Hull deformation monitoring method and device based on time delay inertial measurement

Abstract

The invention provides a hull deformation monitoring method and device based on time delay inertial measurement, and belongs to the technical field of hull deformation monitoring. According to the method, for a ship deformation measurement system composed of two sets of inertial navigation, time delay of gyroscopes in the two sets of inertial navigation is introduced into Kalman filtering to serve as a state quantity, and a dimension expansion Kalman filter is constructed; and operating the dimension expansion Kalman filter, performing joint estimation on the ship deformation angle and the time delay, and realizing estimation and compensation of the time delay error while completing ship deformation estimation. For a ship deformation measurement system composed of two sets of inertial navigation, on the basis of a traditional inertial measurement matching method, the influence of the actual application time delay error on the ship deformation monitoring precision is considered, the dimension expansion filter introducing the time delay amount is designed, accurate estimation and effective compensation of the time delay error are achieved, and the ship deformation monitoring precision is improved. Therefore, the measurement precision of the hull deformation angle is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hull deformation monitoring, and in particular to a hull deformation monitoring method and device based on time-delay inertial measurement. Background Art

[0002] Modern large ships are often equipped with navigation, communication, and detection systems, each of which must coordinate within a unified spatial and temporal reference. However, due to hull deformation, the coordinate references used to coordinate these devices are no longer uniform, severely impacting their performance. Detecting hull deformation and minimizing its impact on these systems has become a pressing issue.

[0003] Currently, a variety of measurement methods have been widely demonstrated and applied, primarily optical measurement, photogrammetry, and inertial measurement. However, optical and photogrammetry methods, due to limitations in measurement methods and installation conditions, cannot be widely used for real-time measurement and compensation under actual ship navigation conditions. Measuring hull deformation using inertial devices not only offers high measurement accuracy and minimal requirements for measurement conditions, but also meets the real-time requirements for hull deformation measurement. Measuring hull deformation angles using inertial devices has become the most promising measurement method in current hull measurement technology.

[0004] However, for the ship deformation measurement system composed of two sets of inertial navigation, the two inertial devices are set at two locations far away from the hull. The lack of synchronization between the two will affect the measurement accuracy of the hull deformation angle. Summary of the Invention

[0005] In view of this, the present invention provides a hull deformation monitoring method and device based on time-delay inertial measurement. For a ship deformation measurement system composed of two groups of inertial navigation, on the basis of the traditional inertial measurement matching method, considering the influence of actual application time delay error on the hull deformation monitoring accuracy, a dimension-expanding filter with time delay is designed to achieve accurate estimation and effective compensation of time delay error, thereby effectively improving the measurement accuracy of the hull deformation angle.

[0006] In order to solve the above technical problems, the present invention is implemented as follows.

[0007] A hull deformation monitoring method based on time-delay inertial measurement, comprising:

[0008] For the ship deformation measurement system composed of two sets of inertial navigation, the time delay of the gyroscopes in the two sets of inertial navigation is set to Introduce Kalman filter as a state quantity to construct an extended dimension Kalman filter;

[0009] Run the extended dimension Kalman filter to transform the hull deformation angle With time delay A joint estimation is performed to estimate and compensate for the time delay error while completing the hull deformation estimation.

[0010] Preferably, the ship deformation measurement system composed of two sets of inertial navigation systems delays the time of the gyroscopes in the two sets of inertial navigation systems. Introducing Kalman filtering as a state variable, constructing an extended dimension Kalman filter includes:

[0011] Construct the hull deformation angle model: the hull deformation angle Described as static deformation angle and dynamic deformation angle sum;

[0012] Construct angular velocity matching equation: Based on the assumption that the hull deformation angle is a small angle, construct mathematical equations between the hull deformation angle and the angular velocity measured by the gyroscope, and the difference in angular velocity measured by the two sets of gyroscopes; construct the time delay of a single gyroscope The expression of the corresponding angular velocity difference is substituted into the mathematical equation, and the angular velocity matching equation of the two groups of gyroscopes is obtained by taking into account the gyroscope drift constant drift and random drift;

[0013] Select static deformation angle , dynamic deformation angle , dynamic angular velocity , constant drift and random drift of the two sets of gyroscopes, time delay Constructing the state vector of the extended dimension Kalman filter ; Based on state vector design, combined with the hull deformation angle model, dynamic deformation angle modeling and gyro drift model, the system state equation of the extended dimension Kalman filter is determined; according to the angular velocity matching equation, the system measurement equation is determined to complete the construction of the extended dimension Kalman filter.

