Waterway ship flow real-time analysis method and system based on big data

Abstract

The application discloses a waterway ship flow real-time analysis method and system based on big data, and relates to the technical field of data analysis, comprising the following steps: according to the relationship between ship speed and flow, combining the maximum information coefficient algorithm, constructing the mapping relationship of ship flow, and calculating the prediction step to predict the ship flow of the water surface to be analyzed. The application can capture the dependent relationship between the ship speed and the flow of any function type, such as nonlinearity, non-monotonicity and time variation, by replacing the linear correlation coefficient or the empirical threshold with the maximum information coefficient, automatically calculate the optimal prediction step through the MIC significance test, shorten the step and improve the sensitivity when the ship density suddenly changes, such as before and after holidays and the tidal window, establish the fast mapping relationship between the effective parameters and the ship flow, improve the acquisition speed of the ship flow, improve the data targeting and save the storage cost.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a method and system for real-time analysis of ship traffic flow in waterways based on big data. Background Technology

[0002] In recent years, real-time ship traffic analysis technology has become increasingly sophisticated, characterized by multi-source data, deep learning models, and refined business scenarios. Low-orbit satellite AIS, VTS radar, CCTV, high-definition infrared, drones, hydro-meteorology, and tidal models are being accessed simultaneously to form daily maritime big data. This discretizes the trajectory of a single ship into micro-features such as water speed, rate of change of heading angle, CPA / TCPA, draft, and tidal range ratio. These features are then overlaid with semantic variables such as port operation plans, berth scheduling, and pilot shifts, significantly expanding the feature dimensions of the analysis.

[0003] Currently, Chinese invention patent CN119624015A discloses a scheduling method and system based on a ship traffic flow prediction model. This method determines the state of a ship queue by performing traffic analysis on the queue under time and queue length factors in the waiting area; it performs dynamic and static data statistics on all ships in the waiting area to obtain the dynamic and static attributes of each ship; it divides the ship queue in the waiting area into intervals for ship traffic flow to obtain initial ship intervals; it performs target optimization scheduling on the ship traffic flow in the intervals to obtain the final interval ship traffic flow; and it generates a ship traffic flow scheduling plan based on a multi-stage double-line lock. However, this related technology does not select similar waters based on relevant information of the current water area, nor does it construct a rapid mapping relationship between traffic flow and effective parameters based on the analysis results of effective historical data of similar waters. This is not conducive to the speed of ship traffic flow acquisition, nor does it quickly obtain relatively accurate ship traffic flow data based on the periodic repetition pattern of ship traffic flow, which is not conducive to the simplicity of ship traffic flow acquisition. It cannot meet the requirements of economy, speed, and convenience for traffic monitoring in large-scale scenarios, and has certain limitations. Summary of the Invention

[0004] The technical problem solved by this invention is that in related technologies, there is no screening of similar waters based on relevant information of the current water area, and no construction of a rapid mapping relationship between flow and effective parameters based on the analysis results of effective historical data of similar waters. This is not conducive to the speed of obtaining ship flow, and there is no rapid and accurate ship flow data obtained based on the periodic repetition pattern of ship flow. This is not conducive to the ease of obtaining ship flow, and it cannot meet the requirements of economy, speed and convenience for large-scale scene flow monitoring, and has certain limitations.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution. In the first aspect, a real-time analysis method for ship flow in waterways based on big data includes the following steps. The real-time analysis method for ship flow in waterways based on big data is characterized by including the following steps: Step S100, based on the relationship between ship speed and flow, combined with the maximum information coefficient algorithm, constructing a mapping relationship for ship flow and calculating the prediction step size.

[0006] The maximum information coefficient algorithm is used to calculate the correlation. Based on the comparison between the correlation and the correlation threshold, the corresponding historical acquisition time is selected to be retained or deleted. Based on the retained historical acquisition time, the average time difference between each group of adjacent historical acquisition time is calculated. The first product is calculated based on the average time difference, and the fourth ratio between the historical acquisition time interval and the average time difference is calculated. The first quotient of the average time difference and the first product is calculated, and the second product of the fourth ratio and the first quotient is calculated. The average value of the second product is set as the prediction step size.

[0007] Step S200: Based on the prediction step size, obtain historical data of the water surface to be analyzed, and predict the ship flow of the water surface to be analyzed based on the mapping relationship and the historical data of the water surface to be analyzed.

[0008] As a preferred embodiment of the real-time analysis method for waterway vessel flow based on big data described in this invention, step S100 includes the following sub-steps: step S101, obtaining the water surface related data to be analyzed, and matching the corresponding historical water surface related data according to the water surface related data to be analyzed.

[0009] Step S102: Substitute the historical water surface data into the expression for the relationship between ship speed and flow rate to obtain the difference between the historical speed and flow rate values;

[0010] Step S103: Using a preset vector construction method, construct a first vector corresponding to any historical water surface data, and construct a second vector corresponding to the historical velocity difference and the historical flow difference.

[0011] Step S104: Calculate the correlation between the first vector and the second vector using the maximum information coefficient algorithm, set the first value as the correlation threshold, and calculate the prediction step size based on the comparison between the correlation and the correlation threshold.

