Port hydrodynamic environment monitoring method and system

Abstract

The invention relates to the technical field of environment monitoring, and discloses a port hydrodynamic environment monitoring method and system, and the method comprises the steps: collecting water quality parameters and / or water flow parameters of a port water area time sequence in real time; preprocessing the water quality parameters and / or water flow parameters of the time sequence, and storing the preprocessed water quality parameters and / or water flow parameters in a database; the preprocessing at least comprises the removal of each water quality parameter abnormal value and each water flow parameter abnormal value under the time sequence; extracting newest input data from the database, inputting the newest input data into the trained numerical model, outputting a hydrodynamic condition, and monitoring a hydrodynamic environment; and generating an early warning report according to the output hydrodynamic condition. Compared with the prior art, the method has the advantages that the abnormal values are eliminated through multiple steps, the data quality, the robustness of the numerical model and the prediction precision of the numerical model are improved, and prevention measures proposed based on the prediction result are effectively realized.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to a port hydrodynamic environment monitoring method and system. Background Art

[0002] With the growth of global trade, ports, as important nodes of international logistics, play a vital role in economic development through safe and efficient operation. However, the complex hydrodynamic environment of port waters and human activities pose challenges to port operations. Therefore, accurate monitoring and evaluation of the port hydrodynamic environment is essential to ensure port safety, promote environmental protection and optimize port planning.

[0003] With the rapid development of science and technology, a series of advanced monitoring technologies have been widely used in the monitoring of port hydrodynamic environment. Existing technologies collect water quality parameters and / or water flow parameters in the hydrodynamic environment through sensors, combine measured data with mathematical models for simulation calculations, predict future hydrodynamic conditions, provide a scientific basis for port planning, design and management, help identify potential risk points, and propose corresponding preventive measures. Numerical simulation technology is highly dependent on data, and the accuracy of numerical simulation is highly dependent on the quality and integrity of input data. If the sensor fails, data transmission is interrupted, or data processing is wrong, it may cause data loss or distortion, which in turn affects the reliability of the simulation results, resulting in large deviations in the simulation results, reducing the accuracy of the prediction of hydrodynamic conditions, causing errors in the identified risk points, and the proposed preventive measures cannot be effectively implemented. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a port hydrodynamic environment monitoring method and system, which eliminates abnormal values ​​and predicts hydrodynamic conditions to realize hydrodynamic condition monitoring.

[0005] The present invention provides a port hydrodynamic environment monitoring method, comprising: Step 1, using a fiber Bragg grating strain sensor to collect water quality parameters and / or water flow parameters of the port water area in real time; Step 2, preprocessing the water quality parameters and / or water flow parameters of the time series, and storing the preprocessed water quality parameters and / or water flow parameters in a database; The preprocessing at least includes: removing abnormal values ​​of each water quality parameter and / or water flow parameter in the time series; the abnormal value removal includes: Step 21, constructing a first threshold value for screening and determining outliers; Step 22, determining whether the water quality parameter and / or water flow parameter in each time series is greater than a first threshold value, if so, defining the water quality parameter and / or water flow parameter in the current time series as a determined abnormal value, and removing the determined abnormal value; if not, defining the water quality parameter and / or water flow parameter in the current time series as a preset normal value, and retaining the preset normal value; Step 23, inspect the abnormality of the preset normal value, optimize the first threshold to obtain a second threshold, remove the preset normal value that fails the inspection, compare the preset normal value that passes the inspection with the second threshold, if the preset normal value is greater than the second threshold, remove the preset normal value; if the preset normal value is less than or equal to the second threshold, retain the preset normal value, and use the water quality parameter and / or water flow parameter corresponding to the retained preset normal value as input data; Step 3, obtain the training set from the database to train the numerical model, obtain the trained numerical model, extract the latest input data from the database, and input it into the trained numerical model, output the hydrodynamic conditions, and monitor the hydrodynamic environment; Step 4: Generate an early warning report based on the output hydrodynamic conditions.

[0006] Furthermore, the water quality parameters include at least the temperature, salinity, and pH value of different monitoring points, and the water flow parameters include at least the water flow velocity of each monitoring point; Constructing the first threshold includes the following steps: Step 211, obtaining the temperature, salinity, pH value, and water flow velocity of each monitoring point in the current time series; Step 212, according to the temperature of each monitoring point in all time series, obtain the average temperature; according to the salinity of each monitoring point in all time series, obtain the average salinity; according to the pH value of each monitoring point in all time series, obtain the average pH; according to the water flow velocity of each monitoring point in all time series, obtain the average water flow velocity; Step 213, obtaining a first threshold value according to temperature, salinity, pH value, water flow velocity, average temperature value, average salinity value, average pH value, and average water flow velocity value.

[0007] Furthermore, the abnormality of the preset normal value is tested, specifically including: testing the time series continuity and spatial correlation of the preset normal value.