[0014] Preferably, in the hull deformation angle model, the dynamic deformation angle is modeled as a second-order Markov process independent in three directions.

[0015] Preferably, the state vector of the extended dimension Kalman filter is Designed to:

[0016] ;

[0017] in, and are the static deformation angle and dynamic deformation angle of the hull deformation angle respectively, is the dynamic angular velocity; and is the constant drift and random drift of the first group of gyroscopes; and is the constant drift and random drift of the second group of gyroscopes;

[0018] According to the hull deformation angle model, dynamic deformation angle modeling and gyro drift model, the system state equation is obtained as follows:

[0019] ;

[0020] In the formula, the state transfer matrix The form is as follows:

[0021] ;

[0022] ;

[0023] ;

[0024] Noise driven array The form is:

[0025] ;

[0026] ;

[0027] Where, is an i×j matrix of all zeros; is the i×j unit matrix; the dynamic deformation angle is modeled using three independent second-order Markov processes. Representing three axes respectively. ; is the irregularity coefficient of the i-axis, which represents the complexity of the dynamic deformation angle; is the main frequency of dynamic deformation of the i-axis; is the correlation time in the first set of gyro random drift models; is the correlation time in the second set of gyro random drift models; is the root mean square error of the dynamic deformation angle, Drive the noise vector for the system.

[0028] Preferably, the system measurement equation determined according to the angular velocity matching equation is:

[0029] ;

[0030] In the formula, the measurement , represents the measurement noise, the measurement matrix for:

[0031] ;

[0032] ;

[0033] Where, It represents the angular velocity difference between the two groups of gyroscopes; and They represent the angular velocities to which the first and second groups of gyroscopes are sensitive, respectively; represents the angular acceleration of the second set of gyroscopes; Represents the vector " " is the antisymmetric matrix of .

[0034] The present invention also provides a hull deformation monitoring device based on time-delay inertial measurement, comprising two sets of inertial navigation systems, an extended-dimensional Kalman filter, and a monitoring module;

[0035] Two sets of inertial navigation systems are set at different positions on the hull, and the gyroscopes in the inertial navigation systems collect angular velocity;

[0036] The state vector of the extended Kalman filter Adding time delay between two gyroscopes , as a state quantity; the hull deformation angle Described as static deformation angle and dynamic deformation angle The sum of the hull deformation angle model is constructed; the system state equation of the extended dimension Kalman filter is constructed based on the hull deformation angle model and the gyro drift model; the time delay is considered to determine the The angular velocity matching equation is then used to construct the system measurement equation of the extended dimension Kalman filter;

[0037] The monitoring module runs the extended dimension Kalman filter to convert the hull deformation angle With time delay A joint estimation is performed to estimate and compensate for the time delay error while completing the hull deformation estimation.

[0038] Preferably, in the hull deformation angle model, the dynamic deformation angle is modeled as a second-order Markov process independent in three directions.