[0012] As a preferred embodiment of the real-time analysis method for waterway vessel flow based on big data described in this invention, the water surface data to be analyzed includes water surface length, water surface width, water depth, water density, water viscosity coefficient, water flow velocity, navigable water area type, average water wave Reynolds number, wind force, and wind speed.

[0013] The length of the water surface is represented as the length of the projection shape of the curved surface corresponding to the water surface to be analyzed on the first plane, which is processed into a rectangle. The width of the water surface is represented as the width of the projection shape of the curved surface to be analyzed on the first plane, which is processed into a rectangle. The water density, water flow velocity, wind force and wind speed are obtained by monitoring through corresponding sensors.

[0014] The first plane satisfies the second constraint condition, which is expressed as follows: the area of ​​the projected shape corresponding to the water surface to be analyzed is the largest. The projection is expressed as the straight line segment connecting any point in the water surface to the corresponding projection point is parallel to the normal of the first plane.

[0015] As a preferred embodiment of the real-time analysis method for waterway vessel flow based on big data described in this invention, the historical water surface related data includes historical water surface length, historical water surface width, historical water depth, historical water viscosity coefficient, historical average water wave Reynolds number, historical vessel flow, historical vessel speed, historical vessel density, historical blockage density, historical vessel weight, historical wind speed, and historical wind pressure.

[0016] As a preferred embodiment of the real-time analysis method for waterway vessel traffic flow based on big data described in this invention, the method for matching historical water surface related data includes: obtaining the water surface length, water surface width, water depth, and navigable water type of the water surface to be analyzed; constructing length intervals, width intervals, and depth intervals based on the water surface length, water surface width, and water depth; retrieving the waterway monitoring database; inputting the length interval, width interval, depth interval, and navigable water type into the waterway monitoring database; and matching historical water surface related data whose length, width, and depth are all distributed within the length interval, width interval, and depth interval, and whose navigable water type is the same as the navigable water type of the water surface to be analyzed.

[0017] As a preferred embodiment of the real-time analysis method for waterway vessel flow based on big data described in this invention, the method for calculating the difference between historical speed and flow rate includes: calculating the average historical water viscosity coefficient and the average historical average water wave Reynolds number; obtaining a free state lookup table; inputting the average historical water viscosity coefficient, the average historical average water wave Reynolds number, the historical water surface length, the historical water surface width, and the historical water depth into the free state lookup table; and obtaining the vessel flow rate, vessel speed, and maximum vessel density in the free flow state corresponding to the average historical water viscosity coefficient, the average historical average water wave Reynolds number, the historical water surface length, the historical water surface width, and the historical water depth, respectively denoted as the vessel flow rate, vessel speed, and maximum vessel density in the historical free flow state.

[0018] Substituting the ship flow rate, ship speed, maximum ship density, historical ship flow rate, historical ship speed, historical ship density, and historical blockage density under historical free-flow conditions into the expression for the relationship between ship speed and flow rate, we obtain the historical speed and flow rate relationship values ​​corresponding to each historical data collection time.

[0019] Calculate the difference between the historical velocity and flow rate values ​​at adjacent historical data collection times, and record it as the third difference. Iterate through the third differences between the historical velocity and flow rate values ​​at each group of adjacent historical data collection times.

[0020] Construct a mapping relationship between the third difference and the corresponding historical ship traffic, historical wind speed, and historical wind pressure.

[0021] As a preferred embodiment of the real-time analysis method for waterway vessel flow based on big data described in this invention, the preset vector construction method, namely the method for constructing the first vector, is expressed as obtaining the historical vessel flow, historical vessel speed, historical vessel density, historical blockage density, historical vessel weight, historical wind speed, and historical wind pressure corresponding to any historical water surface.

[0022] Using historical ship traffic as the independent variable and historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure as dependent variables, a multiple regression analysis was conducted to obtain a multiple regression equation. The coefficients corresponding to historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure were extracted and processed.

[0023] Each character of the processed coefficient is represented using 4-bit binary encoding;

[0024] According to the order of each character in the processed coefficient, the characters of the coefficient after binary encoding are combined to obtain the first binary encoding combination;

[0025] The first binary code combination is converted into a vector, and the vector converted from the first binary code combination is set as the first vector;

[0026] The method for constructing a second vector based on the third difference includes obtaining the third difference, processing the third difference, and representing each character of the processed third difference using 4-bit binary encoding.

[0027] According to the character order of the processed third difference, the characters of the third difference after binary encoding are combined to obtain the second binary encoding combination;

[0028] The second binary code combination is converted into a vector, and the vector converted from the second binary code combination is set as the second vector.

[0029] As a preferred embodiment of the real-time analysis method for waterway vessel flow based on big data described in this invention, the correlation between the first vector and the second vector at any historical data collection time is calculated using the maximum information coefficient algorithm.

[0030] Set the first value as the relevance threshold, and calculate the prediction step size based on the comparison between the relevance and the relevance threshold.

[0031] The results of comparing relevance with the relevance threshold include relevance greater than or equal to the relevance threshold, and relevance less than the relevance threshold.

[0032] When the relevance is greater than or equal to the relevance threshold, the corresponding historical data collection time will be retained.

[0033] When the relevance is less than the relevance threshold, the corresponding historical data collection time will be deleted.