[0008] Furthermore, the continuity of the time series of the preset normal value is tested, including: Step a, calculating a preset difference value based on a preset normal value in the current time series and a preset normal value in the previous time series; Step b, constructing a short-term fluctuation threshold; comparing the preset difference with the short-term fluctuation threshold, if the preset difference is greater than the short-term fluctuation threshold, and the preset difference is restored to within the short-term fluctuation threshold within the next n time series, then marking the preset normal value as short-term fluctuation data and retaining it; if the preset difference is greater than the short-term fluctuation threshold, and the preset difference is not restored to within the short-term fluctuation threshold within the next n time series, then marking the preset normal value as a final abnormal value and removing it; if the difference is less than or equal to the short-term fluctuation threshold, the preset normal value is a normal value and is retained; Step c, calculating the average value based on the preset normal value in the current time series and the preset normal values ​​in the previous m time series; calculating the difference between the preset normal value in the current time series and the average value; Step d, constructing a long-term trend threshold; comparing the gap value with the long-term trend threshold; if the gap value is greater than the long-term trend threshold and the gap value lasts for more than m+1 time series, the preset normal value is marked as the final abnormal value and removed; if the gap value is greater than the long-term trend threshold and the gap value recovers to the long-term trend threshold within the m+1 time series, or the gap value is less than or equal to the long-term trend threshold, it means that the preset normal value has a long-term trend, and the preset normal value is a normal value and is retained.

[0009] Furthermore, the spatial correlation of the preset normal value is tested, including: Step e, obtaining a preset normal value of the current monitoring point; Step f, taking the current monitoring point as the center of the circle, and obtaining preset normal values ​​of neighboring monitoring points in a circular area with a radius x based on the center of the circle; Step g, calculating the average of the preset normal values ​​according to the preset normal value of the current monitoring point and the preset normal values ​​of the adjacent monitoring points; Step h, calculating the difference between the predicted normal value of the current monitoring point and the mean of the preset normal value; Step i, construct a local consistency threshold; compare the difference value with the local consistency threshold, if the difference value is greater than the local consistency threshold, then mark the predicted normal value of the current monitoring point as the final abnormal value and remove it; if the difference value is less than or equal to the local consistency threshold, then the predicted normal value of the current monitoring point is a normal value and is retained; Step j, averaging the preset normal values ​​of all monitoring points to obtain a global average value; Step k, calculate the difference between the predicted normal value of the current monitoring point and the global average value to obtain the global difference; construct a global consistency threshold; compare the global difference with the global consistency threshold, if the global difference is greater than the global consistency threshold, mark the predicted normal value of the current monitoring point as the final abnormal value and remove it; if the global difference is less than or equal to the global consistency threshold, mark the predicted normal value of the current monitoring point as a normal value and retain it.

[0010] Furthermore, a second threshold value is obtained according to the temperature, salinity, pH value, water flow velocity, the average temperature value, the average salinity value, the average pH value, and the average water flow velocity value.

[0011] Further, the water quality parameters and / or water flow parameters corresponding to the preset normal values ​​that have passed the test are used as input data, specifically: the preset normal values ​​obtained through the tests of step a-step d, step e-step k, including: normal values ​​and short-term fluctuation data; abnormal values ​​obtained through the tests of step a-step d, step e-step k, including: final abnormal values; Check whether the normal values ​​or short-term fluctuation data obtained by the test meet the spatial correlation and time series continuity. If not, the normal values ​​or short-term fluctuation data obtained by the test will be eliminated; if so, the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity will be compared with the second threshold. If the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity are greater than the second threshold, the normal values ​​that meet the spatial correlation and time series continuity will be eliminated; if the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity are less than or equal to the second threshold, the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity will be retained, and the corresponding water quality parameters and / or water flow parameters of the retained normal values ​​that meet the spatial correlation and time series continuity will be used as input data.

[0012] Furthermore, the calculation formula of the first threshold is: ; Where T1 represents the first threshold, T k represents the temperature of the kth monitoring point, S k Represents the salinity of the kth monitoring point, pH k represents the pH value of the kth monitoring point, represents the component of the water velocity on the x-axis at the kth monitoring point, represents the component of the water velocity on the y-axis at the kth monitoring point, represents the average temperature, represents the average salinity, Represents the average pH value, represents the average value of the water velocity on the x-axis. represents the average value of the water velocity on the y-axis component, , Represent weight coefficients respectively, z represents the maximum number of monitoring points, Represents the combined deviation of temperature and salinity, Represents the comprehensive deviation of pH value and water flow rate.

[0013] Furthermore, the calculation formula of the second threshold is: ; Where T2 represents the second threshold, T k represents the temperature of the kth monitoring point, S k Represents the salinity of the kth monitoring point, pH k represents the pH value of the kth monitoring point, represents the component of the water velocity on the x-axis at the kth monitoring point, represents the component of the water velocity on the y-axis at the kth monitoring point, represents the average temperature, represents the average salinity, Represents the average pH value, represents the average value of the water velocity on the x-axis. represents the average value of the water velocity on the y-axis component, represents the weight coefficient, and z represents the maximum number of monitoring points.

