Method for measuring spatial angle of drag pipe of dredger

Abstract

A method for measuring a spatial angle of a drag pipe of a dredger, which relates to the technical field of ship engineering. The method comprises: collecting motion data of a hull and motion data of a drag pipe in real time, and using median filtering to remove high-frequency noise; extracting features, and constructing a comprehensive feature vector; using a fuzzy inference system to decouple nonlinear influences between the shaking of the hull and the attitude of the drag pipe; correcting an attitude estimation result on the basis of real-time error feedback; dynamically optimizing the estimation result by means of Bayesian estimation and particle filtering algorithms; and correcting a system deviation. Therefore, the measurement precision is improved. A system further evaluates the overall measurement accuracy on the basis of a machine learning model, and automatically adjusts measurement parameters and processes, so as to improve the adaptability and robustness of the system; and for an inaccurate measurement result, the system can analyze the cause and re-perform spatial angle measurement, thereby ensuring long-term stable operation.

Description

Method for measuring the spatial angle of the dredger's rake pipe Technical Field

[0001] This invention relates to the field of marine engineering technology, specifically to a method for measuring the spatial angle of the rake pipe of a dredger. Background Technology

[0002] Dredgers, as essential equipment for waterway dredging, port construction, and river management, are widely used in complex underwater operations. The scraper pipe system of a dredger is one of its core components, responsible for dredging and transporting sediment. To ensure operational accuracy and efficiency, the spatial angle of the scraper pipe must be accurately measured and monitored in real time. Accurate measurement of the scraper pipe's spatial angle not only helps optimize the operating angle and improve dredging efficiency but also reduces the impact on the aquatic environment and avoids unnecessary resource waste. With the development of modern computing technology, angle measurement methods based on sensor data have gradually become mainstream. Furthermore, with the continuous advancement of signal processing, data fusion, and artificial intelligence technologies, the accuracy and real-time performance of dredger scraper pipe spatial angle measurement have been significantly improved.

[0003] The existing technology has the following shortcomings:

[0004] Due to the dynamic changes of the ship itself, such as hull swaying and wave effects, there may be nonlinear errors in attitude calculation, resulting in inaccurate angle calculation. During operation, the rake pipe may generate complex nonlinear motion due to mud and sand resistance, which makes it difficult for the sensor to accurately track its spatial angle. Summary of the Invention

[0005] The purpose of this invention is to provide a method for measuring the spatial angle of the dredger's scraper pipe, in order to solve the problems mentioned above.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] The method for measuring the spatial angle of the dredger's rake pipe includes the following steps:

[0008] S1: Real-time acquisition of motion data of the hull and the rake pipe, including acceleration, angular velocity and attitude information;

[0009] S2: Based on the motion data of the hull and the rake pipe, nonlinear effects are separated and attitude estimation is optimized;

[0010] S3: Based on the optimized attitude estimation, the difference between the estimation result and the actual measurement value is evaluated through real-time error feedback calculation;

[0011] S4: Based on the evaluation results, the estimation results in the filtering process are corrected in real time using error feedback information;

[0012] S5: Based on the real-time correction results, evaluate the overall accuracy of the dredger's rake pipe spatial angle measurement system, analyze the reasons for inaccurate identification results, and re-measure the spatial angle.

[0013] As a further aspect of the present invention: acquiring motion data of the hull and the rake pipe, and using median filtering to remove high-frequency noise;

[0014] We extract features from the acceleration, angular velocity, and attitude information of the hull and the rake pipe, and construct a comprehensive feature vector.

[0015] As a further aspect of the present invention: the process of constructing the comprehensive feature vector is as follows:

[0016] Acquire motion data of the hull and the rake pipe, and extract time-domain features from acceleration, angular velocity and attitude data, including mean, standard deviation, maximum value, minimum value, peak value and variance;

[0017] Fast Fourier Transform is used to perform frequency domain analysis on acceleration and angular velocity data to extract spectral features such as frequency distribution, spectral density, and frequency components.

[0018] By combining time-frequency analysis techniques, time-frequency domain features are extracted to capture the time-varying characteristics of the data;

[0019] The extracted time-domain, frequency-domain, and time-frequency-domain features are combined to form a preliminary feature vector.

[0020] The gradient boosting tree algorithm is used to train the initial feature vector, automatically selecting the features most relevant to the target prediction. During the gradient boosting tree training process, new feature combinations and cross features are automatically generated to improve the feature expressive power. The influence of each feature on the final target is evaluated, irrelevant features are removed, the most qualified features are retained, and the optimized features are combined to generate the final comprehensive feature vector.

