Buoy offshore safety automatic monitoring method

Abstract

The invention provides a buoy marine safety automatic monitoring method, and belongs to the technical field of monitoring methods. The buoy marine safety automatic monitoring method comprises the following steps: receiving buoy message data, the buoy message data comprises buoy longitude and latitude data, buoy voltage data, buoy anchor lamp data, buoy cabin door data, buoy cabin water inlet data and buoy sensor data; performing multivariate decomposition operation on the longitude and latitude data of the buoy, and establishing a displacement relation function between a buoy displacement variation component and an environment wind speed variation component; constructing a buoy shift matrix based on the shift relation function, calculating a buoy public safety value, a buoy self-safety value and a buoy function safety value, and performing weighted calculation on the buoy public safety value, the buoy self-safety value and the buoy function safety value to obtain a buoy safety evaluation value; and determining the buoy risk level according to the buoy safety evaluation value. The method solves the problem of how to realize accurate identification and evaluation of the offshore safety state of the buoy.

Description

Technical Field

[0001] The present invention belongs to the technical field of monitoring methods, and specifically relates to a method for automatically monitoring the maritime safety of a buoy. Background Art

[0002] Ocean buoys are an important part of the ocean observation network, undertaking multiple tasks such as ocean environmental monitoring, channel marking, and ocean scientific research. With the increase in ocean development activities, the safety threats faced by buoys are increasing day by day, including risks such as dragging anchor, collision, and human damage. Traditional buoy safety monitoring mainly relies on regular cruise inspections and single-parameter threshold monitoring, which have problems such as large monitoring blind spots, lagging response, and high false alarm rates. At present, three main technical means are adopted for buoy safety monitoring: The first is position monitoring based on satellite positioning, which determines the safety status by judging whether the buoy deviates from the predetermined position. However, this method is difficult to distinguish between displacements caused by natural factors and displacements caused by human damage, and the positioning accuracy drops significantly under adverse weather conditions. The second is status monitoring based on sensors, which mainly monitors the working status of key components such as the buoy power supply, anchor light, and hatch. However, each sensor works independently and lacks systematic analysis, making it prone to misjudgment. The third is remote monitoring based on image recognition, which monitors the buoy through shore-based radar or video systems. However, limited by weather, visibility, and monitoring range, it is difficult to achieve all-weather monitoring. In practical applications, these technical means often operate independently and lack an effective data fusion and comprehensive analysis mechanism. Taking the buoy monitoring system in a certain sea area as an example, when using satellite positioning monitoring, due to the inability to distinguish the reasons for the buoy displacement, false alarms often occur, resulting in unnecessary cruise inspections; when using sensor monitoring, due to the lack of support from environmental data, it is difficult to determine whether the abnormal state is caused by adverse weather; when using remote monitoring, the monitoring effect drops significantly at night or in poor visibility conditions. These problems lead to low efficiency in buoy safety management, increase operation and maintenance costs, and affect the stable operation of the ocean observation system. Summary of the Invention

[0003] In view of this, the present invention provides a method for automatically monitoring the maritime safety of a buoy, which solves the problem of how to accurately identify and evaluate the maritime safety status of a buoy.

[0004] The present invention is implemented as follows: The present invention provides a method for automatically monitoring the maritime safety of a buoy. The method includes the following steps: receiving buoy message data, where the buoy message data includes buoy longitude and latitude data, buoy voltage data, buoy anchor light data, buoy hatch data, buoy cabin water ingress data, and buoy sensor data; performing a multivariate decomposition operation on the buoy longitude and latitude data, and establishing a shift relationship function between the buoy displacement variation component and the environmental wind speed variation component; constructing a buoy displacement matrix based on the shift relationship function, calculating the buoy public safety value, the buoy self-safety value, and the buoy functional safety value, and performing a weighted calculation on the buoy public safety value, the buoy self-safety value, and the buoy functional safety value to obtain a buoy safety assessment value; and determining the buoy risk level according to the buoy safety assessment value.

[0005] Based on the above technical solution, the method for automatically monitoring the maritime safety of a buoy according to the present invention can be further improved as follows: Among them, the step of processing the buoy message data is specifically to receive the original message data through a communication receiving module, and process the hierarchical data structure by using a recursive descent algorithm, including message header parsing, data segment extraction, and checksum verification. The state machine design pattern is used to handle the cases of packet loss and out-of-order, and the signal strength threshold is set to 25 decibels, and the signal-to-noise ratio threshold is set to 10 decibels.

[0006] Further, the step of performing a multivariate decomposition operation on the buoy longitude and latitude data is specifically to convert the buoy longitude and latitude data into a time series, use a median filtering method to remove abnormal points, use a cubic spline interpolation method for smoothing processing, and decompose the time series into multiple intrinsic mode functions through an iterative screening process. The iterative stop condition is set to a standard deviation threshold of 0.2, and the buoy displacement stable component and the buoy displacement variation component are obtained.

[0007] Further, the step of establishing the shift relationship function is specifically to establish a preliminary relationship between the buoy displacement variation component and the environmental wind speed variation component by using the least squares method, construct a flow direction influence function through polynomial fitting, and optimize the model parameters by using a genetic algorithm. The population size is set to 100, the number of evolutionary generations is 50 generations, the crossover probability is 0.8, and the mutation probability is 0.1.

[0008] Further, the step of constructing the buoy displacement matrix is specifically to calculate the shift coefficient by using a sliding window method, the window length is set to 6 hours, the sliding step is 1 hour, calculate the displacement components in three orthogonal directions, perform normalization processing by using the maximum and minimum method, and calculate the eigenvalues by using the power iteration method. The convergence threshold is set to 0.001.

[0009] Further, the steps of calculating the public safety value of the buoy specifically include using the eigenvalue decomposition method to extract the main movement direction of the buoy displacement matrix, evaluating the displacement stability in combination with the buoy stability matrix, and using the fuzzy comprehensive evaluation method to evaluate the potential impact degree of the buoy on public safety.

[0010] Further, the steps of calculating the self-safety value of the buoy specifically include using the trend analysis method to evaluate the voltage change trend of the buoy, using the spectral analysis method to identify the abnormal period of the buoy anchor light matching function, and evaluating the buoy hatch data and the buoy cabin water inlet data based on the decision tree method.

