Rotary inertial navigation digital twin system and construction method thereof

Abstract

The invention relates to a rotary inertial navigation digital twinning system and a construction method thereof. The rotary inertial navigation digital twinning system comprises a rotary inertial navigation digital twinning physical model layer, a rotary inertial navigation digital twinning geometric model layer, a multi-source information data acquisition layer, a decision control layer and a diagnosis and prediction analysis layer. Data information collection and transmission of a rotary inertial navigation system entity can be completed, association and interaction between a twin virtual body and the entity are achieved, iterative optimization of model parameters can be achieved through the decision control module, and finally prediction and evaluation of the overall operation state of the system and the performance of the digital twin inertial navigation system are achieved in the diagnosis and prediction analysis module.

Description

Technical Field

[0001] The present invention belongs to the field of inertial navigation technology, and specifically relates to a rotating inertial navigation digital twin system and a construction method thereof. Background Art

[0002] With the rapid development of emerging technologies such as artificial intelligence, big data, and the Internet, the intelligent development of equipment has a reliable foundation. As one of the key technologies for the intelligent development of equipment, digital twin technology can realize the synchronous coexistence of digital models and physical entities or processes in digital space, and complete active intervention and dynamic prediction of the performance status of physical entities.

[0003] At present, digital twins have been widely used in the intelligent manufacturing industry, but have been less used in the field of inertial navigation. For example, Ma Yitong et al. published a study on the application of digital twin technology in inertial systems [J], Missiles and Space Launch Technology, 6: 37-45; Inertial navigation big data analysis and application based on digital twin technology [J]. Cybersecurity and Data Governance, 2023, 42(S1): 76-81) both proposed the application of digital twin technology in inertial systems throughout the product life cycle, but did not give a specific implementation plan for digital twins at the device level.

[0004] To this end, the present invention proposes a method for constructing a rotating inertial navigation digital twin system at the equipment level to achieve the optimization, decision-making, control and maintenance of rotating inertial navigation products, maximize the value of data and models at the equipment level, and promote the digital development of rotating inertial navigation equipment. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention proposes a rotating inertial navigation digital twin system and a construction method thereof.

[0006] One of the above objects of the present invention is achieved by the following technical solution:

[0007] A rotating inertial navigation digital twin system, comprising a rotating inertial navigation digital twin physical model layer, a rotating inertial navigation digital twin geometric model layer, a multi-source information data acquisition layer, a decision control layer, and a diagnosis and prediction analysis layer;

[0008] The rotating inertial navigation digital twin geometric model layer adopts a data-driven approach to achieve dynamic characterization of the inertial navigation operation process, ensuring synchronous symbiosis with the inertial navigation entity;

[0009] The rotating inertial navigation digital twin physical model layer includes a fiber optic gyroscope, an accelerometer, a grating, a servo system and a rotating inertial navigation solution model;

[0010] Multi-source information data acquisition layer: contains a variety of sensors for real-time collection of current, voltage, acceleration, angular velocity, temperature, stress and strain information. At the same time, the internal solution data of the rotating inertial navigation is provided for the interaction of the twin model to form an effective data set to be fed back to the digital twin geometric model and physical model;

[0011] The decision control layer: based on the digital twin physical model, combined with real-time multi-source data, visually detects the operation process of the inertial navigation system; based on the measured data, by establishing a dynamic mapping between the physical entity and the physical model, it makes real-time predictions of the inertial navigation system; based on the performance prediction results, it optimizes the relevant parameters of the inertial navigation system;

[0012] Diagnosis and prediction analysis layer: It is used to extract features based on the original collected information provided by the multi-source data collection layer, and perform frequency domain analysis, time domain analysis, statistical analysis or other characteristics related to the operation status of the rotating inertial navigation system, and compare with the set thresholds or fault characteristics for monitoring and alarm; at the same time, the real-time performance prediction results of the decision control layer are compared with the gyroscope temperature compensation data, accelerometer temperature compensation data, navigation solution data, etc. solved inside the rotating inertial navigation, to evaluate the actual performance of the inertial navigation, and provide future performance prediction results for reference.

