A water flow measurement system and method for unmanned ships and boats based on artificial intelligence

Abstract

The present invention relates to the technical field of flow measurement, and specifically to a water flow measurement system and method for unmanned ships and boats based on artificial intelligence, wherein the system comprises a feature analysis module, a dynamic coding module, a real-time optimization module, an abnormal response module, a data correction module, and an unmanned ship control module. In the present invention, water flow data is collected by a flow velocity sensor of an unmanned ship, the average flow velocity and the maximum flow velocity are analyzed, the flow velocity change trend is extracted, the dynamic change of the water flow direction is judged, flow characteristic data is generated, the periodicity and instantaneous change of the water flow are analyzed based on the flow characteristic data, the coding parameters are adjusted in real time, the coding strategy is optimized, the filter parameters are adjusted according to the optimized coding strategy, the signal processing flow is optimized, the abnormal water flow is detected and responded to, the water flow velocity is monitored, the abnormal data points are judged, the abnormal data are classified and marked, and various types of abnormal water flow data are analyzed by dynamically adjusting the detection parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of flow measurement, and in particular to a water flow measurement system and method for unmanned ships and boats based on artificial intelligence. Background Art

[0002] The field of flow measurement technology focuses on measuring the amount of fluid passing through a specific cross-section in a certain period of time through various instruments and methods. This technology is essential for industrial applications, environmental monitoring, water management and scientific research. There are various types of flow measurement equipment, from traditional mechanical flow meters, including turbine flow meters and volumetric flow meters, to electronic devices that utilize modern technology, including ultrasonic flow meters and electromagnetic flow meters. Provide accurate flow data, optimize process control, ensure system efficiency, and support the assessment of environmental impact. With the development of technology, flow measurement technology is also constantly improving, including wireless sensing technology and real-time data processing, improving the accuracy and convenience of measurement.

[0003] Among them, the water flow measurement system for unmanned ships and boats based on artificial intelligence is a system that combines unmanned boat technology and flow measurement methods. The main purpose of the system is to automatically measure the flow of water bodies without manual operation, especially in inaccessible or dangerous water environments. By using artificial intelligence, the system can autonomously navigate to the measurement point, automatically collect and process data, optimize the measurement process, and predict the impact of environmental changes on flow. It is widely used in environmental monitoring, water resources management, scientific research, commercial fishing and other fields, helping to achieve more accurate and efficient water flow data collection and analysis.

[0004] Existing flow measurement technology relies on traditional flow meters, which are complex to operate and require manual intervention, making them difficult to adapt to complex and changing water flow environments. Mechanical flow meters such as turbine flow meters and volumetric flow meters are easily affected by impurities and sediments during the measurement process, resulting in data distortion. Electronic flow meters such as ultrasonic flow meters and electromagnetic flow meters have reduced accuracy in complex water flow environments and have high installation and maintenance costs. Existing technologies lack autonomous navigation and real-time data processing capabilities, and are unable to achieve effective measurements in inaccessible or dangerous waters, resulting in inefficient environmental monitoring and water resources management, and insufficient timeliness and accuracy in data collection, which affects subsequent data analysis and decision support. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a water flow measurement system and method for unmanned ships and boats based on artificial intelligence.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A water flow measurement system for unmanned ships and boats based on artificial intelligence comprises:

[0007] The feature analysis module analyzes key flow velocity indicators based on the water flow data collected by the flow velocity sensor of the unmanned ship, extracts the change trend of the flow velocity, determines the dynamic change of the water flow direction, and generates flow characteristic data;

[0008] The dynamic coding module analyzes the periodicity and instantaneous changes of the water flow based on the flow characteristic data, automatically adjusts the coding parameters according to the real-time flow velocity changes, and generates an optimized coding strategy;

[0009] The real-time optimization module adjusts the filter parameters based on the optimization coding strategy and various water flow conditions, optimizes the signal processing flow, and generates adjusted signal processing parameters;

[0010] The abnormal response module constructs an artificial intelligence model to detect and respond to abnormal water flow conditions based on the adjusted signal processing parameters, and generates abnormal flow rate response results by dynamically adjusting the detection parameters;

[0011] The data correction module compares normal and abnormal data based on the abnormal flow rate response result, identifies deviation points and records flow rate characteristics, uses a model to monitor and correct flow rate data, and generates corrected flow data;

[0012] The unmanned ship control module analyzes the relationship between the flow data and the navigation route based on the corrected flow data, formulates navigation and measurement plans, adjusts the navigation and measurement methods of the unmanned ship in the water flow environment, and generates optimized navigation parameters.

[0013] As a further solution of the present invention, it is characterized in that: the flow characteristic data include the mean value, variation amplitude and flow direction trend of the water flow velocity, the optimized coding strategy includes a dynamic adjustment scheme of the coding parameters, a periodic change adjustment strategy and instantaneous change response measures, the adjusted signal processing parameters include filter frequency, gain adjustment and signal smoothing parameters, the abnormal flow velocity response results include abnormal data point classification, abnormal cause analysis and real-time response plan, the corrected flow data include flow velocity correction value, deviation point correction record and optimized sampling interval, and the optimized navigation parameters include navigation path optimization, navigation speed adjustment and navigation method.

[0014] As a further solution of the present invention, the feature analysis module includes:

[0015] The data processing submodule filters and normalizes the flow data based on the flow data collected by the flow velocity sensor of the unmanned ship, segments the data and marks the abnormal values, fills the missing data, and generates flow velocity change data;

[0016] The dynamic monitoring submodule analyzes the change trend of the water flow direction based on the flow velocity change data, determines the flow direction of the water flow by calculating the vector change of the real-time flow velocity, and generates water flow direction data;

[0017] The cross-comparison submodule performs a cross-comparison of the flow velocity data and the water temperature data based on the water flow direction data, identifies the key time period by calculating the correlation coefficient between the flow velocity and the water temperature, analyzes the flow velocity and water temperature change patterns within the time period, and generates flow characteristic data.

