Statistical processing intelligent correction method and system for marine hydro meteorological forecast

Abstract

The invention belongs to the technical field of weather forecast management, and particularly relates to a statistical processing intelligent correction method and system for marine hydro weather forecast, and the system comprises a multi-data source transmission processing module, a statistical analysis module, an intelligent correction module, a marine early warning issuing module, a tracing verification module and a weather supervision end. By integrating multi-source data acquisition, preprocessing, statistical analysis and machine learning correction, intelligent correction of marine hydrometeorological forecast data is realized, the forecast accuracy and reliability are remarkably improved, and marine meteorological early warning information is pushed to all users through the meteorological supervision end when the marine meteorological disaster is judged to be generated, so that the accuracy and reliability of the marine meteorological forecast data are improved. And the tracking verification module is used for carrying out tracking verification analysis on the marine meteorological early warning and reminding an administrator to take corresponding optimization measures when a marine meteorological early warning abnormal signal is generated, so that the accuracy of subsequent marine meteorological forecast is further ensured.

Description

Technical Field

[0001] The invention relates to the technical field of meteorological forecast management, and in particular to a statistical processing intelligent correction method and system for marine hydrological and meteorological forecasts. Background Art

[0002] Marine hydrological and meteorological forecasts have an important impact on marine operations, fishing labor, coastal residents' lives and industrial development. The Chinese invention patent with publication number CN118736791A discloses a marine hydrological and meteorological early warning system based on real-time positioning, which collects and analyzes meteorological information through an information collection unit and an analysis and processing unit, and makes early warning judgments based on the analysis results, so as to quickly display the marine meteorological analysis results and provide early warning prompts, thereby improving the safety of sea operations;

[0003] However, in the actual application process of the above invention technical solution, it is impossible to integrate multi-source data collection, preprocessing, statistical analysis and machine learning correction to intelligently correct the marine hydrological and meteorological forecast data, and it is difficult to accurately evaluate the accuracy of marine meteorological warnings and reasonably analyze the timeliness of warning push and the risk of the corresponding marine area when judging that the warning is more accurate, which is not conducive to the effective supervision of marine meteorological forecasts and has a low degree of intelligence.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a statistical processing intelligent correction method and system for marine hydrological and meteorological forecasts, which solves the problems that the prior art cannot integrate multi-source data acquisition, preprocessing, statistical analysis and machine learning correction to perform intelligent correction on marine hydrological and meteorological forecast data, and it is difficult to accurately evaluate the accuracy of marine meteorological warnings and reasonably analyze the timeliness of warning push and the risk of corresponding marine areas when judging that the warning is more accurate, and the degree of intelligence is low.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A statistical processing intelligent correction system for marine hydrological and meteorological forecasts, comprising a multi-data source transmission processing module, a statistical analysis module, an intelligent correction module, a marine warning release module, a traceability verification module and a meteorological supervision terminal; the multi-data source transmission processing module collects marine hydrological and meteorological data from multiple data sources including buoys, satellites, radars and ships in real time, performs preliminary processing on the collected marine hydrological and meteorological data, and sends the preliminary processed marine hydrological and meteorological data to the statistical analysis module;

[0008] The statistical analysis module uses statistical methods to perform trend analysis, correlation analysis and principal component analysis on the pre-processed marine hydrological and meteorological data, revealing the internal relationship and rules between the data, and sends the statistical analysis results to the intelligent correction module; the intelligent correction module constructs an intelligent correction model through machine learning algorithms, and the intelligent correction model performs intelligent correction on the forecast data according to the statistical analysis results, and sends the corrected forecast results to the marine warning release module;

[0009] The marine warning release module determines whether a marine meteorological disaster occurs based on the corrected forecast results. When it is determined that a marine meteorological disaster occurs, it generates marine meteorological warning information, and when the marine meteorological warning information is generated, it sends it to the meteorological supervision end, and the meteorological supervision end pushes the marine meteorological warning information to all users; the traceability verification module performs traceability verification on the marine meteorological warning, generates a marine meteorological warning qualified signal or a marine meteorological warning abnormal signal through analysis, and sends the marine meteorological warning qualified signal or the marine meteorological warning abnormal signal to the meteorological supervision end.