[0013] Moreover, the basic model of fiber optic gyroscope and accelerometer is as follows:

[0014]

[0015] ω out is the angular velocity value output by the gyroscope detection, I is the unit matrix, ΔG ω is the gyro installation error, δK ω is the gyroscope scale factor error, is the theoretical value of angular velocity, ε p is the gyro constant bias; a out is the acceleration value detected and output by the accelerometer, I is the unit matrix, ΔG a is the accelerometer installation error, δK g is the accelerometer scale factor error, is the theoretical value of acceleration, β p is the constant zero bias of the accelerometer.

[0016] Moreover, in step 1, the temperature compensation models of the scale factor error, installation error and zero bias error of the gyroscope and accelerometer all adopt the BP neural network method; the relationship between the input and output of the BP neural network is:

[0017] Q i =tansig(W i,1 P i )+bi,1

[0018] O i =purelin(W i,2 Q i )+b i,2

[0019] Where: i is the scale factor error, installation error or zero bias error; P i is the system input for scale factor error, installation error or zero bias error, i.e. temperature; b i,1 is the hidden layer bias of scale factor error, installation error or zero bias error; b i,2 is the output layer offset of the scale factor error, installation error or zero bias error; W i,1 is the hidden layer weight matrix; W i,2 is the output layer weight matrix; Q i is the hidden layer output; O i is the system output of scale factor error, installation error or zero bias error; tansig is the hyperbolic tangent S-type function; purelin is the linear function.

[0020] Moreover, the servo motor in the servo system generally directly drives the rotating mechanism to rotate, and its basic model is as follows:

[0021]

[0022] In the formula, J0 is the moment of inertia of the servo motor shaft, B0 is the servo motor rotation damping coefficient, K0 is the servo motor rotation stiffness, θ is the motor rotation angle, is the first-order derivative of the motor rotation angle, also called the motor angular velocity, It is the second-order derivative of the motor rotation angle, also called the motor angular acceleration.

[0023] Moreover, the rotational inertial navigation solution model includes attitude differential equations, velocity differential equations and position differential equations; wherein the attitude differential equation is:

[0024]

[0025] In the formula, q0, q1, q2, q3 are the four basic elements of the quaternion equation. is the differential of the quaternion, is the projection of the angular velocity of the carrier coordinate system b relative to the navigation coordinate system n in the three directions of the b system;

[0026] The velocity differential equation is:

[0027]

[0028] In the formula, V nis the velocity of the carrier in the navigation coordinate system, V n The differential of is the attitude matrix of the carrier coordinate system b relative to the navigation coordinate system n, is the projection of the angular velocity of the earth coordinate system e relative to the inertial coordinate system i in the navigation coordinate system n, is the projection of the angular velocity of the navigation coordinate system n relative to the earth coordinate system e in the navigation coordinate system n, g b is the projection of gravitational acceleration in the carrier coordinate system b;

[0029] The position differential equation is:

[0030]

[0031] is the differential of the longitude of the carrier’s location, is the differential of the latitude of the carrier, L is the latitude of the carrier, and h is the height of the carrier. is the differential of the height h of the carrier, R M The radius of curvature of the earth along the meridian where the carrier is located, R N The radius of curvature of the earth along the y-axis at the position of the carrier, V E 、V N 、V U They are respectively the velocity of the carrier in the navigation coordinate system and the components in the three directions of the northeast sky.

[0032] One of the above objects of the present invention is achieved by the following technical solution:

[0033] Step 1: Use 3D modeling software to establish a rotating inertial navigation digital twin geometric model;

[0034] Step 2: Establish and simulate the physical model of the rotating inertial navigation digital twin in a digital way;

[0035] Step 3: In the digital twin platform software, synchronize the geometric model and the physical model to visualize the motion process and input and output results;

[0036] Step 4: Build a data acquisition module to complete data acquisition and transmit it to the digital twin platform for interactive use by the twin model to achieve the association between the twin virtual body and the entity;

[0037] Step 5: Build a decision-making control module and use machine learning and deep learning data-driven modeling methods to achieve iterative optimization of model parameters;

[0038] Step 6: Build a diagnostic and predictive analysis module to evaluate the overall operating status of the system and the performance of the digital twin inertial navigation system, and provide a reference.