[0014] The present invention also discloses a port hydrodynamic environment monitoring system, comprising the following modules: Acquisition module: used to collect the water quality parameters and / or water flow parameters of the port water area in real time through the fiber grating strain sensor; Preprocessing module: connected to the acquisition module, used to preprocess the water quality parameters and / or water flow parameters of the time series, and store the preprocessed water quality parameters and / or water flow parameters in the database; Monitoring module: connected with the preprocessing module, used to obtain the training set from the database to train the numerical model to obtain a trained numerical model, extract the latest input data from the database, input it into the trained numerical model, output the hydrodynamic conditions, and monitor the hydrodynamic environment; Early warning module: connected to the monitoring module, used to generate early warning reports based on the output hydrodynamic conditions.

[0015] The embodiments of the present invention have the following technical effects: Based on the existing method of removing data outliers by a single threshold value and then applying it to a numerical simulation model to predict hydrodynamic conditions, the present invention finely optimizes the process of removing outliers. First, a first threshold value and a second threshold value are constructed by water quality parameters and / or water flow parameters. Based on the first threshold value, obvious outliers in the original data are removed, and preset normal values ​​are retained; then, based on the continuity of the time series and the spatial correlation, the preset normal values ​​are tested, and the normal values ​​that fail the test are removed and the normal values ​​that pass the test are retained. Finally, the normal values ​​that meet both the spatial correlation and the continuity of the time series are compared with the second threshold value, and the final retained normal values ​​are determined and stored in the database. The numerical simulation model and the prediction of hydrodynamic conditions are trained based on the normal values ​​stored in the database. Compared with the prior art, the present invention realizes the removal of outliers through multiple steps, improves the data quality, improves the robustness of the numerical model, improves the prediction accuracy of the numerical model, improves the identification accuracy of the risk points, and effectively implements the preventive measures proposed based on the prediction results.

[0016] The present invention breaks away from the existing method of constructing thresholds through standard deviation and mean to remove obvious outliers, and constructs a first threshold based on the comprehensive deviation degree of temperature and salinity, and the comprehensive deviation degree of pH value and water flow velocity. Compared with the existing threshold constructed by standard deviation and mean, the comprehensive deviation degree of temperature and salinity reflects the changes in the physical properties of the water body, while the comprehensive deviation degree of pH value and water flow velocity reflects the dynamic changes of water quality and water flow. Based on the dynamic changes and physical property changes, the present application comprehensively considers the deviation degree of multiple parameters, more accurately eliminates obvious outliers, reduces false alarms and omissions caused by fluctuations in a single parameter, and improves the quality and reliability of data.

[0017] The present invention optimizes the first threshold to obtain the second threshold, and the second threshold comprehensively considers the comprehensive deviation degree of temperature, salinity, pH, and water flow velocity. Compared with the threshold obtained by considering the comprehensive deviation degree of two parameters, the second threshold can comprehensively evaluate the comprehensive changes of the water body. If the deviation of a certain parameter is not obvious, but the comprehensive deviation of multiple parameters is large, it will be identified as an outlier. Therefore, the second threshold can more accurately screen out outliers, especially those outliers where the change of a single parameter is not obvious. Through multi-parameter comprehensive evaluation, these subtle abnormal changes can be more accurately identified. Through the screening of the second threshold, the quality and reliability of the data can be further improved, ensuring that the data ultimately used for the numerical simulation model is of high quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 is a flow chart of a port hydrodynamic environment monitoring method provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a port hydrodynamic environment monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0021] Figure 1 is a flow chart of a port hydrodynamic environment monitoring method provided by an embodiment of the present invention. Figure 1 , specifically including: Step 1: Using a fiber grating strain sensor, real-time acquisition of water quality parameters and / or water flow parameters of the port water area time series; the water quality parameters at least include temperature, salinity, and pH value at different monitoring points, and the water flow parameters at least include water flow velocity at each monitoring point.

[0022] Step 2: preprocess the water quality parameters and / or water flow parameters of the time series, and store the preprocessed water quality parameters and / or water flow parameters in a database.

[0023] The preprocessing at least includes: removing abnormal values ​​of each water quality parameter and / or abnormal values ​​of water flow parameter in the time series. The preprocessing also includes: missing value processing, repeated value processing, data transformation, etc.

[0024] Outlier removal, including: Step 21, constructing a first threshold for screening and determining outliers.

[0025] Step 211, obtaining the temperature, salinity, pH value, and water flow velocity of each monitoring point in the current time series.

[0026] Step 212, based on the temperature of each monitoring point in all time series, obtain the average temperature; based on the salinity of each monitoring point in all time series, obtain the average salinity; based on the pH value of each monitoring point in all time series, obtain the average pH; based on the water flow velocity of each monitoring point in all time series, obtain the average water flow velocity.

