Construction area sea wave prediction platform based on multi-dimensional data analysis

Abstract

The invention belongs to the technical field of sea wave parameter prediction, and particularly relates to a construction area sea wave prediction platform based on multi-dimensional data analysis, which undertakes a sea wave significant wave height data prediction task of a target construction area in a future preset time period on the current day. On the other hand, the sea wave significant wave height monitoring value of the future preset time period corresponding to each historical reference day is corrected, and the sea wave significant wave height of the future preset time period of the current day in the target construction area is predicted at a time; according to the method, the sea wave significant wave height of the target construction area in the future preset time period on the day is secondarily predicted, the adaptive weight of the primary prediction and the secondary prediction is determined, the comprehensive prediction value of the sea wave significant wave height of the target construction area in the future preset time period on the day is obtained and fed back, effective integration of historical reference and instant prediction is achieved, and the accuracy and reliability of a prediction result are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of ocean wave parameter prediction, and in particular relates to an ocean wave prediction platform for a construction area based on multidimensional data analysis. Background Art

[0002] The change of sea waves has a significant impact on the safety of the construction area, the progress of the project and the economic benefits. Especially in the fields of offshore engineering, port construction, offshore platform installation, etc., factors such as the intensity, frequency and period of the sea waves directly determine the feasibility and safety of the construction. Accurately predicting the sea wave conditions in the construction area is crucial to the smooth progress of the project.

[0003] The effective wave height of waves is one of the key indicators commonly used in the fields of marine engineering, shipping, fisheries, meteorological warning, etc. It not only directly affects the safety of marine operations, ship navigation, offshore engineering construction and other activities, but its prediction accuracy also plays an irreplaceable role in ensuring personnel safety, optimizing resource allocation and reducing operating costs. Therefore, the accurate prediction of the effective wave height of waves is of great significance.

[0004] In the prior art, there are also some solutions related to the prediction of significant wave height of ocean waves, such as the method for predicting significant wave height of ocean waves based on ERA-Interim, which has a Chinese patent publication number of CN104021434A. The method obtains the original data, performs data preprocessing, selects the appropriate sea level pressure field, uses the data of the first time period in ERA-Interim to calibrate the prediction model, uses the data of the second time period in ERA-Interim that is later than the first time period to evaluate the prediction model, and uses the prediction model to predict the significant wave height of ocean waves. By using the long-term stable ERA-Interim reanalysis data source of the European Center for Mesoscale Weather Prediction, data for predicting significant wave height of ocean waves are extracted from it, and then supplemented by the principal component analysis method, not only can the significant wave height of ocean waves for multiple times be predicted, but also the operability is strong and the prediction accuracy is high.

[0005] Another Chinese patent publication number CN104050514A is a method for predicting the long-term trend of significant wave height based on reanalysis data, and its steps include: 1. Collect ERA-Interim meteorological forecast data at each time, 2. Obtain the coordinates of each grid point, 3. Calculate the SLP anomaly value and standard deviation, 4. SLP anomaly value principal component analysis, 5. Box-Cox transformation of sea area data, 6. Calculate the prediction factor of significant wave height, 7. Calculate the standard deviation of significant wave height and prediction factor, 8. Bring the prediction factor into the prediction model, 9. Bring the significant wave height lag value into the model, 10. SLP field prediction based on EOF, 11. Optimize and select prediction factors, 12. Model predicts significant wave height, 13. Evaluate the prediction level, 14. Calculate the long-term trend of significant wave height, and 15. Draw a long-term trend chart of significant wave height. The long-term trend of significant wave height at multiple times can be predicted with high accuracy.

[0006] Although the above scheme proposes some solutions for predicting the effective wave height of waves, the existing technology still has the following limitations, specifically: 1. The existing technology for predicting the effective wave height of waves in a preset period in the future usually relies on the same historical situation as a reference, but ignores the impact of the historical and current differences in seabed topography on wave characteristics, resulting in deviations between the prediction results and the actual situation.

[0007] 2. The existing technology has an instant prediction method for predicting the effective wave height of waves in a preset period of time in the future, which mainly introduces meteorological boundary conditions into the physical model constructed at the prediction location. However, it ignores the problem of meteorological prediction errors, which may reduce the accuracy of the prediction of the effective wave height of waves, and thus affect the safety, economy and feasibility of offshore operations.

[0008] 3. The existing technology has a method of predicting the effective wave height of waves by combining historical reference and instant prediction. However, there is a lack of effective analysis of the weight distribution between history and the present, which may not only lead to the prediction results failing to accurately reflect the actual situation of the waves, but also lead to the lack of adaptability of the prediction results in different sea areas and different time periods. Summary of the invention

[0009] In order to overcome the shortcomings of the background technology, an embodiment of the present invention provides a construction area wave prediction platform based on multi-dimensional data analysis, which can effectively solve the problems involved in the above-mentioned background technology.