[0039] Moreover, in step 1, a geometric model of the rotating inertial navigation system, components and parts is established using 3D modeling software, and assembly and motion relationships are constructed according to the composition of the actual equipment to characterize the relationship between the various components, and finite element models of key components of the rotating system such as motors and bearings are established to characterize their forces and deformations.

[0040] Furthermore, step 5 includes:

[0041] Utilize the graphical display function in the ANSYS Twin Builder and Unity Manufacturing digital twin platforms to display the data information collected by the multi-source information data acquisition layer in a graphically visualized manner; determine the input of the digital twin model as the real-time collected current, voltage, acceleration, angular velocity, temperature, stress-strain information, as well as temperature compensation, axis angle compensation, and navigation solution information; build a twin solution model through the digital twin physical model, and combine it with data-driven modeling methods such as machine learning and deep learning, and output the predicted data of each sensor of the rotating inertial navigation and the predicted data of the navigation solution performance; based on the prediction results of each sensor and the predicted results of the navigation solution performance, and compared with the real-time data collected subsequently, optimize the parameters or structure of each digital twin physical model in the rotating inertial navigation system.

[0042] Moreover, building the diagnosis and prediction analysis module in step 6 includes: first performing model training, preprocessing the original acquisition information provided by the multi-source data acquisition layer and the characteristic information related to the operation status of the rotating inertial navigation system, including processing missing data, correcting abnormal data, etc., to improve data quality and avoid errors in subsequent data analysis; then extracting features from the data, converting the original data into representative feature vectors, and using data dimensionality reduction methods such as principal component analysis and linear discriminant analysis to reduce the dimensionality of the high-dimensional original feature vectors to improve computing efficiency; then passing the reduced feature vectors into the digital twin deep learning model, and obtaining the training and evaluation model.

[0043] Moreover, in step 6, the collected data is input into the trained deep learning model, and the prediction analysis data is output; and it is compared with the set threshold or fault characteristics. If the threshold is exceeded or a fault characteristic appears, an alarm will be given in time; at the same time, the real-time performance prediction results of the decision control layer are compared with the gyroscope temperature compensation data, accelerometer temperature compensation data, navigation solution data, etc. solved inside the rotating inertial navigation, and combined with the analysis results of the original collected data, the performance of the digital twin inertial navigation system is evaluated, and the accuracy of the performance prediction results is judged.

[0044] The advantages and positive effects of the present invention are:

[0045] The present invention provides a rotating inertial navigation digital twin system and a construction method thereof, which can complete the data information collection and transmission of the rotating inertial navigation system entity, realize the association and interaction between the twin virtual body and the entity, and use the decision control module to realize the iterative optimization of the model parameters. Finally, the overall operation status of the system and the performance prediction and evaluation of the digital twin inertial navigation system are realized in the diagnosis and prediction analysis module. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a composition diagram of a rotating inertial navigation system based on digital twin of the present invention;

[0047] Figure 2 It is a schematic diagram of multi-source data composition of the present invention. DETAILED DESCRIPTION

[0048] The structure of the present invention is further described below with reference to the accompanying drawings and by way of examples. It should be noted that the present examples are descriptive rather than restrictive.

[0049] A rotating inertial navigation digital twin system, see Figure 1-Figure 2 The invention points are: including the rotating inertial navigation digital twin physical model layer, the rotating inertial navigation digital twin geometric model layer, the multi-source information data acquisition layer, the decision control layer, and the diagnosis and prediction analysis layer. The specific description of each layer is as follows:

[0050] Rotating inertial navigation digital twin geometric model layer: A data-driven approach is used to achieve dynamic characterization of the inertial navigation operation process, ensuring synchronous symbiosis with the inertial navigation entity. The digital twin geometric model is the basis of the rotating inertial navigation digital twin model.