[0018] As a further solution of the present invention, the dynamic encoding module includes:

[0019] The periodic analysis submodule analyzes the periodicity and instantaneous changes of the water flow based on the flow characteristic data, extracts the periodicity and instantaneous fluctuation characteristics in the water flow data, calculates the statistical indicators of the characteristics, and generates the periodic change trend of the water flow;

[0020] The parameter setting submodule sets the reference value of the initialization coding based on the water flow periodic variation trend, calculates the coding efficiency under the initial conditions, and optimizes the coding parameters according to various water flow conditions to generate coding initialization parameters;

[0021] The automatic adjustment submodule is based on the encoding initialization parameters and the monitored real-time flow rate changes, extracts the instantaneous change value of the current flow rate, performs difference analysis with the initial encoding parameters, identifies the range and direction of the encoding parameter adjustment, automatically updates the encoding parameters in real time, and generates an optimized encoding strategy.

[0022] As a further solution of the present invention, the real-time optimization module includes:

[0023] The filter adjustment submodule adjusts the filter parameters based on the optimized coding strategy, analyzes the noise level and signal strength in the water flow data, detects the noise frequency and amplitude changes in the water flow, and measures the distribution of signal strength at multiple frequencies to generate filter parameter settings;

[0024] The window adjustment submodule dynamically adjusts the filter window size based on the filter parameter setting, measures the signal processing efficiency under various window sizes, calculates the impact of the window size on the signal quality, selects the optimal window size by comparing the signal quality, and generates the window optimization parameters;

[0025] The condition judgment submodule judges the signal characteristics under various flow rate conditions based on the window optimization parameters, extracts the signal change patterns under various flow rate conditions, performs flow rate characteristic comparison analysis, identifies and extracts key signal characteristics, and generates adjusted signal processing parameters.

[0026] As a further solution of the present invention, the abnormal response module includes:

[0027] The flow velocity detection submodule builds an artificial intelligence model based on the adjusted signal processing parameters, uses the model to detect and respond to abnormal water flow conditions, monitors water flow velocity in real time, records water flow velocity values ​​during key time periods, uses the model to detect abnormal change trends, and generates water flow monitoring data;

[0028] The data classification submodule determines abnormal water flow data points based on the water flow monitoring data and marks the abnormal data, extracts the features of the abnormal water flow data points by comparing the monitoring data with the normal flow rate range, and classifies them according to the abnormal type to generate abnormal flow rate classification;

[0029] The dynamic adjustment submodule analyzes various types of water flow abnormality data by dynamically adjusting detection parameters based on the abnormal flow velocity classification, adjusts the sensitivity and discrimination criteria of the detection parameters according to the characteristics of various abnormal types, extracts abnormal flow velocity characteristics, and generates abnormal flow velocity response results.

[0030] As a further solution of the present invention, the data correction module includes:

[0031] The deviation identification submodule compares normal and abnormal data based on the abnormal flow rate response result, identifies the deviation point and records the flow rate characteristics, calculates the difference value, maps the difference value to the time axis, determines the time period range of the deviation point, and generates flow rate deviation data;

[0032] The parameter adjustment submodule adjusts the measurement parameters of the unmanned boat based on the flow deviation data, sets the sensitivity and sampling frequency of the measurement equipment in real time according to the changes in the water flow and the hydrological environment, analyzes the flow acquisition effect according to the adjusted measurement parameters, and generates optimized measurement parameters;

[0033] The artificial intelligence correction submodule uses an artificial intelligence model to monitor and correct the flow rate data based on the optimized measurement parameters, compares the real-time collected flow rate data with the reference data, uses a Kalman filter algorithm, and iteratively adjusts the model parameters to correct errors in the flow rate data, thereby generating corrected flow data;

[0034] The Kalman filter algorithm follows the formula:

[0035] x k =x k-1 +K k (z k -Hx k-1 )+λΔTσ 2

[0036] Among them, x k is the corrected flow rate value, x k-1 is the flow velocity value at the previous moment, K k is the Kalman gain, z kis the measured velocity value at the current moment, H is the observation model, λ is the adjustment coefficient, ΔT is the sampling time interval, σ 2 is the variance of the flow rate data.

[0037] As a further solution of the present invention, the unmanned ship control module includes:

[0038] The path analysis submodule analyzes the correlation between the corrected flow data and the navigation route based on the corrected flow data, calculates the flow changes under various paths, mines the optimal navigation route, and generates path optimization data;

[0039] The navigation optimization submodule analyzes the key nodes and turning points in the path data based on the path optimization data, adjusts the navigation speed, steering angle and acceleration during the navigation process, optimizes the navigation path and speed, and generates navigation control parameters;

[0040] The task adjustment submodule adjusts the navigation and measurement methods of the unmanned ship in various water flow environments based on the navigation control parameters, analyzes the real-time changes of the water flow environment, dynamically adjusts the navigation path and speed of the unmanned ship by updating the navigation tasks and measurement plans in real time, and generates optimized navigation parameters.

[0041] A method for measuring water flow of an unmanned ship or boat based on artificial intelligence, wherein the method for measuring water flow of an unmanned ship or boat based on artificial intelligence is performed based on the above-mentioned water flow measurement system for an unmanned ship or boat based on artificial intelligence, and comprises the following steps:

[0042] S1: Based on the water flow data collected by the flow velocity sensor of the unmanned boat, the average flow velocity and the maximum flow velocity are analyzed, the flow velocity change trend is extracted, the dynamic change of the water flow direction is judged, and the flow characteristic data is obtained;

[0043] S2: Based on the flow characteristic data, analyze the periodicity and instantaneous changes of the water flow, identify the flow velocity frequency characteristics, adjust the encoding parameters according to the real-time flow velocity changes, and generate an optimized encoding strategy;

[0044] S3: Based on the optimized coding strategy, adjust the filter parameters, optimize the signal processing flow, and generate adjusted signal processing parameters;

[0045] S4: Based on the adjusted signal processing parameters, detect and respond to abnormal water flow conditions, monitor water flow velocity, determine abnormal data points, classify and mark abnormal data, and generate abnormal flow velocity response results by dynamically adjusting detection parameters and analyzing various types of abnormal water flow data;

