A real-time analysis method for tide observation data of a marine station

Abstract

The application discloses a kind of real-time analysis methods of ocean station tide level observation data, belong to marine observation field.The method includes the following steps: a, real-time acquisition tidal level data;B, carry out tidal level super extreme value test;Every time interval, extract current tidal level value, compare the tidal level value with the tidal level extreme value of history same period this month;Tidal level extreme value includes history same month maximum value and history same month minimum value;If current tidal level value is higher than history same month maximum value, or lower than history same month minimum value, then trigger alarm;C, carry out tidal level change anomaly test;D, carry out continuity anomaly test;E, carry out blockage siltation test.The application can analyze and test data according to physical law after obtaining data, judge whether data is abnormal or not, whether alarm is needed, and guide watchman to check and maintain observation equipment.The application can alarm in time when tidal level is abnormal, and will not misreport normal tidal level change.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of ocean observation, in particular to a real-time analysis method for tide observation data of an ocean station. BACKGROUND

[0002] Storm surge disaster is the most serious marine disaster in China, and tide observation data records the change of tide level during storm surge process, which is the most basic data for storm surge disaster prediction, loss assessment, disaster prevention and mitigation, etc. With the emphasis and continuous development of China's ocean observation, a large number of ocean stations have been built along the coast to carry out tide observation.

[0003] The tide observation of the ocean station is generally to build a tide well along the coast, to introduce seawater into the tide well after wave dissipation, and to obtain tide data by measuring the liquid level with an instrument. Due to the corrosion of seawater and the impact of strong storm surge disasters on observation facilities, the tide observation facilities are prone to failure, resulting in missing and abnormal observation data.

[0004] Since the tide observation data is prone to missing and abnormal, personnel are needed to inspect and control the quality of the data and observation equipment. However, most of the current coastal tide stations are unattended, and even if they are attended, it is not easy to monitor the data continuously for 24 hours. At the same time, some of the tide observation abnormal problems are more hidden, which cannot be directly distinguished by the on-duty personnel, and need detailed data analysis to find out.

[0005] There are some methods for alarm of abnormal tide observation data at present, which are mainly aimed at data integrity, i.e. mainly detecting whether the observation equipment is working normally, whether the data transmission line is interrupted, etc., and there is no method for checking the physical properties of the data, i.e. whether the data conforms to the physical law, whether it is reasonable, etc. SUMMARY

[0006] In view of the above technical problems, the present application provides a real-time analysis method for tide observation data of an ocean station.

[0007] The technical solution adopted by the present application is as follows:

[0008] A real-time analysis method for tide observation data of an ocean station, comprising the following steps:

[0009] a. Real-time acquisition of tide data;

[0010] b. Tide over-limit value test;

[0011] Every certain period of time, the current tidal level value is extracted, and the tidal level value is compared with the historical same period tidal level extreme value; the tidal level extreme value includes the historical same month maximum value and the historical same month minimum value; if the current tidal level value is higher than the historical same month maximum value or lower than the historical same month minimum value, an alarm is triggered;

[0012] c. Abnormality detection of tidal level change is performed;

[0013] c1. Tidal level values in the past N minutes at the current time are extracted, a tidal level value representing minute data, , a tidal level value representing the past N minutes, a minimum value of the tidal level value representing the past N minutes, a maximum value of the tidal level value representing the past N minutes, and a difference between the maximum value and the minimum value, ;

[0014] c2. If is higher than the upper threshold value, it indicates that the tidal level change has a jitter problem, and an alarm is triggered;

[0015] c3. If is lower than the lower threshold value, and at this time, neither is it within 1 hour of high tide, nor is it within 1 hour of low tide, it indicates that the tidal level change has a fixed problem, and an alarm is triggered;

[0016] d. Continuity abnormality detection is performed;

[0017] d1. Hourly tidal level data at the current time and in the past 4 hours are extracted, , where 0 represents the current time, -4~-1 represents the past 4 hours, a tidal level value representing observation;

[0018] d2. A fitting value of the -2 hour tidal level data is calculated using the tidal level observation values at -4, -3, -1 and 0 hours; the calculation formula of the fitting value is , a fitting value of the -2 hour, and an absolute difference value between the fitting value of the -2 hour and the tidal level observation value, ;

[0019] d3. If is higher than the set threshold value, it indicates that the tidal level data at the current time and in the past 4 hours has a discontinuity problem, and an alarm is triggered;

[0020] e. Blockage and siltation detection is performed;

[0021] e1. Storm surge data in the past 72 hours are extracted, a time series of surge data in the past 72 hours;

[0022] e2, to Fourier decomposition is performed on the time series to obtain the amplitude values of the sequence at different periods, i.e. wherein denotes a Fourier transform function, denotes the amplitude sequence after Fourier transform, denotes the amplitude of a sinusoidal curve with a period of hours;

[0023] e3, extract the amplitude value, compare the amplitude value with a threshold value, and if the amplitude value is higher than the threshold value, trigger an alarm.