[0010] Furthermore, the preliminary processing of the ocean hydrological and meteorological data includes:

[0011] Data cleaning: clean the collected raw data to remove outliers and duplicate values; data denoising: use filtering technology to reduce noise interference in the data; data normalization: normalize data of different dimensions so that the data can be compared and analyzed at the same scale; missing value filling: for missing values ​​in the data, interpolation and mean filling methods are used to fill them.

[0012] Furthermore, the specific analysis process of the traceability verification module includes:

[0013] Set the number of days to be a tracing period of K1. When the number of days reaches K1, obtain all the marine meteorological warning information generated by the marine warning release module within the tracing period, and determine whether to assign the non-optimal forecast symbol XL-1 to the corresponding marine meteorological warning information through warning tracking detection and analysis;

[0014] The proportion of the generated number of marine meteorological warning information corresponding to the non-superior forecast symbol XL-1 during the retrospective period is marked as the non-superior forecast detection value, and the non-superior forecast detection value is numerically compared with the preset non-superior forecast detection threshold; if the non-superior forecast detection value exceeds the preset non-superior forecast detection threshold, a marine meteorological warning abnormal signal is generated.

[0015] Furthermore, if the non-optimal forecast detection value does not exceed the preset non-optimal forecast detection threshold, the tracking verification coefficient corresponding to the corresponding marine meteorological warning information is calculated by ratio with the corresponding preset tracking verification coefficient threshold to obtain the tracking verification value, and all tracking verification values ​​within the tracing period are averaged to obtain the forecast performance value, and the tracking verification value with the largest value within the tracing period is marked as the tracking verification outlier value;

[0016] The tracking evaluation value is obtained by numerically calculating the non-optimal forecast detection value, forecast performance value and tracking verification value, and the tracking evaluation value is numerically compared with the preset tracking evaluation threshold. If the tracking evaluation value exceeds the preset tracking evaluation threshold, a marine meteorological warning abnormal signal is generated; if the tracking evaluation value does not exceed the preset tracking evaluation threshold, a marine meteorological warning qualified signal is generated.

[0017] Furthermore, the specific analysis process of early warning tracking detection analysis is as follows:

[0018] When the marine warning release module generates marine meteorological warning information, warning tracking is performed, and the non-overlapping area ratio of the actual impact area of ​​the marine meteorological disaster involved in the corresponding marine meteorological warning information compared to the predicted impact area is marked as an overlapping anomaly, and the time difference between the predicted occurrence time and the actual occurrence time of the marine meteorological disaster involved in the corresponding marine meteorological warning information is calculated to obtain the predicted time deviation value;

[0019] The predicted intensity and predicted duration of the marine meteorological disasters involved in the corresponding marine meteorological warning information are obtained, and the deviation between the predicted intensity and the actual intensity is marked as the intensity non-accurate value, and the deviation between the predicted duration and the actual duration is marked as the duration detection value;

[0020] The tracking verification coefficient is obtained by numerically calculating the overlapping anomaly values, predicted time deviation values, intensity inaccuracy values ​​and duration deviation values. The tracking verification coefficient is numerically compared with the corresponding preset tracking verification coefficient threshold. If the tracking verification coefficient exceeds the preset tracking verification coefficient threshold, the non-excellent forecast symbol XL-1 is assigned to the corresponding marine meteorological warning information.

[0021] Furthermore, the traceability verification module is communicatively connected to the ocean area risk analysis module, and the traceability verification module sends the ocean meteorological warning qualified signal to the ocean area risk analysis module. When the ocean area risk analysis module receives the ocean meteorological warning qualified signal, it marks the ocean area to be supervised as the target area, and performs regional meteorological risk analysis on the corresponding target area. Through the analysis, it is determined whether a cautious travel signal for the corresponding target area is generated, and when a cautious travel signal for the corresponding target area is generated, it is sent to the meteorological supervision end.