[0051] Rotating inertial navigation digital twin physical model layer: mainly includes models such as fiber optic gyroscope, accelerometer, grating, servo system and rotating inertial navigation solution.

[0052] The basic model of fiber optic gyroscope and accelerometer is as follows:

[0053]

[0054] ω out is the angular velocity value output by the gyroscope detection, I is the unit matrix, ΔG ω is the gyro installation error, δK ω is the gyroscope scale factor error, is the theoretical value of angular velocity, ε p is the gyro constant bias; a out is the acceleration value detected and output by the accelerometer, I is the unit matrix, ΔG a is the accelerometer installation error, δK g is the accelerometer scale factor error, is the theoretical value of acceleration, βp is the constant zero bias of the accelerometer;

[0055] The temperature compensation models for the scale factor error, installation error and bias error of the gyroscope and accelerometer all use the BP neural network method. The BP neural network is an important component of the artificial neural network. It contains multiple hidden layers and has nonlinear mapping capabilities. It is widely used in application solutions such as approximation and regression. The BP neural network contains one or more hidden layers and an output layer. The training parameters are passed from the input node through each hidden layer and the output layer in turn. The network is trained with samples in a known pattern, thereby generating an accurate mapping relationship from output to output.

[0056] The temperature compensation model based on BP neural network for scale factor error, installation error and zero bias error can be expressed as:

[0057]

[0058] Where: i is the scale factor error, installation error or zero bias error; P i is the system input for scale factor error, installation error or zero bias error, i.e. temperature; b i,1 is the hidden layer bias of scale factor error, installation error or zero bias error; b i,2 is the output layer offset of the scale factor error, installation error or zero bias error; W i,1 is the hidden layer weight matrix; W i,2 is the output layer weight matrix; Q i is the hidden layer output; O i is the system output of scale factor error, installation error or zero bias error; tansig is the hyperbolic tangent S-type function; purelin is the linear function.

[0059] The servo motor in the servo system generally directly drives the rotating mechanism to rotate. Its basic model is as follows:

[0060]

[0061] In the formula, J0 is the moment of inertia of the servo motor shaft, B0 is the servo motor rotation damping coefficient, K0 is the servo motor rotation stiffness, θ is the motor rotation angle, is the first-order derivative of the motor rotation angle, also called the motor angular velocity, It is the second-order derivative of the motor rotation angle, also called the motor angular acceleration.

[0062] The rotating inertial navigation solution model includes attitude differential equation, velocity differential equation and position differential equation. Among them, the attitude differential equation is:

[0063]

[0064] In the formula, q0, q1, q2, q3 are the four basic elements of the quaternion equation. is the differential of the quaternion, is the projection of the angular velocity of the carrier coordinate system b relative to the navigation coordinate system n in the three directions of the b system.

[0065] The velocity differential equation is:

[0066]

[0067] In the formula, V n is the velocity of the carrier in the navigation coordinate system, V n The differential of is the attitude matrix of the carrier coordinate system b relative to the navigation coordinate system n, is the projection of the angular velocity of the earth coordinate system e relative to the inertial coordinate system i in the navigation coordinate system n, is the projection of the angular velocity of the navigation coordinate system n relative to the earth coordinate system e in the navigation coordinate system n, g b is the projection of gravitational acceleration in the carrier coordinate system b.

[0068] The position differential equation is:

[0069]

[0070] is the differential of the longitude of the carrier’s location, is the differential of the latitude of the carrier, L is the latitude of the carrier, and h is the height of the carrier. is the differential of the height h of the carrier, R M The radius of curvature of the earth along the meridian where the carrier is located, R N The radius of curvature of the earth along the y-axis at the position of the carrier, V E 、V N 、V U They are respectively the velocity of the carrier in the navigation coordinate system and the components in the three directions of the northeast sky.

