ω wc (n)-ω w5 (n)) / T c ;

[0039] ω w5 (n+1)=ω w5 (n)+Tω dw5 (n+1);

[0040]

[0041] ω w6 (n)=ω w4 (n)-h3ω w5 (n);

[0042] Where ω wc (n+1) represents the gyroscope's rotational angular velocity signal when a random contact failure occurs in the gyroscope's power supply; t c3is a random constant parameter, representing the start time of the random contact failure in the gyroscope power supply; l is a natural number; represents the number of cycles the random contact failure in the gyroscope power supply lasts; T c ω is the differential time constant for random contact failures in the gyroscope power supply, used to describe how quickly the equivalent effect of the contact failure disappears; dw5 (n) represents the equivalent negative angular acceleration signal of the gyroscope under random contact failure, ω w5 (n+1) represents the equivalent negative angular velocity signal of the gyroscope; h3 is the injection function for random contact failure of the gyroscope power supply, ω w6 (n) is the pitch and rotation angular velocity signal of the gyroscope under random contact failure of the gyroscope power supply.

[0043] In one exemplary embodiment of the present invention, a sine function is used to calculate the gyroscope jamming angular acceleration signal; then, it is integrated to obtain the gyroscope jamming angular velocity signal; a gyroscope motor and inner loop jamming fault injection function is then established; and based on the gyroscope pitch rotation angular velocity signal under a random contact failure fault in the gyroscope power supply, the gyroscope jamming angular velocity signal is superimposed to obtain the pitch rotation angular velocity signal under a gyroscope motor and inner loop jamming fault, which is then integrated to obtain the measurement signal of the pitch gyroscope, including:

[0044]

[0045] ω w7 (n+1)=ω w7 (n)+Tω wd (n);

[0046]

[0047] ω w8 (n)=(1-h4)ω w6 (n)+h4ω w7 (n);

[0048] y(n+1)=y(n)+Tω w8 (n);

[0049] Where ω wd (n) represents the angular acceleration signal of the gyroscope jamming and jittering, t c4s t represents the time at which the gyroscope motor and inner ring jamming fault begins. c4e The time when the gyroscope motor and inner ring jamming fault ends; r d ω6 represents the amplitude of the gyroscope's jamming angular acceleration, which is a constant parameter; ω6 represents the frequency of the gyroscope's jamming, which is also a constant parameter; ω6 w7 (n+1) is the angular velocity signal of the gyroscope stuck and jittering; ω w8(n) is the pitch rotation angular velocity signal under the fault of the gyroscope motor and the inner ring jam, and y(n+1) is the measurement signal of the pitch gyroscope.

[0050] In one exemplary embodiment of the present invention, based on the measurement signal of the pitch gyroscope, the platform input signal is set to white noise and compared with the measurement signal of the pitch gyroscope to obtain the platform pitch error signal; an error proportional signal, an error integral signal, and an error lead differential signal are constructed based on the platform pitch error signal; then, the platform pitch comprehensive control signal is superimposed; then, a DC torque motor model is constructed to obtain the platform motor output torque signal; then, a second-order transfer function is used to construct an inertial platform load rotation integral model, and the platform motor output torque signal is input; finally, the platform output angle signal is obtained.

[0051] e2(n) = r2(n) - y(n);

[0052] s1(n) = s1(n) + Te2(n);

[0053]

[0054] u1=k1e2+k2s2+k3e d ;

[0055]

[0056]

[0057] Where r2(n) is the platform input signal, which is a Gaussian-distributed white noise signal; e2(n) is the platform pitch error signal; s1(n) is the error integral signal; k1e2 is the error proportional signal; and e d (n) represents the error leading differential signal; Let T be the transfer function of the leading differential of the error. d2 T d1 is a constant parameter, which is the constant coefficient of the lead differential transfer function; k1, k2, and k3 are constant control parameters; u1 is the platform pitch control signal; The transfer function for the DC torque motor model; k r1 T is the equivalent amplification factor of the DC motor, which is a constant parameter; r1 M1 is the equivalent time constant of the DC motor, which is a constant parameter; M2 is the output torque signal of the platform motor. Let k be the transfer function of the integral model of the load rotation of the inertial platform. r2 T is the equivalent amplification factor for the platform load rotation integral model, and it is a constant parameter; r2 is the equivalent time constant of the platform load rotation integral model, which is a constant parameter; r(n) is the input signal of the platform pitch gyroscope.