[0022] Furthermore, the specific analysis process of regional meteorological risk analysis is as follows:

[0023] The number of marine meteorological warning information generated for the corresponding target area within the tracing period is obtained and marked as the marine meteorological warning value, and the marine meteorological warning value is numerically compared with the preset marine meteorological warning threshold. If the marine meteorological warning value exceeds the preset marine meteorological warning threshold, a cautious travel signal for the corresponding target area is generated.

[0024] Furthermore, if the marine meteorological warning value does not exceed the preset marine meteorological warning threshold, the marine meteorological disaster classification is performed based on all the marine meteorological warning information corresponding to the corresponding target area within the tracing period, and the number of marine meteorological warning information generated corresponding to the corresponding type of marine meteorological disaster is marked as the meteorological disaster inspection value, and each type of marine meteorological disaster is set in advance to correspond to a set of preset weight values, and the product of the meteorological disaster inspection value of the corresponding type of marine meteorological disaster and the corresponding preset weight value is marked as the meteorological disaster risk value;

[0025] The meteorological disaster risk values ​​of all types of marine meteorological disasters corresponding to the corresponding target area within the tracing period are summed up to obtain the marine area risk value, and the marine area risk value is numerically compared with the preset marine area risk threshold. If the marine area risk value exceeds the preset marine area risk threshold, a cautious travel signal is generated for the corresponding target area.

[0026] Furthermore, the traceability verification module is connected to the push timeliness evaluation module in communication, and the traceability verification module sends the qualified marine meteorological warning signal to the push timeliness evaluation module. When receiving the qualified marine meteorological warning signal, the push timeliness evaluation module analyzes the warning push timeliness performance of the meteorological supervision end within the traceability period, and determines whether to generate a push management abnormality signal through analysis, and sends the push management abnormality signal to the meteorological supervision end when it is generated; the specific analysis process is as follows:

[0027] When generating marine meteorological warning information, the time when the meteorological supervision end receives the corresponding marine meteorological warning information is collected and marked as the receiving time, and the time when the meteorological supervision end pushes the corresponding marine meteorological warning information to the user is collected and marked as the pushing time, and the interval between the receiving time and the pushing time is marked as the pushing interval time;

[0028] All the interval times within the tracing period are obtained and their average is calculated to obtain the interval time table value, and the interval time is numerically compared with the preset interval time threshold. If the interval time exceeds the preset interval time threshold, the corresponding interval time is marked as the difference time, and the number of difference times within the tracing period is obtained and the ratio of it to the number of interval times is calculated to obtain the difference time frequency value;

[0029] The push interval table value and the push difference frequency value are numerically compared with the preset push interval table threshold and the preset push difference frequency threshold respectively. If the push interval table value or the push difference frequency value exceeds the corresponding preset threshold, a push management abnormal signal is generated.

[0030] Furthermore, the present invention also proposes a statistical processing intelligent correction method for marine hydrological and meteorological forecasts, comprising the following steps:

[0031] Step 1: Collect ocean hydrological and meteorological data from multiple data sources in real time, and perform preliminary processing on the collected ocean hydrological and meteorological data;

[0032] Step 2: Use statistical methods to conduct trend analysis, correlation analysis and principal component analysis on the pre-processed marine hydrological and meteorological data;

[0033] Step 3: Based on the statistical analysis results, intelligent correction is performed on the forecast data;

[0034] Step 4: judging whether a marine meteorological disaster occurs based on the corrected forecast result, and generating marine meteorological warning information when it is judged that a marine meteorological disaster occurs;

[0035] Step 5: When the marine meteorological warning information is generated, it is sent to the meteorological supervision end, and the meteorological supervision end pushes the marine meteorological warning information to all users.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. In the present invention, by integrating multi-source data collection, preprocessing, statistical analysis and machine learning correction, intelligent correction of marine hydrological and meteorological forecast data is realized, the accuracy and reliability of the forecast are significantly improved, and when it is judged that a marine meteorological disaster occurs, the marine meteorological warning information is pushed to all users, which is convenient for users to take countermeasures in time, and the marine meteorological warning is traced and verified by the traceability verification module. When the marine meteorological warning abnormal signal is generated, the administrator is reminded to take corresponding optimization measures, further ensuring the accuracy of subsequent marine meteorological forecasts, and the degree of intelligence is high;