[0010] The purpose of the present invention can be achieved through the following technical solutions: a construction area wave prediction platform based on multidimensional data analysis, including: a data prediction task acceptance module, a historical reference day retrieval module, a primary prediction module, a secondary prediction module, a comprehensive prediction value feedback module and a cloud database.

[0011] The data prediction task acceptance module is connected to the historical reference day retrieval module and the secondary prediction module respectively, the historical reference day retrieval module is connected to the primary prediction module, the primary prediction module and the secondary prediction module are both connected to the comprehensive prediction numerical feedback module, and the cloud database is connected to the historical reference day retrieval module and the secondary prediction module respectively.

[0012] The data prediction task acceptance module is used to accept the task of predicting the effective wave height data of waves in the target construction area in the future preset period of the day, and define the specified time before the future preset period as the inspection period.

[0013] The historical reference day retrieval module is used to extract the wave height monitoring data and meteorological monitoring data of the target construction area during the inspection period of the day, and to retrieve the reference days with similar data scenarios between the historical inspection period of the target construction area and the inspection period of the day, which are recorded as historical reference days.

[0014] The primary prediction module is used to extract the monitoring values ​​of the effective wave height of waves corresponding to each historical reference day in the preset future period, consider the difference in seabed topography between each historical reference day and the current day in the target construction area, correct the monitoring values ​​of the effective wave height of waves corresponding to each historical reference day in the preset future period, and then predict the effective wave height of waves in the preset future period of the target construction area on that day.

[0015] The secondary prediction module is used to construct a physical model of the seabed topography in the target construction area on the day, import the weather forecast data for the preset period in the future of the day to make an immediate prediction of the effective wave height, correct the forecast deviation of the weather conditions for the preset period in the future of the day, and then make a secondary prediction of the effective wave height in the target construction area for the preset period in the future of the day.

[0016] The comprehensive prediction value feedback module is used to determine the adaptation weights of the first and second predictions, and to accumulate the products of the effective wave heights of waves in the target construction area in the preset time period in the future on the same day with the first and second predictions and their corresponding adaptation weights, so as to obtain the comprehensive prediction value of the effective wave height of waves in the target construction area in the preset time period in the future on the same day and give feedback.

[0017] The cloud database is used to store preset overall data scenario similarity thresholds, preset permitted meteorological deviation coefficient thresholds, and preset meteorological parameter grade scoring rules.

[0018] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention retrieves reference days with similar data scenarios between the historical inspection period of the target construction area and the inspection period of the current day, and conducts a forecast based on them, thereby helping to more accurately capture the laws and trends of wave changes, making it closer to the actual wave changes, thereby improving the accuracy of historical reference forecasts.

[0019] (2) The present invention considers the difference in seabed topography between each historical reference day and the current day in the target construction area, and makes a refined correction to the monitoring value of the effective wave height of waves in the future preset period corresponding to each historical reference day, thereby realizing a one-time prediction of the effective wave height of waves in the future preset period in the target construction area on the current day, significantly improving the accuracy of the prediction and providing strong support for the safety planning of construction activities.

[0020] (3) The present invention is based on the real-time prediction of the effective wave height of ocean waves, corrects the forecast deviation of meteorological conditions in a preset period of time in the future on the same day, and then realizes the secondary prediction of the effective wave height of ocean waves in the preset period of time in the future on the same day in the target construction area. It not only improves the prediction accuracy and dynamic adaptability, but also optimizes the prediction process and decision support, and reduces the construction risk and cost.

[0021] (4) The present invention determines the adaptation weights of the primary prediction and the secondary prediction, combines the primary and secondary predictions of the effective wave heights of the target construction area in the preset period of time in the future on the same day, obtains and feeds back the comprehensive prediction value of the effective wave height of the target construction area in the preset period of time in the future on the same day, and effectively integrates the historical reference with the real-time prediction, thereby improving the accuracy and reliability of the prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0023] Figure 1 It is a schematic diagram of module connection of the present invention.

[0024] Figure 2 A logical schematic diagram for specifically retrieving each historical reference date for the present invention.

[0025] Figure 3 The present invention is a logical schematic diagram for correcting the forecast deviation of meteorological conditions for a preset time period in the future on the same day. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 are within the scope of protection of the present invention.

[0027] Reference Figure 1 As shown, the present invention provides a construction area wave prediction platform based on multidimensional data analysis, including: a data prediction task acceptance module, a historical reference day retrieval module, a primary prediction module, a secondary prediction module, a comprehensive prediction value feedback module and a cloud database.