[0058] Beneficial effects

[0059] This invention discloses a soft fault injection method for inertial platform gyroscopes, with the following four main innovations: First, it employs a soft fault injection approach to model and simulate five typical gyroscope faults, resulting in a comprehensive and detailed analysis of the gyroscope fault injection process. Furthermore, the soft fault injection method is concise, novel, and practical. Second, it simultaneously utilizes difference equations and transfer functions to simulate multiple physical components of the inertial platform, such as the DC motor, gyroscope, platform proportional-integral-derivative integrated controller, and inertial platform rotation integral model. This simplifies the solution process without sacrificing accuracy. Third, it uses the multiplication of the fault injection time function with the fault signal. Building upon the original series connection of each fault level, it achieves convenient, direct, and system-wide superposition and series processing of various faults without conflict through different addition and subtraction methods. Fourth, it cleverly employs the concept of equivalence, establishing a differential model for random contact failure faults in the gyroscope power supply. It calculates the equivalent negative angular acceleration signal of the gyroscope under random contact failure faults, then integrates it to obtain the equivalent negative angular velocity signal of the gyroscope, thus simulating random contact failure faults.

[0060] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0061] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0062] Figure 1 This is a flowchart of a soft fault injection method for an inertial platform gyroscope provided by the present invention;

[0063] Figure 2 This is the pitch rotation angular velocity signal curve of the gyroscope (unit: radians per second) provided by the embodiment of the present invention;

[0064] Figure 3 This is the equivalent gyroscope negative angular velocity signal curve (unit: radians per second) of the method provided in the embodiments of the present invention;

[0065] Figure 4 It is the gyroscope rotation angular velocity signal (unit: radians per second) when a random contact failure occurs in the gyroscope power supply provided in the embodiments of the present invention.

[0066] Figure 5 This is the gyroscope jamming jitter angular velocity signal curve (unit: radians per second) of the method provided in the embodiments of the present invention;

[0067] Figure 6 This is the measurement signal curve (unit: degrees) of the pitch gyroscope provided in the embodiments of the present invention;

[0068] Figure 7 This is the platform motor output torque signal curve (unitless) of the method provided in the embodiments of the present invention;

[0069] Figure 8 This is the input signal curve (unit: degrees) of the platform pitch gyroscope of the method provided in the embodiments of the present invention. Detailed Implementation

[0070] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make the invention more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of the invention. However, those skilled in the art will recognize that the technical solutions of the invention may be practiced with one or more of these specific details omitted, or other methods, components, apparatus, steps, etc., may be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of the invention.

[0071] This invention provides a soft fault injection method for inertial platform gyroscopes. It employs the transfer function method to establish a gyroscope measurement model, an inertial platform DC torque motor model, and an inertial platform load rotation integral model. Then, it establishes a soft fault differential simulation model for five types of gyroscope faults: high-frequency noise amplification faults, gyroscope power supply failure faults, gyroscope power supply instantaneous anomaly faults, gyroscope power supply random contact failures, and gyroscope motor and inner loop jamming faults. This model yields the gyroscope pitch and rotation speed signals under fault injection. Based on this, a fault injection function is used to superimpose the fault signals to obtain the gyroscope measurement signals under fault injection. These signals are then compared with the inertial platform input signals to obtain the error signal. A proportional-integral-lead-derivative control model for the gyroscope platform is established to achieve dynamic calculation of the entire gyroscope platform under soft fault injection, realizing closed-loop calculation of fault injection and simulation. This method has the advantages of detailed and precise gyroscope fault injection modeling and high accuracy in fault modeling and simulation.

[0072] The following will further explain and illustrate the present invention's method for injecting soft faults into an inertial platform gyroscope, with reference to the accompanying drawings. (Reference) Figure 1 As shown, this method for injecting soft faults into an inertial platform gyroscope may include the following steps:

[0073] Step S10: Set the initial value of the input signal of the platform pitch gyroscope to 0. Establish the measurement dynamic model of the pitch gyroscope using a difference equation. Compare the input signal of the pitch gyroscope with the measurement signal of the pitch gyroscope to obtain the measurement error signal of the pitch gyroscope. Then, establish a measurement high-frequency noise amplification fault model, design a high-frequency noise amplification fault injection function, and obtain the measurement high-frequency noise amplification fault signal. Then, superimpose the measurement high-frequency noise amplification fault signal on the measurement error signal. Solve the calculation to obtain the measurement error signal of the gyroscope under fault injection. Perform low-pass operation and input it into the measurement dynamic model of the gyroscope to obtain the pitch angular acceleration signal of the gyroscope. Then, perform integration to obtain the pitch rotation angular velocity signal of the gyroscope.