[0039] Preferably, the state vector of the extended Kalman filter is Designed to:

[0040] ;

[0041] in, and are the static deformation angle and dynamic deformation angle of the hull deformation angle respectively, is the dynamic angular velocity; and is the constant drift and random drift of the first group of gyroscopes; and is the constant drift and random drift of the second group of gyroscopes;

[0042] The system state equation constructed based on the hull deformation angle model and gyro drift model is:

[0043] ;

[0044] In the formula, the state transfer matrix The form is as follows:

[0045] ;

[0046] ;

[0047] ;

[0048] Noise driven array The form is:

[0049] ;

[0050] ;

[0051] Where, is an i×j matrix of all zeros; is the i×j unit matrix; the dynamic deformation angle is modeled using three independent second-order Markov processes. Representing three axes respectively. , is the irregularity coefficient of the i-axis, which characterizes the complexity of the dynamic deformation angle. is the main frequency of dynamic deformation of the i-axis; is the correlation time in the first set of gyro random drift models; is the correlation time in the second set of gyro random drift models; is the root mean square error of the dynamic deformation angle, Drive noise vector for the system;

[0052] The system measurement equation constructed based on the angular velocity matching equation is:

[0053] ;

[0054] In the formula, the measurement , represents the measurement noise, the measurement matrix for:

[0055] ;

[0056] ;

[0057] Where, It represents the angular velocity difference between the two groups of gyroscopes; and They represent the angular velocities to which the first and second groups of gyroscopes are sensitive, respectively; represents the angular acceleration of the second set of gyroscopes; Represents the vector " " is the antisymmetric matrix of .

[0058] Beneficial effects:

[0059] (1) The present invention proposes a hull deformation monitoring method based on time-delay inertial measurement. For a ship deformation measurement system composed of two sets of inertial navigation, the time delay of the gyroscopes in the two sets of inertial navigation is introduced into the Kalman filter as a state quantity, and an extended dimension Kalman filter is constructed. The extended dimension Kalman filter is run to jointly estimate the hull deformation angle and the time delay. While completing the hull deformation estimation, the time delay error is estimated and compensated. For a ship deformation measurement system composed of two sets of inertial navigation, the present invention, based on the traditional inertial measurement matching method, considers the influence of the actual application time delay error on the hull deformation monitoring accuracy, designs an extended dimension filter that introduces the time delay, realizes the accurate estimation and effective compensation of the time delay error, and thus effectively improves the measurement accuracy of the hull deformation angle.

[0060] (2) The present invention can not only effectively estimate the hull deformation angle and gyro data delay, but also improve the accuracy of hull deformation angle estimation.

[0061] (3) The hull deformation monitoring method based on time-delay inertial measurement proposed in this invention has strong versatility and can be extended to other carriers. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 Schematic diagram of the hull deformation monitoring method based on time-delay inertial measurement of the present invention.

[0063] Figure 2 Schematic diagram of a hull deformation monitoring device based on time-delay inertial measurement according to the present invention. DETAILED DESCRIPTION

[0064] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0065] For ships, due to the large size of the hull, the hull deformation cannot be ignored, and the demand for high-precision deformation monitoring technology is quite urgent.

[0066] For a ship deformation measurement system consisting of two sets of inertial navigation systems, it is necessary to synchronize inertial devices that are located far apart. This invention selects the time delay errors of the two sets of inertial navigation systems as compensation dimensions from the multiple dimensions that affect monitoring accuracy. Rather than directly compensating for the time delay errors, this invention designs an extended-dimensional Kalman filter that introduces time delays. This time delay is used as a state variable in the filter, and is jointly estimated and optimized with the hull deformation. While simultaneously estimating the hull deformation, this method achieves accurate estimation and effective compensation for the time delay errors, effectively improving the measurement accuracy of the hull deformation angle. This method can be widely applied in the field of deformation monitoring.

[0067] Figure 1 FIG. 1 shows a flow chart of a method for monitoring ship deformation based on time-delay inertial measurement according to the present invention, as shown in FIG. Figure 1 As shown, its execution process includes the following steps:

[0068] Step 1: Design the hull deformation angle model and establish the mathematical relationship between the hull deformation angle and inertial measurement.