[0046] S5: Based on the abnormal flow velocity response results, compare normal and abnormal data, identify deviation points and record flow velocity characteristics, use artificial intelligence models to monitor and correct flow velocity data, analyze the relationship between flow data and navigation routes based on the corrected flow data, optimize the navigation path and speed, and generate optimized navigation parameters.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are:

[0048] In the present invention, water flow data is collected by the flow velocity sensor of the unmanned boat, the average flow velocity and the maximum flow velocity are analyzed, the flow velocity change trend is extracted, the dynamic change of the water flow direction is judged, and the flow characteristic data is generated. Based on the flow characteristic data, the periodicity and instantaneous changes of the water flow are analyzed, the coding parameters are adjusted in real time, and the coding strategy is optimized. According to the optimized coding strategy, the filter parameters are adjusted to optimize the signal processing flow. Detect and respond to abnormal water flow conditions, monitor the water flow velocity, judge abnormal data points, classify and mark abnormal data, and analyze various types of abnormal water flow data by dynamically adjusting the detection parameters. Compare normal and abnormal data, identify deviation points and record flow velocity characteristics, use artificial intelligence to monitor and correct flow velocity data, and generate corrected flow data. Analyze the relationship between flow data and navigation routes, adjust the navigation and measurement methods of unmanned boats in various water flow environments, and generate optimized navigation parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a system flow chart of the present invention;

[0050] Figure 2 It is a schematic diagram of the system framework of the present invention;

[0051] Figure 3 It is a flow chart of the feature analysis module of the present invention;

[0052] Figure 4 It is a flow chart of the dynamic encoding module of the present invention;

[0053] Figure 5 It is a flow chart of the real-time optimization module of the present invention;

[0054] Figure 6 is a flow chart of the abnormal response module of the present invention;

[0055] Figure 7 It is a flow chart of the data correction module of the present invention;

[0056] Figure 8 This is a flow chart of the unmanned ship control module of the present invention;

[0057] Fig. 9 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0060] Embodiment 1

[0061] See also Figure 1 to Figure 2 The present invention provides a technical solution: a water flow measurement system for unmanned ships and boats based on artificial intelligence, comprising:

[0062] The characteristic analysis module analyzes the flow data collected by the flow sensor of the unmanned boat, analyzes the flow velocity, flow direction, and water temperature indicators, extracts the change trend of the flow velocity, determines the dynamic change of the water flow direction, and cross-compares the flow velocity and water temperature to generate flow characteristic data;

[0063] The dynamic coding module analyzes the periodicity and instantaneous changes of water flow based on flow characteristic data, sets initial coding parameters, automatically adjusts coding parameters according to real-time flow rate changes, and generates optimized coding strategies;

[0064] The real-time optimization module is based on the optimization coding strategy, adjusts the filter parameters according to various water flow conditions, dynamically adjusts the filter window size with reference to the water flow characteristic parameters, determines the signal characteristics under various flow rate conditions, and generates adjusted signal processing parameters;

[0065] The abnormal response module builds an artificial intelligence model based on the adjusted signal processing parameters, detects and responds to abnormal water flow conditions, monitors water flow velocity in real time, determines abnormal data points, classifies and marks abnormal data points, and analyzes various types of abnormal water flow data by dynamically adjusting detection parameters to generate abnormal flow velocity response results;

[0066] The data correction module is based on the abnormal flow rate response results. By comparing normal and abnormal data, it identifies deviation points and records flow rate characteristics, adjusts the measurement parameters of the unmanned ship, corrects abnormal data points, adjusts the sampling interval in real time according to the flow rate changes, and uses artificial intelligence models to monitor and correct flow rate data to generate corrected flow data;

[0067] Based on the corrected flow data, the unmanned ship control module analyzes the relationship between the corrected flow data and the navigation route, formulates navigation and measurement plans, optimizes the navigation path and speed, adjusts the navigation and measurement methods of the unmanned ship in various water flow environments, and generates optimized navigation parameters.

[0068] Flow characteristic data include the mean value, variation amplitude and flow direction trend of water velocity. The optimized coding strategy includes the dynamic adjustment scheme of coding parameters, periodic change adjustment strategy and instantaneous change response measures. The adjusted signal processing parameters include filter frequency, gain adjustment and signal smoothing parameters. The abnormal flow velocity response results include abnormal data point classification, abnormal cause analysis and real-time response plan. The corrected flow data includes flow velocity correction value, deviation point correction record and optimized sampling interval. The optimized navigation parameters include navigation path optimization, navigation speed adjustment and navigation method.

[0069] See also Figure 3 and Figure 2 , the feature analysis module includes:

[0070] The data processing submodule filters and normalizes the flow data based on the flow data collected by the flow velocity sensor of the unmanned ship, segments the data and marks the abnormal values, fills the missing data, and generates flow velocity change data;

[0071] The data processing submodule is based on the water flow data collected by the flow velocity sensor of the unmanned ship. The low-frequency disturbance in the flow velocity data is eliminated through a high-pass filter to ensure the consistency of data quality. The flow velocity data is normalized using the Z score method, and the value of each data point is converted into a standard score relative to the mean and standard deviation of the entire data set. The abnormal points in the data are identified. Specifically, the data points that differ from the mean by more than two times the standard deviation are marked as abnormal. For missing data points, linear interpolation is used to calculate and fill in the values ​​of adjacent data points to ensure the integrity of the data. The complete data set is divided into segments of fixed length, and each segment is marked by the start and end time points to provide a basis for subsequent flow velocity change analysis and generate flow velocity change data. The formula is:

[0072]

[0073] Among them, F(i) represents the flow rate value after the i-th filtration, v i is the original flow velocity data point, α is the attenuation coefficient, and n is the filter window size.