[0071] Multi-source information data acquisition layer: data acquisition is the basis for building a digital twin framework. The rotating inertial navigation system contains a variety of sensors, such as gyroscopes, accelerometers, grating encoders, temperature sensors, current sensors, voltage sensors, pressure sensors, etc., which can collect current, voltage, acceleration, angular velocity, temperature, stress and strain information in real time. At the same time, the internal solution data of the rotating inertial navigation, such as gyroscope temperature compensation data, accelerometer temperature compensation data, navigation data, etc., are provided for the interaction of the twin model to form an effective data set to be fed back to the digital twin geometric model and physical model. Data acquisition runs through all aspects of the digital twin inertial navigation and is the data foundation of the digital twin inertial navigation system.

[0072] Decision control layer: Decision control is based on the digital twin physical model, and combines real-time multi-source data to visually detect the operation process of the inertial navigation system; based on measured data, by establishing a dynamic mapping between physical entities and physical models, real-time prediction of the inertial navigation system is performed; based on the performance prediction results, the relevant parameters of the inertial navigation system, such as servo control parameters and inertial component parameters, are optimized. Decision control will evolve in real time with the changes in the digital twin model and data acquisition. It is a complex dynamic process and a key factor in the successful implementation of the digital twin inertial navigation system.

[0073] Diagnosis and prediction analysis layer: This layer mainly extracts features based on the original information collected by the multi-source data collection layer, such as current, voltage, acceleration, angular velocity, temperature, stress and strain, and performs frequency domain analysis, time domain analysis, statistical analysis or other characteristics related to the operation of the rotating inertial navigation system, and compares it with the set threshold or fault characteristics for monitoring and alarm. At the same time, the real-time performance prediction results of the decision control layer are compared with the gyroscope temperature compensation data, accelerometer temperature compensation data, navigation solution data, etc. solved inside the rotating inertial navigation to evaluate the actual performance of the inertial navigation and provide future performance prediction results for reference.

[0074] The specific implementation process of the above-mentioned rotating inertial navigation digital twin system and the corresponding module layers are as follows:

[0075] 1. Establish a digital twin geometric model: Use 3D modeling software such as Solidworks, ProE, Unity3D, etc. to establish geometric models of rotating inertial navigation equipment, components (rotating system, inertial components, etc.) and parts (motors, bearings, fiber optic gyroscopes, accelerometers, etc.), assemble and construct motion relationships according to the composition of the actual equipment to characterize the relationship between the various components, and establish finite element models of key components of the rotating system such as motors and bearings to characterize their forces and deformations.

[0076] 2. Establish a digital twin physical model: The physical model is also called a mechanism model. It is used to establish and simulate the control logic and input-output relationship of the actual system. It can characterize the performance of the rotating inertial navigation entity. The main models include fiber optic gyroscopes, accelerometers, gratings, servo systems, and rotating inertial navigation solutions. The establishment of the physical model is mainly to accurately characterize the performance of the rotating inertial navigation entity and to ensure the real-time nature of the synchronous evolution of the physical model and the entity.

[0077] 3. In digital twin platforms such as ANSYS Twin Builder and Unity Manufacturing, the integrated environment is used to synchronize the geometric model and the physical model to ensure the exchange and synchronization of data between different modules. At the same time, the platform can be used to build behavioral models and realize the visualization of the motion process and input and output results.

[0078] 4. Build data collection module: Figure 2 As shown in the figure, the information related to the rotating inertial navigation is collected and transmitted to the digital twin platform. The basic components of the rotating inertial navigation include gyroscopes, accelerometers, grating encoders, temperature sensors, etc., and electrical data acquisition units such as current sensors, voltage sensors, and stress sensors are added to collect current, voltage, acceleration, angular velocity, temperature, stress and strain information in real time, and collect temperature compensation, shaft angle compensation, navigation solution information, etc. inside the rotating inertial navigation, and transmit them to the digital twin platform through industrial buses such as serial ports, CAN or OPC UA for interactive use by the twin model, so as to realize the association between the twin virtual body and the entity.