[0028] The data prediction task acceptance module is connected to the historical reference day retrieval module and the secondary prediction module respectively, the historical reference day retrieval module is connected to the primary prediction module, the primary prediction module and the secondary prediction module are both connected to the comprehensive prediction numerical feedback module, and the cloud database is connected to the historical reference day retrieval module and the secondary prediction module respectively.

[0029] The data prediction task undertaking module is used to undertake the task of predicting the effective wave height data of waves in the target construction area in the future preset period of the day, and define the specified time before the future preset period as the inspection period.

[0030] The historical reference day retrieval module is used to extract the wave height monitoring data and meteorological monitoring data of the target construction area during the inspection period of the day, and retrieve each reference day with similar data context between the historical inspection period of the target construction area and the inspection period of the day, and record them as each historical reference day.

[0031] Specifically, the wave height monitoring data includes the wave height monitoring values ​​at each unit time point.

[0032] The meteorological monitoring data include average air pressure value, average temperature value, wind direction angle range, average wind speed value and total precipitation.

[0033] It should be noted that the above-mentioned wave height monitoring data can be obtained through shore wave measurement radar or buoy wave measurement, and the meteorological monitoring data and the meteorological forecast data mentioned below can be obtained through the meteorological station in the target construction area or meteorological satellite.

[0034] Reference Figure 2 As shown, specifically, the specific retrieval process of each historical reference day includes: taking each diary of the same season as the current day in each historical year as each historical concurrent day, extracting the wave height monitoring data and meteorological monitoring data in the inspection period of each historical concurrent day in the target construction area, respectively analyzing the similarity of the wave height data context and meteorological data context of each historical concurrent day inspection period relative to the current day inspection period, multiplying the similarity of the wave height data context with the similarity of the meteorological data context, and obtaining the overall data context similarity of each historical concurrent day inspection period relative to the current day inspection period; if the overall data context similarity of a historical concurrent day inspection period relative to the current day inspection period is greater than or equal to the preset overall data context similarity threshold stored in the cloud database, it means that the data context of the historical concurrent day inspection period of the target construction area is similar to that of the current day inspection period, and the historical concurrent day is used as the historical reference day to retrieve each historical reference day.

[0035] Specifically, the analysis of the similarity of the wave height data scenarios of the observation periods of the same historical days relative to the observation period of the current day includes: performing cosine similarity calculation on the wave height monitoring values ​​of the same historical days and each unit time point within the observation period of the current day, and obtaining the similarity of the wave height change trend of the same historical days relative to the observation period of the current day.

[0036] It should be noted that the specific calculation process for obtaining the similarity of the wave height change trend of each historical day relative to the current day's observation period is as follows: the wave height monitoring value of each unit time point in the observation period of each historical day is recorded as ,in is the number of each historical day in the same period, , is the number of each unit time point during the inspection period, , the monitored wave height values ​​at each unit time point during the day’s inspection period are recorded as , according to the formula Analyze the similarity of wave height change trends during the same period in history relative to the current day's observation period.

[0037] The survey period is divided into sub-periods, and the effective wave heights of each historical day and each survey sub-period on the current day are obtained. The effective wave height sequences of each historical day and the current day are constructed according to the chronological order of the survey sub-periods. The autocorrelation coefficients of the effective wave height sequences of each historical day and the current day at each preset lag order are calculated, and the similarity of the changing trends of the effective wave heights of each historical day relative to the current day during the survey period is compared and analyzed.

[0038] It should be noted that the specific process of obtaining the effective wave height of waves on the above-mentioned historical days and each investigation sub-period on the day is as follows: taking a certain investigation sub-period on a certain historical day as an example, the wave height monitoring values ​​of waves at each unit time point in the investigation sub-period on the historical day are arranged in order from small to large, and the wave height monitoring values ​​of waves in the last third of the order are averaged to obtain the effective wave height of waves in the investigation sub-period on the historical day. Similarly, the effective wave height of waves on each historical day and each investigation sub-period on the day is obtained.

[0039] It should be noted that the calculation process of the autocorrelation coefficients of the above-mentioned wave effective wave height series at each preset lag order for each historical day and the current day is as follows: taking a certain historical day as an example, for each element in the wave effective wave height series of the historical day Calculate the mean, is the number of each element in the effective wave height sequence, , and obtain the average effective wave height of the sub-period of the historical daily observation period , the significant wave height sequence of the same historical day is at the preset lag order The specific calculation formula of the autocorrelation coefficient is: ,in is the length of the effective wave height sequence, that is, the number of elements in the effective wave height sequence. The preset lag orders include , substitute into the above formula to obtain the autocorrelation coefficient of the significant wave height sequence of the historical same period at each preset lag order. Similarly, the autocorrelation coefficient of the significant wave height sequence of each historical same period and the current day at each preset lag order can be calculated.