[0074] Specifically, this can be broken down into the following four steps. Step 1: Set the initial value of the input signal of the platform's pitch gyroscope to 0. Compare the input signal of the pitch gyroscope with the measurement signal of the pitch gyroscope to obtain the measurement error signal of the pitch gyroscope.

[0075] r(1) = 0;

[0076] e(n) = r(n) - y(n);

[0077] Where r(n) is the input signal of the pitch gyroscope; r(1) is the initial value of the input signal of the pitch gyroscope; e(n) is the measurement error signal of the pitch gyroscope; and y(n) is the measurement signal of the pitch gyroscope.

[0078] The second step involves establishing a high-frequency noise amplification fault model, designing a high-frequency noise amplification fault injection function, obtaining the high-frequency noise amplification fault signal, and superimposing the high-frequency noise amplification fault signal onto the measurement error signal. The resulting measurement error signal of the gyroscope under fault injection is then calculated as follows:

[0079] e wi (n)=a i sin(iω w t);

[0080]

[0081]

[0082] Where ω wg Here, is a constant parameter, and is the fundamental frequency of the high-frequency noise; a i g is the amplitude of the i-th high-frequency noise amplification;i The fault injection function for the i-th high-frequency noise amplification; t is The start time for injecting faults that amplify high-frequency noise; t ie The fault injection end time is the i-th high-frequency noise amplification fault; j is the total number of high-frequency noise amplification faults; e wi (n) represents the measurement of high-frequency noise amplification fault signal, e a (n) represents the measurement error signal of the gyroscope under fault injection;

[0083] The third step involves performing a low-pass operation on the gyroscope's measurement error signal under fault injection to obtain the pitch error low-pass rate signal; then, after integration, the pitch error low-pass signal is obtained as follows:

[0084] e bd (n)=(e a (n)-e b (n)) / T1;

[0085] e b (n+1)=e b (n)+e bd (n)T;

[0086] Where T1 is the low-pass time constant of the gyroscope; e bd (n) represents the pitch error low-pass rate signal; e b (n) represents the pitch error low-pass signal, with an initial value of 0; T is a constant integration parameter.

[0087] The fourth step involves substituting the low-pass pitch error signal into the gyroscope's measurement dynamic model to obtain the gyroscope's pitch angular acceleration signal; then, integration is performed to obtain the gyroscope's pitch rotation angular velocity signal as follows:

[0088]

[0089] ω w (n)=ω w (n+1)+Tω dd (n);

[0090] Where ω n ε is the natural frequency parameter of the gyroscope, which is a constant; a ω is the damping ratio parameter of the gyroscope, which is a constant; dd (n) represents the pitch acceleration signal of the gyroscope; ω w (n) represents the pitch and rotation angular velocity signal of the gyroscope.

[0091] Step S20: Based on the pitch rotation angular velocity signal of the gyroscope, establish a gyroscope power supply fault model and a gyroscope power outage fault model. First, capture the instantaneous angular velocity signal of the gyroscope power outage fault to obtain the initial angular velocity signal of the gyroscope power outage fault. Then, use a transfer function to establish a gyroscope speed decay model caused by power outage to obtain the decay speed signal of the gyroscope power outage fault. Next, establish a gyroscope power outage fault injection function and superimpose the pitch rotation angular velocity signal of the gyroscope to obtain the pitch rotation angular velocity signal under the injection of the gyroscope power outage fault.

[0092] Specifically, this can be broken down into the following three steps. The first step is to capture the angular velocity signal at the instant of the gyroscope power failure, obtaining the initial angular velocity signal of the gyroscope power failure as follows:

[0093]

[0094] Where t c1 ω represents the start time of the gyroscope power supply failure. wa This is the initial angular velocity at which the gyroscope power supply fails.

[0095] The second step involves establishing a gyroscope speed decay model caused by a power outage using a transfer function. The decay speed signal due to the power outage fault is as follows:

[0096]

[0097] Where T a ω is the time constant for the gyroscope's rotational speed to dissipate after a power outage; s is the differential operator of the transfer function; w1 (n) is the fading speed signal when the gyroscope power supply fails.