[0021] As a further aspect of the present invention: the separation of nonlinear effects and optimization of attitude estimation specifically includes:

[0022] A comprehensive feature vector is obtained, and singular spectral decomposition is used to transform the comprehensive feature vector of the hull and the rake pipe into input features for the fuzzy inference system. Fuzzy rules are formulated to describe the relationship between the hull sway and the rake pipe attitude. An adaptive method is adopted to adjust the membership function and rule weight of the fuzzy rules based on real-time data, and the performance of the inference system is dynamically optimized.

[0023] The collected motion data is input into the fuzzy inference system, and the nonlinear influence between the hull sway and the rake tube attitude is separated based on the output of the fuzzy inference system.

[0024] By using the membership values ​​of fuzzy inference, the effects of hull sway and the estimated values ​​of the rake pipe attitude are calculated, thus achieving dynamic separation of nonlinear effects.

[0025] Based on real-time feedback data, the fuzzy rules and membership functions in the fuzzy inference system are dynamically updated. Through the output of the fuzzy inference system, combined with real-time data and the optimized model, accurate attitude estimation results of the hull and the rake pipe are provided.

[0026] As a further aspect of the present invention: the difference between the evaluation estimation result and the actual measured value specifically includes:

[0027] Based on the optimized attitude estimation results, the actual measured values ​​and the estimated values ​​are compared and analyzed to calculate the estimation deviation. The calculation process is as follows:

[0028] The difference between the estimated value and the actual measured value is calculated to obtain the estimation bias. The estimation bias is then used to obtain the posterior distribution by combining the prior distribution and the likelihood function, and the bias estimate is updated. The calculation expression is as follows:

[0029] In the formula, A represents the actual measured value, δ represents the estimation bias, P(δ|A) represents the update of the estimation bias δ ​​given the actual measured value A, P(A|δ) represents the probability of the actual measured value A given the estimation bias δ, P(δ) represents the probability distribution of the estimation bias, and P(A) represents the standardization constant to ensure that the total probability of the posterior distribution is 1.

[0030] The estimation bias δ ​​follows a normal distribution, and the prior distribution is:

[0031] In the formula, Σ0 represents the prior variance of the estimation bias, Σ0 represents the covariance matrix of the estimated values, T represents the transpose operation, and d represents the dimension of the data.

[0032] Substituting the prior distribution and likelihood function into Bayes' formula, we obtain the posterior distribution, which can be calculated as follows:

[0033] In the formula, Indicates an estimated value;

[0034] By maximizing the posterior distribution, the maximum a posteriori estimate of the estimation bias is obtained, and the calculation expression is as follows:

[0035] In the formula, argmax represents the maximum a posteriori estimate. δ Represents the maximum value function;

[0036] By differentiating the posterior distribution and setting it to 0, the estimated bias value is obtained, and the calculation expression is as follows:

[0037] In the formula, This represents the estimated deviation value. This represents the accuracy matrix of the actual measurement data. This represents the accuracy matrix of the prior estimate.

[0038] As a further aspect of the present invention: the real-time correction of the estimation results during the filtering process using error feedback information specifically includes:

[0039] Obtain the estimation deviation value, calculate the correction coefficient based on the estimation deviation value and the particle filter algorithm, adjust the estimation result at the current time according to the correction coefficient, and obtain the corrected estimation state.

[0040] As a further aspect of the present invention: the process of obtaining the correction coefficient is as follows:

[0041] Obtain the estimated deviation value, and calculate the correction factor based on the estimated deviation value. The calculation expression is as follows:

[0042] In the formula, α k This represents the correction factor. The expression represents the estimated deviation value, T represents the transpose operation, i represents the i-th particle during the particle operation, N represents the total number of particles, k represents the time step, and P represents the time step. k The covariance matrix represents the state estimate. Particle weights.

[0043] As a further aspect of the present invention: the evaluation of the overall accuracy of the dredging vessel's scraper pipe spatial angle measurement system specifically includes:

[0044] Based on the real-time correction results, a combined feature vector is constructed from the estimated deviation value and correction coefficient during the measurement of the dredger's scraper pipe spatial angle. This vector serves as the input to a machine learning model. Combined with historical data, the machine learning model is trained. Based on the training results, the overall accuracy evaluation result of the system is output. The machine learning model is a deep learning model, which includes an input layer for receiving the combined feature vector, a hidden layer composed of multiple fully connected neural networks for extracting nonlinear features, and an output layer that outputs the overall accuracy evaluation result of the system. The deep learning model is trained using supervised learning with historical data, and the accuracy labels in the historical data serve as the overall accuracy score of the system.