[0040] It should also be noted that the specific process of the above-mentioned comparative analysis of the similarity of the trend of change of the effective wave height of waves in the same period of each historical day relative to the current day during the inspection period is as follows: subtract the autocorrelation coefficients of the effective wave height sequences of waves in the same period of each historical day from those of the current day at each preset lag order, convert the calculated difference into an absolute value and further perform a ratio analysis with the autocorrelation coefficients of the effective wave height sequence of waves in the current day at the corresponding preset lag order, and obtain the absolute deviation rate of the autocorrelation coefficients of the effective wave height sequence of waves in the same period of each historical day relative to the current day at each preset lag order, and obtain the average deviation rate of the trend of change of the effective wave height of waves in the same period of each historical day relative to the current day during the inspection period through mean calculation, and take its negative value and substitute it into the natural exponential function to obtain the similarity of the trend of change of the effective wave height of waves in the same period of each historical day relative to the current day during the inspection period.

[0041] The similarity of the wave height change trend is accumulated with the similarity of the wave effective wave height change trend to obtain the similarity of the wave height data scenarios of each historical daily observation period relative to the current day's observation period.

[0042] Specifically, the analysis of the similarity of the meteorological data contexts of each historical period of observation relative to the current day's observation period includes: calculating the relative deviation rates of the average air pressure value, average temperature value, wind direction angle range, average wind speed value, and total precipitation of each historical period of observation relative to the current day's observation period and accumulating them, taking the negative value of the accumulated value and substituting it into the natural exponential function to obtain the similarity of the meteorological data contexts of each historical period of observation relative to the current day's observation period.

[0043] It should be noted that the specific calculation process of the relative deviation rate of the above-mentioned each historical period of observation relative to the average air pressure value, average temperature value, wind direction angle range, average wind speed value, and total precipitation in the observation period of the same day is as follows: calculate the absolute difference between the average air pressure value of each historical period of observation relative to the inspection period of the same day, and perform ratio analysis on the absolute difference with the average air pressure value in the inspection period of the day to obtain the relative deviation rate of each historical period of observation relative to the average air pressure value in the inspection period of the day.

[0044] The calculation process of the relative deviation rate of the average temperature, average wind speed and total precipitation in each historical period relative to the average temperature, average wind speed and total precipitation in the current period is consistent with the calculation process of the relative deviation rate of the average air pressure in each historical period relative to the current period, and will not be elaborated here.

[0045] Obtain the non-overlapping intervals between the wind direction angle intervals in the observation periods of the same historical days and the wind direction angle intervals in the observation period of the day, and perform ratio analysis on the length of the non-overlapping intervals and the length of the wind direction angle intervals in the observation period of the day, so as to obtain the relative deviation rate of the wind direction angle intervals in the observation periods of the same historical days relative to the wind direction angle intervals in the observation period of the day.

[0046] The embodiment of the present invention retrieves reference days with similar data scenarios between the historical inspection period of the target construction area and the inspection period of the current day, and conducts a forecast based on this, so as to help more accurately capture the laws and trends of wave changes, make it closer to the actual wave changes, and thus improve the accuracy of historical reference forecasts.

[0047] The primary prediction module is used to extract the monitoring values ​​of the effective wave height of waves corresponding to the preset time period in the future on each historical reference day, take into account the difference in seabed topography between each historical reference day and the current day in the target construction area, correct the monitoring values ​​of the effective wave height of waves corresponding to the preset time period in the future on each historical reference day, and then predict the effective wave height of waves in the preset time period in the future on the target construction area.

[0048] Specifically, the correction of the monitoring values ​​of the effective wave height of each historical reference day corresponding to the preset period in the future includes: is the independent variable, is the number of each seafloor topographic parameter, , based on the effective wave height of the future preset period Construct a multiple linear regression model for the dependent variable, ,in For the The regression coefficients of the seafloor topography parameters are is the intercept, is the error value.

[0049] The monitoring values ​​of each seabed topography parameter on each historical reference day in the target construction area are collected, and combined with the corresponding effective wave height in the future preset time period, they are substituted into the multivariate linear regression model for fitting, and the intercept, error value, and regression coefficient corresponding to each seabed topography parameter are determined. In this way, the multivariate linear regression model is used as the correlation model between the seabed topography of the target construction area and the effective wave height of waves.