[0098] The third step involves establishing a gyroscope power outage injection function based on the fading rotation speed signal of the gyroscope power outage fault. This function is then superimposed with the pitch rotation angular velocity signal of the gyroscope to obtain the pitch rotation angular velocity signal under the gyroscope power outage fault injection, as follows:

[0099]

[0100] ω w2 (n)=ω w (n)-h1ω w1 (n);

[0101] At this time t h1 t represents the start time of the gyroscope power supply failure. h2 h1 is the end time of the gyroscope power failure fault; h1 is the gyroscope power failure fault injection function; ω w2 (n) is the pitch rotation angular velocity signal injected under the power failure of the gyroscope.

[0102] Step S30: Establish a differential model for the instantaneous fault of the gyroscope power supply. First, capture the gyroscope angular velocity when the instantaneous fault occurs. Then, calculate the angular acceleration signal of the instantaneous fault based on the differential model of the instantaneous fault, and then integrate it to obtain the angular velocity signal of the instantaneous fault. Next, design the injection function for the instantaneous fault. Then, based on the pitch rotation angular velocity signal under the injection of the gyroscope power supply power failure fault, obtain the pitch rotation angular velocity signal of the gyroscope under the injection of the instantaneous fault.

[0103] Specifically, this can be broken down into the following three steps. The first step is to extract the gyroscope angular velocity at the moment of the instantaneous abnormal fault in the gyroscope power supply based on the gyroscope pitch and rotation angular velocity signal, as follows:

[0104]

[0105] Where ω wb t represents the gyroscope angular velocity at the moment of instantaneous abnormal failure of the gyroscope power supply. c2 This represents the start time of the instantaneous abnormal fault in the gyroscope power supply.

[0106] The second step involves calculating the angular acceleration signal of the instantaneous anomaly of the gyroscope power supply using the differential model of the instantaneous anomaly fault model, and then integrating it to obtain the angular velocity signal of the instantaneous anomaly of the gyroscope power supply as follows:

[0107] ω dw3 (n)=c d (ω wb (n)-ω w3 (n)) / T b ;

[0108] ω w3 (n+1)=ω w3 (n)+Tω dw3 (n+1);

[0109] Where ω dw3 This refers to the instantaneous abnormal angular acceleration signal of the gyroscope power supply; c d T represents the intensity parameter of the instantaneous anomaly in the gyroscope power supply; it is a constant parameter. b ω is the time constant for instantaneous anomalies in the gyroscope power supply; it is a constant parameter. w3 This is the instantaneous abnormal angular velocity signal of the gyroscope power supply.

[0110] The third step is to design a function to inject instantaneous faults into the gyroscope power supply. Then, based on the pitch rotation angular velocity signal injected under the power failure fault of the gyroscope power supply, the instantaneous abnormal angular velocity signal of the gyroscope power supply is superimposed to obtain the gyroscope pitch rotation angular velocity signal under the instantaneous abnormal fault of the gyroscope power supply as follows:

[0111]

[0112] ω w4 (n)=ω w2 (n)+h2ω w3 (n);

[0113] Where h2 is the instantaneous fault injection function for the gyroscope power supply; ω w4 The pitch and rotation angular velocity signal of the gyroscope is injected during a momentary abnormal fault in the gyroscope power supply.

[0114] Step S40: Establish a differential model for the random contact failure of the gyroscope power supply. First, capture the gyroscope rotation angular velocity signal when the random contact failure of the gyroscope power supply occurs. Then, calculate the equivalent negative angular acceleration signal of the gyroscope under the random contact failure fault according to the differential model of the random contact failure of the gyroscope power supply, and then integrate it to obtain the equivalent negative angular velocity signal of the gyroscope. Then, establish an injection function for the random contact failure of the gyroscope power supply. Based on the pitch rotation angular velocity signal of the gyroscope under the instantaneous abnormal fault injection of the gyroscope power supply, superimpose the equivalent negative angular velocity signal of the gyroscope to obtain the pitch rotation angular velocity signal of the gyroscope under the random contact failure of the gyroscope power supply.

[0115] Specifically, this can be broken down into the following three steps. The first step is to capture the gyroscope's rotational angular velocity signal when a random contact failure occurs in the gyroscope's power supply, as follows:

[0116]

[0117] Where ω wc (n+1) represents the gyroscope's rotational angular velocity signal when a random contact failure occurs in the gyroscope's power supply; t c3 is a random constant parameter, which represents the start time of the random contact failure of the gyroscope power supply; l is a natural number, which represents the number of cycles of the random contact failure of the gyroscope power supply.