[0074] The dynamic monitoring submodule analyzes the changing trend of water flow direction based on the flow velocity change data, determines the flow direction of water flow by calculating the vector change of real-time flow velocity, and generates water flow direction data;

[0075] The dynamic monitoring submodule is based on the flow velocity change data. It extracts the velocity vector at each time point from the processed flow velocity change data, and uses the vector difference method to calculate the vector change between two consecutive time points to determine the real-time change trend of the water flow direction. The specific method is to calculate the angle and size between each pair of continuous vectors, describe the dynamic flow direction of the water flow through the vector change, and use the statistical analysis of the angle change to determine the main water flow direction. Whenever the difference between the measured vector angle and the previous one exceeds the preset threshold, it is recorded as a direction change event and generates water flow direction data. The formula is:

[0076]

[0077] in, represents the velocity vector difference between two consecutive measurement points, and are the velocity vectors of two consecutive time points respectively.

[0078] The cross-comparison submodule performs a cross-comparison of the flow velocity data and the water temperature data based on the water flow direction data. By calculating the correlation coefficient between the flow velocity and the water temperature, the key time period is identified, the flow velocity and water temperature change patterns within the time period are analyzed, and the flow characteristic data is generated.

[0079] The cross-comparison submodule collects flow velocity data and water temperature data in the same time period based on the water flow direction data to ensure the synchronization and comparability of the two data. The correlation coefficient analysis method is used to calculate the correlation between flow velocity and water temperature, standardize the flow velocity and water temperature data, and then calculate the ratio of covariance to each variance to determine the degree of linear correlation between the two, identify the key time period with high correlation between flow velocity change and water temperature change, and further analyze the change pattern of flow velocity and water temperature in the key time period, including rising and falling trends, volatility and periodicity, to generate flow characteristic data. The formula is:

[0080]

[0081] Where x and y are flow velocity and water temperature, ρ xy is the correlation coefficient between flow velocity and water temperature, x i and i are the observed values ​​of flow velocity and water temperature at a single time point, and is the average value of flow rate and water temperature.

[0082] See also Figure 4 and Figure 2, the dynamic encoding module includes:

[0083] The periodic analysis submodule analyzes the periodic and instantaneous changes of water flow based on the flow characteristic data, extracts the periodic changes and instantaneous fluctuation characteristics in the water flow data, calculates the statistical indicators of the characteristics, and generates the trend of periodic changes in water flow;

[0084] The periodic analysis submodule analyzes the water flow data based on the flow characteristic data to identify and extract its periodic changes, performs Fourier transform on the data, converts the time series data into frequency domain representation, thereby clearly identifying the main periodic components, evaluating the instantaneous fluctuation characteristics, and using short-time Fourier transform to transform the data in the local time window, which enables effective analysis of the instantaneous changes of non-stable water flow. Based on the analysis, key statistical indicators such as amplitude, frequency, phase and fluctuation intensity are calculated to fully understand the dynamic characteristics of the water flow, and the results of the periodic and instantaneous analysis are combined to generate the trend of water flow periodic changes. The formula is:

[0085]

[0086] Among them, F(ω) represents the complex function in the frequency domain, f(t) is the water flow velocity or water temperature data in the time domain, ω is the frequency transformed to the frequency domain, and t is the time.

[0087] The parameter setting submodule sets the reference value of the initialization code based on the change trend of the water flow cycle, calculates the coding efficiency under the initial conditions, and optimizes the coding parameters according to various water flow conditions to generate the coding initialization parameters;

[0088] The parameter setting submodule sets the reference value of the initialization coding based on the trend of the water flow cycle change, determines the reference coding rate according to the main cycle and amplitude characteristics of the water flow, calculates the coding efficiency under the initial conditions, evaluates the data compression ratio and error rate during the coding process to ensure the optimization of coding efficiency, monitors the data performance under different water flow conditions, such as the impact of shortened cycles or increased amplitudes, and adjusts the coding parameters according to the changes, including the coding window size and step length, to adapt to different water flow dynamics, and generates the coding initialization parameters according to the tuned parameters. The formula is:

[0089] P new =P init ×(1+β(ΔA+ΔT))

[0090] Among them, P new represents the optimized encoding parameters, P init is the initial encoding parameter, β is the tuning coefficient, ΔA and ΔT represent the amplitude and period changes respectively.

[0091] The automatic adjustment submodule extracts the instantaneous change value of the current flow rate based on the encoding initialization parameters and the monitored real-time flow rate changes, and performs difference analysis with the initial encoding parameters to identify the range and direction of encoding parameter adjustment, automatically updates the encoding parameters in real time, and generates an optimized encoding strategy;

[0092] The automatic adjustment submodule monitors and extracts the current flow rate data based on the encoding initialization parameters, especially focusing on the instantaneous change value of the flow rate, performs difference analysis with the initial encoding parameters, compares the deviation between the current flow rate change and the expected encoding efficiency set initially, and identifies the specific range and direction of the encoding parameter adjustment based on the difference analysis results. It adjusts the size of the encoding window or modifies the data compression ratio to adapt to the actual data flow requirements, automatically updates the encoding parameters in real time, and generates an optimized encoding strategy. The formula is:

[0093] P adjust =P current +γ×(V current -V target )

[0094] Among them, P adjust is the adjusted encoding parameter, P current is the current encoding parameter, γ is the adjustment sensitivity coefficient, V current is the instantaneous change value of the flow rate currently monitored, V target It is the target flow rate value set based on the initial encoding parameters.

[0095] See also Figure 5 and Figure 2 , the real-time optimization module includes:

[0096] The filter adjustment submodule adjusts the filter parameters based on the optimized coding strategy, analyzes the noise level and signal strength in the water flow data, detects the noise frequency and amplitude changes in the water flow, and measures the distribution of signal strength at multiple frequencies to generate filter parameter settings;

[0097] The filter adjustment submodule evaluates the noise level and signal strength in the current water flow data based on the optimization coding strategy, and uses the spectrum analysis tool to determine the noise frequency and amplitude change of the water flow data. By analyzing the data, the system detects the signal strength distribution at different frequencies. The key step is to determine the signal-to-noise ratio in order to optimize the performance of the filter. According to the detailed analysis results of the signal strength and noise frequency, the filter parameters are adjusted, including setting the filter cutoff frequency and bandwidth to adapt to the current noise conditions and maintain the integrity of the signal, and generate the filter parameter settings. The formula is:

[0098] F new =F base ·e -α(S / N)

[0099] Among them, F new is the adjusted filter cutoff frequency, F base is the base cutoff frequency, α is the adjustment factor based on the signal-to-noise ratio, and S / N is the signal-to-noise ratio.