[0038] 2. In the present invention, the qualified marine meteorological warning signal is sent to the marine area risk analysis module and the push timeliness assessment module through the traceability verification module. The marine area risk analysis module performs regional meteorological risk analysis on the corresponding target area, and reminds users to avoid going to the corresponding target area when generating a cautious travel signal for the corresponding target area, thereby ensuring user safety. The push timeliness assessment module analyzes the warning push timeliness performance of the meteorological supervision end within the traceability period, and reminds the administrator to conduct cause investigation and analysis and make reasonable improvement measures when generating a push management abnormality signal, thereby ensuring the subsequent rapid push of marine meteorological warning information and realizing effective supervision of marine meteorological forecasts. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0040] Figure 1 This is a system block diagram of Embodiment 1 of the present invention;

[0041] Figure 2 It is a system block diagram of Embodiment 2 and Embodiment 3 of the present invention;

[0042] Figure 3 This is a flow chart of the method of Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0043] 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.

[0044] Embodiment 1: Figure 1 As shown, the statistical processing intelligent correction system for marine hydrological and meteorological forecasts proposed by the present invention includes a multi-data source transmission processing module, a statistical analysis module, an intelligent correction module, a marine warning release module, a traceability verification module and a meteorological supervision terminal;

[0045] The multi-data source transmission processing module collects ocean hydrological and meteorological data in real time from multiple data sources including buoys, satellites, radars and ships (i.e., collects various marine environmental parameters, including wind speed, wind direction, temperature, humidity, rainfall, wave height, current speed, etc.), and uses automated collection technology to obtain ocean hydrological and meteorological data from various data sources in real time to ensure the timeliness and accuracy of the data, and performs preliminary processing on the collected ocean hydrological and meteorological data to improve the data quality, and sends the preliminary processed ocean hydrological and meteorological data to the statistical analysis module; the processing process of preliminary processing of ocean hydrological and meteorological data is as follows:

[0046] Data cleaning: clean the collected raw data, remove outliers and duplicate values, etc., to ensure the accuracy and consistency of the data; data denoising: use filtering technology or other denoising methods to reduce noise interference in the data and improve data quality; data normalization: normalize data of different dimensions so that the data can be compared and analyzed at the same scale; missing value filling: for missing values ​​in the data, use interpolation and mean filling methods to fill them to ensure data integrity.

[0047] The statistical analysis module uses statistical methods to perform trend analysis on the preprocessed marine hydrological and meteorological data (using time series analysis technology to perform trend analysis on the preprocessed data to reveal the changing patterns of the data over time), correlation analysis (calculating the correlation coefficients between different parameters and analyzing the correlation and dependency between them) and principal component analysis (using the principal component analysis method to reduce the dimensionality of the data and extract the main characteristic variables to provide a basis for subsequent machine learning correction), revealing the inherent relationships and patterns between the data, providing a basis for intelligent correction, and sending the statistical analysis results to the intelligent correction module.

[0048] The intelligent correction module builds an intelligent correction model through machine learning algorithms (such as random forests, support vector machines, neural networks, etc.). The intelligent correction model intelligently corrects the forecast data based on the statistical analysis results, and sends the corrected forecast results to the marine early warning release module, realizing the intelligent correction of marine hydrological and meteorological forecast data, and significantly improving the accuracy and reliability of the forecast.

[0049] The marine warning release module determines whether a marine meteorological disaster will occur based on the corrected forecast results. When it is determined that a marine meteorological disaster will occur, it generates marine meteorological warning information (including the type of marine meteorological disaster generated, the predicted generation time, the intensity, and other information). When the marine meteorological warning information is generated, it is sent to the meteorological supervision end. The meteorological supervision end pushes the marine meteorological warning information to all users to remind users to take timely response measures.