[0024] The beneficial technical effects of the present application are as follows:

[0025] After obtaining the data, the present application can analyze and test the data according to physical laws, determine whether the data is abnormal and whether it needs to be alarmed, and guide the on-duty personnel to check and maintain the observation equipment. The present application can alarm in time when the tidal level is abnormal, and will not misreport the normal tidal level change.

[0026] Specifically, the present method has the following advantages:

[0027] 1) Unlike the conventional method of verifying data integrity only, the present application analyzes the characteristics of tidal level data based on physical laws, and can realize abnormal value detection and automatically trigger an alarm mechanism.

[0028] 2) The present application designs a method for judging the abnormality of tidal level change, which calculates the difference between the maximum and minimum values of the tidal level within 10 minutes to determine whether the test is abnormal.

[0029] 3) The present application designs a method for judging the abnormality of tidal level continuity, which calculates the difference between the fitting value and the observed value to determine whether the test is abnormal.

[0030] 4) The present application designs a method for judging the blockage and siltation of the tidal well, which calculates the amplitudes of 12-hour and 24-hour periods by Fourier transform to determine whether the test is abnormal.

[0031] 5) The present method has the advantages of high accuracy, simple and convenient operation, etc. DETAILED DESCRIPTION

[0032] Figure 1 is a flowchart of the real-time analysis method for tidal level observation data of the marine station of the present application;

[0033] Figure 2 is an example curve of tidal level jitter abnormality in a specific application example of the present application;

[0034] Figure 3For the specific application examples of the present application, the tide level is fixed and the example curve is shown in the figure;

[0035] Figure 4 For the specific application examples of the present application, the tide level is fixed and the example curve is shown in the figure;

[0036] Figure 5 For the specific application examples of the present application, the tide level is fixed and the example curve is shown in the figure;

[0037] Figure 6 For the specific application examples of the present application, the tide level is fixed and the example curve is shown in the figure; DETAILED DESCRIPTION

[0038] The tide level data includes:

[0039] (1) Tide level extreme value: the tide level value is higher than the maximum value of the historical same period, or lower than the minimum value of the historical same period. The possibility of tide level exceeding the extreme value is small, and when it appears, the instrument should be checked to determine whether it is real.

[0040] (2) Tide level abnormal change: one is that the tide level appears abnormal high-frequency up and down jitter, which may be caused by wave jitter due to insufficient wave breaking of the tide well; one is that the tide level is fixed at a value and does not change, which may be that the instrument measuring the liquid surface is stuck and cannot change with the tide level.

[0041] (3) Tide level continuity anomaly: the tide level curve is not smooth and continuous, and there is a systematic deviation, which may be that the instrument base surface has a problem.

[0042] (4) Tide well blockage and siltation: the tide well is blocked and silted, causing poor tide water inflow and outflow, which is manifested as tide level change lag, low tide not going down, and high tide not coming up.

[0043] The previous tide level detection method of the marine station only considers data integrity, only focuses on whether the tide level observation equipment is working and whether the data transmission is interrupted, and cannot check whether the observation facility is abnormal and whether the observation data is reasonable. In the unattended marine station, if the problem of tide level data cannot be found in time, it will lead to the loss of correct observation data. In addition, some observation equipment abnormal problems cannot be directly judged by the on-duty personnel observation, and need to be analyzed combined with historical data to distinguish.

[0044] Based on this, the present invention proposes a real-time analysis method for tide level observation data at marine stations. This method performs real-time analysis of tide level observation data by checking the data every minute, hour, and day. It makes real-time judgments on problems such as extreme tide levels, abnormal tide level changes, abnormal tide level continuity, and blockage and siltation of tide gauges. Furthermore, an identification algorithm can be designed to automatically detect data anomalies and promptly remind marine station staff to inspect and repair the equipment to restore normal data observation as soon as possible.

[0045] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0046] like Figure 1 As shown, a method for real-time analysis of tide level observation data at a marine station includes the following steps:

[0047] a. Obtain tide level data in real time;

[0048] The tidal data includes tidal values, astronomical tidal values, and storm surge data. Tidal values ​​are obtained from tide gauge instruments, astronomical tidal values ​​are calculated based on historical data harmonic analysis, and storm surge data is the difference between tidal values ​​and astronomical tidal values.