[0034] Obtain the retained historical acquisition times, calculate the time difference between adjacent historical acquisition times in the retained historical acquisition times, traverse each group of adjacent historical acquisition times, obtain the time difference corresponding to each group of adjacent historical acquisition times, calculate the average value of the time difference corresponding to adjacent historical acquisition times, and record it as the average time difference.

[0035] Based on the average time difference, the prediction step size is calculated. The calculation method for the prediction step size includes calculating the water surface length, water surface width, and water depth, and the first, second, and third ratios of these ratios with the historical water surface length, historical water surface width, and historical water depth, respectively; calculating the first product of the first, second, and third ratios; obtaining the historical data collection time interval for this historical water surface data; and calculating the fourth ratio of the historical data collection time interval to the average time difference.

[0036] Calculate the first quotient of the product of the average time difference and the first product, and calculate the second product of the fourth ratio and the first quotient;

[0037] Iterate through the second product corresponding to each matching historical water surface data, calculate the average of the second product, and set the average of the second product as the prediction step size.

[0038] As a preferred embodiment of the real-time analysis method for waterway vessel flow based on big data described in this invention, the method involves obtaining the current time point, calculating the difference between the current time point and the predicted step size, recording it as a historical time point, and obtaining historical data for the historical time point.

[0039] Historical data includes the third difference, historical wind speed, and historical wind pressure;

[0040] By inputting the third difference, historical wind speed, and historical wind pressure into the mapping relationship, the historical ship traffic flow of the water area to be analyzed at the historical time point can be obtained;

[0041] Set the historical vessel flow rate of the water surface to be analyzed at a historical time point as the vessel flow rate after the predicted step size of the water surface to be analyzed.

[0042] Secondly, a real-time analysis system for waterway vessel traffic flow based on big data, including a construction module and a prediction module;

[0043] The construction module constructs a mapping relationship between ship speed and flow rate based on the relationship between ship speed and flow rate, combined with the maximum information coefficient algorithm, and calculates the prediction step size.

[0044] The prediction module obtains historical data of the water surface to be analyzed based on the prediction step size, and predicts the ship traffic flow of the water surface to be analyzed based on the mapping relationship and the historical data of the water surface to be analyzed.

[0045] The beneficial effects of this invention are as follows: By using the maximum information coefficient instead of the linear correlation coefficient or empirical threshold, it can capture any functional dependence between ship speed and flow, such as nonlinear, non-monotonic, and time-varying ones. The optimal prediction step size is automatically calculated through the MIC significance test. When there are sudden changes in ship density, the step size is shortened to improve sensitivity, such as before and after holidays and during tidal windows. A rapid mapping relationship between effective parameters and ship flow is established. While improving the speed of ship flow acquisition, the targeting of data is improved and storage costs are saved. Attached Figure Description

[0046] Figure 1 This is a basic flowchart illustrating a real-time analysis method for waterway vessel flow based on big data, provided as an embodiment of the present invention. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0048] It should be understood that the step numbers used herein are for ease of description only and are not intended to limit the order in which the steps are performed. It should also be understood that the terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention.

[0049] Unless the context clearly indicates otherwise, the singular forms “one,” “a,” and “the” are intended to include the plural forms.

[0050] The terms "comprising" and "including" indicate the presence of the described feature, whole, step, operation, element, and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or collections thereof.

[0051] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0052] The real-time analysis method for ship traffic flow in waterways based on big data provided in this application is specifically illustrated through the following embodiments. First, the real-time analysis method for ship traffic flow in waterways based on big data in this application embodiment is described.

[0053] The real-time analysis method for waterway vessel traffic flow based on big data provided in this application relates to the field of data analysis. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the real-time analysis method for waterway vessel traffic flow based on big data, but is not limited to the above forms.

[0054] This application can also be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0055] Example, refer to Figure 1As an embodiment of the present invention, a real-time analysis method for ship flow in waterways based on big data is provided, including the following steps: Step S100, based on the relationship between ship speed and flow, combined with the maximum information coefficient algorithm, a mapping relationship for ship flow is constructed, and the prediction step size is calculated.

[0056] The maximum information coefficient algorithm is used to calculate the correlation. Based on the comparison between the correlation and the correlation threshold, the corresponding historical acquisition time is selected to be retained or deleted. Based on the retained historical acquisition time, the average time difference between each group of adjacent historical acquisition time is calculated. The first product is calculated based on the average time difference, and the fourth ratio between the historical acquisition time interval and the average time difference is calculated. The first quotient of the average time difference and the first product is calculated, and the second product of the fourth ratio and the first quotient is calculated. The average value of the second product is set as the prediction step size.

[0057] Step S200: Based on the prediction step size, obtain historical data of the water surface to be analyzed, and predict the ship flow of the water surface to be analyzed based on the mapping relationship and the historical data of the water surface to be analyzed.

[0058] More preferably, the present invention uses the maximum information coefficient instead of the linear correlation coefficient or empirical threshold, which can capture any functional dependence between ship speed and flow, such as nonlinear, non-monotonic, and time-varying. The optimal prediction step size is automatically calculated through the MIC significance test. The step size is shortened to improve sensitivity when there are sudden changes in ship density, such as before and after holidays and during tidal windows. The effective parameters and ship flow are established to quickly map the relationship, which improves the speed of ship flow acquisition, enhances the targeting of data and saves storage costs.