[0050] The traceability verification module performs traceability verification on the marine meteorological warning. It generates a qualified marine meteorological warning signal or a marine meteorological warning abnormal signal through analysis, and sends the qualified marine meteorological warning signal or the marine meteorological warning abnormal signal to the meteorological supervision end. It can verify the accuracy of the forecast and issue a warning in time to remind the administrator to take corresponding optimization measures, thereby ensuring the accuracy of subsequent marine meteorological forecasts. The specific analysis process of the traceability verification module is as follows:

[0051] When the marine warning release module generates marine meteorological warning information, warning tracking is performed, and the non-overlapping area ratio of the actual impact area of ​​the marine meteorological disaster involved in the corresponding marine meteorological warning information compared to the predicted impact area is marked as an overlapping anomaly, and the time difference between the predicted occurrence time and the actual occurrence time of the marine meteorological disaster involved in the corresponding marine meteorological warning information is calculated to obtain the predicted time deviation value;

[0052] The predicted intensity and predicted duration of the marine meteorological disasters involved in the corresponding marine meteorological warning information are obtained, and the deviation between the predicted intensity and the actual intensity is marked as the intensity non-accurate value, and the deviation between the predicted duration and the actual duration is marked as the duration detection value;

[0053] The tracking verification coefficient ZX is obtained by numerically calculating the overlap anomaly value FY, the prediction time deviation value XW, the intensity non-accurate value NW and the time detection deviation value SP through the formula ZX=(hy×FY+tp×XW+tu×NW+re×SP) / 4; wherein hy, tp, tu, and re are preset weight coefficients with values ​​greater than zero, and the larger the value of the tracking verification coefficient ZX, the less accurate the corresponding marine meteorological forecast result;

[0054] The tracking verification coefficient ZX is numerically compared with the corresponding preset tracking verification coefficient threshold. If the tracking verification coefficient ZX exceeds the preset tracking verification coefficient threshold, it indicates that the corresponding marine meteorological forecast result is inaccurate, and the non-excellent forecast symbol XL-1 is assigned to the corresponding marine meteorological warning information;

[0055] A tracing period with a number of days set to K1, preferably, K1=15; when the number of days reaches K1, all marine meteorological warning information generated by the marine warning issuance module within the tracing period is obtained, and the proportion of the generated number of marine meteorological warning information corresponding to the non-superior forecast symbol XL-1 within the tracing period is marked as a non-superior forecast detection value, and the non-superior forecast detection value is numerically compared with a preset non-superior forecast detection threshold; if the non-superior forecast detection value exceeds the preset non-superior forecast detection threshold, indicating that the accuracy of the marine meteorological forecast within the tracing period is poor, a marine meteorological warning abnormality signal is generated.

[0056] Furthermore, if the non-optimal forecast detection value does not exceed the preset non-optimal forecast detection threshold, the tracking verification coefficient corresponding to the corresponding marine meteorological warning information is calculated by ratio with the corresponding preset tracking verification coefficient threshold to obtain the tracking verification value, and all tracking verification values ​​within the tracing period are averaged to obtain the forecast performance value, and the tracking verification value with the largest value within the tracing period is marked as the tracking verification outlier value;

[0057] The non-optimal forecast detection value LP, forecast performance value WF and tracking verification value QN are numerically calculated by the formula GX=kp×LP+(wq×WF+mg×QN) / 2 to obtain the tracking evaluation value GX; wherein kp, wq, mg are preset weight coefficients with values ​​greater than zero, kp>wq>mg; and the larger the value of the tracking evaluation value GX, the worse the overall performance of the marine meteorological forecast accuracy during the tracing period;

[0058] The tracking evaluation value GX is numerically compared with the preset tracking evaluation threshold. If the tracking evaluation value GX exceeds the preset tracking evaluation threshold, it indicates that the accuracy of the marine meteorological forecast during the tracing period is generally poor, and a marine meteorological warning abnormality signal is generated; if the tracking evaluation value GX does not exceed the preset tracking evaluation threshold, it indicates that the accuracy of the marine meteorological forecast during the tracing period is generally good, and a marine meteorological warning qualified signal is generated.