[0069] Step 1.1: Based on the analysis of the causes of deformation angle, model it into two components: static deformation angle and dynamic deformation angle:

[0070] (1)

[0071] Where, represents the deformation angle, represents the static deformation angle, Represents the dynamic deformation angle. The static deformation angle is caused by slow-changing factors and can be regarded as a constant for short-term observations. Therefore, the derivative of the static deformation angle is is 0:

[0072] (2)

[0073] The dynamic deformation angle comes from external dynamic disturbances, and the model is established using three independent second-order Markov processes:

[0074] (3)

[0075] Where, Representing three axes respectively. Represents Gaussian white noise with mean 0 and variance 1. ; is the i-axis irregularity coefficient, which represents the complexity of the dynamic deformation angle; is the main frequency of dynamic deformation of the i-axis; represents the dynamic deformation angle of the i-axis, . and They are The first and second derivatives of .

[0076] Step 1.2: Based on the assumption that the hull deformation angle is small, the mathematical equation between the hull deformation angle and inertial measurement is constructed as follows:

[0077] (4)

[0078] Where, represents the angular velocity difference between the two groups of gyroscopes, Indicates the angular velocity to which the first set of gyroscopes are sensitive. Indicates the angular velocity to which the second set of gyroscopes are sensitive. Deformation angle The derivative of .

[0079] Step 2: Considering the time delay between the two sets of gyroscope data in actual applications, obtain the angular velocity matching equation after considering the time delay.

[0080] Step 2.1: Assume that the first set of gyros is The output of the second gyro at the same time The output at the moment corresponds to:

[0081] (5)

[0082] Step 2.2: The second set of gyros and There is an angular velocity difference at all times, that is:

[0083] (6)

[0084] Where, represents the angular acceleration of the second set of gyroscopes;

[0085] Step 2.3: Substitute formula (6) into the mathematical equation in step 1.2 to obtain:

[0086] (7)

[0087] Step 2.4: According to formula (4), replace Replace with , and further consider gyro drift - , then (7) is modified as follows:

[0088] (8)

[0089] Where, 、 Respectively represent the gyro drift of the two groups of gyros, including constant drift and random drift:

[0090] (9)

[0091] Where, 、 Respectively represent the constant drift and random drift of the first group of gyroscopes, 、 They represent the constant drift and random drift of the second group of gyroscopes respectively. The gyroscope drift model can be expressed as:

[0092] (10)

[0093] in, and is a constant, and its differential is 0; and It is constructed as a first-order Markov model. and are the noises of the two gyro drift models, is the correlation time in the first set of gyro random drift models; is the correlation time in the second set of gyro random drift models;

[0094] Step 3: Expand the time delay into the Kalman filter state quantity and introduce the time delay into the design The Kalman filter is used to effectively estimate the time delay and deformation angle.

[0095] Step 3.1: Select the static deformation angle , dynamic deformation angle , dynamic angular velocity , constant drift and random drift of the two sets of gyroscopes, time delay As a state quantity, the state vector Set it to the following 22-dimensional form:

[0096] (11)

[0097] Step 3.2: Based on the hull deformation angle model and gyro random drift model, the system state equation can be obtained:

[0098] (12)

[0099] In the formula, the state transfer matrix The form is as follows:

[0100] (13)

[0101] (14)

[0102] (15)

[0103] In the above formula, It is a 22×22 matrix, and the elements in the 22nd column correspond to the expanded time delay terms. , all 0.

[0104] Noise driven array The form is:

[0105] (16)

[0106] (17)

[0107] Step 3.3: According to the angular velocity matching equation (Formula 7), the system measurement equation can be obtained:

[0108] (18)

[0109] In the formula, the measurement , represents the measurement noise, the measurement matrix for:

[0110] (19)

[0111] (20)

[0112] The above measurement matrix The last item was added , which corresponds to the time delay introduced .

[0113] Step 3.4: Execute the Kalman filter equation to estimate the dynamic deformation angle and time delay.

[0114] This concludes the process.

[0115] Based on the above method, the present invention also provides a hull deformation monitoring device based on time-delay inertial measurement, such as Figure 2 As shown, it includes two sets of inertial navigation, extended dimension Kalman filter and monitoring module.