[0011] Further, the steps of calculating the functional safety value of the buoy specifically include using the data quality control method to identify the abnormal values and missing values of the buoy sensor data, calculating the time continuity index and the spatial consistency index of the data, and using the analytic hierarchy process to determine the index weights.

[0012] Further, the calculation steps of the buoy safety evaluation value specifically include using the Delphi method to determine the initial weights of each safety index, optimizing the weight coefficients through the analytic hierarchy process and performing a consistency test, and using weighted summation to obtain the final buoy safety evaluation value.

[0013] Further, the steps of determining the risk level of the buoy specifically include using the clustering analysis method to determine the level boundary values, and using the fuzzy rule inference to judge the risk level of the buoy.

[0014] Compared with the prior art, the beneficial effects of a buoy maritime safety automatic monitoring method provided by the present invention are as follows: The present invention realizes the accurate identification and evaluation of the buoy safety state by establishing a buoy safety monitoring model based on multi-source data fusion. First, the empirical mode decomposition method is used to perform multi-scale analysis on the buoy displacement data, effectively separating the displacement characteristics caused by natural factors and human factors, and improving the accuracy of abnormal state identification. Second, by constructing a displacement relationship function, a quantitative association is established between the buoy displacement and environmental factors, enabling the system to accurately judge whether the displacement belongs to the normal state. Third, the time-frequency analysis method is used to process the state data such as voltage and anchor lights, and combined with environmental data for comprehensive judgment, reducing the false alarm rate. Finally, a safety evaluation system based on multi-index weighting is established to realize the quantitative evaluation of the safety state. The solution of the present invention overcomes the limitations of traditional monitoring methods and establishes a complete set of buoy safety monitoring technical systems. This solution can not only accurately identify the abnormal state of the buoy, but also predict potential safety risks, providing a scientific basis for buoy safety management. The system has strong adaptability, can automatically adjust parameters according to the environmental characteristics of different sea areas, and has strong applicability. Through a multi-level data analysis and evaluation mechanism, the comprehensive monitoring of the buoy safety state is realized, providing a strong guarantee for the stable operation of the ocean observation network. Brief Description of the Drawings

[0015] Figure 1 It is a flowchart of an automatic monitoring method for buoy maritime safety. Detailed Implementation Manner

[0016] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0017] As Figure 1 shown, it is a flowchart of an automatic monitoring method for buoy maritime safety provided by the present invention. This method includes the following steps: S01. Receive buoy message data, where the buoy message data includes buoy longitude and latitude data, buoy voltage data, buoy anchor light data, buoy hatch data, buoy cabin water ingress data, and buoy sensor data; The specific implementation manner of step S01 is to use a communication receiving module to obtain buoy message data, which involves a data parsing process. First, the original message data is received through a satellite communication module. At this time, the message data contains various sensor data and communication protocol headers, and protocol parsing is required. After protocol parsing, buoy longitude and latitude data, buoy voltage data, buoy anchor light data, buoy hatch data, buoy cabin water ingress data, and buoy sensor data are obtained. Among them, the communication protocol adopts the General Oceanographic Data Format Specification, and the data parsing is implemented using a recursive descent algorithm to process the hierarchical data structure. The parsing process includes three stages: message header parsing, data segment extraction, and checksum verification. The data parsing module adopts a state machine design pattern, which can effectively handle the loss and out-of-order of data packets. In terms of data reception quality, the signal strength threshold is set to 25 dB, and the signal-to-noise ratio threshold is set to 10 dB. Packets below the threshold will be marked as pending confirmation status. The purpose of this step is to ensure the integrity and accuracy of data reception and provide a reliable data basis for subsequent analysis.

[0018] S02. Perform multivariate decomposition operation on the buoy longitude and latitude data to obtain a buoy displacement stable component and a buoy displacement variation component; The specific implementation of step S02 is to process the buoy longitude and latitude data based on the empirical mode decomposition principle. First, the longitude and latitude data are converted into a time series, and data preprocessing is performed on the time series data, including outlier processing and data smoothing. The median filtering method is used for data preprocessing to remove outliers, and the filter window size is set to 5 data points. Then, the cubic spline interpolation method is used to smooth the data, and the interpolation node interval is set to 30 minutes. Next, empirical mode decomposition is performed on the preprocessed data, and the time series is decomposed into multiple intrinsic mode functions through an iterative screening process. During the decomposition process, the iterative stop condition is set to a standard deviation threshold of 0.2, and the maximum number of iterations is 10 times. Finally, the obtained intrinsic mode functions are classified into two categories: the buoy displacement stable component and the buoy displacement variable component. Among them, the components with a frequency lower than 0.1 Hz are classified as stable components, and the components with a frequency higher than this frequency are classified as variable components. The purpose of this step is to separate the stable trend and fluctuation characteristics of the buoy displacement, providing a basis for subsequent displacement state judgment.

[0019] S03. Obtain environmental monitoring data, where the environmental monitoring data includes environmental wind speed data, environmental wind direction data, and environmental flow direction data, and perform multivariate decomposition operations on the environmental wind speed data to obtain the environmental wind speed stable component and the environmental wind speed variable component; The specific implementation of step S03 is to obtain and process environmental monitoring data. First, obtain the environmental wind speed data, environmental wind direction data, and environmental flow direction data from a meteorological station, and the data sampling interval is 10 minutes. Perform quality control on the obtained environmental data, including data range inspection and time continuity inspection. The effective range of the wind speed data is set to 0 to 60 meters per second, and the range of the wind direction data is 0 to 360 degrees. The data outside the range is corrected using the interpolation method. Then, apply the wavelet decomposition method to the environmental wind speed data for multi-scale analysis, select the Debauch wavelet as the basis function, and set the decomposition level to 4 layers. The environmental wind speed stable component and the environmental wind speed variable component are obtained through reconstruction. Among them, the threshold denoising method is used in the reconstruction process, and the parameter of the soft threshold function is set to 3 times the standard deviation. The purpose of this step is to obtain high-quality environmental data and extract the main characteristics of the wind speed through signal processing methods.