[0027] Step 213, obtaining a first threshold value according to temperature, salinity, pH value, water flow velocity, average temperature value, average salinity value, average pH value, and average water flow velocity value.

[0028] ; Where T1 represents the first threshold, T k represents the temperature of the kth monitoring point, S k Represents the salinity of the kth monitoring point, pH k represents the pH value of the kth monitoring point, represents the component of the water velocity on the x-axis at the kth monitoring point, represents the component of the water velocity on the y-axis at the kth monitoring point, represents the average temperature, represents the average salinity, Represents the average pH value, represents the average value of the water velocity on the x-axis. represents the average value of the water velocity on the y-axis component, , Represent weight coefficients respectively, z represents the maximum number of monitoring points, Represents the combined deviation of temperature and salinity, Represents the comprehensive deviation of pH value and water flow rate.

[0029] The design of the first threshold is based on the concept of Euclidean distance, which is a commonly used multidimensional data distance measurement method used to measure the distance between a data point and a reference point (the mean of each parameter), so as to obtain the comprehensive deviation degree of the data point in the multidimensional space.

[0030] Among them, temperature and salinity are the main factors affecting the density of seawater. Rising temperature leads to a decrease in density, while increasing salinity leads to an increase in density. The combined changes in temperature and salinity significantly affect the density of seawater, and thus affect the stability and dynamic characteristics of water flow. Therefore, changes in temperature and salinity will affect the thermodynamic stability of water bodies. The pH value reflects the acidity and alkalinity of water quality, has an important impact on aquatic organisms and chemical reactions, and is very important for port environmental monitoring. The water flow velocity directly affects the flow characteristics and material transport of the water body. The change in water flow velocity can reflect the intensity and direction of the water flow, and has a direct impact on the safety of ship navigation and port facilities. Therefore, the change in pH value and water flow velocity affects the water body dynamic effect. In this embodiment, the comprehensive deviation degree of temperature and salinity, the comprehensive deviation degree of pH value and water flow velocity, construct the first threshold, based on dynamic changes and physical property changes, comprehensively consider the deviation degree of multiple parameters, more accurately eliminate obvious abnormal values, reduce false alarms and omissions caused by fluctuations in a single parameter, and improve the quality and reliability of data.

[0031] The first threshold value constructed based on the water quality parameters and / or water flow parameters acquired in real time in this embodiment can immediately reflect the changes in the current environment, respond quickly when an abnormal event occurs, reduce delays, and improve the real-time response capability of the system. Therefore, this embodiment uses the constructed first threshold value as an evaluation criterion to preliminarily screen obvious abnormal values.

[0032] Step 22, determine whether the water quality parameter and / or water flow parameter in each time series is greater than the first threshold value. If so, define the water quality parameter and / or water flow parameter in the current time series as a determined abnormal value, and remove the determined abnormal value; if not, define the water quality parameter and / or water flow parameter in the current time series as a preset normal value, and retain the preset normal value.

[0033] When the water quality parameter and / or water flow parameter is greater than the first threshold, it means that the parameter far exceeds the evaluation standard of this embodiment, which means that the comprehensive deviation degree between the parameter and temperature and salinity, and the comprehensive deviation degree between pH value and water flow velocity are large. This embodiment defines the parameter deviating from the first threshold as an obvious abnormal value, and removes it through this step. The definition of not removed is the preset normal value.

[0034] Step 23, check the abnormal situation of the preset normal value, optimize the first threshold to obtain a second threshold, eliminate the preset normal value that fails the test, compare the preset normal value that passes the test with the second threshold, if the preset normal value is greater than the second threshold, eliminate the preset normal value; if the preset normal value is less than or equal to the second threshold, retain the preset normal value, and use the water quality parameter and / or water flow parameter corresponding to the retained preset normal value as input data.

[0035] The abnormality of the preset normal value is tested, specifically including: testing the time series continuity and spatial correlation of the preset normal value.

[0036] Test the continuity of the time series of preset normal values, including: Step a, calculating a preset difference value based on a preset normal value in the current time series and a preset normal value in the previous time series.

[0037] Step b, constructing a short-term fluctuation threshold; comparing the preset difference with the short-term fluctuation threshold; if the preset difference is greater than the short-term fluctuation threshold, and the preset difference is restored to within the short-term fluctuation threshold within the next n time series, then the preset normal value is marked as short-term fluctuation data and retained; if the preset difference is greater than the short-term fluctuation threshold, and the preset difference is not restored to the short-term fluctuation threshold within the next n time series, then the preset normal value is marked as the final abnormal value and removed; if the preset difference is less than or equal to the short-term fluctuation threshold, the preset normal value is a normal value.

[0038] Since the calculated preset difference may be a negative value, this embodiment preferably uses the absolute value of the preset difference to compare with the short-term fluctuation threshold.