[0079] 5. Build a decision-making control module: Use the graphical display function in digital twin platforms such as ANSYS Twin Builder and Unity Manufacturing to display the data information collected by the multi-source information data acquisition layer in a graphically visualized manner. Determine the input of the digital twin model as the current, voltage, acceleration, angular velocity, temperature, stress strain and other information collected in real time by gyroscopes, accelerometers, grating encoders, temperature sensors, current sensors, voltage sensors, stress sensors, etc., as well as temperature compensation, axis angle compensation, navigation solution and other information. Through the digital twin physical model, combined with data-driven modeling methods such as machine learning and deep learning, build a twin solution model, and output the prediction data of each sensor of the rotating inertial navigation and the prediction data of the navigation solution performance. Based on the prediction results of each sensor and the prediction results of the navigation solution performance, and compared with the real-time data collected subsequently, optimize the parameters or structures of each digital twin physical model in the rotating inertial navigation system, such as the fiber optic gyroscope and accelerometer model, servo control model, rotating inertial navigation solution model, etc.

[0080] 6. Build the diagnosis and prediction analysis module: First, conduct model training, that is, pre-process the original acquisition information provided by the multi-source data acquisition layer and the characteristic information related to the operation status of the rotating inertial navigation system, including processing missing data, correcting abnormal data, etc., to improve data quality and avoid errors in subsequent data analysis; by extracting features from the data, convert the original data into representative feature vectors, and use data dimensionality reduction methods such as principal component analysis and linear discriminant analysis to reduce the dimension of the high-dimensional original characteristic vectors to improve computational efficiency; pass the reduced-dimensional feature vectors into the digital twin deep learning model, and obtain the training and evaluation model. After the new data is collected and input, the trained deep learning model will output the prediction analysis data; on the one hand, it will be compared with the set threshold or fault characteristics, and an alarm will be given in time if it is found that the threshold is exceeded or the fault characteristics appear. At the same time, the real-time performance prediction results of the decision control layer are compared with the gyroscope temperature compensation data, accelerometer temperature compensation data, navigation solution data, etc. solved inside the rotating inertial navigation, and combined with the analysis results of the original acquisition data, the performance of the digital twin inertial navigation system is evaluated, and the accuracy of the performance prediction results is judged.

[0081] 7. Integration of rotating inertial navigation digital twin system: Through digital twin platforms such as ANSYS Twin Builder and Unity Manufacturing, the geometric model and physical model are synchronized to ensure data exchange and synchronization between different modules. The data acquisition module uses communication with the twin platform to synchronize entity information to the twin geometric model and physical model, and realize the association between the twin virtual body and the entity; the decision-making control module realizes iterative optimization of each model by learning the data; the diagnosis and predictive analysis module evaluates the overall operating status of the system and the performance of the digital twin inertial navigation system, and provides references.

[0082] Although the embodiments and drawings of the present invention are disclosed for illustrative purposes, those skilled in the art will appreciate that various substitutions, changes and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the scope of the present invention is not limited to the contents disclosed in the embodiments and drawings.