[0050] It should be noted that the above-mentioned seabed topography parameters include but are not limited to seabed slope, seabed roughness, etc. If the target construction area is a shallow sea area, it can be monitored by sensors installed at the bottom of the buoy. If the target construction area is a deep sea area, it can be monitored by ultrasonic detection equipment.

[0051] The monitoring values ​​of various seabed topography parameters in the target construction area on the day are collected, and the monitoring differences between various seabed topography parameters in the target construction area on each historical reference day and the current day are calculated. The differences are substituted into the correlation model between the seabed topography and the effective wave height of the waves in the target construction area to output the corrected value of the effective wave height of the waves under the influence of the difference in seabed topography between each historical reference day and the current day. The corrected value of the effective wave height of the waves under the influence of the difference in seabed topography between each historical reference day and the current day is output, and the corrected value of the effective wave height of the waves corresponding to the historical reference day in the preset future time period is accumulated, so as to realize the correction of the effective wave height monitoring value of the waves corresponding to the preset future time period on each historical reference day.

[0052] Specifically, the once predicted effective wave height of waves in the target construction area in the preset future time period on the same day is the calculated average of the corrected effective wave height monitoring values ​​of waves in the target construction area corresponding to the preset future time period on each historical reference day.

[0053] The embodiment of the present invention takes into account the differences in seabed topography between each historical reference day and the current day in the target construction area, and makes a fine correction to the monitoring values ​​of the effective wave height of waves corresponding to the future preset time period on each historical reference day, thereby realizing a one-time prediction of the effective wave height of waves in the target construction area in the future preset time period on the same day, significantly improving the accuracy of the prediction and providing strong support for the safety planning of construction activities.

[0054] The secondary prediction module is used to construct a physical model of the seabed topography of the target construction area on the day, import the weather forecast data for a preset period in the future of the day to make an immediate prediction of the effective wave height of the waves, correct the forecast deviation of the weather conditions for the preset period in the future of the day, and then secondary predict the effective wave height of the waves in the target construction area for the preset period in the future of the day.

[0055] It should be noted that the above-mentioned real-time prediction of effective wave height specifically adopts the computational fluid dynamics method, which is an existing mature technology and can be obtained through computational fluid dynamics software. The specific prediction process is: the meteorological forecast data for the preset time period in the future of the day is imported as the boundary condition into the constructed physical model of the seabed topography of the target construction area on that day, and the flow of seawater and the generation, propagation and deformation of waves are simulated by solving the Navier-Stokes equations, so as to obtain the real-time predicted wave height at each unit time point in the preset time period in the future of the day in the target construction area, and thus the real-time predicted wave effective wave height is calculated.

[0056] Reference Figure 3As shown, specifically, the correction of the forecast deviation of meteorological conditions for a preset time period in the future on the same day includes: randomly selecting meteorological monitoring data of a preset number of days in a preset historical period corresponding to a preset time period in the future to perform real-time forecast processing of effective wave height of waves, and obtaining the error values ​​of the real-time forecast of a preset number of days in the preset historical period relative to the actual monitored effective wave height of waves, which are recorded as the real-time forecast error values ​​of each historical reference day.

[0057] Meteorological monitoring data for each historical reference day corresponding to a preset time period in the future is collected. According to the preset meteorological parameter grade scoring rules stored in the cloud database, the average air pressure value, average temperature value, wind direction angle range, average wind speed value and total precipitation in the meteorological monitoring data are scored according to their corresponding levels and the scored values ​​are added up to obtain the meteorological monitoring score for each historical reference day corresponding to the preset time period in the future.

[0058] It should be noted that the above-mentioned preset meteorological parameter grade scoring rules specifically include the meteorological parameter monitoring value range and scoring value corresponding to each meteorological parameter grade, and each meteorological parameter specifically refers to air pressure parameter, temperature parameter, wind direction angle parameter, wind speed parameter and precipitation parameter.

[0059] Similarly, based on the weather forecast data for the preset time period in the future of the day, the weather forecast data includes the predicted average air pressure value, average temperature value, wind direction angle range, average wind speed value and total precipitation, to obtain the weather forecast score for the preset time period in the future of the day.

[0060] Calculate the deviation coefficient between the meteorological monitoring score of each historical reference day corresponding to the future preset time period and the meteorological forecast score of the future preset time period of the day, select the historical reference diaries whose deviation coefficient is less than or equal to the preset permitted meteorological deviation coefficient threshold stored in the cloud database as each historical reference day, extract the instantaneous prediction error value of each historical reference day for average calculation to obtain the prediction error value of the meteorological conditions for the future preset time period of the day, and add it to the instantaneous prediction of the effective wave height of the wave for the future preset time period of the day to correct the prediction deviation of the meteorological conditions for the future preset time period of the day.