[0039] Step c, calculating an average value based on the preset normal value in the current time series and the preset normal values ​​in the previous m time series; calculating the difference between the preset normal value in the current time series and the average value.

[0040] In this step, a representative average value is calculated based on the parameters in the current time series and multiple past time series. The difference between the parameter of each time series and the average value can determine whether the parameter deviates from the average value, which serves as a criterion for subsequent judgment of whether there is a long-term trend.

[0041] Step d, constructing a long-term trend threshold; comparing the gap value with the long-term trend threshold; if the gap value is greater than the long-term trend threshold and the gap value lasts for more than m+1 time series, the preset normal value is marked as the final abnormal value and removed; if the gap value is greater than the long-term trend threshold and the gap value recovers to the long-term trend threshold within the m+1 time series, or the gap value is less than or equal to the long-term trend threshold, it means that the preset normal value has a long-term trend, and the preset normal value is a normal value and is retained.

[0042] If the gap value is greater than the long-term trend threshold, and the gap value has not been reduced from the past time series to the current time series and has not been reduced to the long-term trend threshold range, this embodiment considers that the preset normal value does not have a long-term trend and determines it as an abnormal value, so it needs to be removed. If the gap value is greater than the long-term trend threshold and the gap value has recovered to the long-term trend threshold before the current time series, or the gap value is less than or equal to the long-term trend threshold, this embodiment considers that the preset normal value has a long-term trend, determines it as a normal value, and needs to be retained.

[0043] The spatial correlation of the preset normal values ​​is tested, including: Step e, obtaining the preset normal value of the current monitoring point.

[0044] Step f, taking the current monitoring point as the center of the circle, and obtaining preset normal values ​​of neighboring monitoring points in a circular area with a radius x based on the center of the circle.

[0045] Step g, calculating the average of the preset normal values ​​according to the preset normal value of the current monitoring point and the preset normal values ​​of the adjacent monitoring points; Step h, calculating the difference between the predicted normal value of the current monitoring point and the mean of the preset normal value; Step i, construct a local consistency threshold; compare the difference value with the local consistency threshold. If the difference value is greater than the local consistency threshold, the predicted normal value of the current monitoring point is marked as the final abnormal value and removed; if the difference value is less than or equal to the local consistency threshold, the predicted normal value of the current monitoring point is the normal value and is retained.

[0046] The local consistency check mainly focuses on the difference between the data of the current monitoring point and its surrounding monitoring points. It can effectively capture the changes in the local environment and reduce false alarms caused by changes in the global environment, especially when the local environment changes significantly.

[0047] The circular area with radius x obtained based on the center of the circle is considered to be a local area in this embodiment. The monitoring points in the local area are monitored by multiple sensors. The continuity of the environment, the synchronization of data collection, the calibration and standardization of sensors, and the joint influence of environmental factors make the data detected by multiple sensors in the same time series have a certain correlation. If the data of a certain sensor deviates significantly from the data of other sensors, it is an outlier. In this regard, this embodiment constructs a local consistency threshold. This embodiment believes that the data within the threshold range is correlated with the data of other monitoring points in the local area. Therefore, if the difference value is greater than the local consistency threshold, the predicted normal value of the current monitoring point is marked as the final outlier and eliminated; if the difference value is less than or equal to the local consistency threshold, the predicted normal value of the current monitoring point is a normal value and is retained.

[0048] Step j, average the preset normal values ​​of all monitoring points to obtain a global average value.

[0049] Step k, calculate the difference between the predicted normal value of the current monitoring point and the global average value to obtain the global difference; construct a global consistency threshold; compare the global difference with the global consistency threshold, if the global difference is greater than the global consistency threshold, mark the predicted normal value of the current monitoring point as the final abnormal value and remove it; if the global difference is less than or equal to the global consistency threshold, mark the predicted normal value of the current monitoring point as a normal value and retain it.

[0050] The global consistency judgment takes into account the data of all monitoring points in the entire water area, which can reflect the changes in the overall environment. Due to the large amount of data involved, it is less affected by local environmental changes, the results are more stable, and it can capture global abnormal events. In a larger range, the data of multiple monitoring points have certain similarities. If the data of a certain monitoring point or the global difference deviates significantly from the data in other areas, it is an outlier. Therefore, if the global difference is greater than the global consistency threshold, the predicted normal value of the current monitoring point is marked as the final outlier and removed; if the global difference is less than or equal to the global consistency threshold, the predicted normal value of the current monitoring point is marked as a normal value and retained.

[0051] The second threshold value is obtained according to the temperature, salinity, pH value, water flow velocity, the average temperature value, the average salinity value, the average pH value, and the average water flow velocity value.