[0020] S04. Establish a displacement relationship function between the buoy displacement variable component and the environmental wind speed variable component, and construct a buoy displacement matrix according to the displacement relationship function; The specific implementation of step S04 is to establish a correlation model between the buoy displacement and environmental factors. First, the least squares method is used to establish a preliminary relationship between the buoy displacement change component and the environmental wind speed change component. A weighting strategy is adopted during the fitting process, and the weights of recent data are set to larger values. Then, the environmental flow direction data is introduced to construct a flow direction influence function, which is obtained by polynomial fitting, and the polynomial order is set to the third order. After that, the genetic algorithm is used to optimize the model parameters. The population size is set to 100, the number of evolutionary generations is 50, the crossover probability is 0.8, and the mutation probability is 0.1. Finally, a displacement relationship function is obtained, which includes the comprehensive influence of environmental factors such as wind speed and flow direction. The purpose of this step is to construct an accurate buoy displacement prediction model to achieve an accurate judgment of the buoy displacement state.

[0021] S05. Calculate the buoy stability matrix based on the buoy displacement matrix, and determine the buoy displacement state according to the buoy stability matrix; The specific implementation of step S05 is to calculate the buoy stability matrix based on the buoy displacement matrix and determine the displacement state. First, the displacement coefficient is calculated by the sliding window method. The window length is set to 6 hours, and the sliding step is 1 hour. For each time window, the displacement components in three orthogonal directions are calculated, and the weighted average method is used to obtain the displacement coefficient. Then, a buoy stability matrix is established, and the matrix elements are obtained through normalization. The maximum-minimum method is used during the normalization process. Finally, the buoy displacement state is determined based on the eigenvalues of the stability matrix. The power iteration method is used for eigenvalue calculation, and the convergence threshold is set to 0.001. The purpose of this step is to achieve a quantitative judgment of the buoy displacement state through matrix analysis methods.

[0022] S06. Calculate the two-dimensional transformation value of the buoy voltage data, and judge the buoy voltage state according to the two-dimensional transformation value, which is recorded as the first buoy voltage determination result; The specific implementation of step S06 is to perform time-frequency analysis on the buoy voltage data. First, the time series of the voltage data is reconstructed. The delay coordinate method is used, and the delay time is determined by the mutual information method, and the embedding dimension is determined by the false nearest neighbor method. Then, a two-dimensional projection of the reconstructed phase space is performed, and the principal component analysis method is used for the projection. After that, the spectral characteristics of the projection data are calculated. The fast Fourier transform algorithm is used, and the number of transformation points is set to 1024. The voltage state is judged based on the spectral characteristics, and the judgment basis includes the amplitude and frequency distribution characteristics of the main frequency component. The purpose of this step is to achieve early identification of abnormal buoy voltage states through time-frequency analysis methods.

[0023] S07. If the first buoy voltage determination result is abnormal, calculate the horizontal value and vertical value of the buoy voltage, and calculate the buoy voltage fluctuation value according to the horizontal value and vertical value of the buoy voltage; The specific implementation of step S07 is to calculate the quantitative index of the voltage state. First, calculate the change rates of the voltage data in the time dimension and the amplitude dimension. The central difference method is adopted, and the difference step size is set to the sampling period. Then, calculate the spatial distribution characteristics of the change rates, including statistical quantities such as the mean, variance, and skewness. After that, combine these characteristic quantities to form the voltage fluctuation value, and the combination process adopts the weighted summation method, and the weight coefficients are determined by principal component analysis. The purpose of this step is to establish a quantitative evaluation index for the voltage state and provide an objective basis for abnormal judgment.

[0024] S08. Calculate the lateral change value and the longitudinal change value of the buoy anchor light data, establish a buoy anchor light matching function, and determine the seasonal compliance of the buoy anchor light data according to the buoy anchor light matching function; The specific implementation of step S08 is to analyze the time series characteristics of the buoy anchor light data. First, calculate the theoretical sunrise and sunset times based on the astronomical algorithm, and the algorithm takes into account the influence of geographical location and seasonal changes. Then, construct a theoretical model of the anchor light working cycle. The model uses trigonometric functions to describe the law of day and night changes, and the function parameters are obtained by least squares fitting of historical data. After that, calculate the matching degree between the actual anchor light state and the theoretical model, and the matching degree calculation adopts the correlation analysis method. The purpose of this step is to judge whether the buoy anchor light system is working properly by analyzing the working state of the anchor light.

[0025] S09. Establish a safety judgment function for the buoy hatch data and the buoy hatch water inlet data, and continuously monitor for 3 hours according to the safety judgment function to determine the buoy intrusion state; The specific implementation of step S09 is to monitor the buoy hatch and the hatch water inlet state. First, digitize the hatch data and the hatch water inlet data, and use the threshold judgment method to convert the analog signal into a discrete state. Then, construct a state transition model. The model is based on the Markov chain theory, and the state transition probability is estimated by statistical learning methods. After that, update the state evaluation result in real time, and the evaluation process adopts a sliding time window, and the window length is set to 3 hours. The purpose of this step is to realize the timely detection and alarm of buoy intrusion events.

[0026] S10. Calculate the buoy public safety value according to the buoy displacement matrix and the buoy stability matrix; The specific implementation of step S10 is to calculate the buoy public safety value. First, calculate the displacement trend based on the buoy displacement matrix, and use the eigenvalue decomposition method to extract the main movement direction. Then, combine the buoy stability matrix to evaluate the stability of the displacement, and the evaluation process adopts the fuzzy comprehensive evaluation method. After that, calculate the public safety value according to the evaluation result, and the calculation process takes into account factors such as the distance from the distance sensitive area and the moving speed. The purpose of this step is to quantify the potential impact degree of the buoy on public safety.

[0027] S11. Calculate the buoy's own safety value based on the buoy voltage fluctuation value, the buoy anchor light matching function, and the safety determination function; The specific implementation of step S11 is to calculate the buoy's own safety value. First, analyze the time series characteristics of the voltage fluctuation value, and use the trend analysis method to evaluate the voltage change trend. Then, evaluate the periodic characteristics of the anchor light matching function, and use the spectral analysis method to identify abnormal periods. After that, based on the output result of the safety determination function, use the decision tree method to comprehensively evaluate the safety status of the buoy itself. The purpose of this step is to evaluate the working status of each subsystem of the buoy and judge the safety level of the buoy itself.