Claims

1. A rotating inertial navigation digital twin system, characterized in that: It includes the rotating inertial navigation digital twin physical model layer, the rotating inertial navigation digital twin geometric model layer, the multi-source information data acquisition layer, the decision-making control layer, and the diagnosis and prediction analysis layer; The rotating inertial navigation digital twin geometric model layer adopts a data-driven approach to achieve dynamic characterization of the inertial navigation operation process, ensuring synchronous symbiosis with the inertial navigation entity; The rotating inertial navigation digital twin physical model layer includes a fiber optic gyroscope, an accelerometer, a grating, a servo system and a rotating inertial navigation solution model; The multi-source information data acquisition layer includes a variety of sensors for real-time acquisition of current, voltage, acceleration, angular velocity, temperature, stress and strain information. At the same time, the solution data inside the rotating inertial navigation system is provided for the interaction of the twin model to form an effective data set to be fed back to the digital twin geometric model and physical model. The decision control layer: based on the digital twin physical model, combines real-time multi-source data to perform visual detection of the operation process of the inertial navigation system; Based on the measured data, the inertial navigation system is predicted in real time by establishing a dynamic mapping between the physical entity and the physical model; the relevant parameters of the inertial navigation system are optimized based on the performance prediction results; The diagnosis and prediction analysis layer is used to extract features based on the original collected information provided by the multi-source data collection layer, and perform frequency domain analysis, time domain analysis, statistical analysis or other characteristics related to the operation status of the rotating inertial navigation system, and compare with the set threshold or fault characteristics for monitoring and alarm; at the same time, the real-time performance prediction results of the decision control layer are compared with the gyroscope temperature compensation data, accelerometer temperature compensation data, navigation solution data, etc. solved inside the rotating inertial navigation, to evaluate the performance of the actual inertial navigation, and provide future performance prediction results for reference.

2. The rotating inertial navigation digital twin system according to claim 1, characterized in that: The basic model of fiber optic gyroscope and accelerometer is as follows: ω out is the angular velocity value output by the gyroscope detection, I is the unit matrix, ΔG ω is the gyro installation error, δK ω is the gyroscope scale factor error, is the theoretical value of angular velocity, ε p is the gyro constant bias; a out is the acceleration value detected and output by the accelerometer, I is the unit matrix, ΔG a is the accelerometer installation error, δK g is the accelerometer scale factor error, is the theoretical value of acceleration, β p is the constant zero bias of the accelerometer.

3. The rotating inertial navigation digital twin system according to claim 2, characterized in that: The temperature compensation models for the scale factor error, installation error and zero bias error of the gyroscope and accelerometer all adopt the BP neural network method; the relationship between the input and output of the BP neural network is: Q i =tansig(W i,1 P i )+b i,1 O i =purelin(W i,2 Q i )+b i,2 Where: i represents scale factor error, installation error or zero bias error; P i is the system input for scale factor error, installation error or zero bias error, i.e. temperature; b i,1 is the hidden layer bias of scale factor error, installation error or zero bias error; b i,2 is the output layer offset of the scale factor error, installation error or zero bias error; W i,1 is the hidden layer weight matrix; W i,2 is the output layer weight matrix; Q i is the hidden layer output; O i is the system output of scale factor error, installation error or zero bias error; tansig is the hyperbolic tangent S-type function; purelin is the linear function.

4. The rotating inertial navigation digital twin system according to claim 1, characterized in that: The servo motor in the servo system generally directly drives the rotating mechanism to rotate. Its basic model is as follows: In the formula, J0 is the moment of inertia of the servo motor shaft, B0 is the servo motor rotation damping coefficient, K0 is the servo motor rotation stiffness, θ is the motor rotation angle, is the first-order derivative of the motor rotation angle, also called the motor angular velocity, It is the second-order derivative of the motor rotation angle, also called the motor angular acceleration.

5. The rotating inertial navigation digital twin system according to claim 1, characterized in that: The rotational inertial navigation solution model includes attitude differential equation, velocity differential equation and position differential equation; among which, the attitude differential equation is: In the formula, q0, q1, q2, q3 are the four basic elements of the quaternion equation. is the differential of the quaternion, is the projection of the angular velocity of the carrier coordinate system b relative to the navigation coordinate system n in the three directions of the b system; The velocity differential equation is: In the formula, V n is the velocity of the carrier in the navigation coordinate system, V n The differential of is the attitude matrix of the carrier coordinate system b relative to the navigation coordinate system n, is the projection of the angular velocity of the earth coordinate system e relative to the inertial coordinate system i in the navigation coordinate system n, is the projection of the angular velocity of the navigation coordinate system n relative to the earth coordinate system e in the navigation coordinate system n, g b is the projection of gravitational acceleration in the carrier coordinate system b; The position differential equation is: is the differential of the longitude of the carrier’s location, is the differential of the latitude of the carrier, L is the latitude of the carrier, and h is the height of the carrier. is the differential of the height h of the carrier, R M The radius of curvature of the earth along the meridian where the carrier is located, R N The radius of curvature of the earth along the y-axis at the position of the carrier, V E 、V N 、V U They are respectively the velocity of the carrier in the navigation coordinate system and the components in the three directions of the northeast sky.