[0049] b. Conduct extreme tide level tests;

[0050] Every minute, the current tide level is retrieved and compared with the historical extreme tide levels for the same month. Extreme tide levels include the historical maximum and minimum tide levels for the same month. If the current tide level is higher than the historical maximum or lower than the historical minimum, an alarm is triggered; the on-duty personnel are notified with the message: "Tide level is higher than the historical maximum or lower than the historical minimum."

[0051] c. Conduct anomaly detection of tidal level changes;

[0052] This is performed once per minute. There are two types of abnormal tide level changes: one is a rapid fluctuation in the tide level within a short period, and the other is a stable tide level over a period of time. The specific steps are as follows:

[0053] c1. Extract the tide level values ​​of the past 10 minutes from the current moment. This represents the tide level value for each minute. , This indicates the tide level over the past 10 minutes. This represents the minimum tide level over the past 10 minutes. This represents the maximum tide level over the past 10 minutes. The difference between the maximum and minimum values ​​is calculated. .

[0054] c2, if If the tide level is higher than the upper limit threshold, it indicates that there is a problem with the tide level fluctuation, triggering an alarm; the on-duty personnel will be notified and a message will be displayed: "Abnormal fluctuation in tide level has occurred."

[0055] c3, if If the tide level is below the lower threshold and is neither within one hour of high tide nor within one hour of low tide, it indicates that the tide level is fixed and triggers an alarm; the on-duty personnel are notified and prompted that the tide level is fixed.

[0056] Specifically, when determining whether a tide level is fixed or fluctuating abnormally, further criteria must be established, namely, it must be distinguished from normal conditions. Normally, the tide level remains constant at the peak of high and low tides, and during slack tides; this is a normal phenomenon. The key distinction lies in the timing, which must occur outside of high and low tides.

[0057] Therefore, if ∆Tmin is lower than the lower limit threshold, it indicates that there may be a problem of fixed tide level. However, it should be distinguished from the normal high tide, low tide and slack tide periods. Therefore, there should be a judgment condition. That is, if it is neither within 1 hour of high tide nor within 1 hour of low tide, it indicates that the tide level change is fixed and triggers an alarm.

[0058] The upper and lower thresholds are determined based on historical tide data statistics; the historical tide data is calculated every minute. To construct a cumulative probability curve, select the 0.5% and 99.5% quantiles as the upper and lower thresholds, respectively.

[0059] d. Perform continuity anomaly testing;

[0060] This is performed hourly. The test primarily targets hourly tidal data to check for rapid changes or discontinuities. The specific steps are as follows:

[0061] d1. Extract hourly tide data for the current time and the previous 4 hours. , 0 represents the current time, and -4 to -1 represents the previous 4 hours. This indicates the observed tide level value.

[0062] d2. Using the tide level observations at -4, -3, -1, and 0 hours, calculate the fitted value of the tide level data for the -2nd hour; the formula for calculating the fitted value is: , This represents the fitted value at hour -2. Calculate the absolute difference between the fitted value at hour -2 and the observed tide level. .

[0063] d3, if If the value is higher than the set threshold, it indicates that there is a discontinuity problem in the current time and the tidal level data of the previous 4 hours, triggering an alarm; informing the on-duty personnel and prompting: the hourly tidal level value is discontinuous.

[0064] The set threshold is determined according to historical tidal level data statistics; the hourly of the historical tidal level data is calculated, the cumulative probability curve is made, and the 99.5% quantile is selected as the set threshold.

[0065] e. Perform the blockage and siltation test;

[0066] Select a fixed time every day for testing. When the tide well appears blockage and siltation, the storm surge of the tidal level will appear unreasonable periodic changes, and this place mainly tests whether the problem occurs. The specific steps are as follows:

[0067] e1. Extract the storm surge data of the previous 72 hours, which represents the time series of the surge data in the past 72 hours.

[0068] e2. Fourier decomposition is performed on the time series to obtain the amplitude values of the sequence at different periods, that is, where represents the Fourier transform function, represents the amplitude sequence after Fourier transform, represents the amplitude of the sinusoidal curve with a period of hours.

[0069] e3. Extract the amplitude values of 12 hours and 24 hours, and . Compare and with the threshold value, if one of them is higher than the threshold value, an alarm is triggered, informing the on-duty personnel and prompting: there is a blockage and siltation problem in the tide well.

[0070] The threshold value at this place can be determined according to experience, and the threshold value is generally selected as 10 cm.

[0071] The present application can analyze and test the tidal level data in real time, can be continuously operated for 24 hours, and can be applied to unmanned marine stations, greatly reducing the workload of the on-duty personnel and improving the data quality.

[0072] The present application will be further described below in combination with specific application examples. Because the discrimination method of the tidal level over-limit value problem is relatively clear and intuitive, only other problems are exemplified here.