[0028] S12. Calculate the buoy functional safety value based on the lateral change value and the longitudinal change value of the buoy sensor data; The specific implementation of step S12 is to calculate the buoy functional safety value. First, conduct a quality assessment of the buoy sensor data, and use the data quality control method to identify outliers and missing values. Then, calculate the time continuity index and the spatial consistency index of the data, and use the statistical test method in the calculation process. After that, obtain the functional safety value by integrating multiple indicators, and use the analytic hierarchy process to determine the index weights in the integration process. The purpose of this step is to evaluate the reliability of the buoy observation function.

[0029] S13. Perform weighted calculation on the buoy public safety value, the buoy's own safety value, and the buoy functional safety value to obtain the buoy safety assessment value; The specific implementation of step S13 is to calculate the buoy safety assessment value. First, use the expert scoring method to determine the initial weights of each safety index, and use the Delphi method in the scoring process. Then, optimize the weight coefficients through the analytic hierarchy process, construct a judgment matrix and conduct a consistency test. After that, perform weighted summation on each safety value to obtain the final safety assessment value. The purpose of this step is to obtain a quantitative index of the overall safety status of the buoy.

[0030] S14. Determine the buoy risk level according to the buoy safety assessment value; The specific implementation of step S14 is to determine the buoy risk level. First, establish a risk level classification standard, and use the clustering analysis method to determine the boundary values of the levels. Then, judge the risk level of the buoy based on the safety assessment value, and use the fuzzy rule reasoning in the judgment process. After that, automatically generate corresponding disposal suggestions according to the risk level. The purpose of this step is to provide decision support for buoy safety management.

[0031] The specific implementation of the above steps is described in detail below: The specific implementation of step S01 is to use a communication receiving module to obtain buoy message data. First, the original message data is received through a satellite communication module. At this time, the message data includes buoy latitude and longitude data, buoy voltage data, buoy anchor light data, buoy hatch data, buoy cabin water ingress data, and buoy sensor data. During the data reception process, quality assessment is carried out through signal quality indicators where is the signal strength, is the noise strength, is the signal amplitude. When is lower than 25 decibels, the data retransmission mechanism needs to be triggered. The data parsing process uses a recursive descent algorithm to implement the processing of the hierarchical data structure, including three stages: message header parsing, data segment extraction, and checksum verification. During the verification process, the cyclic redundancy check method is used, where is the th byte in the data packet to ensure the integrity of data transmission. The data parsing module uses a state machine design pattern, which can effectively handle the loss and disorder of data packets. The purpose of this step is to ensure the integrity and accuracy of data reception and provide a reliable data basis for subsequent analysis.

[0032] The specific implementation of step S02 is to process the buoy latitude and longitude data based on the empirical mode decomposition principle. First, the latitude and longitude data is converted into a time series, and data preprocessing is performed on the time series data, including outlier processing and data smoothing. The median filtering method is used for data preprocessing to remove outliers, and the filter window size is set to 5 data points. Outliers are removed by calculating . Then, the cubic spline interpolation method is used to smooth the data, and the interpolation node interval is set to 30 minutes. Next, empirical mode decomposition is performed on the preprocessed data, and the decomposition expression is . Through the iterative screening process, the time series is decomposed into multiple intrinsic mode functions. During the decomposition process, the iterative stop condition is set to a standard deviation threshold of 0.2, and the maximum number of iterations is 10 times. Finally, the obtained intrinsic mode functions are classified into two categories: the stable component of buoy displacement and the variable component of buoy displacement. The components with a frequency lower than 0.1 Hz are classified as stable components, and the components higher than this frequency are classified as variable components. The purpose of this step is to separate the stable trend and fluctuation characteristics of buoy displacement and provide a basis for subsequent shift state judgment.

[0033] For the multivariate decomposition operation in step S02, it is specifically expressed as follows: ; where is the time series of buoy latitude and longitude data; is the stable component of buoy displacement; is the component of the buoy displacement change; is the number of decomposition levels, and the value range is 3 to 5.

[0034] The specific implementation of step S03 is to obtain and process environmental monitoring data. First, obtain environmental wind speed data, environmental wind direction data, and environmental flow direction data from a weather station, and the data sampling interval is 10 minutes. Perform quality control on the obtained environmental data, including data range check and time continuity check. The effective range of wind speed data is set to 0 to 60 meters per second, and the range of wind direction data is 0 to 360 degrees. Data outside the range is corrected using the interpolation method. Then, apply the wavelet decomposition method to the environmental wind speed data for multi-scale analysis. The decomposition expression is , select the Debauch wavelet as the basis function, and set the number of decomposition levels to 4. The wavelet transform coefficients are calculated by , where is the wavelet basis function. Obtain the stable component and the change component of the environmental wind speed through reconstruction. The reconstruction process uses the threshold denoising method, and the parameter of the soft threshold function is set to 3 times the standard deviation. The purpose of this step is to obtain high-quality environmental data and extract the main features of the wind speed through signal processing methods.

[0035] For the multi-component decomposition of wind speed in step S03, it is specifically expressed as follows: ; In the formula, is the time series of environmental wind speed data; is the stable component of the environmental wind speed; is the component of the environmental wind speed change; is the number of decomposition levels, and the value range is 3 to 5.

[0036] The specific implementation of step S04 is to establish a correlation model between the buoy displacement and environmental factors. First, use the least squares method to establish a preliminary relationship between the buoy displacement change component and the environmental wind speed change component. A weighted strategy is adopted during the fitting process, and the weight of recent data is set to a larger value. Then, introduce the environmental flow direction data to construct the flow direction influence function , which is obtained by polynomial fitting, and the polynomial order is set to 3. Then, establish the shift relationship function , use the genetic algorithm to optimize the model parameters, set the population size to 100, the number of evolutionary generations to 50, the crossover probability to 0.8, and the mutation probability to 0.1. The fitness function of the algorithm uses the root mean square error, and the optimal combination of weight coefficients is obtained through iterative optimization. The purpose of this step is to construct an accurate buoy shift prediction model to achieve an accurate judgment of the buoy shift state.

[0037] For the shift relationship function in step S04, it is specifically expressed as follows: ; In the formula, is the buoy displacement function; is the variation component of the buoy displacement; is the variation component of the environmental wind speed; is the flow direction influence function; is the weight coefficient, and its value range is 0 - 1; is the error term, and its value range is 0.01 - 0.1.