6. A method for constructing a rotating inertial navigation digital twin system based on any one of claims 1 to 5, characterized in that: include: Step 1: Use 3D modeling software to establish a rotating inertial navigation digital twin geometric model; Step 2: Establish and simulate the physical model of the rotating inertial navigation digital twin in a digital way; Step 3: In the digital twin platform software, synchronize the geometric model and the physical model to visualize the motion process and input and output results; Step 4: Build a data acquisition module to complete data acquisition and transmit it to the digital twin platform for interactive use by the twin model to achieve the association between the twin virtual body and the entity; Step 5: Build a decision-making control module and use machine learning and deep learning data-driven modeling methods to achieve iterative optimization of model parameters; Step 6: Build a diagnostic and predictive analysis module to evaluate the overall operating status of the system and the performance of the digital twin inertial navigation system, and provide a reference.

7. The method for constructing a rotating inertial navigation digital twin system according to claim 6, characterized in that: In step 1, a 3D modeling software is used to establish the geometric model of the rotating inertial navigation system, components and parts. The assembly and motion relationship are constructed according to the composition of the actual equipment to characterize the relationship between the various components. In addition, a finite element model of the key components of the rotating system, such as motors and bearings, is established to characterize their forces and deformations.

8. The method for constructing a rotating inertial navigation digital twin system according to claim 6, characterized in that: Step 5 includes: Utilize the graphical display function in the ANSYS Twin Builder and Unity Manufacturing digital twin platforms to display the data information collected by the multi-source information data acquisition layer in a graphically visualized manner; determine the input of the digital twin model as the real-time collected current, voltage, acceleration, angular velocity, temperature, stress-strain information, as well as temperature compensation, axis angle compensation, and navigation solution information; build a twin solution model through the digital twin physical model, and combine it with data-driven modeling methods such as machine learning and deep learning, and output the predicted data of each sensor of the rotating inertial navigation and the predicted data of the navigation solution performance; based on the prediction results of each sensor and the predicted results of the navigation solution performance, and compared with the real-time data collected subsequently, optimize the parameters or structure of each digital twin physical model in the rotating inertial navigation system.

9. The method for constructing a rotating inertial navigation digital twin system according to claim 6, characterized in that: Building the diagnosis and prediction analysis module in step 6 includes: first performing model training, preprocessing the original acquisition information provided by the multi-source data acquisition layer and the characteristic information related to the operation status of the rotating inertial navigation system, including processing missing data, correcting abnormal data, etc., to improve data quality and avoid errors in subsequent data analysis; then extracting features from the data, converting the original data into representative feature vectors, and using data dimensionality reduction methods such as principal component analysis and linear discriminant analysis to reduce the dimensionality of the high-dimensional original feature vectors to improve computing efficiency; then passing the reduced feature vectors into the digital twin deep learning model, and obtaining the training and evaluation models.

10. The method for constructing a rotating inertial navigation digital twin system according to claim 6, characterized in that: In step 6, the collected data is input into the trained deep learning model, and the prediction analysis data is output; it is compared with the set threshold or fault characteristics. If the threshold is exceeded or a fault characteristic appears, an alarm will be given in time; at the same time, the real-time performance prediction results of the decision control layer are compared with the gyroscope temperature compensation data, accelerometer temperature compensation data, navigation solution data, etc. solved inside the rotating inertial navigation system, and combined with the analysis results of the original collected data, the performance of the digital twin inertial navigation system is evaluated, and the accuracy of the performance prediction results is judged.

Images