[0045] As a further aspect of the present invention: based on the overall accuracy evaluation results of the system, the quality identification results of the system are divided into accurate identification results and inaccurate identification results, specifically including:

[0046] Obtain the accuracy score of the dredger's rake pipe spatial angle measurement system, and determine whether the accuracy score is greater than or equal to a preset threshold. If yes, it is classified as an accurate identification result; otherwise, it is classified as an inaccurate identification result.

[0047] As a further aspect of the present invention: the step of analyzing the causes of inaccurate identification results and re-measuring the spatial angle specifically includes:

[0048] By comparing the similarity between inaccurate identification results and historical measurement data, if the similarity is lower than a preset threshold, it is determined that the current measurement value is inconsistent with the characteristics of historical data. Based on the cause of inaccurate identification, the measurement parameters and processes are dynamically adjusted, including: recalibrating the sensor and adjusting the sampling frequency parameters; re-measuring the rake tube spatial angle to obtain new measurement values, correcting the measurement values ​​based on correction coefficients, calculating new corrected estimates, reconstructing the comprehensive feature vector, and combining the corrected estimates and measurement values ​​with historical data to update the overall accuracy assessment results of the system.

[0049] The beneficial effects of this invention are:

[0050] (1) This invention optimizes the measurement process of the spatial angle of the dredging vessel's rake pipe by combining multiple advanced technologies. In terms of real-time attitude estimation, the system collects the acceleration, angular velocity, and attitude information of the hull and rake pipe, and uses median filtering to remove high-frequency noise, ensuring the smoothness and accuracy of the input data. During error feedback correction, the system uses Bayesian estimation and particle filtering algorithms to dynamically adjust the estimation results, correcting estimation deviations in real time, thereby eliminating interference from external environmental changes (such as hull swaying and waves) on the measurement results. Through continuous error feedback and adaptive correction, the system can accurately separate nonlinear effects, ensuring that the measurement results are highly consistent with the actual state. Furthermore, by combining a fuzzy inference system to model the complex relationship between the hull and rake pipe attitudes, and dynamically updating the optimized model, the robustness and accuracy of angle measurement are further improved, enabling the method to provide stable and reliable measurement results in complex and ever-changing environments.

[0051] (2) This invention integrates a mechanism of real-time correction and machine learning model feedback, enabling the dredging vessel scraper pipe spatial angle measurement system to possess a high degree of adaptability. The system first analyzes the real-time correction results and continuously updates the estimation model to ensure the accuracy and stability of the measurement results. When inaccurate identification results occur during the measurement process, the system analyzes historical data using a deep learning model, automatically identifies the error source, and adjusts the measurement parameters and procedures. For example, the system can recalibrate the sensor, adjust the sampling frequency, or change the filtering strategy to cope with the impact of environmental changes. Through continuous optimization and adaptive adjustment, the system can maintain the stability of measurement accuracy under different working environments, operating conditions, and complex dynamic scenarios. Furthermore, the system can dynamically evaluate its overall accuracy, promptly feedback and correct inaccurate results, thereby ensuring long-term stable operation and avoiding the performance degradation or accuracy decrease problems that traditional methods easily encounter under complex conditions, thus enhancing the robustness and reliability of the system. Attached Figure Description

[0052] The invention will now be further described with reference to the accompanying drawings.

[0053] Figure 1 is a flowchart of the specific steps of the method for measuring the spatial angle of the dredger's rake pipe according to the present invention.

[0054] Figure 2 is a flowchart of the calculation process for estimating the deviation value in this invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Please refer to Figure 1. This invention is a method for measuring the spatial angle of the dredger's scraper pipe, comprising the following steps:

[0057] S1: Real-time acquisition of motion data of the hull and the rake pipe, including acceleration, angular velocity and attitude information;

[0058] S2: Based on the motion data of the hull and the rake pipe, nonlinear effects are separated and attitude estimation is optimized;

[0059] S3: Based on the optimized attitude estimation, the difference between the estimation result and the actual measurement value is evaluated through real-time error feedback calculation;

[0060] S4: Based on the evaluation results, the estimation results in the filtering process are corrected in real time using error feedback information;

[0061] S5: Based on the real-time correction results, evaluate the overall accuracy of the dredger's rake pipe spatial angle measurement system, analyze the reasons for inaccurate identification results, and re-measure the spatial angle.