[0038] The specific representation of the buoy displacement matrix is as follows: ; In the formula, represents at the moment of the displacement coefficient in the direction of.

[0039] The specific implementation of step S05 is to calculate the buoy stability matrix based on the buoy displacement matrix and determine the displacement state. First, calculate the displacement coefficient through the sliding window method. The window length is set to 6 hours, and the sliding step is 1 hour. For each time window, calculate the displacement components in three orthogonal directions, and use the weighted average method to obtain the displacement coefficient and construct the displacement matrix . Then establish the buoy stability matrix, and the matrix elements are obtained through normalization. The normalization process uses the maximum - minimum method. Finally, determine the buoy displacement state based on the eigenvalues of the stability matrix. The eigenvalue calculation uses the power iteration method , and the convergence threshold is set to 0.001. The purpose of this step is to quantitatively judge the buoy displacement state through matrix analysis methods.

[0040] The specific implementation of step S06 is to perform time - frequency analysis on the buoy voltage data. First, reconstruct the voltage data in the time series. The delay - coordinate method is used, and the delay time is determined by the mutual information method, and the embedding dimension is determined by the false nearest neighbor method. Then, perform two - dimensional projection on the reconstructed phase space, using the two - dimensional transformation , where is the original voltage time series. The projection method uses principal component analysis, and the number of transformation points is set to 1024. Then calculate the spatial distribution characteristics of the voltage data , where is the correction coefficient, and its value range is 0.1 to 0.3. The purpose of this step is to early identify the abnormal state of the buoy voltage through time - frequency analysis methods.

[0041] For the calculation of the two - dimensional transformation value in step S06, the specific representation is as follows: ; In the formula, is the two - dimensional transformation value of the buoy voltage data; is the original voltage time series; is the imaginary unit.

[0042] The specific implementation of step S07 is to calculate the quantitative index of the voltage state. First, calculate the change rates of the voltage data in the time dimension and the amplitude dimension, using the central difference method with the difference step set to the sampling period. Then, calculate the spatial distribution characteristics of the change rates, including statistics such as mean, variance, and skewness. After that, combine these characteristic quantities to form the voltage fluctuation value, which is calculated based on the spatial gradients of the lateral voltage value and the longitudinal voltage value. The purpose of this step is to establish a quantitative evaluation index for the voltage state and provide an objective basis for anomaly judgment.

[0043] For the calculation of the fluctuation value in step S07, it is specifically expressed as follows: ; In the formula, is the buoy voltage fluctuation value; is the lateral voltage value; is the longitudinal voltage value; is the correction coefficient, and its value range is 0.1 to 0.3.

[0044] The specific implementation of step S08 is to analyze the timing characteristics of the buoy anchor light data. First, calculate the theoretical sunrise and sunset times based on the astronomical algorithm, which takes into account the influence of geographical location and seasonal changes. Then, construct the theoretical model of the anchor light working cycle , where is the 24-hour cycle, is the seasonal adjustment coefficient, is the error term. After that, calculate the matching degree between the actual anchor light state and the theoretical model, and the matching degree is calculated using the correlation analysis method. The purpose of this step is to judge whether the buoy anchor light system is working properly by analyzing the working state of the anchor light.

[0045] For the anchor light matching function in step S08, it is specifically expressed as follows: ; In the formula, is the buoy anchor light matching function; is the day-night cycle, taking 24 hours; is the seasonal adjustment coefficient; is the error term, and its value range is 0.05 to 0.15.

[0046] The specific implementation of step S09 is to monitor the buoy hatch and hatch water ingress status. First, digitize the hatch data and hatch water ingress data to establish the safety judgment function , where is the hatch state function, is the water inlet state function, is the weight coefficient. The state evaluation result is updated in real time, and the sliding time window is adopted in the evaluation process, and the window length is set to 3 hours. The purpose of this step is to realize the timely detection and alarm of the buoy intrusion event.

[0047] For the safety judgment function in step S09, it is specifically expressed as follows: ; In the formula, is the safety judgment function; is the hatch door state function; is the water inlet state function; is the weight coefficient, and the value range is 01; is the error term, and the value range is 0.010.1.

[0048] The specific implementation manners of steps S10 to S14 involve the comprehensive safety assessment of the buoy. First, calculate the buoy safety assessment value , where is the weight coefficient. The buoy displacement prediction is calculated by , and the safety assessment is carried out in combination with the actual position of the buoy. The final offshore safety assessment score is obtained by calculation. The system divides the risk level according to the assessment score into: above 100 points is Class A risk, 40 to 100 points is Class B risk, and 0 to 40 points is Class C risk. Based on the risk level, the clustering analysis method is used to determine the grade boundary value, and the fuzzy rule reasoning is used to judge the risk level of the buoy. Finally, the corresponding disposal suggestions are automatically generated according to the risk level. The purpose of this stage is to realize the quantitative assessment and risk early warning of the buoy safety state, and provide decision-making support for the buoy safety management.

[0049] For the weighted calculation in step S13, it is specifically expressed as follows: ; In the formula, is the buoy safety assessment value; is the buoy public safety value; is the buoy own safety value; is the buoy functional safety value; is the weight coefficient, and satisfies ; is the system error, and the value range is 0.01~0.05.

[0050] The construction principles of these equations are as follows: 1. The multivariate decomposition operation adopts the empirical mode decomposition principle, which can effectively separate the stable component and the variable component, facilitating subsequent analysis; 2. The shift relationship function takes into account multiple influences such as wind speed and flow direction, adopts the principle of linear superposition, and introduces an error term to improve robustness; 3. The two-dimensional transformation value calculation adopts the Fourier transform principle, which can better reflect the frequency domain characteristics of voltage data; 4. The anchor light matching function is based on the periodic change characteristics of trigonometric functions and can accurately describe the law of day and night changes; 5. The safety judgment function adopts the form of weighted summation, considering the comprehensive influence of multiple safety factors; 6. The calculation of the buoy safety evaluation value adopts the linear weighting method, which can flexibly adjust the importance of each index.

[0051] The following details the process of the derivation and establishment of each equation.

[0052] 1. Derivation of the buoy longitude and latitude multi - decomposition equation : First, screen the extreme points based on the original longitude and latitude sequence; Secondly, perform cubic spline interpolation on the extreme points to obtain the upper envelope and the lower envelope ; Then, calculate the mean function ; Next, separate from the original data; Finally, iterate the above process to obtain each component.