[0061] It should be noted that the specific process of calculating the deviation coefficient between the meteorological monitoring score for the future preset time period corresponding to each historical reference day and the meteorological forecast score for the future preset time period on the current day is: subtract the absolute value of the meteorological monitoring score for the future preset time period corresponding to each historical reference day from the meteorological forecast score for the future preset time period on the current day, and perform a ratio analysis on the absolute difference with the preset reference score to obtain the deviation coefficient between the meteorological monitoring score for the future preset time period corresponding to each historical reference day and the meteorological forecast score for the future preset time period on the current day.

[0062] Specifically, the secondary prediction of the effective wave height of waves in the target construction area in the future preset period of the day is the cumulative value of the forecast error value of the meteorological conditions in the future preset period of the day and the effective wave height of waves predicted in real time in the future preset period of the day.

[0063] The embodiment of the present invention corrects the forecast deviation of meteorological conditions in a preset time period in the future on the same day based on the real-time forecast of the effective wave height of waves, and then realizes the secondary forecast of the effective wave height of waves in the preset time period in the future on the same day in the target construction area, which not only improves the forecast accuracy and dynamic adaptability, but also optimizes the forecast process and decision support, and reduces construction risks and costs.

[0064] The comprehensive prediction value feedback module is used to determine the adaptation weights of the primary prediction and the secondary prediction, accumulate the products of the primary and secondary predictions of the effective wave heights of waves in the target construction area in the preset time period in the future on the same day and their corresponding adaptation weights, obtain the comprehensive prediction value of the effective wave height of waves in the target construction area in the preset time period in the future on the same day and give feedback.

[0065] Specifically, the determination of the adaptation weights of the primary prediction and the secondary prediction includes: the analysis process of the effective wave heights of waves in the preset future time period of the same and secondary prediction target construction areas is consistent, the effective wave heights of waves in the preset future time period of each day in the preset historical period of the primary and secondary prediction target construction areas are compared with the effective wave heights of waves actually monitored in the preset future time period of each day in the preset historical period of the target construction area, the adaptation weights of the primary prediction and the secondary prediction of the preset future time period of each day in the preset historical period of the target construction area are determined, the adaptation weights of the primary prediction and the secondary prediction of the preset future time period of each day in the preset historical period of the target construction area are outlined with the date serial number as the horizontal coordinate and the adaptation weight as the vertical coordinate, and the best fitting function of the adaptation weight change curve is further obtained, and the best fitting function of the adaptation weight change curve is substituted into the date serial number of the day to determine the adaptation weights of the primary prediction and the secondary prediction of the day.

[0066] It should be noted that the specific process of determining the adaptation weights of the primary prediction and the secondary prediction corresponding to the future preset time period on each day within the historical preset period of the target construction area is: setting the adaptation weight of the primary prediction and the adaptation weight of the secondary prediction to satisfy the constraint condition that the cumulative value is 1, thereby generating a series of test groups, which contain the test adaptation weights corresponding to the primary prediction and the secondary prediction.

[0067] The products of the effective wave heights of waves in the preset future period on each day of the preset historical period of the target construction area predicted by the first and second predictions are respectively accumulated with the corresponding test adaptation weights set by each test group to obtain the comprehensive prediction values ​​of the preset future period on each day of the preset historical period of the target construction area calculated by each test group, and the absolute values ​​are subtracted from the effective wave heights of waves actually monitored on the corresponding days of the preset future period of the target construction area. The test adaptation weights of the first and second predictions set by the test group with the smallest absolute difference are selected to determine the adaptation weights of the first and second predictions of the preset future period on each day of the preset historical period of the target construction area.

[0068] It should also be noted that the best fitting function of the above-mentioned adaptation weight change curve can be obtained through the best fitting tool of Matlab software.

[0069] The embodiment of the present invention determines the adaptation weights of the primary prediction and the secondary prediction, combines the primary and secondary predictions of the effective wave heights of waves in the target construction area for a preset period of time in the future on the same day, obtains and feeds back a comprehensive prediction value of the effective wave height of waves in the target construction area for a preset period of time in the future on the same day, and effectively integrates historical references with real-time predictions, thereby improving the accuracy and reliability of the prediction results.

[0070] The cloud database is used to store preset overall data context similarity thresholds, preset permitted meteorological deviation coefficient thresholds, and preset meteorological parameter grade scoring rule contents.

[0071] The data sources in the cloud database of this embodiment are shown in Table 1 below.