[0059] More preferably, the expression for the relationship between ship speed and flow rate is:

[0060] ;

[0061] Where q is the ship flow rate, q f Let q be the ship flow rate under free-flow conditions. max Let v(q) be the maximum ship flow in the channel, and v(q) be the average ship speed under the ship flow. In other words, v(q) represents the relationship between ship speed and flow. f Let ρ be the ship's speed in free flow, and ρ be the ship's density. f ρ represents the maximum ship density in free flow. j Let k be the blockage density, k be the wave velocity coefficient, i.e., k is a constant, and k is distributed in the double-closed interval from 0 to 1.

[0062] More preferably, the vessel flow rate is expressed as the number of vessels counted on the water surface area per unit hour.

[0063] Free flow is defined as water flow with zero air resistance.

[0064] The ship flow rate under free flow conditions is expressed as the number of ships counted on the water surface area per unit hour under free flow conditions.

[0065] Free-flow velocity is expressed as the maximum average velocity per hour of free-flowing water under ship flow conditions and ship movement conditions.

[0066] Ship density is expressed as the average ratio of the water surface area to the number of ships on the water per unit hour;

[0067] The maximum ship density in free flow is expressed as the ratio of the maximum number of ships that can be accommodated on the water surface per unit hour to the water surface area under free flow velocity conditions, where the maximum number satisfies the first constraint condition, which means that no two ships collide or rear-end each other.

[0068] Congestion density is expressed as the average of the ratios of the number of ships staying on the water surface for more than a first hour per unit hour under the ship flow conditions. The ratio of the number of ships staying on the water surface for more than a first hour to the water surface area is set as the congestion density.

[0069] More preferably, the ship flow model is constructed based on wave theory, queuing theory, and traffic flow characteristics. The ship flow model construction technology is existing technology and will not be elaborated here. The ship flow model is used as the technical background.

[0070] Step S100 includes the following sub-steps: Step S101, obtain the water surface related data to be analyzed, and match the corresponding historical water surface related data according to the water surface related data to be analyzed.

[0071] Step S102: Substitute the historical water surface data into the expression for the relationship between ship speed and flow rate to obtain the difference between the historical speed and flow rate values;

[0072] Step S103: Using a preset vector construction method, construct a first vector corresponding to any historical water surface data, and construct a second vector corresponding to the historical velocity difference and the historical flow difference.

[0073] Step S104: Calculate the correlation between the first vector and the second vector using the maximum information coefficient algorithm, set the first value as the correlation threshold, and calculate the prediction step size based on the comparison between the correlation and the correlation threshold.

[0074] The water surface data to be analyzed include water surface length, water surface width, water depth, water density, water viscosity coefficient, water flow velocity, navigable water area type, average water wave Reynolds number, wind force, and wind speed.

[0075] The length of the water surface is represented as the length of the projection shape of the curved surface corresponding to the water surface to be analyzed on the first plane, which is processed into a rectangle. The width of the water surface is represented as the width of the projection shape of the curved surface to be analyzed on the first plane, which is processed into a rectangle. The water density, water flow velocity, wind force and wind speed are obtained by monitoring the corresponding sensors, and the water density, water flow velocity, wind force and wind speed are the time intervals between the acquisition.

[0076] The first plane satisfies the second constraint condition, which is expressed as follows: the area of ​​the projected shape corresponding to the water surface to be analyzed is the largest. The projection is expressed as the straight line segment connecting any point in the water surface to the corresponding projection point is parallel to the normal of the first plane.

[0077] More preferably, for the sensors corresponding to water density, water flow velocity, water viscosity coefficient, wind force, and wind speed, a monitoring method is configured. The monitoring method includes setting the monitoring time interval to a second duration, and collecting the corresponding data every second duration. The second duration is dynamically updated based on the rate of change between the collected data at the current collection time and the collected data at the previous collection time.

[0078] The rate of change between the data collected at the current acquisition time and the data collected at the previous acquisition time is expressed as follows: calculate the difference between the data collected at the current acquisition time and the data collected at the previous acquisition time, and denote it as the first difference. Calculate the ratio of the first difference to the data collected at the previous acquisition time, and set the ratio of the first difference to the data collected at the previous acquisition time as the rate of change between the data collected at the current acquisition time and the data collected at the previous acquisition time.

[0079] Set the second value as the rate of change threshold, compare the rate of change with the rate of change threshold, and when the rate of change is less than or equal to the rate of change threshold, do not update the second duration. When the rate of change is greater than the rate of change threshold, use the third duration as the change amount, calculate the difference between the second duration and the third duration, record it as the second difference, and update the second difference as the new monitoring time interval.

[0080] When using the new monitoring time interval as the next updated monitoring time interval, the minuend is calculated by subtracting it from the third duration.

[0081] Among them, the dynamic update satisfies the third constraint condition, and the monitoring time interval after the update is greater than or equal to the fourth duration.

[0082] More preferably, the types of navigable waterways include rivers, lakes, reservoirs, canals, and specific sea areas.