[0062] In S1, motion data of the hull and the rake pipe are collected in real time, including acceleration, angular velocity, and attitude information, specifically including:

[0063] Multiple high-precision inertial measurement units (IMUs), including accelerometers, gyroscopes, and attitude sensors, are installed on the hull and rake pipe of the dredger. The IMUs on the hull are positioned near the center of gravity to ensure that the acquired data accurately reflects the overall motion state; the IMUs on the rake pipe are installed in key moving parts, such as joint nodes and ends, to facilitate the capture of local motion characteristics. Each sensor is securely fixed to the measurement surface using industrial-grade mounting devices to avoid data deviation caused by loose installation.

[0064] Sensors collect three-dimensional acceleration, three-dimensional angular velocity, and attitude (e.g., Euler angles) information of the hull and the rake tube. The data acquisition frequency is designed to be set to, for example, 100Hz to ensure the capture of high-frequency details of dynamic motion. The collected raw data is transmitted in real time to the central processing unit (CPU) via a high-speed communication protocol (e.g., wireless network) for unified processing, and the data streams of each sensor are synchronized using timestamps to ensure the alignment of motion information between the hull and the rake tube.

[0065] The raw data received in real time undergoes preprocessing, including noise filtering, zero-bias correction, and drift compensation. Acceleration and angular velocity data are fused using a Kalman filter to eliminate the influence of random noise and obtain more stable attitude information. Simultaneously, based on the known sensor installation location and reference coordinate system, the acquired data is uniformly transformed into a global coordinate system, providing accurate input data for subsequent spatial angle measurements and corrections.

[0066] In S2, based on the motion data of the hull and the rake pipe, nonlinear effects are separated and attitude estimation is optimized, specifically including:

[0067] Acquire motion data of the hull and rake pipe, and use median filtering to remove high-frequency noise;

[0068] Features are extracted from the acceleration, angular velocity, and attitude information of the hull and the rake pipe, and a comprehensive feature vector is constructed.

[0069] The process of constructing the comprehensive feature vector is as follows:

[0070] First, motion data of the hull and the rake tube are collected in real time using IMU, accelerometer, and gyroscope sensors. The collected data includes acceleration (Ax, Ay, Az), angular velocity (ωx, ωy, ωz), and attitude angles (pitch, roll, yaw).

[0071] The acquired raw data is processed by median filtering to remove high-frequency noise. Median filtering can effectively filter out instantaneous spike noise, maintain the smoothness of the signal, and avoid interference from high-frequency noise in subsequent feature extraction.

[0072] Feature extraction includes:

[0073] Time-domain features: Extracting basic statistical features from acceleration, angular velocity, and attitude data, such as mean, standard deviation, maximum, minimum, peak, and variance. By calculating these features, the basic laws and fluctuation characteristics of motion can be captured from the raw data.

[0074] Frequency domain features: Fast Fourier Transform (FFT) is used to perform frequency domain analysis on acceleration and angular velocity data to extract spectral features such as frequency distribution, spectral density, and main frequency components. Frequency domain features can reveal the motion patterns of the hull and the rake at different frequencies, helping to identify the laws of nonlinear motion.

[0075] Time-frequency domain combined characteristics: By using time-frequency analysis techniques (such as short-time Fourier transform and waveform transform), the combined characteristics of the time domain and frequency domain are extracted to capture the time-varying characteristics of the hull and the rake pipe under dynamic changes.

[0076] Constructing a comprehensive feature vector includes:

[0077] The time-domain features, frequency-domain features, and time-frequency-domain features extracted from acceleration, angular velocity, and attitude data will be merged to form a preliminary feature set. Each feature vector will include:

[0078] Acceleration characteristics (such as mean, standard deviation, maximum value)

[0079] Angular velocity characteristics (such as frequency components and variance)

[0080] Attitude characteristics (such as statistics on pitch, roll, and yaw angles)

[0081] These multidimensional features are concatenated into a large feature vector, which contains all the important dynamic information;

[0082] Gradient boosting trees are used for feature selection to evaluate the influence of each feature on the prediction target. The gradient boosting tree algorithm automatically selects the most informative features and removes irrelevant features, thereby improving the expressive power of the features.

[0083] During the training of gradient boosting trees, the cross-effects of different features are considered. Gradient boosting trees adaptively handle feature cross-effects through their tree structure, generating new composite features. For example, different axial components of acceleration are combined with angular velocity to create new features, thereby enhancing the ability to capture complex dynamic patterns.