[0053] Parameter description: Obtained through iterative calculation, and two conditions need to be met: The difference between the number of extreme points and the number of zero - crossing points does not exceed 1; At any point, the mean of the mean envelope is approximately 0.

[0054] 2. Derivation of the shift relationship function : Step 1: Establish an initial model by fitting historical data using the least - squares method; Step 2: Introduce the flow - direction influence function to correct the model; Step 3: Optimize the weight coefficients through the genetic algorithm .

[0055] Parameter acquisition method: Obtained by experimental methods. The specific steps are as follows: (1) Measure the buoy offset under different flow - direction conditions; (2) Establish the mapping relationship between the flow direction and the offset; (3) Obtain the flow direction influence function through polynomial fitting.

[0056] 3. Buoy displacement matrix Construction process: The first step: Calculate the displacement vectors at adjacent times; The second step: Decompose the displacement vectors into three orthogonal directions; The third step: Calculate the displacement coefficients in each direction through the sliding window method.

[0057] Matrix element Calculation formula: ; In the formula: is the window length; is the weight of the th sampling point; is the th sampling point in the direction displacement component.

[0058] 4. Two-dimensional transformation value of voltage Derivation: The first step: Perform time-frequency domain conversion on the original voltage sequence; the second step: Introduce a complex exponential kernel function for two-dimensional transformation; the third step: Optimize the calculation efficiency through fast Fourier transform.

[0059] Specific implementation steps: (1) Rearrange the voltage data into a two-dimensional matrix; (2) Perform Fourier transform on the row direction; (3) Perform Fourier transform on the column direction; (4) Extract the transformation coefficients.

[0060] 5. Buoy voltage fluctuation value Derivation: The first step: Calculate the voltage spatial gradient: ; ; The second step: Introduce a correction coefficient to compensate for numerical calculation errors.

[0061] Parameter Determination method: Determine the optimal correction coefficient range through Monte Carlo simulation.

[0062] 6. Anchor light matching function Derivation: The first step: Determine the sunrise and sunset times based on astronomical calculations; the second step: Establish a reference period function; the third step: Introduce a seasonal adjustment coefficient.

[0063] Method for obtaining seasonal adjustment coefficient: ; ; In the formula: is the day of the year; is the fitting coefficient.

[0064] 7. Derivation of safety judgment function : The first step: Construct the hatch status indication function: ; The second step: Construct the water inlet status function: ; The third step: Optimize the weight coefficient by machine learning method.

[0065] 8. Derivation of buoy safety evaluation value : The first step: Determine the initial weight by the analytic hierarchy process; the second step: Correct the weight using the expert scoring method; the third step: Introduce the systematic error term to improve the robustness of the model.

[0066] Method for determining weight coefficient: (1) Establish a judgment matrix; (2) Calculate the eigenvector; (3) Conduct a consistency test.

[0067] Specifically, the principle of the present invention is as follows: The technical solution of the present invention is based on the principles of multi-source data fusion and multi-scale analysis. By establishing a mathematical model, the displacement characteristics, state parameters, and environmental factors of the buoy are organically combined. In terms of displacement analysis, the empirical mode decomposition principle is adopted. This method has the characteristics of strong self-adaptability and can handle non-linear and non-stationary signals, and can effectively separate displacement characteristics at different scales. By analyzing the frequency characteristics of each scale component, the periodic displacement caused by natural factors and the sudden displacement under abnormal conditions can be distinguished. In terms of environmental correlation analysis, a displacement relationship function is established based on the principle of fluid mechanics. This function takes into account the influence of environmental factors such as wind force and water flow on the movement of the buoy. By optimizing the weight coefficients through the genetic algorithm, the precise correlation between environmental factors and the buoy displacement is achieved. In terms of state monitoring, time-frequency analysis methods are used to process sensor data. By analyzing the time-domain and frequency-domain characteristics of the signals, characteristic quantities that can reflect the working state of the buoy are extracted. The system also establishes a state evaluation model based on fuzzy rules. This model can handle the uncertainty of data and improve the reliability of the evaluation results. All these analysis results determine the weights through the analytic hierarchy process to form the final safety evaluation value. The innovation of this technical solution lies in: First, through multi-scale analysis, an accurate description of the buoy movement characteristics is achieved; second, a quantitative relationship between environmental factors and the buoy state is established; finally, a complete safety evaluation system is formed. These technological innovations support each other and jointly constitute a logically rigorous technical solution, which can effectively solve the problems of buoy safety state identification and evaluation.

[0068] The following provides a specific Embodiment 1 of the present invention. The specific implementation methods of each step in this Embodiment 1 are described in detail as follows: An intelligent buoy with automatic safety monitoring function is deployed in a certain sea area. The deployment location is at 21°15′ north latitude and 114°32′ east longitude. The water depth of this sea area is about 150 meters, belonging to the offshore environment, adjacent to important shipping lanes and fishing areas. This intelligent buoy has functions such as automatic meteorological observation and ocean hydrological monitoring, and the designed service life is 3 years. The buoy has a mass of 2000 kilograms, the anchor chain length is 180 meters, and the turning radius is designed to be 25 meters.

[0069] The buoy was launched on January 10, 2024. In this embodiment, the monitoring data on January 15, 2024 is selected for analysis and illustration. First, the buoy message data is received through the satellite communication system, and the data reception quality index is decibels, higher than the threshold requirement of 25 decibels, and the data transmission is stable and reliable. The parsing results of the original message are shown in Table 1 below:

[0070] The environmental monitoring data obtained during the same period is shown in Table 2:

[0071] Perform multivariate decomposition on the longitude and latitude data, and use the empirical mode decomposition method to obtain 3 intrinsic mode functions. The first-layer component represents short-term fluctuations with a frequency of approximately 0.2 Hz; the second-layer component represents medium-term changes with a frequency of approximately 0.05 Hz; the third-layer component represents long-term trends with a frequency below 0.01 Hz. The analysis results show that the buoy displacement is within the turning radius range, which belongs to the normal state.

[0072] After wavelet decomposition of the environmental data, the stable component of the wind speed is 8.5 m / s, and the maximum amplitude of the fluctuating component is 1.6 m / s. Through calculation using the shift relationship function, the weight coefficients , , . The maximum eigenvalue of the shift matrix is 0.82, which is lower than the warning threshold of 0.9, indicating that the buoy mooring state is stable.