[0100] The window adjustment submodule dynamically adjusts the filter window size based on the filter parameter settings, measures the signal processing efficiency under various window sizes, calculates the impact of window size on signal quality, selects the optimal window size by comparing signal quality, and generates window optimization parameters;

[0101] The window adjustment submodule is based on the setting of filtering parameters. By setting filtering windows of different sizes, the efficiency of signal processing is measured for each window size, including a comprehensive evaluation of the signal recovery and noise suppression capabilities. The specific impact of each window size on the signal quality is calculated using statistical methods, mainly by comparing the signal-to-noise ratio (SNR) and distortion. According to the comparison results of signal quality, the performance of each window size is compared, and the optimal window size that can maximize signal processing efficiency while minimizing the impact of noise is selected to generate window optimization parameters. The formula is:

[0102]

[0103] Among them, W opt is the optimal window size, w i is the size of the i-th window, D i is the signal distortion of the ith window, β is the attenuation coefficient, and N is the number of windows considered.

[0104] The condition judgment submodule judges the signal characteristics under various flow rate conditions based on the window optimization parameters, extracts the signal change patterns under various flow rate conditions, and performs comparative analysis of flow rate characteristics, identifies and extracts key signal characteristics, and generates adjusted signal processing parameters;

[0105] The conditional judgment submodule is based on the window optimization parameters. It analyzes the signal characteristics under various flow rate conditions, including the basic statistical description of the signal such as mean, variance, peak value, etc., and then identifies the signal change mode under different flow rates, such as periodic changes, mutations or stability, etc., and conducts comparative analysis of flow rate characteristics. It calculates the similarity and difference of signal characteristics under different flow rate conditions. Through comparison, it can identify the key flow rate characteristics that affect the signal processing effect. According to the key characteristics, it extracts and optimizes the signal processing parameters, adjusts the signal amplification, filtering limit, etc., and generates the adjusted signal processing parameters. The formula is:

[0106]

[0107] Among them, S optis the optimized signal processing parameter, s j is the original signal parameter observed under the jth flow rate condition, C j is the signal characteristic at the jth flow rate, C ref is the signal characteristic at the reference flow rate, λ is the attenuation coefficient, and M is the number of different flow rate conditions considered.

[0108] See also Figure 6 and Figure 2 , the exception response module includes:

[0109] The flow velocity detection submodule builds an artificial intelligence model based on the adjusted signal processing parameters, uses the model to detect and respond to abnormal water flow conditions, monitors water flow velocity in real time, records water flow velocity values ​​during key time periods, uses the model to detect abnormal change trends, and generates water flow monitoring data;

[0110] The velocity detection submodule builds an artificial intelligence model designed specifically for water velocity monitoring based on the adjusted signal processing parameters. The model is trained in combination with historical water flow data and current signal processing parameters to optimize its prediction accuracy. The trained model is used to monitor water velocity in real time, paying special attention to abnormal velocity changes that may occur in the water flow, such as sudden increases or decreases in flow velocity, and recording water velocity values ​​in all key time periods. Key time periods are usually periods when the model predicts high risk or high variability. The model is also used to detect abnormal change trends in water flow, and to identify potential abnormal events and generate water flow monitoring data by analyzing the differences between model output and actual observations. The formula is:

[0111]

[0112] Among them, ΔV represents the trend measurement of abnormal changes, V k is the measured flow velocity value at the kth detection point, V pred,k is the flow rate value predicted by the model, and N is the total number of detection points in the critical time period.

[0113] The data classification submodule determines abnormal water flow data points based on water flow monitoring data and marks the abnormal data. By comparing the monitoring data with the normal flow rate range, the features of abnormal water flow data points are extracted, and they are classified according to the abnormal type to generate abnormal flow rate classification.

[0114] The data classification submodule is based on water flow monitoring data. It determines abnormal water flow data points by setting a predetermined flow rate threshold range. The threshold is carefully set according to the typical behavior of the water flow and environmental conditions. When the flow rate data exceeds the threshold, the data point is marked as abnormal and further analyzed. By comparing the abnormal data point with the normal flow rate range, the characteristics of the abnormal data point are extracted. The characteristics include the maximum deviation of the flow rate, frequency, duration, etc. According to the characteristics, the abnormal type is classified, such as instantaneous surge, continuous low flow rate, etc., to generate abnormal flow rate classification. The formula is:

[0115]

[0116] Among them, E represents the quantitative value of the abnormality degree, V i is the velocity value of each data point marked as abnormal, V norm is the average value of the normal flow rate range, α i is the weight factor assigned according to the severity of the flow rate anomaly, and M is the number of abnormal data points.

[0117] The dynamic adjustment submodule is based on abnormal flow velocity classification. By dynamically adjusting the detection parameters, it analyzes various types of water flow abnormality data. According to the characteristics of various abnormal types, it adjusts the sensitivity and discrimination criteria of the detection parameters, extracts the abnormal flow velocity characteristics, and generates abnormal flow velocity response results.

[0118] The dynamic adjustment submodule is based on the abnormal flow rate classification and comprehensively evaluates various types of water flow abnormality data, including instantaneous surges, continuous low flow rates, etc. According to the specific characteristics of each type of abnormality, such as abnormal duration, change amplitude and frequency of occurrence, the system dynamically adjusts the detection parameters, including the adjustment of sensitivity and discrimination criteria, to improve the ability to identify abnormalities under complex water flow conditions. The adjusted detection parameters are used to more accurately identify and extract abnormal flow velocity characteristics, ensuring that the detection system can adapt to the changing water flow environment and generate abnormal flow velocity response results. The formula is:

[0119]

[0120] Among them, P adj is the adjusted detection parameter, P init is the initial detection parameter, δ j is the increment adjusted for the jth abnormality type, f(C j ) is a function of the anomaly feature adjustment increment, and J is the total number of anomaly types.