[0052] ; Where T2 represents the second threshold, T k represents the temperature of the kth monitoring point, S k Represents the salinity of the kth monitoring point, pH k represents the pH value of the kth monitoring point, represents the component of the water velocity on the x-axis at the kth monitoring point, represents the component of the water velocity on the y-axis at the kth monitoring point, represents the average temperature, represents the average salinity, Represents the average pH value, represents the average value of the water velocity on the x-axis. represents the average value of the water velocity on the y-axis component, represents the weight coefficient, and z represents the maximum number of monitoring points.

[0053] The design of the second threshold is also based on the concept of Euclidean distance, which is used to measure the distance between all parameters and the reference point (the mean of each parameter), so as to obtain the comprehensive deviation degree of all parameter points in the multidimensional space.

[0054] The first threshold is optimized to obtain the second threshold. The second threshold comprehensively considers the comprehensive deviation of temperature, salinity, pH, and water flow velocity. Compared with the threshold obtained by considering the comprehensive deviation of two parameters, the second threshold can comprehensively evaluate the comprehensive changes of the water body. If the deviation of a certain parameter is not obvious, but the comprehensive deviation of multiple parameters is large, it will be identified as an outlier. Therefore, the second threshold can more accurately screen out outliers, especially those with insignificant changes in a single parameter. Through comprehensive multi-parameter evaluation, these subtle abnormal changes can be more accurately identified. Through the screening of the second threshold, the quality and reliability of the data can be further improved, ensuring that the data ultimately used for the numerical simulation model is of high quality.

[0055] It is worth noting that the short-term fluctuation threshold, long-term trend threshold, local consistency threshold, and global consistency threshold are all obtained by calculating the standard deviation of the data within the corresponding time range.

[0056] The standard deviation of the short-term fluctuation threshold is calculated by the preset normal values ​​of the current time series and the previous time series. The standard deviation of the long-term trend threshold is calculated by the preset normal values ​​of the current time series and the previous m time series. The standard deviation of the local consistency threshold is calculated by the preset normal values ​​of the current monitoring point and the neighboring monitoring points. The standard deviation of the global consistency threshold is calculated by the preset normal values ​​of all monitoring points.

[0057] The water quality parameters and / or water flow parameters corresponding to the preset normal values ​​that have passed the test are used as input data, specifically: the preset normal values ​​obtained through the tests of steps a to d, and steps e to k, including: normal values ​​and short-term fluctuation data; the abnormal values ​​obtained through the tests of steps a to d, and steps e to k, including: the final abnormal value; Check whether the normal values ​​or short-term fluctuation data obtained by the test meet the spatial correlation and time series continuity. If not, the normal values ​​or short-term fluctuation data obtained by the test will be eliminated; if so, the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity will be compared with the second threshold. If the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity are greater than the second threshold, the normal values ​​that meet the spatial correlation and time series continuity will be eliminated; if the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity are less than or equal to the second threshold, the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity will be retained, and the corresponding water quality parameters and / or water flow parameters of the retained normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity will be used as input data.

[0058] Step 3: Obtain the training set from the database to train the numerical model, obtaining a trained numerical model. Extract the latest input data from the database and input it into the trained numerical model to output the hydrodynamic conditions and monitor the hydrodynamic environment.

[0059] Use 70% of the input data in the database as the training set, 15% as the validation set, and 15% as the test set. Train the model with the training set, and validate and test the model with the validation set and the test set. The model that passes the validation and testing is the trained numerical model.

[0060] The training set in this embodiment includes water quality parameters (T, S) and water flow parameters (pH, V x , V y ) under different time series, as well as the hydrodynamic conditions corresponding to different water quality parameters and water flow parameters. Use the hydrodynamic conditions as the training labels.

[0061] Obtain water quality parameters and / or water flow parameters in real time, preprocess them according to this embodiment and store them in the database. Extract the latest input data from the database and input it into the trained numerical model to obtain the future hydrodynamic conditions and realize the monitoring of the hydrodynamic environment.

[0062] Step 4: Generate a warning report according to the output hydrodynamic conditions.

[0063] First, set the warning thresholds: The warning thresholds can be divided into multiple levels, such as the warning threshold, the alarm threshold, and the emergency threshold. Specifically: Temperature: Normal range (15 - 25°C), warning (25°C < Temperature T < 30°C or 10°C ≤ Temperature T < 15°C), alarm (Temperature T ≥ 30°C or Temperature T < 10°C); Salinity: Normal range (0 - 35 PSU), warning (35 PSU < Salinity S < 40 PSU or Salinity S < 0 PSU), alarm (Salinity S ≥ 40 PSU) pH value: Normal range (6.5 - 8.5), warning (8.5 ≤ pH < 9 or pH < 6.5), alarm (pH ≥ 9) Water flow velocity: Normal range (0 - 2 m / s), warning (2 m / s < V ≤ 3 m / s), alarm (V > 3 m / s) Then, perform threshold judgment: Compare the predicted hydrodynamic conditions with the set thresholds to determine whether the thresholds are exceeded. Consider the abnormal conditions of multiple parameters comprehensively and generate a comprehensive warning report: When the temperature, salinity, pH value, and water flow velocity are all within the normal range thresholds, continue monitoring without warning.