[0073] The voltage data analysis shows that the average voltage during the daytime (07:00 - 17:00) is 13.0 V, and the average voltage during the nighttime (17:00 - 07:00 the next day) is 12.7 V. The maximum voltage fluctuation value is 0.15, which is lower than the abnormal threshold of 0.2, indicating that the power supply system is working properly. The status of the anchor light matches well with the theoretical sunrise (06:58) and sunset (17:42) times, with a deviation within 5 minutes.

[0074] The calculation results of the safety assessment indicators are shown in Table 3 below:

[0075] The comparison between the present invention and the traditional buoy safety monitoring method is shown in Table 4 below:

[0076] The specific application effect of this embodiment shows that this method can realize the all-weather automatic monitoring of the buoy safety state, and has the following advantages compared with the traditional method: 1. By using the multivariate decomposition and correlation analysis methods, the accuracy of abnormal state recognition is improved, and false alarms and missed alarms are reduced. 2. A quantitative safety assessment system is established, making safety management more objective and standardized. 3. Predictive early warning is realized, providing sufficient response time for safety handling. 4. The system has good self-adaptability and can automatically adjust parameters according to environmental changes.

[0077] In practical applications, this method has been popularized and applied in multiple buoy monitoring stations in the South China Sea, achieving good results and providing strong guarantees for the safe operation of the buoys.

[0078] To better understand and implement the present invention, the following provides Example 2 of a specific application scenario of the present invention, the buoy offshore safety automatic monitoring and assessment system: 1.1. Buoy displacement monitoring method: After the buoy's longitude and latitude data hint changes occur, it is advisable to determine whether the buoy's position status is abnormal based on the speed and direction of the buoy's movement, combined with the measured wind, wave, and current data or corresponding forecast data. This mainly includes: Slight position changes of the moored buoy within the mooring turning radius are normal; The buoy does not send back positioning data, including the situation where there is no positioning data in the message data, the original message cannot be parsed, and the independent positioning system does not send back resolvable positioning data simultaneously; If the buoy's longitude and latitude change significantly, calculate the displacement distance and displacement direction between two adjacent time instances, and combine with the measured (or forecast) wind speed, wind direction, and current direction at the same time instance for determination. The method can be seen in Table 5; If the buoy's longitude and latitude change significantly and it is determined to be a false alarm, but the movement occurs continuously and the speed is stable, it can be determined as vessel towing.

[0079] Table 5 Buoy Displacement Judgment Table

[0080] The displacement distance formula is as follows: The displacement distance of the buoy from point A to point B , which is the spherical distance between points A and B, that is, the minor arc of the great circle passing through points A and B. The longitudes and latitudes of points A and B are A( , ), B( , ), and the calculation formula is as follows: ; ; In the formula: represents the Earth's radius of 6371 km; represents the central angle between points A and B; represents latitude; represents longitude.

[0081] 1.2. Buoy Voltage Safety Monitoring Method: Carry out buoy voltage safety monitoring by analyzing the voltage data in the buoy message, specifically including: Under relatively consistent weather conditions, the buoy voltage starts to rise 2 - 3 hours after sunrise every day, and the buoy voltage starts to drop 2 - 3 hours after sunset every day. The voltage returned in the buoy status information should be higher than the minimum rated voltage of each buoy sensor; The maximum voltage of the buoy battery should remain stable. If the maximum voltage decays, it is necessary to judge whether it is caused by bad weather through data such as wind, temperature, air pressure, relative humidity, and visibility; If the voltage is normal for many days, but the voltage returned by the buoy status information is lower than the minimum rated voltage of each buoy sensor, excluding bad weather, it can be determined that the power system is faulty; The voltage alarm will be lifted the next time after it is given. If there is no voltage alarm for 24 consecutive hours, it can be determined that the buoy voltage is normal. When monitoring the buoy voltage, the influence of the buoy's longitude and latitude, sunlight angle, season, weather, obstruction (ice and snow), etc. should be considered.

[0082] 1.3. Safety monitoring method of buoy anchor light: By analyzing the anchor light data in the buoy message, the safety monitoring of the buoy anchor light is carried out, including: The anchor buoy light should be light-sensitive, kept "off" during the day and "on" at night. The switching process between "on" and "off" should be continuous and in accordance with physical laws, and the switching process should not occur multiple times a day; There should be no significant change in the "on" and "off" times of the buoy anchor light on adjacent dates, and the working hours should be consistent with the seasonal characteristics of the area where the buoy is located.

[0083] 1.4. Buoy hatch safety monitoring method: The door sensor should be installed at the entry and exit doors to carry out buoy door safety monitoring based on the door alarm information, including: After receiving the alarm from the hatch sensor, continuously monitor the working status of the buoy for 3 hours. If the alarm is lifted or the system is working normally, it is a false alarm or the sensor is loose; Within 3 hours of monitoring, if the entire system fails and is determined to be caused by the hatch being open and an outsider entering and damaging the buoy system, a joint judgment should be made based on the water ingress into the buoy compartment.

[0084] 1.5. Buoy tank water ingress monitoring method: The tank water inflow sensor should be installed at the bottom of the buoy instrument. According to the tank water inflow alarm information, the buoy water inflow safety monitoring should be carried out, including: After the water inflow sensor alarms, the buoy status and water inflow situation are analyzed by combining the buoy voltage data and hatch alarm information; After the water inflow sensor alarms, the maximum wave height data at the corresponding time is used to analyze whether a ship collision has occurred.

[0085] 1.6. Buoy message status monitoring method: By analyzing the buoy parsed message and the original message status monitoring, it includes: If the buoy currently displays no data, it is advisable to check the original message to determine whether the buoy has no data to send back or has data but cannot be parsed; In the original message record, by analyzing the internal code of the buoy, the communication method number, the receiving time and the message time, whether the current time message can be manually parsed and whether the longitude and latitude information can be manually read.

[0086] 2 Buoy Marine Safety Assessment: 2.1 Buoy Marine Safety Assessment Indexes: 2.1.1 Indexes Affecting Public Safety: Factors affecting public safety mainly include factors that the safety monitoring buoy has shifted and may or have entered the disputed sea area or may cause public safety to marine facilities, marine aquaculture, channel safety, etc.