[0059] Embodiment 2: Figure 2 As shown, the difference between this embodiment and the first embodiment is that the traceability verification module is communicatively connected to the marine area risk analysis module, and the traceability verification module sends the marine meteorological warning qualified signal to the marine area risk analysis module. When the marine area risk analysis module receives the marine meteorological warning qualified signal, it marks the marine area to be supervised as the target area, and performs regional meteorological risk analysis on the corresponding target area;

[0060] Through analysis, it is determined whether to generate a cautious signal for the corresponding target area. When the cautious signal for the corresponding target area is generated, it is sent to the meteorological supervision end. The meteorological supervision end pushes the cautious signal and the corresponding target area to all users, which can accurately feedback the high-risk areas on the ocean and push them to users in time to remind users to avoid going to the corresponding target area, thereby ensuring user safety. The specific analysis process of regional meteorological risk analysis is as follows:

[0061] The number of marine meteorological warning information generated corresponding to the corresponding target area during the tracing period is obtained and marked as the marine meteorological warning value, and the marine meteorological warning value is numerically compared with the preset marine meteorological warning threshold. If the marine meteorological warning value exceeds the preset marine meteorological warning threshold, it indicates that the meteorological risk of the corresponding target area during the tracing period is high, and a cautious travel signal for the corresponding target area is generated.

[0062] Furthermore, if the marine meteorological warning value does not exceed the preset marine meteorological warning threshold, the marine meteorological disaster classification (including storm surge, sea fog disaster, tsunami, etc.) is performed based on all the marine meteorological warning information corresponding to the corresponding target area within the tracing period, and the number of marine meteorological warning information corresponding to the corresponding type of marine meteorological disaster is marked as the meteorological disaster inspection value, and each type of marine meteorological disaster is set in advance to correspond to a set of preset weight values ​​greater than zero, and the higher the security risk brought by the corresponding type of marine meteorological disaster, the greater the value of the preset weight value that matches it; the product of the meteorological disaster inspection value of the corresponding type of marine meteorological disaster and the corresponding preset weight value is marked as the meteorological disaster risk value;

[0063] The ocean area risk value is calculated by summing up the meteorological disaster risk values ​​of all types of marine meteorological disasters corresponding to the corresponding target area during the tracing period, where the larger the ocean area risk value, the higher the overall meteorological risk of the corresponding target area; the ocean area risk value is numerically compared with the preset ocean area risk threshold. If the ocean area risk value exceeds the preset ocean area risk threshold, it indicates that the overall meteorological risk of the corresponding target area during the tracing period is high, and a cautious travel signal is generated for the corresponding target area.

[0064] Embodiment 3: Figure 2As shown, the difference between this embodiment and the first and second embodiments is that the traceability verification module is connected to the push timeliness evaluation module in communication, and the traceability verification module sends the qualified marine meteorological warning signal to the push timeliness evaluation module. When the qualified marine meteorological warning signal is received, the push timeliness evaluation module analyzes the warning push timeliness performance of the meteorological supervision end within the traceability period, and determines whether to generate a push management abnormality signal through analysis;

[0065] When a push management abnormality signal is generated, it is sent to the meteorological supervision end, which can reasonably analyze and accurately feedback the timeliness of the warning push for marine meteorological warning information within the traceability period, so as to remind the administrator to conduct cause investigation and analysis and make reasonable improvement measures to ensure the rapid subsequent push of marine meteorological warning information; the specific analysis process is as follows:

[0066] When generating marine meteorological warning information, the time when the meteorological supervision end receives the corresponding marine meteorological warning information is collected and marked as the receiving time, and the time when the meteorological supervision end pushes the corresponding marine meteorological warning information to the user is collected and marked as the pushing time, and the interval between the receiving time and the pushing time is marked as the pushing interval time; wherein, the larger the value of the pushing interval time, the less timely the push of the corresponding marine meteorological warning information is;

[0067] All the push interval times within the tracing period are obtained and their average is calculated to obtain the push interval time table value, and the push interval time is compared with the preset push interval time threshold. If the push interval time exceeds the preset push interval time threshold, it indicates that the corresponding push of the corresponding marine meteorological warning information is not timely, then the corresponding push interval time is marked as the push difference time, and the number of push difference times within the tracing period is obtained and the ratio of it to the number of push interval times is calculated to obtain the push difference time frequency value;

[0068] The push interval table value and the push difference frequency value are numerically compared with the preset push interval table threshold and the preset push difference frequency threshold. If the push interval table value or the push difference frequency value exceeds the corresponding preset threshold, it indicates that the push performance of marine meteorological warning information during the tracing period is worse overall, and it is more unfavorable for users to take corresponding response measures in time, then a push management abnormality signal is generated.