[0116] Two sets of inertial navigation systems are installed at different locations on the hull, with the gyroscopes in the inertial navigation systems collecting angular velocity. In practice, inertial navigation system 1 is installed at the center of the ship and fixed to it, while inertial navigation system 2 is installed on or near the payload on board, with its coordinate system aligned with the payload's coordinate system.

[0117] The state vector of the extended Kalman filter Adding time delay between two gyroscopes , as a state quantity; the hull deformation angle Described as static deformation angle and dynamic deformation angle The dynamic deformation angle is modeled as a second-order Markov process independent of three directions, thus constructing the hull deformation angle model; based on the hull deformation angle model and the gyro drift model, the system state equation of the extended dimension Kalman filter is constructed; the time delay is considered to determine The subsequent angular velocity matching equation is used to construct the system measurement equation of the expanded dimension Kalman filter.

[0118] The monitoring module runs the extended dimension Kalman filter to convert the hull deformation angle With time delay A joint estimation is performed to estimate and compensate for the time delay error while completing the hull deformation estimation.

[0119] In a preferred embodiment, the state vector of the extended dimension Kalman filter is Designed to:

[0120]

[0121] Then, the system state equation is constructed as shown in equations (12) to (17); the system measurement equation is constructed as shown in equations (18) to (20), which will not be repeated here.

[0122] The above specific embodiments merely illustrate the design principles of the present invention. The shapes and names of the components described herein may vary and are not limiting. Therefore, those skilled in the art may modify or substitute equivalents for the technical solutions described in the above embodiments. Such modifications and substitutions, without departing from the inventive spirit and technical solutions of the present invention, shall fall within the scope of protection of the present invention.

Claims

1. A method for monitoring hull deformation based on time-delay inertial measurement, characterized in that: include: For the ship deformation measurement system composed of two sets of inertial navigation, the time delay of the gyroscopes in the two sets of inertial navigation is set to Introduce Kalman filter as a state quantity to construct an extended dimension Kalman filter; Run the extended dimension Kalman filter to transform the hull deformation angle With time delay A joint estimation is performed to estimate and compensate for the time delay error while completing the hull deformation estimation.

2. The method for monitoring hull deformation based on time-delay inertial measurement according to claim 1, characterized in that: The ship deformation measurement system composed of two sets of inertial navigation is to delay the time of the gyroscopes in the two sets of inertial navigation. Introducing Kalman filtering as a state variable, constructing an extended dimension Kalman filter includes: Construct the hull deformation angle model: the hull deformation angle Described as static deformation angle and dynamic deformation angle sum; Construct angular velocity matching equation: Based on the assumption that the hull deformation angle is a small angle, construct mathematical equations between the hull deformation angle and the angular velocity measured by the gyroscope, and the difference in angular velocity measured by the two sets of gyroscopes; construct the time delay of a single gyroscope The expression of the corresponding angular velocity difference is substituted into the mathematical equation, and the angular velocity matching equation of the two groups of gyroscopes is obtained by taking into account the gyroscope drift constant drift and random drift; Select static deformation angle , dynamic deformation angle , dynamic angular velocity , constant drift and random drift of the two sets of gyroscopes, time delay Constructing the state vector of the extended dimension Kalman filter ; Based on state vector design, combined with the hull deformation angle model, dynamic deformation angle modeling and gyro drift model, the system state equation of the extended dimension Kalman filter is determined; according to the angular velocity matching equation, the system measurement equation is determined to complete the construction of the extended dimension Kalman filter.

3. The method for monitoring hull deformation based on time-delay inertial measurement according to claim 2, characterized in that: In the hull deformation angle model, the dynamic deformation angle is modeled as a second-order Markov process independent of three directions.