[0087] 2.1.2 Indexes Affecting the Safety of the Buoy Itself Factors affecting the safety of the buoy itself mainly include factors that affect the safety of the buoy itself such as voltage faults, hatch opening, buoy water ingress, anchor light faults, dragging anchor, etc. found in marine safety monitoring.

[0088] 2.1.3 Indexes Affecting the Functional Safety of the Buoy Factors affecting the functional safety of the buoy mainly include factors that affect the safety of the buoy's observation activities due to message parsing, data transmission, sensor faults, etc. found in marine safety monitoring.

[0089] 2.1.4 Grading of Marine Safety Assessment Indexes The grading of the buoy marine safety assessment indexes can be seen in Table 6.

[0090] Table 6 Grading of Buoy Marine Safety Assessment Indexes

[0091] 2.2 Buoy Marine Safety Assessment Calculation: ; In the formula: represents the buoy marine safety assessment score, , , are the index assignments corresponding to the situations affecting public safety, affecting the safety of the buoy itself, and affecting the functional safety of the buoy in the marine safety assessment indexes respectively.

[0092] 2.3 Buoy Safety Status Assessment: Table 7 Buoy Marine Safety Disposal Methods

[0093] 3. The buoy shift prediction formula is as follows: ; In the formula: represents the speed of the buoy, which can be calculated by referring to the following formula; represents the sum of all driving forces acting on the buoy; represents the object mass; Indicates the exercise time.

[0094] The hydrodynamic mathematical model is established based on the Eulerian field, and the Lagrangian viewpoint is needed to describe the motion of particles. In the Eulerian field, for two-dimensional planar problems, the velocity at any spatial point can be expressed as: ; Adopting the Lagrangian viewpoint, the velocity of any particle can be expressed as: ; The motion trajectory of the particle can be obtained through the following integral: ; If the at each moment is known, the motion trajectory of the particle can be obtained from the above formula by means of numerical integration.

[0095] The variables and explanations of the present invention are shown in Table 8:

[0096] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for automatic monitoring of buoy marine safety, characterized in that: The method comprises the following steps: receiving buoy message data, wherein the buoy message data comprises buoy latitude and longitude data, buoy voltage data, buoy anchor light data, buoy hatch data, buoy cabin water inflow data, and buoy sensor data; performing multivariate decomposition operation on the buoy latitude and longitude data, and establishing a displacement relationship function between the buoy displacement change component and the ambient wind speed change component; constructing a buoy displacement matrix based on the displacement relationship function, calculating the buoy public safety value, the buoy self-safety value, and the buoy functional safety value, and performing weighted calculation on the buoy public safety value, the buoy self-safety value, and the buoy functional safety value to obtain a buoy safety assessment value; The buoy risk level is determined according to the buoy safety assessment value.

2. A method for automatic monitoring of buoy marine safety according to claim 1, characterized in that: The step of processing the buoy message data is specifically to receive the original message data through the communication receiving module, and use a recursive descent algorithm to process the hierarchical data structure, including message header parsing, data segment extraction, checksum verification, and use a state machine design pattern to handle data packet loss and disorder, and set the signal strength threshold to 25 decibels and the signal-to-noise ratio threshold to 10 decibels.

3. A method for automatic monitoring of marine safety of buoys according to claim 2, characterized in that: The step of performing multivariate decomposition operation on the buoy longitude and latitude data is specifically to convert the buoy longitude and latitude data into a time series, use the median filtering method to remove abnormal points, use the cubic spline interpolation method to perform smoothing, decompose the time series into multiple intrinsic mode functions through an iterative screening process, set the iteration stop condition to the standard deviation threshold of 0.2, and obtain the buoy displacement stability component and the buoy displacement change component.

4. A method for automatic monitoring of marine safety of buoys according to claim 3, characterized in that: The steps of establishing the displacement relationship function are as follows: using the least squares method to establish the preliminary relationship between the buoy displacement change component and the ambient wind speed change component, constructing the flow direction influence function through polynomial fitting, and optimizing the model parameters by genetic algorithm. The population size is set to 100, the evolutionary generation is 50 generations, the crossover probability is 0.8, and the mutation probability is 0.

1.

5. A method for automatic monitoring of buoy marine safety according to claim 4, characterized in that: The steps of constructing the buoy shift matrix are as follows: specifically, the shift coefficient is calculated by the sliding window method, the window length is set to 6 hours, the sliding step is 1 hour, the displacement components in three orthogonal directions are calculated, the maximum and minimum method is used for normalization, the eigenvalue is calculated by the power iteration method, and the convergence threshold is set to 0.

001.

6. A method for automatic monitoring of buoy marine safety according to claim 5, characterized in that: The steps of calculating the public safety value of the buoy are to extract the main movement direction of the buoy displacement matrix by using the eigenvalue decomposition method, evaluate the displacement stability in combination with the buoy stability matrix, and evaluate the potential impact of the buoy on public safety by using the fuzzy comprehensive evaluation method.

7. A method for automatic monitoring of marine safety of buoys according to claim 6, characterized in that: The steps of calculating the safety value of the buoy itself are specifically to use a trend analysis method to evaluate the buoy voltage change trend, use a spectrum analysis method to identify the abnormal period of the buoy anchor light matching function, and evaluate the buoy door data and buoy cabin water inflow data based on a decision tree method.

8. A method for automatic monitoring of marine safety of buoys according to claim 7, characterized in that: The steps for calculating the functional safety value of the buoy are specifically to use a data quality control method to identify abnormal values ​​and missing values ​​of the buoy sensor data, calculate the time continuity index and spatial consistency index of the data, and use a hierarchical analysis method to determine the index weights.

9. A method for automatic monitoring of buoy marine safety according to claim 8, characterized in that: The calculation steps of the buoy safety assessment value are as follows: the Delphi method is used to determine the initial weights of various safety indicators, the weight coefficients are optimized by the hierarchical analysis method and consistency test is performed, and the final buoy safety assessment value is obtained by weighted summation.

10. A method for automatic monitoring of marine safety of buoys according to claim 9, characterized in that: The step of determining the risk level of the buoy is specifically to use a cluster analysis method to determine the level boundary value, and to use fuzzy rule reasoning to determine the risk level of the buoy.

Images