[0121] See also Figure 7 and Figure 2 , the data correction module includes:

[0122] The deviation identification submodule compares normal and abnormal data based on the abnormal flow rate response results, identifies the deviation point and records the flow rate characteristics, calculates the difference value, maps the difference value to the time axis, determines the time period range of the deviation point, and generates flow rate deviation data;

[0123] The deviation identification submodule compares the normal flow rate range and the detected abnormal flow rate data based on the abnormal flow rate response results, including statistically analyzing the flow rate average and standard deviation under normal conditions and the flow rate records during abnormal conditions, identifying the deviation points that deviate from the normal range, and recording the flow rate characteristics of the deviation points in detail, such as the peak value, frequency and change trend of the flow rate, calculating the difference between the normal flow rate data and the abnormal data, using absolute value or square difference and other methods to quantify the degree of deviation, mapping the difference value to the time axis, and determining the time period range of the deviation point by associating the deviation value with its corresponding time point, and generating flow rate deviation data. The formula is:

[0124] D(t)=∫|v norm (t)-v ex (t)|dt

[0125] Where D(t) represents the deviation measure at time t, v norm (t) is the normal flow velocity value at time t, v ex (t) is the abnormal flow velocity value.

[0126] The parameter adjustment submodule adjusts the measurement parameters of the unmanned boat based on the flow deviation data, sets the sensitivity and sampling frequency of the measurement equipment in real time according to the changes in the water flow and hydrological environment, analyzes the flow acquisition effect according to the adjusted measurement parameters, and generates optimized measurement parameters;

[0127] The parameter adjustment submodule is based on the velocity deviation data. It analyzes the deviation data to evaluate the measurement requirements under the current water flow and hydrological environment conditions, and identifies whether there are measurement errors or deficiencies caused by environmental changes. Based on the analysis, the measurement equipment settings of the unmanned boat are adjusted, including increasing or decreasing the sensitivity of the equipment and adjusting the sampling frequency to adapt to the current water flow characteristics and dynamic changes. Through experiments and simulation tests, the impact of the adjusted measurement parameters on the flow velocity collection effect is evaluated to ensure the accuracy and real-time nature of the measurement data. Based on the adjustment and evaluation results, the optimized measurement parameters are generated. The formula is:

[0128]

[0129] Among them, P new is the adjusted measurement parameter, P init is the initial measurement parameter, k is the adjustment coefficient, D i is the ith deviation value, is the average of the deviation values ​​and n is the number of deviation values.

[0130] The artificial intelligence correction submodule uses artificial intelligence models to monitor and correct flow rate data based on optimized measurement parameters, compares the real-time collected flow rate data with reference data, uses the Kalman filter algorithm, and iteratively adjusts the model parameters to correct errors in the flow rate data and generate corrected flow data;

[0131] The Kalman filter algorithm follows the formula:

[0132] x k =x k-1 +K k (z k -Hx k-1 )+λΔTσ 2

[0133] Among them, x k is the corrected flow rate value, x k-1 is the flow velocity value at the previous moment, K k is the Kalman gain, z k is the measured velocity value at the current moment, H is the observation model, λ is the adjustment coefficient, ΔT is the sampling time interval, σ 2 is the variance of the flow rate data.

[0134] The execution flow is as follows:

[0135] Get the current measured flow rate value z from the flow rate sensor k , and get the flow velocity value x at the previous moment k-1 The observation model H converts the velocity value at the previous moment into a predicted value with the same unit as the measured value, and calculates the Kalman gain K k , add the adjustment coefficient λ, dynamically adjust the system, and enhance the adaptability of the model to rapid changes. ΔT represents the time between two measurements, and the correction term λΔTσ is added 2 , where σ 2 Represents the variance of the flow rate data, generating the corrected flow rate value x k .

[0136] See also Figure 8 and Figure 2 , the unmanned ship control module includes:

[0137] The path analysis submodule analyzes the relationship between the corrected flow data and the navigation route based on the corrected flow data, calculates the flow changes under various paths, explores the optimal navigation route, and generates path optimization data;

[0138] Based on the corrected flow data, the path analysis submodule integrates the flow data and the current navigation route information of the unmanned ship, evaluates the impact of flow data on navigation efficiency, calculates and compares the flow changes under multiple preset navigation paths, analyzes the flow velocity characteristics of each path and its potential impact on navigation time and resource consumption, and uses data analysis methods to mine the optimal navigation route under the current hydrological and flow conditions. The route will be the path with the smallest flow change and the most stable under the premise of ensuring safety and efficiency, generate path optimization data, and optimize the decision support of the navigation system. The formula is:

[0139]

[0140] Among them, R opt is the optimal navigation route, R k is the kth navigation path considered, w j is the weight factor assigned according to the importance of the flight segment, ΔF j,k is on path R k The flow change of the jth segment on the path R k number of segments on.

[0141] The navigation optimization submodule analyzes the key nodes and turning points in the path optimization data based on the path optimization data, adjusts the navigation speed, steering angle and acceleration during the navigation process, optimizes the navigation path and speed, and generates navigation control parameters;

[0142] The navigation optimization submodule is based on the path optimization data and analyzes the key nodes and turning points in the path data in detail. The nodes and turning points are places where special attention needs to be paid to adjusting the course or speed during navigation. According to the location and nature of the key points, the speed, steering angle and acceleration during navigation are adjusted to adapt to various water flow conditions and path characteristics. Through simulation and real-time data feedback, the navigation path and speed setting are optimized to ensure that the unmanned ship can minimize energy consumption and time consumption while maintaining efficient navigation. Based on optimization and adjustment, accurate navigation control parameters are generated. The formula is:

[0143]

[0144]

[0145] Among them, V new is the adjusted sailing speed, V init is the initial velocity, γ i is the speed adjustment coefficient based on the i-th node, θ i is the turning angle of the path at the node, n is the total number of key nodes, θ new is the adjusted steering angle, θ init is the initial steering angle, δ iis the steering adjustment factor, φ i It is the adjustment value corresponding to the steering angle.