[0083] A calculation method is configured for the average water wave Reynolds coefficient. The calculation method includes: monitoring wave height, wave period, and wave number based on wave sensors; obtaining the wave amplitude of the Airy wave based on the wave height; obtaining the circular frequency of the Airy wave based on the wave period; obtaining the expression for the Airy wave; substituting the wave amplitude, circular frequency, wave number, and wave period into the expression for the Airy wave; plotting a cosine wave image based on the expression for the Airy wave after substituting the wave amplitude, circular frequency, wave number, and wave period; calculating the first distance between any pair of adjacent wave crests and troughs in the cosine wave image at any acquisition time; and setting the first distance as the wavelength at that acquisition time.

[0084] Retrieve the formula for calculating the Reynolds coefficient of water waves, select any sampling time, obtain the water density, water flow velocity, water viscosity coefficient, and wavelength corresponding to that sampling time, input the water density, water flow velocity, water viscosity coefficient, and wavelength corresponding to that sampling time into the formula for calculating the Reynolds coefficient of water waves, and obtain the Reynolds coefficient of water waves corresponding to the water density, water flow velocity, water viscosity coefficient, and wavelength corresponding to that sampling time;

[0085] By iterating through each acquisition time, the water wave Reynolds coefficient corresponding to each acquisition time is obtained;

[0086] Calculate the average value of the water wave Reynolds coefficient and set the average value of the water wave Reynolds coefficient as the average water wave Reynolds number.

[0087] More preferably, the expression for the Airy wave is, , where x is the abscissa of any point on the water surface in the world coordinate system, η is the Airy wave amplitude, Airy wave is represented as microwave, ω is the circular frequency, t is the wave period, A is the wave amplitude value, and λ is the wavelength.

[0088] More preferably, the expression for calculating the amplitude of the Airy wave is: Where H is the wave height;

[0089] Based on the wave period, the expression for calculating the circular frequency of the Airy wave is: .

[0090] More preferably, the formula for calculating the Reynolds coefficient of water is Re=ρvλ / μ, where Re is the Reynolds coefficient of water, ρ is the water density, v is the water flow velocity, and μ is the water viscosity coefficient.

[0091] Historical water surface data include historical water surface length, historical water surface width, historical water depth, historical water viscosity coefficient, historical mean water wave Reynolds number, historical ship flow, historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure.

[0092] More preferably, the historical ship speed is expressed as the average ship speed under the historical ship flow at any historical monitoring moment within the historical monitoring period; the historical ship flow is expressed as the historical ship flow at any historical monitoring moment within the historical monitoring period; the historical ship density is expressed as the historical ship density at any historical monitoring moment within the historical monitoring period; the historical blockage density is expressed as the historical blockage density at any historical monitoring moment within the historical monitoring period; the historical ship weight is expressed as the average historical ship weight at any historical monitoring moment within the historical monitoring period; and the historical wind speed, historical wind pressure, historical water viscosity coefficient, and historical mean water wave Reynolds number are expressed as the historical wind speed and historical wind pressure at any historical monitoring moment within the historical monitoring period.

[0093] The matching method for historical water surface related data includes: obtaining the water surface length, width, depth, and navigable water type of the water surface to be analyzed; constructing length intervals, width intervals, and depth intervals based on the water surface length, width, and depth; retrieving the waterway monitoring database; inputting the length interval, width interval, depth interval, and navigable water type into the waterway monitoring database; and matching historical water surface related data whose length, width, and depth are all distributed within the length interval, width interval, and depth interval, and whose navigable water type is the same as that of the water surface to be analyzed.

[0094] More preferably, the distribution is represented as greater than or equal to the lower limit of the corresponding interval and less than or equal to the upper limit of the corresponding interval.

[0095] More preferably, the method for constructing the length interval, width interval, and depth interval includes setting the first length to a length allowable value, setting the first width to a width allowable value, and setting the first depth to a depth allowable value.

[0096] Calculate the sum of the water surface length and the first length, and the difference between the water surface length and the first length, respectively. Set the sum of the water surface length and the first length, and the difference between the water surface length and the first length, as the upper limit and lower limit of the length, respectively. Construct a length interval based on the upper limit and lower limit of the length.

[0097] Calculate the sum of the water surface width and the first width, and the difference between the water surface width and the first width, respectively. Set the sum of the water surface width and the first width, and the difference between the water surface width and the first width, respectively as the upper limit and lower limit of the width. Construct the width interval based on the upper limit and lower limit of the width.

[0098] Calculate the sum of the water surface depth and the first depth, and the difference between the water surface depth and the first depth, respectively. Set the sum of the water surface depth and the first depth, and the difference between the water surface depth and the first depth, as the upper limit and lower limit of the depth, respectively. Construct a depth interval based on the upper limit and lower limit of the depth.

[0099] The method for calculating the difference between historical speed and flow rate values ​​includes: calculating the average historical water viscosity coefficient and the average historical average water wave Reynolds number; obtaining a free state lookup table; inputting the average historical water viscosity coefficient, the average historical average water wave Reynolds number, the historical water surface length, the historical water surface width, and the historical water depth into the free state lookup table; and obtaining the ship flow rate, ship speed, and maximum ship density in the free flow state corresponding to the average historical water viscosity coefficient, the average historical average water wave Reynolds number, the historical water surface length, the historical water surface width, and the historical water depth, respectively, which are denoted as the ship flow rate, ship speed, and maximum ship density in the historical free flow state.