[0084] The specific steps for separating nonlinear effects are as follows:

[0085] Real-time acquisition of motion data of the hull and the rake pipe, including acceleration (Ax, Ay, Az), angular velocity (ωx, ωy, ωz), and attitude angles (pitch angle, roll angle, yaw angle);

[0086] The collected raw data is preprocessed by using median filtering to remove high-frequency noise and normalizing the data to ensure a uniform scale for the input data.

[0087] By using the singular spectrum decomposition method in "nonlinear dynamics analysis", potential nonlinear dynamic characteristics are identified from acceleration, angular velocity and attitude data;

[0088] By decomposing the signal using the SSD method, we can identify the main periodic components and abrupt change points, and understand the behavior patterns of the hull and rake pipe in a nonlinear environment.

[0089] The acceleration, angular velocity, and attitude data of the hull and the rake pipe are mapped to a dynamic nonlinear feature space to extract dynamic features in the time and frequency domains. These features will be used for analysis in subsequent fuzzy inference systems.

[0090] Design an adaptive fuzzy inference system (FIS) to handle nonlinear dynamic features using a fuzzy rule base. The inputs to the fuzzy system are acceleration, angular velocity, attitude angle, and features extracted from nonlinear dynamic analysis.

[0091] Based on the nonlinear relationships observed in the data, appropriate fuzzy rules are formulated to describe the interaction between the hull and the rake pipe. These include: "When the hull acceleration is large and the angular velocity is small, the rake pipe attitude changes slowly."

[0092] An adaptive fuzzy inference method is used to automatically adjust the membership functions and rule weights of fuzzy rules. These adjustments are based on real-time data feedback, optimizing the system's response to nonlinear effects.

[0093] Using a fuzzy inference system, the nonlinear effects between hull rolling and rake pipe attitude are automatically identified and dynamically separated. Based on real-time data input, the system calculates the output of fuzzy rules, providing the membership value for each state, thereby estimating the nonlinear dynamics of the hull and rake pipe separately.

[0094] The influence values ​​of hull sway and the estimated values ​​of the rake pipe attitude are obtained through the output of fuzzy inference, and the nonlinear effects are separated. This process does not rely on traditional nonlinear models, but rather dynamically adapts to the changing motion state through the rules of the fuzzy inference system.

[0095] Based on real-time data collection, the system automatically updates fuzzy rules and membership functions. Through a real-time feedback mechanism, the fuzzy inference system can adaptively adjust the inference process according to changes in the environment (such as wave changes and ship load changes), avoiding error accumulation caused by improper initial settings.

[0096] Real-time control algorithms are used to optimize the performance of the fuzzy system, gradually improving the system's accuracy in identifying and separating the nonlinear relationship between hull sway and rake tube attitude.

[0097] The output of the fuzzy inference system will serve as the input for the final attitude estimation, combining real-time feedback information to provide the precise attitude angles of the hull and the rake. The final attitude estimation result is output through a data fusion method for the dynamic control and positioning of the hull and the rake.

[0098] Please refer to Figure 2. In S3, based on the optimized attitude estimation, the difference between the estimation result and the actual measurement value is evaluated through real-time error feedback calculation. Specifically, this includes:

[0099] Based on the optimized attitude estimation results, the actual measured values ​​and the estimated values ​​are compared and analyzed to calculate the estimation deviation. The calculation process is as follows:

[0100] The difference between the estimated value and the actual measured value is calculated to obtain the estimation bias. The estimation bias is then used to obtain the posterior distribution by combining the prior distribution and the likelihood function, and the bias estimate is updated. The calculation expression is as follows:

[0101] In the formula, A represents the actual measured value, δ represents the estimation bias, P(δ|A) represents the update of the estimation bias δ ​​given the actual measured value A, P(A|δ) represents the probability of the actual measured value A given the estimation bias δ, P(δ) represents the probability distribution of the estimation bias, and P(A) represents the standardization constant to ensure that the total probability of the posterior distribution is 1.

[0102] The estimation bias δ ​​follows a normal distribution, and the prior distribution is:

[0103] In the formula, Σ0 represents the prior variance of the estimation bias, Σ0 represents the covariance matrix of the estimated values, T represents the transpose operation, and d represents the dimension of the data.