[0064] When the temperature and / or salinity and / or pH value and / or water flow rate reaches the alarm threshold or warning threshold, comprehensively check the abnormality of each parameter, take measures to adjust the abnormal parameters, and continuously monitor the changes in abnormal parameters, record data, analyze the causes of the abnormalities, notify maintenance personnel and relevant departments, and take emergency measures.

[0065] like Figure 2 As shown, this embodiment also discloses a port hydrodynamic environment monitoring system, including the following modules: Acquisition module: used to collect the water quality parameters and / or water flow parameters of the port water area in real time through the fiber grating strain sensor; Preprocessing module: connected to the acquisition module, used to preprocess the water quality parameters and / or water flow parameters of the time series, and store the preprocessed water quality parameters and / or water flow parameters in the database; Monitoring module: connected with the preprocessing module, used to obtain the training set from the database to train the numerical model to obtain a trained numerical model, extract the latest input data from the database, input it into the trained numerical model, output the hydrodynamic conditions, and monitor the hydrodynamic environment; Early warning module: connected to the monitoring module, used to generate early warning reports based on the output hydrodynamic conditions.

[0066] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0067] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship 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 a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring port hydrodynamic environment, characterized in that: include: Step 1, using a fiber Bragg grating strain sensor to collect water quality parameters and / or water flow parameters of the port water area in real time; Step 2, preprocessing the water quality parameters and / or water flow parameters of the time series, and storing the preprocessed water quality parameters and / or water flow parameters in a database; The preprocessing at least includes the elimination of abnormal values ​​of each water quality parameter and / or abnormal value of water flow parameter in the time series: Step 21, constructing a first threshold value for screening and determining outliers; Step 22, determining whether the water quality parameter and / or water flow parameter in each time series is greater than a first threshold value, if so, defining the water quality parameter and / or water flow parameter in the current time series as a determined abnormal value, and removing the determined abnormal value; if not, defining the water quality parameter and / or water flow parameter in the current time series as a preset normal value, and retaining the preset normal value; Step 23, testing the abnormality of the preset normal value, specifically including: testing the time series continuity and spatial correlation of the preset normal value, optimizing the first threshold to obtain a second threshold, eliminating the preset normal value that fails the test, comparing the preset normal value that passes the test with the second threshold, if the preset normal value is greater than the second threshold, eliminating the preset normal value; if the preset normal value is less than or equal to the second threshold, retaining the preset normal value, and using the water quality parameter and / or water flow parameter corresponding to the retained preset normal value as input data; Step 3, obtain the training set from the database to train the numerical model, obtain the trained numerical model, extract the latest input data from the database, and input it into the trained numerical model, output the hydrodynamic conditions, and monitor the hydrodynamic environment; Step 4: Generate an early warning report based on the output hydrodynamic conditions.

2. A method for monitoring port hydrodynamic environment according to claim 1, characterized in that: The water quality parameters include at least the temperature, salinity and pH value at different monitoring points, and the water flow parameters include at least the water flow velocity at each monitoring point; Constructing the first threshold includes the following steps: Step 211, obtaining the temperature, salinity, pH value, and water flow velocity of each monitoring point in the current time series; Step 212, according to the temperature of each monitoring point in all time series, obtain the average temperature; according to the salinity of each monitoring point in all time series, obtain the average salinity; according to the pH value of each monitoring point in all time series, obtain the average pH; according to the water flow velocity of each monitoring point in all time series, obtain the average water flow velocity; Step 213, obtaining a first threshold value according to temperature, salinity, pH value, water flow velocity, average temperature value, average salinity value, average pH value, and average water flow velocity value.

3. A method for monitoring port hydrodynamic environment according to claim 2, characterized in that: The calculation formula of the first threshold is: ； Where T1 represents the first threshold, T k represents the temperature of the kth monitoring point, S k Represents the salinity of the kth monitoring point, pH k represents the pH value of the kth monitoring point, represents the component of the water velocity on the x-axis at the kth monitoring point, represents the component of the water velocity on the y-axis at the kth monitoring point, represents the average temperature, represents the average salinity, Represents the average pH value, represents the average value of the water velocity on the x-axis. represents the average value of the water velocity on the y-axis component, , Represent weight coefficients respectively, z represents the maximum number of monitoring points, Represents the combined deviation of temperature and salinity, Represents the comprehensive deviation of pH value and water flow rate.