[0100] More preferably, the free state lookup table is obtained from the ship model test database based on the ITTC standard. In ship model tests, free state usually refers to the free self-propulsion method, which is a test method that simulates the free navigation conditions of a real ship. The ship model moves autonomously under the propeller propulsion without the application of external force, so that the speed of the ship model is naturally synchronized with the speed of the trailer, thereby reflecting the propulsion performance of the ship.

[0101] The free state lookup table contains free flow state data under different combinations of hydrodynamic parameters.

[0102] Substituting the ship flow rate, ship speed, maximum ship density, historical ship flow rate, historical ship speed, historical ship density, and historical blockage density under historical free-flow conditions into the expression for the relationship between ship speed and flow rate, we obtain the historical speed and flow rate relationship values ​​corresponding to each historical data collection time.

[0103] Calculate the difference between the historical velocity and flow rate values ​​at adjacent historical data collection times, and record it as the third difference. Iterate through the third differences between the historical velocity and flow rate values ​​at each group of adjacent historical data collection times.

[0104] Construct a mapping relationship between the third difference and the corresponding historical ship traffic, historical wind speed, and historical wind pressure.

[0105] More preferably, by inputting the third difference, historical wind speed, and historical wind pressure into the mapping relationship, the historical wind speed corresponding to the third difference, historical wind speed, and historical wind pressure is obtained.

[0106] The preset vector construction method, namely the first vector construction method, is expressed as obtaining the historical ship flow, historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure corresponding to any historical water surface.

[0107] Using historical ship traffic as the independent variable and historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure as dependent variables, a multiple regression analysis was conducted to obtain a multiple regression equation. The coefficients corresponding to historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure were extracted and processed.

[0108] Each character of the processed coefficient is represented using 4-bit binary encoding;

[0109] According to the order of each character in the processed coefficient, the characters of the coefficient after binary encoding are combined to obtain the first binary encoding combination;

[0110] The first binary code combination is converted into a vector, and the vector converted from the first binary code combination is set as the first vector.

[0111] More preferably, the method for processing the coefficients corresponding to historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure includes removing the decimal point from the coefficients corresponding to historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure, and performing fixed-length processing to obtain integers with the same number of characters. The fixed-length processing means counting the number of characters of each coefficient after removing the decimal point, denoted as the first quantity, selecting the first quantity with the largest value, and padding the remaining coefficients after removing the decimal point with 0s, so that the number of characters of each coefficient after removing the decimal point after padding with 0s is the first quantity with the largest value.

[0112] More preferably, since each character is distributed between 0 and 9, all types of characters can be covered by 4-bit binary encoding.

[0113] The method for constructing a second vector based on the third difference includes obtaining the third difference, processing the third difference, and representing each character of the processed third difference using 4-bit binary encoding.

[0114] According to the character order of the processed third difference, the characters of the third difference after binary encoding are combined to obtain the second binary encoding combination;

[0115] The second binary code combination is converted into a vector, and the vector converted from the second binary code combination is set as the second vector.

[0116] More preferably, the method for processing the third difference includes removing the decimal point from the third difference and padding the entire third difference with zeros after removing the decimal point, so that the number of characters in the third difference after the zero-padding operation is equal to the first number with the largest value.

[0117] More preferably, the means of converting the first binary code into a vector and the means of converting the second binary code into a vector are existing technologies and will not be described in detail here.

[0118] The correlation between the first and second vectors at any historical acquisition time is calculated using the maximum information coefficient algorithm.

[0119] Set the first value as the relevance threshold, and calculate the prediction step size based on the comparison between the relevance and the relevance threshold.

[0120] More preferably, the maximum information coefficient algorithm is as follows:

[0121] ;

[0122] Where a and b are the largest first vector modulus and the largest second vector modulus, respectively, X is the first vector, Y is the second vector, B is a constant, and MIC(X,Y) is the maximum information coefficient of the first and second vectors.

[0123] The results of comparing relevance with the relevance threshold include relevance greater than or equal to the relevance threshold, and relevance less than the relevance threshold.

[0124] When the relevance is greater than or equal to the relevance threshold, the corresponding historical data collection time will be retained.

[0125] When the relevance is less than the relevance threshold, the corresponding historical data collection time will be deleted.

[0126] Obtain the retained historical acquisition times, calculate the time difference between adjacent historical acquisition times in the retained historical acquisition times, traverse each group of adjacent historical acquisition times to obtain the time difference corresponding to each group of adjacent historical acquisition times, calculate the average value of the time difference corresponding to adjacent historical acquisition times, and record it as the average time difference value.

[0127] Based on the average time difference, the prediction step size is calculated. The calculation method for the prediction step size includes calculating the water surface length, water surface width, and water depth, and the first, second, and third ratios of these ratios with the historical water surface length, historical water surface width, and historical water depth, respectively; calculating the first product of the first, second, and third ratios; obtaining the historical data collection time interval for this historical water surface data; and calculating the fourth ratio of the historical data collection time interval to the average time difference.

[0128] Calculate the first quotient of the product of the average time difference and the first product, and calculate the second product of the fourth ratio and the first quotient;

[0129] Iterate through the second product corresponding to each matching historical water surface data, calculate the average of the second product, and set the average of the second product as the prediction step size.