[0072] Table 1 Detailed description of data sources in cloud database The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A construction area wave prediction platform based on multidimensional data analysis, characterized in that: include: The data prediction task acceptance module is used to accept the task of predicting the effective wave height data of the target construction area in the future preset period of the day, and define the specified time before the future preset period as the inspection period; The historical reference day retrieval module is used to extract the sea wave height monitoring data and meteorological monitoring data of the target construction area during the inspection period of the day, and retrieve the reference days with similar data scenarios between the historical inspection period of the target construction area and the inspection period of the day, and record them as historical reference days; The primary prediction module is used to extract the monitoring values ​​of the effective wave height of the waves corresponding to the preset time period in the future on each historical reference day, and to correct the monitoring values ​​of the effective wave height of the waves corresponding to the preset time period in the future on each historical reference day by considering the difference in the seabed topography between the target construction area on each historical reference day and the current day, so as to predict the effective wave height of the waves in the preset time period in the future on the target construction area on the current day. The secondary prediction module is used to construct a physical model of the seabed topography of the target construction area on the day, import the weather forecast data for the preset period in the future of the day to make an immediate prediction of the effective wave height, correct the forecast deviation of the weather conditions for the preset period in the future of the day, and then secondary predict the effective wave height of the target construction area for the preset period in the future of the day; The comprehensive prediction value feedback module is used to determine the adaptation weights of the first and second predictions, and to accumulate the products of the first and second predictions of the effective wave heights of the target construction area in the preset period in the future on the same day and the corresponding adaptation weights, so as to obtain the comprehensive prediction value of the effective wave height of the target construction area in the preset period in the future on the same day and give feedback; The cloud database is used to store preset overall data scenario similarity thresholds, preset permitted meteorological deviation coefficient thresholds, and preset meteorological parameter grade scoring rules.

2. The construction area wave prediction platform based on multidimensional data analysis according to claim 1, characterized in that: The wave height monitoring data includes the wave height monitoring values ​​at each unit time point; The meteorological monitoring data include average air pressure value, average temperature value, wind direction angle range, average wind speed value and total precipitation.

3. The construction area wave prediction platform based on multidimensional data analysis according to claim 2, characterized in that: The specific retrieval process of each historical reference day includes: taking each diary of the same season as the current day in each historical year as each historical concurrent day, extracting the wave height monitoring data and meteorological monitoring data in the inspection period of each historical concurrent day in the target construction area, respectively analyzing the similarity of the wave height data context and meteorological data context of each historical concurrent day inspection period relative to the current day inspection period, multiplying the similarity of the wave height data context with the similarity of the meteorological data context, and obtaining the overall data context similarity of each historical concurrent day inspection period relative to the current day inspection period; if the overall data context similarity of a certain historical concurrent day inspection period relative to the current day inspection period is greater than or equal to the preset overall data context similarity threshold stored in the cloud database, it means that the data context of the historical concurrent day inspection period of the target construction area is similar to that of the current day inspection period, and the historical concurrent day is used as the historical reference day to retrieve each historical reference day.

4. The construction area wave prediction platform based on multidimensional data analysis according to claim 3 is characterized by: The analysis of the similarity of the wave height data scenarios of the observation periods of the same historical days relative to the observation period of the current day includes: performing cosine similarity calculation on the wave height monitoring values ​​of the same historical days and each unit time point in the observation period of the current day to obtain the similarity of the wave height change trend of the same historical days relative to the observation period of the current day; The survey period is divided into sub-periods, and the effective wave heights of the waves on the historical days and the sub-periods of the current day are obtained. The effective wave height sequences of the waves on the historical days and the current day are constructed according to the time sequence of the sub-periods. The autocorrelation coefficients of the effective wave height sequences of the historical days and the current day at the preset lag orders are calculated, and the similarity of the changing trends of the effective wave heights of the waves on the historical days relative to the current day in the survey period is compared and analyzed. The similarity of the wave height change trend is accumulated with the similarity of the wave effective wave height change trend to obtain the similarity of the wave height data scenarios of each historical daily observation period relative to the current day's observation period.

5. The construction area wave prediction platform based on multidimensional data analysis according to claim 3 is characterized by: The analysis of the similarity of the meteorological data contexts of each historical period of observation relative to the current day's observation period includes: calculating the relative deviation rates of the average air pressure value, average temperature value, wind direction angle range, average wind speed value, and total precipitation of each historical period of observation relative to the current day's observation period and accumulating them, taking the negative value of the accumulated value and substituting it into the natural exponential function to obtain the similarity of the meteorological data contexts of each historical period of observation relative to the current day's observation period.