4. The method for monitoring hull deformation based on time-delay inertial measurement according to claim 3, characterized in that: The state vector of the extended dimension Kalman filter Designed to: ； in, and are the static deformation angle and dynamic deformation angle of the hull deformation angle respectively, is the dynamic angular velocity; and is the constant drift and random drift of the first group of gyroscopes; and is the constant drift and random drift of the second group of gyroscopes; According to the hull deformation angle model, dynamic deformation angle modeling and gyro drift model, the system state equation is obtained as follows: ； In the formula, the state transfer matrix The form is as follows: ； ； ； Noise driven array The form is: ； ； Where, is an i×j matrix of all zeros; is the i×j unit matrix; the dynamic deformation angle is modeled using three independent second-order Markov processes. Representing three axes respectively. ; is the irregularity coefficient of the i-axis, which represents the complexity of the dynamic deformation angle; is the main frequency of dynamic deformation of the i-axis; is the correlation time in the first set of gyro random drift models; is the correlation time in the second set of gyro random drift models; is the root mean square error of the dynamic deformation angle, Drive the noise vector for the system.

5. The method for monitoring hull deformation based on time-delay inertial measurement according to claim 3, characterized in that: The system measurement equation determined according to the angular velocity matching equation is: ； In the formula, the measurement , represents the measurement noise, the measurement matrix for: ； ； Where, It represents the angular velocity difference between the two groups of gyroscopes; and They represent the angular velocities to which the first and second groups of gyroscopes are sensitive, respectively; represents the angular acceleration of the second set of gyroscopes; express" " is the antisymmetric matrix of .

6. A hull deformation monitoring device based on time-delay inertial measurement, characterized in that: Includes two sets of inertial navigation, extended dimension Kalman filter and monitoring module; Two sets of inertial navigation systems are set at different positions on the hull, and the gyroscopes in the inertial navigation systems collect angular velocity; The state vector of the extended Kalman filter Adding time delay between two gyroscopes , as a state quantity; the hull deformation angle Described as static deformation angle and dynamic deformation angle The sum of the hull deformation angle model is constructed; the system state equation of the extended dimension Kalman filter is constructed based on the hull deformation angle model and the gyro drift model; the time delay is considered to determine the The angular velocity matching equation is then used to construct the system measurement equation of the extended dimension Kalman filter; The monitoring module runs the extended dimension Kalman filter to convert the hull deformation angle With time delay A joint estimation is performed to estimate and compensate for the time delay error while completing the hull deformation estimation.

7. The hull deformation monitoring device based on time-delay inertial measurement according to claim 6, characterized in that: In the hull deformation angle model, the dynamic deformation angle is modeled as a second-order Markov process independent of three directions.

8. The hull deformation monitoring device based on time-delay inertial measurement according to claim 7, characterized in that: State vector of the extended Kalman filter Designed to: ； in, and are the static deformation angle and dynamic deformation angle of the hull deformation angle respectively, is the dynamic angular velocity; and is the constant drift and random drift of the first group of gyroscopes; and is the constant drift and random drift of the second group of gyroscopes; The system state equation constructed based on the hull deformation angle model and gyro drift model is: ； In the formula, the state transfer matrix The form is as follows: ； ； ； Noise driven array The form is: ； ； Where, is an i×j matrix of all zeros; is the i×j unit matrix; the dynamic deformation angle is modeled using three independent second-order Markov processes. Representing three axes respectively. , is the irregularity coefficient of the i-axis, which characterizes the complexity of the dynamic deformation angle. is the main frequency of dynamic deformation of the i-axis; is the correlation time in the first set of gyro random drift models; is the correlation time in the second set of gyro random drift models; is the root mean square error of the dynamic deformation angle, Drive noise vector for the system; The system measurement equation constructed based on the angular velocity matching equation is: ； In the formula, the measurement , represents the measurement noise, the measurement matrix for: ； ； Where, It represents the angular velocity difference between the two groups of gyroscopes; and They represent the angular velocities to which the first and second groups of gyroscopes are sensitive, respectively; represents the angular acceleration of the second set of gyroscopes; express" " is the antisymmetric matrix of .

Images