[0118] The second step involves calculating the equivalent negative angular acceleration signal of the gyroscope under random contact failure faults using the differential model of the gyroscope power supply, and then integrating the results to obtain the equivalent negative angular velocity signal of the gyroscope as follows:

[0119] ω dw5 (n)=(ω wc (n)-ω w5 (n)) / T c ;

[0120] ω w5 (n+1)=ω w5 (n)+Tω dw5 (n+1);

[0121] Where T c ω is the differential time constant for random contact failures in the gyroscope power supply, used to describe how quickly the equivalent effect of the contact failure disappears; dw5 (n) represents the equivalent negative angular acceleration signal of the gyroscope under random contact failure, ω w5 (n+1) represents the equivalent negative angular velocity signal of the gyroscope.

[0122] The third step is to establish an injection function for a random contact failure in the gyroscope power supply. Based on the pitch and rotation angular velocity signal of the gyroscope under the instantaneous abnormal fault injection of the gyroscope power supply, an equivalent negative angular velocity signal of the gyroscope is superimposed to obtain the gyroscope pitch and rotation angular velocity signal under the random contact failure in the gyroscope power supply as follows:

[0123]

[0124] ω w6 (n)=ω w4 (n)-h3ω w5 (n);

[0125] Where h3 is the injection function for random contact failure of the gyroscope power supply, ω w6 (n) is the pitch and rotation angular velocity signal of the gyroscope under random contact failure of the gyroscope power supply.

[0126] Step S50: Establish a gyroscope motor and inner ring jamming fault model. At this time, the gyroscope experiences jamming and stalling jittering. First, use a sine function to calculate the gyroscope jamming jittering angular acceleration signal; then integrate it to obtain the gyroscope jamming jittering angular velocity signal; then establish a gyroscope motor and inner ring jamming fault injection function; then, based on the gyroscope pitch rotation angular velocity signal under the random poor contact fault of the gyroscope power supply, superimpose the gyroscope jamming jittering angular velocity signal to obtain the pitch rotation angular velocity signal under the gyroscope motor and inner ring jamming fault, and then integrate it to obtain the measurement signal of the pitch gyroscope.

[0127] Specifically, this can be broken down into the following three steps. First, use a sine function to calculate the gyroscope jamming angular acceleration signal; then integrate it to obtain the gyroscope jamming angular velocity signal as follows:

[0128]

[0129] ω w7 (n+1)=ω w7 (n)+Tω wd (n);

[0130] Where ω wd (n) represents the angular acceleration signal of the gyroscope jamming and jittering, t c4st represents the time at which the gyroscope motor and inner ring jamming fault begins. c4e The time when the gyroscope motor and inner ring jamming fault ends; r d ω6 is the amplitude of the angular acceleration of the gyroscope jamming, which is a constant parameter; ω6 is the frequency of the gyroscope jamming, which is a constant parameter.

[0131] The second step is to establish a fault injection function for the gyroscope motor and inner ring jamming. Then, based on the gyroscope pitch rotation angular velocity signal under a random poor contact fault in the gyroscope power supply, the gyroscope jamming jitter angular velocity signal is superimposed to obtain the pitch rotation angular velocity signal under the gyroscope motor and inner ring jamming fault as follows:

[0132]

[0133] ω w8 (n)=(1-h4)ω w6 (n)+h4ω w7 (n);

[0134] Where ω w7 (n+1) is the angular velocity signal of the gyroscope stuck and jittering; ω w8 (n) is the pitch rotation angular velocity signal under the fault of the gyroscope motor and the inner ring jam.

[0135] The third step involves integrating the pitch rotation angular velocity signal under the condition of gyroscope motor and inner ring jamming to obtain the following measurement signal from the pitch gyroscope:

[0136] y(n+1)=y(n)+Tω w8 (n);

[0137] Where y(n+1) is the measurement signal of the pitch gyroscope.

[0138] Step S60: Based on the measurement signal from the pitch gyroscope, the platform input signal is set to white noise and compared with the measurement signal from the pitch gyroscope to obtain the platform pitch error signal. Based on the platform pitch error signal, error proportional signal, error integral signal, and error lead differential signal are constructed. Then, the platform pitch comprehensive control signal is superimposed. Next, a DC torque motor model is constructed to obtain the platform motor output torque signal. Then, a second-order transfer function is used to construct an inertial platform load rotation integral model, which is input to the platform motor output torque signal. The platform output angle signal is obtained and simultaneously used as the input signal for the platform pitch gyroscope, thus completing the closed-loop calculation of the entire system and realizing the simulation of gyroscope soft fault injection and dynamic output results for the inertial platform.