[0069] Embodiment 4: Figure 3 As shown, the difference between this embodiment and the first, second and third embodiments is that the statistical processing intelligent correction method for ocean hydrological and meteorological forecasts proposed by the present invention includes the following steps:

[0070] Step 1: Collect ocean hydrological and meteorological data from multiple data sources in real time, and perform preliminary processing on the collected ocean hydrological and meteorological data;

[0071] Step 2: Use statistical methods to conduct trend analysis, correlation analysis and principal component analysis on the pre-processed marine hydrological and meteorological data;

[0072] Step 3: Based on the statistical analysis results, intelligent correction is performed on the forecast data;

[0073] Step 4: judging whether a marine meteorological disaster occurs based on the corrected forecast result, and generating marine meteorological warning information when it is judged that a marine meteorological disaster occurs;

[0074] Step 5: When the marine meteorological warning information is generated, it is sent to the meteorological supervision end, and the meteorological supervision end pushes the marine meteorological warning information to all users.

[0075] The working principle of the present invention is as follows: when in use, the marine hydrological and meteorological data are collected from multiple data sources in real time through the multi-data source transmission processing module and preliminarily processed; the statistical analysis module performs trend analysis, correlation analysis and principal component analysis on the pre-processed marine hydrological and meteorological data; the intelligent correction module performs intelligent correction on the forecast data according to the statistical analysis results through the intelligent correction model, which is conducive to ensuring the accuracy of the marine meteorological forecast results; the marine early warning release module determines whether a marine meteorological disaster occurs based on the corrected forecast results, and generates marine meteorological early warning information when it is determined that a marine meteorological disaster occurs; the meteorological supervision end pushes the marine meteorological early warning information to all users, reminding users to take timely countermeasures to reduce losses and ensure user safety; and the marine meteorological early warning is traced and verified by the traceability verification module and the accuracy of the marine meteorological forecast is reasonably evaluated; when the marine meteorological early warning abnormal signal is generated, the administrator is reminded to take corresponding optimization measures, so as to further ensure the accuracy of subsequent marine meteorological forecasts.

[0076] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions. The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the contents of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that technicians in the relevant technical field can understand and use the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A statistical processing intelligent correction system for marine hydrological and meteorological forecasts, characterized in that: It includes a multi-data source transmission and processing module, a statistical analysis module, an intelligent correction module, a marine warning release module, a traceability verification module and a meteorological supervision terminal; the multi-data source transmission and processing module collects marine hydrological and meteorological data from multiple data sources in real time, and performs preliminary processing on the collected marine hydrological and meteorological data; The statistical analysis module uses statistical methods to perform trend analysis, correlation analysis and principal component analysis on the pre-processed marine hydrological and meteorological data, and sends the statistical analysis results to the intelligent correction module; The intelligent correction module builds an intelligent correction model through machine learning algorithms. The intelligent correction model performs intelligent correction on the forecast data based on the statistical analysis results. The marine warning release module determines whether a marine meteorological disaster occurs based on the corrected forecast results. When a marine meteorological disaster occurs, it generates marine meteorological warning information, and the meteorological supervision end pushes the marine meteorological warning information to all users. The traceability verification module performs traceability verification on the marine meteorological warning, generates a marine meteorological warning qualified signal or a marine meteorological warning abnormal signal through analysis, and sends the marine meteorological warning qualified signal or the marine meteorological warning abnormal signal to the meteorological supervision end.

2. The statistical processing intelligent correction system for ocean hydrological and meteorological forecast according to claim 1 is characterized in that: The processing of preliminary processing of ocean hydrological and meteorological data includes: Data cleaning: clean the collected raw data to remove outliers and duplicate values; data denoising: use filtering technology to reduce noise interference in the data; data normalization: normalize data of different dimensions so that the data can be compared and analyzed at the same scale; missing value filling: for missing values ​​in the data, interpolation and mean filling methods are used to fill them.