[0146] The task adjustment submodule adjusts the navigation and measurement methods of the unmanned ship in various water flow environments based on the navigation control parameters, analyzes the real-time changes of the water flow environment, dynamically adjusts the navigation path and speed of the unmanned ship by updating the navigation tasks and measurement plans in real time, and generates optimized navigation parameters;

[0147] The task adjustment submodule evaluates the real-time changes of the current water flow environment, including water flow speed, direction and potential obstacles, based on the navigation control parameters. According to the real-time data, the navigation path and speed of the unmanned ship are dynamically adjusted to adapt to the changing water flow conditions and optimize the measurement effect. The navigation tasks and measurement plans are updated in real time to ensure that the unmanned ship can collect data and navigate the path under the best conditions. Through adjustment and update, the optimized navigation parameters are generated. The formula is:

[0148]

[0149] Among them, P new is the adjusted navigation parameter, P current is the current navigation parameter, κ is the adjustment coefficient, λ j is the weight factor assigned according to the change of water flow, ΔE j is the change value of the jth environmental factor, and m is the total number of environmental factors considered.

[0150] See also Fig. 9 A method for measuring water flow of an unmanned ship or boat based on artificial intelligence is provided. The method for measuring water flow of an unmanned ship or boat based on artificial intelligence is performed based on the water flow measurement system of the unmanned ship or boat based on artificial intelligence, and comprises the following steps:

[0151] S1: Based on the water flow data collected by the flow velocity sensor of the unmanned boat, the average flow velocity and the maximum flow velocity are analyzed. By performing segmented statistics on the flow velocity data in the continuous time window, the flow velocity change trend is extracted. The flow direction change analysis is used to quantitatively describe the dynamic change of the water flow direction and obtain the flow characteristic data.

[0152] S2: Based on the flow characteristic data, the periodicity and instantaneous changes of the water flow are analyzed through periodic detection technology, and the key frequency characteristics of the flow velocity are identified using frequency analysis tools. The compression ratio and data resolution of the data encoding are adjusted according to the real-time monitored flow velocity fluctuations to generate an optimized encoding strategy;

[0153] S3: Based on the optimized coding strategy, the filter parameters are adjusted, the noise-to-signal ratio of the velocity data is compared in real time, the filter design is optimized to match the frequency characteristics of the data, the data transmission frequency is adjusted to match the real-time changes of the signal, and the adjusted signal processing parameters are generated;

[0154] S4: Based on the adjusted signal processing parameters, use anomaly detection technology to detect and respond to abnormal water flow conditions, monitor water flow velocity, identify abnormal data points through real-time data analysis, classify abnormal data and mark abnormal categories, analyze various types of abnormal water flow data by adjusting data sampling frequency and detection sensitivity, and generate abnormal flow velocity response results;

[0155] S5: Based on the abnormal flow velocity response results, compare normal and abnormal data, identify deviation points and record flow velocity characteristics through difference analysis, use data correction process to correct flow velocity measurement errors, compare the dependency between flow data and navigation routes through correlation analysis, optimize navigation path and speed, and generate optimized navigation parameters.

[0156] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A water flow measurement system for unmanned ships and boats based on artificial intelligence, characterized in that: The system comprises: The feature analysis module analyzes key flow velocity indicators based on the water flow data collected by the flow velocity sensor of the unmanned ship, extracts the change trend of the flow velocity, determines the dynamic change of the water flow direction, and generates flow characteristic data; The feature analysis module comprises: The data processing submodule filters and normalizes the flow data based on the flow data collected by the flow velocity sensor of the unmanned ship, segments the data and marks the abnormal values, fills the missing data, and generates flow velocity change data; The dynamic monitoring submodule analyzes the change trend of the water flow direction based on the flow velocity change data, determines the flow direction of the water flow by calculating the vector change of the real-time flow velocity, and generates water flow direction data; The cross comparison submodule performs a cross comparison of the flow velocity data and the water temperature data based on the water flow direction data, identifies the key time period by calculating the correlation coefficient between the flow velocity and the water temperature, analyzes the flow velocity and water temperature change pattern within the time period, and generates flow characteristic data; The dynamic coding module analyzes the periodicity and instantaneous changes of the water flow based on the flow characteristic data, automatically adjusts the coding parameters according to the real-time flow velocity changes, and generates an optimized coding strategy; The real-time optimization module adjusts the filter parameters based on the optimization coding strategy and various water flow conditions, optimizes the signal processing flow, and generates adjusted signal processing parameters; The abnormal response module constructs an artificial intelligence model based on the adjusted signal processing parameters, detects and responds to abnormal water flow conditions, and generates abnormal flow rate response results by dynamically adjusting the detection parameters; The data correction module compares normal and abnormal data based on the abnormal flow rate response result, identifies deviation points and records flow rate characteristics, uses a model to monitor and correct flow rate data, and generates corrected flow data; The unmanned ship control module analyzes the relationship between the flow data and the navigation route based on the corrected flow data, formulates navigation and measurement plans, adjusts the navigation and measurement methods of the unmanned ship in the water flow environment, and generates optimized navigation parameters; The real-time optimization module includes: The filter adjustment submodule adjusts the filter parameters based on the optimized coding strategy, analyzes the noise level and signal strength in the water flow data, detects the noise frequency and amplitude changes in the water flow, and measures the distribution of signal strength at multiple frequencies to generate filter parameter settings; The window adjustment submodule dynamically adjusts the filter window size based on the filter parameter setting, measures the signal processing efficiency under various window sizes, calculates the impact of the window size on the signal quality, selects the optimal window size by comparing the signal quality, and generates the window optimization parameters; The condition judgment submodule judges the signal characteristics under various flow rate conditions based on the window optimization parameters, extracts the signal change patterns under various flow rate conditions, performs flow rate characteristic comparison analysis, identifies and extracts key signal characteristics, and generates adjusted signal processing parameters; The abnormal response module includes: The flow velocity detection submodule builds an artificial intelligence model based on the adjusted signal processing parameters, uses the model to detect and respond to abnormal water flow conditions, monitors water flow velocity in real time, records water flow velocity values ​​during key time periods, uses the model to detect abnormal change trends, and generates water flow monitoring data; The data classification submodule determines abnormal water flow data points based on the water flow monitoring data and marks the abnormal data, extracts the features of the abnormal water flow data points by comparing the monitoring data with the normal flow rate range, and classifies them according to the abnormal type to generate abnormal flow rate classification; The dynamic adjustment submodule analyzes various types of water flow abnormality data by dynamically adjusting detection parameters based on the abnormal flow velocity classification, adjusts the sensitivity and discrimination criteria of the detection parameters according to the characteristics of various abnormal types, extracts abnormal flow velocity characteristics, and generates abnormal flow velocity response results.