[0139] Specifically, this can be broken down into the following four steps. Step 1: Based on the measurement signal from the pitch gyroscope, set the platform input signal to a white noise signal and compare it with the measurement signal from the pitch gyroscope to obtain the platform pitch error signal as follows:

[0140] e2(n) = r2(n) - y(n);

[0141] Where r2(n) is the platform input signal, which is a Gaussian distributed white noise signal, and e2(n) is the platform pitch error signal.

[0142] The second step involves constructing the proportional error signal, integral error signal, and differential error lead signal based on the platform pitch error signal; then, these are superimposed to form the overall platform pitch control signal as follows:

[0143] s1(n) = s1(n) + Te2(n);

[0144]

[0145] u1=k1e2+k2s2+k3e d ;

[0146] Where s1(n) is the error integral signal, k1e2 is the error proportional signal, and e d (n) represents the error leading differential signal; Let T be the transfer function of the leading differential of the error. d2 T d1 is a constant parameter, which is the constant coefficient of the lead differential transfer function; k1, k2, and k3 are constant control parameters; u1 is the platform pitch control signal.

[0147] The third step is to construct a DC torque motor model and obtain the platform motor output torque signal as follows:

[0148]

[0149] in The transfer function for the DC torque motor model; k r1 T is the equivalent amplification factor of the DC motor, which is a constant parameter; r1 M1 is the equivalent time constant of the DC motor, which is a constant parameter; M2 is the output torque signal of the platform motor.

[0150] The fourth step involves constructing an integral model of the inertial platform load rotation using a second-order transfer function, inputting the platform motor output torque signal; this yields the platform's output angle signal, which is then used as the input signal for the platform's pitch gyroscope, thus completing the closed-loop solution of the entire system as follows:

[0151]

[0152] in Let k be the transfer function of the integral model of the load rotation of the inertial platform. r2 T is the equivalent amplification factor for the platform load rotation integral model, and it is a constant parameter; r2 is the equivalent time constant of the platform load rotation integral model, which is a constant parameter; r(n) is the input signal of the platform pitch gyroscope.

[0153] Case Implementation and Computer Simulation Results Analysis

[0154] In step S10, j = 1, a1 = 0.01, ω wg =100, ω n =40, ε a =0.65, t 1s =6,t 1e =6.8, the gyroscope pitch and rotation angular velocity signals are as follows: Figure 2 As shown.

[0155] In step S20, it was determined that no power failure of the gyroscope power supply occurred.

[0156] In step S30, it is determined that no instantaneous abnormal fault occurred in the gyroscope power supply.

[0157] In step S40, t is selected. c3 =13, l=3; the equivalent gyroscope negative angular velocity signal is obtained as follows Figure 3 As shown, when a random contact failure occurs in the gyroscope power supply, the gyroscope's rotational angular velocity signal is as follows: Figure 4 As shown.

[0158] In step S50, r is selected d =0.01, ω6=10; t c4s =20,t c4e =24, the angular velocity signal of the gyroscope stuck and jittering is obtained as follows Figure 5 As shown, the measurement signal of the pitch gyroscope is as follows: Figure 6 As shown.

[0159] In step S60, k is selected. r1 =235, k r2 =0.01, T r1 =0.001, T r2 =1.23, thus obtaining the platform motor output torque signal as follows: Figure 7 As shown, the input signal of the platform's pitch gyroscope is as follows: Figure 8 As shown.

[0160] First, it should be noted that the probability of all five major faults occurring simultaneously is very small. Therefore, the fault selection experiments in steps S20 and S30 were not conducted, as the five faults would interfere with each other, which would be detrimental to fault analysis. Meanwhile, the experimental results show that the impact of the brief contact failure is minimal and can be ignored. However, the jamming and jittering caused by the gyroscope motor and inner ring jamming fault has a significant impact on the stability of the entire inertial platform. The jitter angle peaks at approximately 5 degrees, resulting in an impact of approximately 0.2 degrees on the gyroscope measurement, and an impact of approximately 0.15 degrees on the overall angular accuracy of the stable platform. Simultaneously, the impact of the high-frequency noise amplification fault on the overall platform's angular accuracy becomes apparent at approximately 6 seconds, with an impact of approximately 0.01 degrees.

[0161] The above experimental results show that the gyroscope soft fault injection method provided by the present invention is very effective, especially in solving the complex dynamic characteristics of the entire inertial platform under the action of gyroscope faults. It has high simulation accuracy and reasonable reliability, thus having high engineering application value.