[0130] More preferably, the ship flow rate at the time point corresponding to the previous prediction step is approximately the same as the ship flow rate at the time point corresponding to the next prediction step. This application is applied to large-scale water surfaces. Since the ship flow rate is relatively large, in this application, the ship flow rate at the time point corresponding to the previous prediction step is set to be equal to the ship flow rate at the time point corresponding to the next prediction step.

[0131] Get the current time point, calculate the difference between the current time point and the prediction step size, record it as the historical time point, and obtain the historical data of the historical time point;

[0132] Historical data includes the third difference, historical wind speed, and historical wind pressure;

[0133] By inputting the third difference, historical wind speed, and historical wind pressure into the mapping relationship, the historical ship traffic flow of the water area to be analyzed at the historical time point can be obtained;

[0134] Set the historical vessel flow rate of the water surface to be analyzed at a historical time point as the vessel flow rate after the predicted step size of the water surface to be analyzed.

[0135] More preferably, the present invention uses the maximum information coefficient instead of the linear correlation coefficient or empirical threshold, which can capture any functional dependence between ship speed and flow, such as nonlinear, non-monotonic, and time-varying. The optimal prediction step size is automatically calculated through the MIC significance test. The step size is shortened to improve sensitivity when there are sudden changes in ship density, such as before and after holidays and during tidal windows. The effective parameters and ship flow are established to quickly map the relationship, which improves the speed of ship flow acquisition, enhances the targeting of data and saves storage costs.

[0136] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0137] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the protection scope of the present invention.

Claims

1. A real-time analysis method for waterway ship flow based on big data, characterized in that, The process includes the following steps: Step S100, based on the relationship between ship speed and flow rate, combined with the maximum information coefficient algorithm, constructing a mapping relationship for ship flow rate and calculating the prediction step size; Based on the historical water surface data matched with the water surface data to be analyzed, and based on the expression of the relationship between historical water surface data and ship speed and flow rate, the difference between historical speed and flow rate is obtained. Then, the first vector and the second vector are constructed using a preset vector construction method. The correlation between the first and second vectors is calculated using the maximum information coefficient algorithm. Based on the comparison between the correlation and the correlation threshold, the corresponding historical acquisition time is selected to be retained or deleted. Based on the retained historical acquisition time, the average time difference between each group of adjacent historical acquisition time is calculated. The first product is calculated based on the average time difference, and the fourth ratio between the historical acquisition time interval and the average time difference is calculated. The first quotient of the average time difference and the first product is calculated, and the second product of the fourth ratio and the first quotient is calculated. The average value of the second product is set as the prediction step size. Step S200: Based on the prediction step size, historical water surface data, the expression of the relationship between ship speed and flow rate, and the historical data of the water surface to be analyzed, obtain the historical speed and flow rate relationship value corresponding to each historical acquisition time, calculate the difference between the historical speed and flow rate relationship values ​​of adjacent historical acquisition times, and record it as the third difference value, and construct the mapping relationship between the third difference value, the corresponding historical ship flow rate, historical wind speed, and historical wind pressure. Based on the mapping relationship and historical data of the water surface to be analyzed, the ship traffic flow of the water surface to be analyzed is predicted; The method for predicting ship traffic on the water surface to be analyzed includes: obtaining the current time point, calculating the difference between the current time point and the prediction step size, recording it as the historical time point, and obtaining historical data for the historical time point; Historical data includes the third difference, historical wind speed, and historical wind pressure; By inputting the third difference, historical wind speed, and historical wind pressure into the mapping relationship, the historical ship traffic flow of the water area to be analyzed at the historical time point can be obtained; Set the historical vessel flow rate of the water surface to be analyzed at a historical time point as the vessel flow rate after the predicted step size of the water surface to be analyzed.

2. The big data based waterway vessel traffic flow real time analysis method as claimed in claim 1, wherein, The first vector represents the vector corresponding to any historical water surface data, and the second vector represents the vector corresponding to the historical velocity difference and the historical flow difference. The correlation between the first vector and the second vector is calculated using the maximum information coefficient algorithm. The first value is set as the correlation threshold. The prediction step size is calculated based on the comparison between the correlation and the correlation threshold.

3. The big data based waterway vessel traffic flow real time analysis method of claim 2, wherein, The water surface data to be analyzed include water surface length, water surface width, water depth, water density, water viscosity coefficient, water flow velocity, navigable water area type, average water wave Reynolds number, wind force, and wind speed. The length of the water surface is represented as the length of the projection shape of the curved surface corresponding to the water surface to be analyzed on the first plane, which is processed into a rectangle. The width of the water surface is represented as the width of the projection shape of the curved surface to be analyzed on the first plane, which is processed into a rectangle. The water density, water flow velocity, wind force and wind speed are obtained by monitoring through corresponding sensors. The first plane satisfies the second constraint condition, which is expressed as follows: the area of ​​the projected shape corresponding to the water surface to be analyzed is the largest. The projection is expressed as the straight line segment connecting any point in the water surface to the corresponding projection point is parallel to the normal of the first plane.

4. The big data based waterway vessel traffic flow real time analysis method as claimed in claim 1, wherein, Historical water surface data include historical water surface length, historical water surface width, historical water depth, historical water viscosity coefficient, historical mean water wave Reynolds number, historical ship flow, historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure.