2. The water flow measurement system for unmanned ships and boats based on artificial intelligence according to claim 1 is characterized by: The flow characteristic data include the mean value, variation amplitude and flow direction trend of water flow velocity; the optimized coding strategy includes a dynamic adjustment scheme of coding parameters, a periodic change adjustment strategy and instantaneous change response measures; the adjusted signal processing parameters include filter frequency, gain adjustment and signal smoothing parameters; the abnormal flow velocity response results include abnormal data point classification, abnormal cause analysis and real-time response scheme; the corrected flow data include flow velocity correction value, deviation point correction record and optimized sampling interval; the optimized navigation parameters include navigation path optimization, navigation speed adjustment and navigation method.

3. The water flow measurement system for unmanned ships and boats based on artificial intelligence according to claim 1 is characterized in that: The dynamic encoding module comprises: The periodic analysis submodule analyzes the periodicity and instantaneous changes of the water flow based on the flow characteristic data, extracts the periodicity and instantaneous fluctuation characteristics in the water flow data, calculates the statistical indicators of the characteristics, and generates the periodic change trend of the water flow; The parameter setting submodule sets the reference value of the initialization coding based on the water flow periodic variation trend, calculates the coding efficiency under the initial conditions, and optimizes the coding parameters according to various water flow conditions to generate coding initialization parameters; The automatic adjustment submodule is based on the encoding initialization parameters and the monitored real-time flow rate changes, extracts the instantaneous change value of the current flow rate, performs difference analysis with the initial encoding parameters, identifies the range and direction of the encoding parameter adjustment, automatically updates the encoding parameters in real time, and generates an optimized encoding strategy.

4. The water flow measurement system for unmanned ships and boats based on artificial intelligence according to claim 1 is characterized in that: The data correction module comprises: The deviation identification submodule compares normal and abnormal data based on the abnormal flow rate response result, identifies the deviation point and records the flow rate characteristics, calculates the difference value, maps the difference value to the time axis, determines the time period range of the deviation point, and generates flow rate deviation data; The parameter adjustment submodule adjusts the measurement parameters of the unmanned boat based on the flow deviation data, sets the sensitivity and sampling frequency of the measurement equipment in real time according to the changes in the water flow and the hydrological environment, analyzes the flow acquisition effect according to the adjusted measurement parameters, and generates optimized measurement parameters; The artificial intelligence correction submodule is based on the optimized measurement parameters and uses the artificial intelligence model to monitor and correct the flow rate data, compares the real-time collected flow rate data with the reference data, adopts the Kalman filtering algorithm, and iteratively adjusts the model parameters to correct the errors in the flow rate data and generate corrected flow data.

5. The water flow measurement system for unmanned ships and boats based on artificial intelligence according to claim 1 is characterized in that: The unmanned ship control module includes: The path analysis submodule analyzes the correlation between the corrected flow data and the navigation route based on the corrected flow data, calculates the flow changes under various paths, mines the optimal navigation route, and generates path optimization data; The navigation optimization submodule analyzes the key nodes and turning points in the path data based on the path optimization data, adjusts the navigation speed, steering angle and acceleration during the navigation process, optimizes the navigation path and speed, and generates navigation control parameters; The task adjustment submodule adjusts the navigation and measurement methods of the unmanned ship in various water flow environments based on the navigation control parameters, analyzes the real-time changes of the water flow environment, dynamically adjusts the navigation path and speed of the unmanned ship by updating the navigation tasks and measurement plans in real time, and generates optimized navigation parameters.

6. A method for measuring water flow of unmanned ships and boats based on artificial intelligence, characterized in that: The water flow measurement system for unmanned ships and boats based on artificial intelligence according to any one of claims 1 to 5 comprises the following steps: Based on the water flow data collected by the flow velocity sensor of the unmanned boat, the average flow velocity and the maximum flow velocity are analyzed, the flow velocity change trend is extracted, the dynamic change of the water flow direction is judged, and the flow characteristic data is obtained; Based on the flow characteristic data, the periodicity and instantaneous changes of the water flow are analyzed, the flow velocity frequency characteristics are identified, the encoding parameters are adjusted according to the real-time flow velocity changes, and an optimized encoding strategy is generated; Based on the optimized coding strategy, filter parameters are adjusted, the signal processing flow is optimized, and adjusted signal processing parameters are generated; Based on the adjusted signal processing parameters, detect and respond to abnormal water flow conditions, monitor water flow velocity, determine abnormal data points, classify and mark abnormal data, and generate abnormal flow velocity response results by dynamically adjusting detection parameters and analyzing various types of abnormal water flow data; Based on the abnormal flow velocity response results, normal and abnormal data are compared, deviation points are identified and flow velocity characteristics are recorded, and the flow velocity data is monitored and corrected using an artificial intelligence model. Based on the corrected flow data, the relationship between the flow data and the navigation route is analyzed, the navigation path and speed are optimized, and optimized navigation parameters are generated.

Images