Claims

1. A method for injecting soft faults into an inertial platform gyroscope, characterized by the following steps: Step S10: Set the initial value of the input signal of the platform pitch gyroscope to 0, establish the measurement dynamic model of the pitch gyroscope using the difference equation, and obtain the measurement error signal of the pitch gyroscope by comparing the input signal of the pitch gyroscope with the measurement signal of the pitch gyroscope. A high-frequency noise amplification fault model is then established, and a high-frequency noise amplification fault injection function is designed to obtain the high-frequency noise amplification fault signal. The high-frequency noise amplification fault signal is then superimposed on the measurement error signal. The measurement error signal of the gyroscope under fault injection is calculated, and after low-pass processing, it is input into the gyroscope's measurement dynamic model to obtain the gyroscope's pitch acceleration signal. Finally, integration is performed to obtain the gyroscope's pitch rotation angular velocity signal as follows: ； ； ； ； ； ； ； ； ； in This is the input signal for the pitch gyroscope; The initial value of the input signal for the pitch gyroscope; This is the measurement error signal of the pitch gyroscope; The measurement signal is from the pitch gyroscope; where is a constant parameter, and is the fundamental frequency of the high-frequency noise; For the first The amplitude of the high-frequency noise amplification; For the first A fault injection function that amplifies high-frequency noise; For the first The fault injection start time with amplified high-frequency noise; For the first The fault injection end time of a high-frequency noise amplification; This represents the total number of high-frequency noise amplification faults. To measure high-frequency noise amplification fault signals, The measurement error signal of the gyroscope under fault injection; where This is the low-pass time constant of the gyroscope; This is the pitch error low-pass rate signal; The pitch error low-pass signal is initialized to 0. These are constant integration parameters; This is the natural frequency parameter of the gyroscope, and it is a constant value. Here is the damping ratio parameter of the gyroscope, which is a constant. This is the pitch acceleration signal from the gyroscope; This is the pitch and rotation angular velocity signal of the gyroscope; Step S20: Based on the pitch rotation angular velocity signal of the gyroscope, a gyroscope power supply fault model and a gyroscope power outage fault model are established. First, the instantaneous angular velocity signal of the gyroscope power outage fault is captured to obtain the initial angular velocity signal of the gyroscope power outage fault. Then, a gyroscope speed decay model caused by power outage is established using a transfer function to obtain the decay speed signal of the gyroscope power outage fault. Next, a gyroscope power outage fault injection function is established, and the pitch rotation angular velocity signal of the gyroscope is superimposed to obtain the pitch rotation angular velocity signal under the injection of the gyroscope power outage fault as follows: ； ； ； ； in This is the start time of the gyroscope power supply failure. The initial angular velocity of the gyroscope in the event of a power outage; in The time constant for the gyroscope's rotational speed to decrease after a power outage; For the differential operator of the transfer function; This is the fading speed signal for a gyroscope power supply failure; where... This is the start time of the gyroscope power supply failure. This is the end time of the gyroscope power supply failure. Inject a function to handle power outage faults in the gyroscope power supply; The pitch rotation angular velocity signal injected under power failure of the gyroscope power supply; Step S30: Establish a differential model for the instantaneous fault of the gyroscope power supply. First, capture the gyroscope angular velocity at the time of the instantaneous fault. Then, calculate the angular acceleration signal of the instantaneous fault based on the differential model of the instantaneous fault, and integrate it to obtain the angular velocity signal of the instantaneous fault. Next, design the injection function for the instantaneous fault. Finally, superimpose the angular velocity signal of the instantaneous fault onto the pitch rotation angular velocity signal under the injection of the gyroscope power supply failure fault to obtain the gyroscope pitch rotation angular velocity signal under the injection of the instantaneous fault, as follows: ； ； ； ； ； in The angular velocity of the gyroscope when a momentary abnormal fault occurs in the gyroscope power supply; The start time of the instantaneous abnormal fault in the gyroscope power supply; This is the instantaneous abnormal angular acceleration signal of the gyroscope power supply; This is the intensity parameter of the instantaneous anomaly of the gyroscope power supply; it is a constant parameter. This is the time constant for instantaneous anomalies in the gyroscope power supply; it is a constant parameter. This is the instantaneous abnormal angular velocity signal of the gyroscope power supply; Inject functions to handle transient abnormal faults in the gyroscope power supply; The pitch and rotation angular velocity signal of the gyroscope is injected under instantaneous abnormal fault of the gyroscope power supply; Step S40: Establish a differential