5. The big data based waterway vessel traffic flow real time analysis method of claim 4, wherein, The matching method for historical water surface related data includes: obtaining the water surface length, width, depth, and navigable water type of the water surface to be analyzed; constructing length intervals, width intervals, and depth intervals based on the water surface length, width, and depth; retrieving the waterway monitoring database; inputting the length interval, width interval, depth interval, and navigable water type into the waterway monitoring database; and matching historical water surface related data whose length, width, and depth are all distributed within the length interval, width interval, and depth interval, and whose navigable water type is the same as that of the water surface to be analyzed.

6. The big data based waterway vessel traffic flow real time analysis method of claim 5, wherein, The method for calculating the difference between historical speed and flow rate values ​​includes: calculating the average historical water viscosity coefficient and the average historical average water wave Reynolds number; obtaining a free state lookup table; inputting the average historical water viscosity coefficient, the average historical average water wave Reynolds number, the historical water surface length, the historical water surface width, and the historical water depth into the free state lookup table; and obtaining the ship flow rate, ship speed, and maximum ship density in the free flow state corresponding to the average historical water viscosity coefficient, the average historical average water wave Reynolds number, the historical water surface length, the historical water surface width, and the historical water depth, respectively, which are denoted as the ship flow rate, ship speed, and maximum ship density in the historical free flow state. Substituting the ship flow rate, ship speed, maximum ship density, historical ship flow rate, historical ship speed, historical ship density, and historical blockage density under historical free-flow conditions into the expression for the relationship between ship speed and flow rate, we obtain the historical speed and flow rate relationship values ​​corresponding to each historical data collection time. Calculate the difference between the historical velocity and flow rate values ​​at adjacent historical data collection times, and record it as the third difference. Iterate through the third differences between the historical velocity and flow rate values ​​at each group of adjacent historical data collection times. Construct a mapping relationship between the third difference and the corresponding historical ship traffic, historical wind speed, and historical wind pressure.

7. The big data based waterway vessel traffic flow real time analysis method as claimed in claim 2, wherein, The preset vector construction method, namely the first vector construction method, is expressed as obtaining the historical ship flow, historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure corresponding to any historical water surface. Using historical ship traffic as the independent variable and historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure as dependent variables, a multiple regression analysis was conducted to obtain a multiple regression equation. The coefficients corresponding to historical ship speed, historical ship density, historical blockage density, historical ship weight, historical wind speed, and historical wind pressure were extracted and processed. Each character of the processed coefficient is represented using 4-bit binary encoding; According to the order of each character in the processed coefficient, the characters of the coefficient after binary encoding are combined to obtain the first binary encoding combination; The first binary code combination is converted into a vector, and the vector converted from the first binary code combination is set as the first vector; The method for constructing a second vector based on the third difference includes obtaining the third difference, processing the third difference, and representing each character of the processed third difference using 4-bit binary encoding. According to the character order of the processed third difference, the characters of the third difference after binary encoding are combined to obtain the second binary encoding combination; The second binary code combination is converted into a vector, and the vector converted from the second binary code combination is set as the second vector.

8. The big data based waterway vessel traffic flow real time analysis method as claimed in claim 2, wherein, The correlation between the first and second vectors at any historical acquisition time is calculated using the maximum information coefficient algorithm. Set the first value as the relevance threshold, and calculate the prediction step size based on the comparison between the relevance and the relevance threshold. The results of comparing relevance with the relevance threshold include relevance greater than or equal to the relevance threshold, and relevance less than the relevance threshold. When the relevance is greater than or equal to the relevance threshold, the corresponding historical data collection time will be retained. When the relevance is less than the relevance threshold, the corresponding historical data collection time will be deleted. Obtain the retained historical acquisition times, calculate the time difference between adjacent historical acquisition times in the retained historical acquisition times, traverse each group of adjacent historical acquisition times, obtain the time difference corresponding to each group of adjacent historical acquisition times, calculate the average value of the time difference corresponding to adjacent historical acquisition times, and record it as the average time difference. Based on the average time difference, the prediction step size is calculated. The calculation method for the prediction step size includes calculating the water surface length, water surface width, and water depth, and the first, second, and third ratios of these ratios with the historical water surface length, historical water surface width, and historical water depth, respectively; calculating the first product of the first, second, and third ratios; obtaining the historical data collection time interval for this historical water surface data; and calculating the fourth ratio of the historical data collection time interval to the average time difference. Calculate the first quotient of the product of the average time difference and the first product, and calculate the second product of the fourth ratio and the first quotient; Iterate through the second product corresponding to each matching historical water surface data, calculate the average of the second product, and set the average of the second product as the prediction step size.

9. A big data-based waterway ship flow real-time analysis system for performing the big data-based waterway ship flow real-time analysis method of claim 1, characterized in that, Includes a building module and a prediction module; The construction module constructs a mapping relationship between ship speed and flow rate based on the relationship between ship speed and flow rate, combined with the maximum information coefficient algorithm, and calculates the prediction step size. The prediction module obtains historical data of the water surface to be analyzed based on the prediction step size, and predicts the ship traffic flow of the water surface to be analyzed based on the mapping relationship and the historical data of the water surface to be analyzed.

Images