[0104] Substituting the prior distribution and likelihood function into Bayes' formula, we obtain the posterior distribution, which can be calculated as follows:

[0105] In the formula, Indicates an estimated value;

[0106] By maximizing the posterior distribution, the maximum a posteriori estimate of the estimation bias is obtained, and the calculation expression is as follows:

[0107] In the formula, argmax represents the maximum a posteriori estimation. δ Represents the maximum value function;

[0108] By differentiating the posterior distribution and setting it to 0, the estimated bias value is obtained, and the calculation expression is as follows:

[0109] In the formula, This represents the estimated deviation value. This represents the accuracy matrix of the actual measurement data. This represents the accuracy matrix of the prior estimate.

[0110] It should be noted that this embodiment corrects the deviation of the optimized attitude estimation result based on Bayesian theory, and calculates the estimated deviation value by comparing the actual measurement value and the estimated value. First, the initial deviation is obtained by calculating the difference, and the posterior distribution is constructed by combining the prior distribution and the likelihood function. The prior distribution describes the initial belief of the estimation deviation, usually assuming that it follows a normal distribution, and the variance is calculated based on the covariance matrix of the estimated value; the likelihood function represents the probability distribution of the actual measurement value under a given estimation deviation. By substituting the prior distribution and the likelihood function into Bayes' formula, the posterior distribution of the estimation deviation is obtained, and the maximum posterior estimate is obtained by maximizing the posterior distribution. By differentiating the posterior distribution and setting it to zero, the accurate estimation deviation value is obtained. The definition of the accuracy matrix is ​​used to combine the measurement data with the prior estimate to ensure that the corrected deviation value is more consistent with the actual situation, thereby improving the accuracy and robustness of the estimation.

[0111] In S4, based on the evaluation results, error feedback information is used to correct the estimation results in real time during the filtering process, specifically including:

[0112] Obtain the estimation deviation value, calculate the correction coefficient based on the estimation deviation value and the particle filter algorithm, adjust the estimation result at the current time according to the correction coefficient, and obtain the corrected estimation state.

[0113] The process of obtaining the correction coefficient is as follows:

[0114] Obtain the estimated deviation value, and calculate the correction factor based on the estimated deviation value. The calculation expression is as follows:

[0115] In the formula, α k This represents the correction factor. The value represents the estimated deviation, T represents the transpose operation, i represents the i-th particle during the particle operation, N represents the total number of particles, k represents the time step, and P represents the time step. k The covariance matrix represents the state estimate. Particle weights.

[0116] The correction process is as follows: Adjust the current estimation state by combining the correction factor and the estimation deviation value, and calculate the corrected estimation state. The calculation expression is:

[0117] In the formula, This indicates the estimated state before the correction. Indicates the corrected estimated state, α k This represents the correction factor. This indicates the estimated deviation value.

[0118] It should be noted that this technical solution improves estimation accuracy by using evaluation results and error feedback information to correct the estimation results in real time during the filtering process. First, the estimation deviation is obtained by comparing the current estimated value with the actual measured value. Next, a correction coefficient is calculated based on the deviation value using a particle filtering algorithm. This correction coefficient integrates the covariance matrix of the particle states and weight information, expressing the degree of influence of the current deviation on the estimation state correction. Finally, the estimation state at the current moment is adjusted by combining the correction coefficient and the deviation value to obtain the corrected estimation state. This dynamic correction mechanism ensures that the difference between the estimation result and the actual measured value can be adjusted in real time during the filtering process, thereby improving the system's accuracy and robustness.

[0119] In S5, based on the real-time correction results, the overall accuracy of the dredger's scraper pipe spatial angle measurement system is evaluated. For inaccurate identification results, the causes of inaccurate identification are analyzed, and spatial angle measurements are performed again. Specifically, this includes:

[0120] Based on the real-time correction results, a combined feature vector is constructed from the estimated deviation value and correction coefficient during the measurement of the dredger's scraper pipe spatial angle. This vector serves as the input to a machine learning model. Combined with historical data, the machine learning model is trained. Based on the training results, the overall accuracy evaluation result of the system is output. The machine learning model is a deep learning model, which includes an input layer for receiving the combined feature vector, a hidden layer composed of multiple fully connected neural networks for extracting nonlinear features, and an output layer that outputs the overall accuracy evaluation result of the system. The deep learning model is trained using supervised learning with historical data, and the accuracy labels in the historical data serve as the overall accuracy score of the system.

[0121] Obtain the accuracy score of the dredger's rake pipe spatial angle measurement system, and determine whether the accuracy score is greater than or equal to a preset threshold. If yes, it is classified as an accurate identification result; otherwise, it is classified as an inaccurate identification result.