model for a random contact failure in the gyroscope power supply. First, capture the gyroscope rotation angular velocity signal when the random contact failure occurs. Then, calculate the equivalent negative angular acceleration signal of the gyroscope under the random contact failure fault based on the differential model, and then integrate to obtain the equivalent negative angular velocity signal of the gyroscope. Next, establish an injection function for the random contact failure of the gyroscope power supply. Based on the pitch rotation angular velocity signal of the gyroscope under the instantaneous abnormal fault injection of the gyroscope power supply, superimpose the equivalent negative angular velocity signal of the gyroscope to obtain the gyroscope pitch rotation angular velocity signal under the random contact failure of the gyroscope power supply as follows: ； ； ； ； in This is the gyroscope rotational angular velocity signal when a random contact failure occurs in the gyroscope power supply. This is a random constant parameter, representing the start time of a random contact failure in the gyroscope power supply. It is a natural number; it represents the number of cycles in which the random contact failure of the gyroscope power supply lasts. The differential time constant is the random contact failure of the gyroscope power supply, which is used to describe how fast the equivalent effect of the contact failure disappears. This is the equivalent negative angular acceleration signal of the gyroscope under random contact failure. This is the equivalent negative angular velocity signal of a gyroscope; This is the injection function for random contact failures in the gyroscope power supply. This is the pitch and rotational angular velocity signal of the gyroscope under random contact failure of the gyroscope power supply. Step S50: Establish a gyroscope motor and inner ring jamming fault model. At this point, the gyroscope experiences jamming and stalling. First, a sine function is used to calculate the gyroscope jamming and stalling angular acceleration signal; then, this is integrated to obtain the gyroscope jamming and stalling angular velocity signal; next, an injection function for the gyroscope motor and inner ring jamming fault is established; then, based on the gyroscope pitch rotation angular velocity signal under a random contact failure fault in the gyroscope power supply, the gyroscope jamming and stalling angular velocity signal is superimposed to obtain the pitch rotation angular velocity signal under the gyroscope motor and inner ring jamming fault. This signal is then integrated to obtain the following measurement signal from the pitch gyroscope: ； ； ； ； ； in This is the angular acceleration signal for the gyroscope stuck and jittering. This is the time when the gyroscope motor and inner ring jamming fault begins. The time when the gyroscope motor and inner ring jamming fault ends; This is the amplitude of the angular acceleration due to the gyroscope jamming and jittering; it is a constant parameter. This is the frequency of the gyroscope's jitter when it jams; it is a constant parameter. This is the angular velocity signal for the gyroscope to jitter when stuck. This is the pitch rotation angular velocity signal under the condition of a gyroscope motor and inner ring jamming fault. The measurement signal is from the pitch gyroscope; Step S60: Based on the measurement signal from the pitch gyroscope, the platform input signal is set to white noise and compared with the measurement signal from the pitch gyroscope to obtain the platform pitch error signal. Based on the platform pitch error signal, error proportional signal, error integral signal, and error lead differential signal are constructed. Then, the platform pitch comprehensive control signal is superimposed. Next, a DC torque motor model is constructed to obtain the platform motor output torque signal. Then, a second-order transfer function is used to construct an inertial platform load rotation integral model, which is input to the platform motor output torque signal. The platform output angle signal is obtained and simultaneously used as the input signal for the platform pitch gyroscope, thus completing the closed-loop calculation of the entire system and realizing the simulation of gyroscope soft fault injection and dynamic output results for the inertial platform. ； ； ； ； ； ； in The input signal to the platform is a Gaussian-distributed white noise signal. This is the platform pitch error signal; For the error integral signal, This is the error ratio signal. The leading differential signal is the error signal; Let the transfer function be the leading differential of the error. , These are constant parameters, which are the constant coefficients of the lead differential transfer function; , , These are constant control parameters; This is the platform pitch control signal; This is the transfer function for the DC torque motor model; This is the equivalent amplification factor of the DC motor, which is a constant parameter; is the equivalent time constant of the DC motor, which is a constant parameter; Output torque signal to the platform motor; Let be the transfer function of the integral model of the inertial platform load rotation. This is the equivalent amplification factor for the platform load rotation integral model, and it is a constant parameter. This is the equivalent time constant of the platform load rotation integral model, and it is a constant parameter; This is the input signal for the platform's pitch gyroscope.

Images