4. A method for monitoring port hydrodynamic environment according to claim 3, characterized in that: Test the continuity of the time series of preset normal values, including: Step a, calculating a preset difference value based on a preset normal value in the current time series and a preset normal value in the previous time series; Step b, constructing a short-term fluctuation threshold; comparing the preset difference with the short-term fluctuation threshold, if the preset difference is greater than the short-term fluctuation threshold, and the preset difference is restored to within the short-term fluctuation threshold within the next n time series, then marking the preset normal value as short-term fluctuation data and retaining it; if the preset difference is greater than the short-term fluctuation threshold, and the preset difference is not restored to within the short-term fluctuation threshold within the next n time series, then marking the preset normal value as a final abnormal value and removing it; if the difference is less than or equal to the short-term fluctuation threshold, the preset normal value is a normal value and is retained; Step c, calculating the average value based on the preset normal value in the current time series and the preset normal values ​​in the previous m time series; calculating the difference between the preset normal value in the current time series and the average value; Step d, constructing a long-term trend threshold; comparing the gap value with the long-term trend threshold; if the gap value is greater than the long-term trend threshold and the gap value lasts for more than m+1 time series, the preset normal value is marked as the final abnormal value and removed; if the gap value is greater than the long-term trend threshold and the gap value recovers to the long-term trend threshold within the m+1 time series, or the gap value is less than or equal to the long-term trend threshold, it means that the preset normal value has a long-term trend, and the preset normal value is a normal value and is retained.

5. A method for monitoring port hydrodynamic environment according to claim 4, characterized in that: The calculation formula of the second threshold is: ； Where T2 represents the second threshold, T k represents the temperature of the kth monitoring point, S k Represents the salinity of the kth monitoring point, pH k represents the pH value of the kth monitoring point, represents the component of the water velocity on the x-axis at the kth monitoring point, represents the component of the water velocity on the y-axis at the kth monitoring point, represents the average temperature, represents the average salinity, Represents the average pH value, represents the average value of the water velocity on the x-axis. represents the average value of the water velocity on the y-axis component, represents the weight coefficient, and z represents the maximum number of monitoring points.

6. A method for monitoring port hydrodynamic environment according to claim 5, characterized in that: The spatial correlation of the preset normal values ​​is tested, including: Step e, obtaining a preset normal value of the current monitoring point; Step f, taking the current monitoring point as the center of the circle, and obtaining preset normal values ​​of neighboring monitoring points in a circular area with a radius x based on the center of the circle; Step g, calculating the average of the preset normal values ​​according to the preset normal values ​​of the current monitoring point and the preset normal values ​​of the adjacent monitoring points; Step h, calculating the difference between the predicted normal value of the current monitoring point and the mean of the preset normal value; Step i, construct a local consistency threshold; compare the difference value with the local consistency threshold, if the difference value is greater than the local consistency threshold, then mark the predicted normal value of the current monitoring point as the final abnormal value and remove it; if the difference value is less than or equal to the local consistency threshold, then the predicted normal value of the current monitoring point is a normal value and is retained; Step j, averaging the preset normal values ​​of all monitoring points to obtain a global average value; Step k, calculate the difference between the predicted normal value of the current monitoring point and the global average value to obtain the global difference; construct a global consistency threshold; compare the global difference with the global consistency threshold, if the global difference is greater than the global consistency threshold, mark the predicted normal value of the current monitoring point as the final abnormal value and remove it; if the global difference is less than or equal to the global consistency threshold, mark the predicted normal value of the current monitoring point as a normal value and retain it.

7. A method for monitoring port hydrodynamic environment according to claim 6, characterized in that: The water quality parameters and / or water flow parameters corresponding to the preset normal values ​​that have passed the test are used as input data, specifically: the preset normal values ​​obtained through the tests of steps a to d, and steps e to k, including: normal values ​​and short-term fluctuation data; the abnormal values ​​obtained through the tests of steps a to d, and steps e to k, including: the final abnormal value; Check whether the normal values ​​or short-term fluctuation data obtained by the test meet the spatial correlation and time series continuity. If not, the normal values ​​or short-term fluctuation data obtained by the test will be eliminated; if so, the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity will be compared with the second threshold. If the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity are greater than the second threshold, the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity will be eliminated; if the normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity are less than or equal to the second threshold, the normal values ​​that meet the spatial correlation and time series continuity are retained, and the corresponding water quality parameters and / or water flow parameters of the retained normal values ​​or short-term fluctuation data that meet the spatial correlation and time series continuity are used as input data.

8. A port hydrodynamic environment monitoring system, used to implement a port hydrodynamic environment monitoring method according to any one of claims 1 to 7, characterized in that: Includes the following modules: Acquisition module: used to collect the water quality parameters and / or water flow parameters of the port water area in real time through the fiber grating strain sensor; Preprocessing module: connected to the acquisition module, used to preprocess the water quality parameters and / or water flow parameters of the time series, and store the preprocessed water quality parameters and / or water flow parameters in the database; Monitoring module: connected with the preprocessing module, used to obtain the training set from the database to train the numerical model to obtain a trained numerical model, extract the latest input data from the database, input it into the trained numerical model, output the hydrodynamic conditions, and monitor the hydrodynamic environment; Early warning module: connected to the monitoring module, used to generate early warning reports based on the output hydrodynamic conditions.

Images