3. The statistical processing intelligent correction system for ocean hydrological and meteorological forecast according to claim 1 is characterized in that: The specific analysis process of the traceability verification module includes: All the marine meteorological warning information generated by the marine warning issuance module within the tracing period is obtained, and the warning tracking detection analysis is used to determine whether to assign the non-superior forecast symbol XL-1 to the corresponding marine meteorological warning information; the proportion of the generated number of marine meteorological warning information corresponding to the non-superior forecast symbol XL-1 within the tracing period is marked as the non-superior forecast detection value. If the non-superior forecast detection value exceeds the preset non-superior forecast detection threshold, a marine meteorological warning abnormal signal is generated.

4. The statistical processing intelligent correction system for ocean hydrological and meteorological forecast according to claim 3 is characterized in that: If the non-optimal forecast detection value does not exceed the preset non-optimal forecast detection threshold, the tracking evaluation value is obtained by numerically calculating the non-optimal forecast detection value, the forecast performance value and the tracking verification value. If the tracking evaluation value exceeds the preset tracking evaluation threshold, a marine meteorological warning abnormality signal is generated; If the tracking evaluation value does not exceed the preset tracking evaluation threshold, a qualified marine meteorological warning signal is generated.

5. The statistical processing intelligent correction system for ocean hydrological and meteorological forecast according to claim 3 is characterized in that: The specific analysis process of early warning tracking detection analysis is as follows: When the marine warning release module generates marine meteorological warning information, warning tracking is carried out. The tracking verification coefficient is obtained by numerically calculating the overlapping anomaly values, predicted time deviation values, intensity inaccuracy values ​​and duration deviation values. If the tracking verification coefficient exceeds the preset tracking verification coefficient threshold, the non-excellent forecast symbol XL-1 is assigned to the corresponding marine meteorological warning information.

6. The statistical processing intelligent correction system for ocean hydrological and meteorological forecasts according to claim 3 is characterized in that: The traceability verification module is communicatively connected to the ocean area risk analysis module. The traceability verification module sends the ocean meteorological warning qualified signal to the ocean area risk analysis module. When the ocean area risk analysis module receives the ocean meteorological warning qualified signal, it conducts regional meteorological risk analysis on the corresponding target area, and when a cautious travel signal for the corresponding target area is generated, it is sent to the meteorological supervision end.

7. The statistical processing intelligent correction system for ocean hydrological and meteorological forecasts according to claim 6 is characterized in that: The specific analysis process of regional meteorological risk analysis is as follows: The number of marine meteorological warning information generated for the corresponding target area within the tracing period is obtained and marked as the marine meteorological warning value. If the marine meteorological warning value exceeds the preset marine meteorological warning threshold, a cautious travel signal for the corresponding target area is generated.

8. The statistical processing intelligent correction system for ocean hydrological and meteorological forecasts according to claim 7 is characterized in that: If the marine meteorological warning value does not exceed the preset marine meteorological warning threshold, the meteorological disaster risk values ​​of all types of marine meteorological disasters corresponding to the corresponding target area within the tracing period will be summed up to obtain the marine area risk value. If the marine area risk value exceeds the preset marine area risk threshold, a cautious travel signal will be generated for the corresponding target area.

9. The statistical processing intelligent correction system for ocean hydrological and meteorological forecast according to claim 3 is characterized in that: The traceability verification module is communicated with the push timeliness assessment module. When receiving the qualified marine meteorological warning signal, the push timeliness assessment module analyzes the warning push timeliness performance of the meteorological supervision end within the traceability period. If the push interval table value or the push frequency value exceeds the corresponding preset threshold, a push management exception signal is generated and sent to the meteorological supervision end.

10. A statistical processing intelligent correction method for marine hydrological and meteorological forecasts, characterized in that: The method adopts the statistical processing intelligent correction system for ocean hydrological and meteorological forecasts as described in any one of claims 1-9.

Images