6. The construction area wave prediction platform based on multidimensional data analysis according to claim 1, characterized in that: The correction of the monitoring values ​​of the effective wave height of each historical reference day corresponding to the preset period in the future includes: is the independent variable, is the number of each seafloor topographic parameter, , based on the effective wave height of the future preset period Construct a multiple linear regression model for the dependent variable, ,in For the The regression coefficients of the seafloor topography parameters are is the intercept, is the error value; Collect the monitoring values ​​of each seabed topography parameter of the target construction area on each historical reference day, and substitute them into the multivariate linear regression model for fitting in combination with the effective wave height of the corresponding future preset period, and determine the intercept, error value, and the regression coefficient corresponding to each seabed topography parameter. In this way, the multivariate linear regression model is used as the correlation model between the seabed topography of the target construction area and the effective wave height of the wave; The monitoring values ​​of various seabed topography parameters in the target construction area on the day are collected, and the monitoring differences between various seabed topography parameters in the target construction area on each historical reference day and the current day are calculated. The differences are substituted into the correlation model between the seabed topography and the effective wave height of the waves in the target construction area to output the corrected value of the effective wave height of the waves under the influence of the difference in seabed topography between each historical reference day and the current day. The corrected value of the effective wave height of the waves under the influence of the difference in seabed topography between each historical reference day and the current day is output, and the corrected value of the effective wave height of the waves corresponding to the historical reference day in the preset future time period is accumulated, so as to realize the correction of the effective wave height monitoring value of the waves corresponding to the preset future time period on each historical reference day.

7. The construction area wave prediction platform based on multidimensional data analysis according to claim 6, characterized in that: The once-predicted effective wave height of waves in the target construction area for a preset period in the future on that day is the calculated average of the corrected effective wave height monitoring values ​​of waves in the target construction area for the preset period in the future on each historical reference day.

8. The construction area wave prediction platform based on multidimensional data analysis according to claim 2, characterized in that: The correction of the forecast deviation of meteorological conditions for a preset period of time in the future on the same day includes: randomly selecting meteorological monitoring data of a preset number of days in a preset historical period corresponding to a preset period of time in the future to perform real-time forecast processing of effective wave height of waves, and obtaining the error values ​​of the real-time forecast of the effective wave height of waves actually monitored relative to the actual monitoring on the preset number of days in the preset historical period, and recording them as the real-time forecast error values ​​of each historical reference day; Collect meteorological monitoring data for each historical reference day corresponding to a preset period in the future, and assign points to the average air pressure value, average temperature value, wind direction angle range, average wind speed value, and total precipitation in the meteorological monitoring data according to the preset meteorological parameter grade scoring rules stored in the cloud database, and add them up to obtain the meteorological monitoring score for each historical reference day corresponding to the preset period in the future; Similarly, based on the weather forecast data for the preset time period in the future of the day, the weather forecast score for the preset time period in the future of the day is obtained; Calculate the deviation coefficient between the meteorological monitoring score of each historical reference day corresponding to the future preset time period and the meteorological forecast score of the future preset time period of the day, select the historical reference diaries whose deviation coefficient is less than or equal to the preset permitted meteorological deviation coefficient threshold stored in the cloud database as each historical reference day, extract the instantaneous prediction error value of each historical reference day for average calculation to obtain the prediction error value of the meteorological conditions for the future preset time period of the day, and add it to the instantaneous prediction of the effective wave height of the wave for the future preset time period of the day to correct the prediction deviation of the meteorological conditions for the future preset time period of the day.

9. The construction area wave prediction platform based on multidimensional data analysis according to claim 8, characterized in that: The effective wave height of the waves in the target construction area for the second prediction in the preset time period in the future on the same day is the cumulative value of the forecast error value of the meteorological conditions in the preset time period in the future on the same day and the effective wave height of the waves predicted in real time in the preset time period in the future on the same day.

10. The construction area wave prediction platform based on multidimensional data analysis according to claim 1, characterized in that: The determination of the adaptation weights of the primary prediction and the secondary prediction includes: the analysis process of the effective wave heights of waves in the future preset time period of the same and secondary predicted target construction areas is consistent, the effective wave heights of waves corresponding to the future preset time period on each day within the historical preset period of the primary and secondary predicted target construction areas are compared with the effective wave heights of waves actually monitored on each day within the historical preset period of the target construction area to determine the adaptation weights of the primary prediction and the secondary prediction for the future preset time period on each day within the historical preset period of the target construction area, the adaptation weight change curve of the primary prediction and the secondary prediction for the future preset time period of the target construction area is outlined with the date serial number as the horizontal coordinate and the adaptation weight as the vertical coordinate, and the best fitting function of the adaptation weight change curve is further obtained, and the best fitting function is substituted into the date serial number of the day to determine the adaptation weights of the primary prediction and the secondary prediction for the day.

Images