[0122] By comparing the similarity between inaccurate identification results and historical measurement data, if the similarity is lower than a preset threshold, it is determined that the current measurement value is inconsistent with the characteristics of historical data. Based on the cause of inaccurate identification, the measurement parameters and processes are dynamically adjusted, including: recalibrating the sensor and adjusting the sampling frequency parameters; re-measuring the rake tube spatial angle to obtain new measurement values, correcting the measurement values ​​based on correction coefficients, calculating new corrected estimates, reconstructing the comprehensive feature vector, and combining the corrected estimates and measurement values ​​with historical data to update the overall accuracy assessment results of the system.

[0123] The working principle of this invention is as follows: By acquiring real-time motion data of the hull and the rake pipe, and employing various data processing techniques such as median filtering, feature extraction, fuzzy inference, and particle filtering, the rake pipe attitude estimation is optimized. This method gradually improves the system's measurement accuracy through dynamic feedback and error correction. Specifically, a machine learning model is used to evaluate the system's accuracy, and adjustments are made automatically based on the evaluation results. For inaccurate identification results, the system can analyze and locate the cause of the error, re-measure the spatial angle, ensuring the accuracy and stability of the measurement results, and adapting to complex working environments and dynamic changes.

[0124] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0125] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave) means. The computer-readable storage medium can be any available medium that a computer can access or a server or data center data storage device containing one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0126] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0127] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0128] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for measuring the spatial angle of a dredger's harrow pipe, characterized in that, Includes the following steps: S1: Real-time acquisition of motion data of the hull and the rake pipe, including acceleration, angular velocity and attitude information; S2: Based on the motion data of the hull and the rake pipe, nonlinear effects are separated and attitude estimation is optimized; S3: Based on the optimized attitude estimation, the difference between the estimation result and the actual measurement value is evaluated through real-time error feedback calculation; S4: Based on the evaluation results, the estimation results in the filtering process are corrected in real time using error feedback information; S5: Based on the real-time correction results, evaluate the overall accuracy of the dredger's rake pipe spatial angle measurement system, analyze the reasons for inaccurate identification results, and re-measure the spatial angle.

2. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 1, characterized in that, Acquire motion data of the hull and rake pipe, and use median filtering to remove high-frequency noise; We extract features from the acceleration, angular velocity, and attitude information of the hull and the rake pipe, and construct a comprehensive feature vector.

3. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 2, characterized in that, The process of constructing the comprehensive feature vector is as follows: Acquire motion data of the hull and the rake pipe, and extract time-domain features from acceleration, angular velocity and attitude data, including mean, standard deviation, maximum value, minimum value, peak value and variance; Fast Fourier Transform is used to perform frequency domain analysis on acceleration and angular velocity data to extract spectral features such as frequency distribution, spectral density, and frequency components. By combining time-frequency analysis techniques, time-frequency domain features are extracted to capture the time-varying characteristics of the data; The extracted time-domain, frequency-domain, and time-frequency-domain features are combined to form a preliminary feature vector. The gradient boosting tree algorithm is used to train the initial feature vectors and automatically select the features most relevant to the target prediction. During gradient boosting tree training, new feature combinations and cross features are automatically generated to improve feature representation capabilities; the influence of each feature on the final target is evaluated, irrelevant features are removed, the most qualified features are retained, and the optimized features are combined to generate the final comprehensive feature vector.

4. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 1, characterized in that, The separation of nonlinear effects and optimization of attitude estimation specifically includes: A comprehensive feature vector is obtained, and singular spectral decomposition is used to transform the comprehensive feature vector of the hull and the rake pipe into input features for the fuzzy inference system. Fuzzy rules are formulated to describe the relationship between the hull sway and the rake pipe attitude. An adaptive method is adopted to adjust the membership function and rule weight of the fuzzy rules based on real-time data, and the performance of the inference system is dynamically optimized. The collected motion data is input into the fuzzy inference system, and the nonlinear influence between the hull sway and the rake tube attitude is separated based on the output of the fuzzy inference system. By using the membership values ​​of fuzzy inference, the effects of hull sway and the estimated values ​​of the rake pipe attitude are calculated, thus achieving dynamic separation of nonlinear effects. Based on real-time feedback data, the fuzzy rules and membership functions in the fuzzy inference system are dynamically updated. Through the output of the fuzzy inference system, combined with real-time data and the optimized model, accurate attitude estimation results of the hull and the rake pipe are provided.

5. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 1, characterized in that, The differences between the estimated assessment results and the actual measured values ​​specifically include: Based on the optimized attitude estimation results, the actual measured values ​​and the estimated values ​​are compared and analyzed to calculate the estimation deviation. The calculation process is as follows: The difference between the estimated value and the actual measured value is calculated to obtain the estimation bias. The estimation bias is then used to obtain the posterior distribution by combining the prior distribution and the likelihood function, and the bias estimate is updated. The calculation expression is as follows: In the formula, A represents the actual measured value, δ represents the estimation bias, P(δ|A) represents the update of the estimation bias δ ​​given the actual measured value A, P(A|δ) represents the probability of the actual measured value A given the estimation bias δ, P(δ) represents the probability distribution of the estimation bias, and P(A) represents the standardization constant to ensure that the total probability of the posterior distribution is 1. The estimation bias δ ​​follows a normal distribution, and the prior distribution is: In the formula, Σ0 represents the prior variance of the estimation bias, Σ0 represents the covariance matrix of the estimated values, T represents the transpose operation, and d represents the dimension of the data. Substituting the prior distribution and likelihood function into Bayes' formula, we obtain the posterior distribution, which can be calculated as follows: In the formula, Indicates an estimated value; By maximizing the posterior distribution, the maximum a posteriori estimate of the estimation bias is obtained, and the calculation expression is as follows: In the formula, argmax represents the maximum a posteriori estimation. δ Represents the maximum value function; By differentiating the posterior distribution and setting it to 0, the estimated bias value is obtained, and the calculation expression is as follows: In the formula, This represents the estimated deviation value. This represents the accuracy matrix of the actual measurement data. This represents the accuracy matrix of the prior estimate.

6. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 1, characterized in that, The method of using error feedback information to correct the estimation results in real time during the filtering process specifically includes: Obtain the estimation deviation value, calculate the correction coefficient based on the estimation deviation value and the particle filter algorithm, adjust the estimation result at the current time according to the correction coefficient, and obtain the corrected estimation state.

7. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 6, characterized in that, The process of obtaining the correction coefficient is as follows: Obtain the estimated deviation value, and calculate the correction factor based on the estimated deviation value. The calculation expression is as follows: In the formula, α k This represents the correction factor. The value represents the estimated deviation, T represents the transpose operation, i represents the i-th particle during the particle operation, N represents the total number of particles, k represents the time step, and P represents the time step. k The covariance matrix represents the state estimate. Particle weights.

8. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 1, characterized in that, The evaluation of the overall accuracy of the dredger's scraper pipe spatial angle measurement system specifically includes: Based on the real-time correction results, a combined feature vector is constructed from the estimated deviation value and correction coefficient during the measurement of the dredger's scraper pipe spatial angle. This vector serves as the input to a machine learning model. Combined with historical data, the machine learning model is trained. Based on the training results, the overall accuracy evaluation result of the system is output. The machine learning model is a deep learning model, which includes an input layer for receiving the combined feature vector, a hidden layer composed of multiple fully connected neural networks for extracting nonlinear features, and an output layer that outputs the overall accuracy evaluation result of the system. The deep learning model is trained using supervised learning with historical data, and the accuracy labels in the historical data serve as the overall accuracy score of the system.

9. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 8, characterized in that, Based on the overall accuracy assessment results of the system, the system's quality identification results are divided into accurate identification results and inaccurate identification results, specifically including: Obtain the accuracy score of the dredger's rake pipe spatial angle measurement system, and determine whether the accuracy score is greater than or equal to a preset threshold. If yes, it is classified as an accurate identification result; otherwise, it is classified as an inaccurate identification result.

10. The method for measuring the spatial angle of the dredger's scraper pipe according to claim 1, characterized in that, The process of analyzing the causes of inaccurate identification results and re-measuring the spatial angle specifically includes: By comparing the similarity between inaccurate identification results and historical measurement data, if the similarity is lower than a preset threshold, it is determined that the current measurement value is inconsistent with the characteristics of historical data. Based on the cause of inaccurate identification, the measurement parameters and processes are dynamically adjusted, including: recalibrating the sensor and adjusting the sampling frequency parameters; re-measuring the rake tube spatial angle to obtain new measurement values, correcting the measurement values ​​based on correction coefficients, calculating new corrected estimates, reconstructing the comprehensive feature vector, and combining the corrected estimates and measurement values ​​with historical data to update the overall accuracy assessment results of the system.

Images