[0073] I. Tidal level change anomaly test;

[0074] A marine station calculates the ​, the cumulative probability curve is made, and the quantiles of 0.5% and 99.5% are 5 cm and 20 cm respectively.

[0075] (1) As shown in Figure 2 , the marine station assumes that the current time is 15:03, and conducts a minute-by-minute test of abnormal changes in tidal level. The tidal level data for the past ten minutes (14:54-15:03) are 172, 173, 174, 174, 175, 176, 208, 480, 545, and 489, respectively, so , At this time , the on-duty personnel are notified and prompted that the tidal level has appeared abnormal jitter.

[0076] (2) As shown in Figure 3 , the marine station assumes that the current time is 15:38, and conducts a minute-by-minute test of abnormal changes in tidal level. The tidal level data for the past ten minutes (15:29-15:38) are 323, 323, 324, 324, 324, 324, 324, 324, 324, and 325, respectively, so , , .

[0077] According to the astronomical tide calculation, the high and low tide times on the current day are 00:03, 05:30, 12:13, and 17:13, and the current time is 95 minutes away from the nearest time. Therefore, at this time and more than 1 hour away from the high tide time / low tide time, the alarm is triggered, the on-duty personnel are notified, and it is prompted that the tidal level value is fixed and unchanged.

[0078] At the same time, it should be noted that at 05:24 of the marine station , but at this time, it is only 6 minutes away from the high tide time and less than 1 hour, so the alarm will not be triggered.

[0079] Second, the continuity test of tidal level;

[0080] As shown in Figure 4 , at 17:00 of a certain day, the marine station conducts a one-hour test of tidal level continuity. At this time, the tidal levels from 13:00 to 17:00 are , , , , The fitted value of the third hour is , so the alarm is triggered, the on-duty personnel are notified, and it is prompted that the hourly tidal level value is discontinuous.

[0081] Three, clogging and siltation test;

[0082] As shown in Figure 5As shown, a marine station conducted a blockage and siltation inspection at 00:00 on September 2nd. The time series of water level increases for a total of 72 hours, from August 30th to September 1st, was obtained. ,right Fourier decomposition To obtain amplitude sequences with different periods Plot the sequence curve. The curve shows that... 10, 10. If the conditions are met, an alarm will be triggered, and the on-duty personnel will be notified with a message indicating that the tide gauge well is blocked or silted up.

[0083] like Figure 6 As shown, a certain marine station began its siltation and blockage inspection at 00:00 on November 7th. The amplitude curve shows that... 10, 10. This curve represents a normal storm surge process and does not trigger an alarm.

[0084] For any parts not mentioned above, existing technologies can be adopted or referenced.

[0085] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for real-time analysis of tide level observation data at marine stations, characterized in that... Includes the following steps: a. Obtain tide level data in real time; b. Conduct extreme tide level tests; Every so often, the current tide level is extracted and compared with the historical extreme tide levels for the same period. The extreme tide levels include the historical maximum and minimum tide levels for the same month. If the current tide level is higher than the historical maximum or lower than the historical minimum tide level for the same month, an alarm is triggered. c. Conduct anomaly detection of tidal level changes; c1. Extract the tide level values ​​from the past N minutes at the current time. This represents the tide level value for each minute. , This represents the tide level over the past N minutes. This represents the minimum tide level over the past N minutes. This represents the maximum tide level over the past N minutes. The difference between the maximum and minimum tide levels is calculated. ; c2, if If the water level is higher than the upper limit threshold, it indicates that there is a fluctuation in the tide level, triggering an alarm. c3, if If the water level is below the lower threshold and is neither within one hour of high tide nor within one hour of low tide, it indicates that the water level is changing in a fixed manner, triggering an alarm. d. Perform continuity anomaly testing; d1. Extract hourly tide data for the current time and the previous 4 hours. , 0 represents the current time, and -4 to -1 represents the previous 4 hours. Indicates the observed tide level value; d2. Using the tide level observations for the previous -4, -3, -1, and 0 hours, calculate the fitted value of the tide level data for the -2nd hour; the formula for calculating the fitted value is as follows: , This represents the fitted value at hour -2. Calculate the absolute difference between the fitted value at hour -2 and the observed tide level. ; d3, if If the tide level is higher than the set threshold, it indicates that there is a discontinuity in the tide level data for the current time and the previous 4 hours, triggering an alarm. e. Conduct a blockage and siltation inspection; e1. Extract storm surge data from the previous 72 hours. This represents a time series of water level increase data over the past 72 hours. e2, to The time series is subjected to Fourier decomposition to obtain the amplitude values ​​of the series at different periods, i.e. ,in Represents the Fourier transform function. This represents the amplitude sequence after the Fourier transform. The period is represented as The amplitude of the sine curve in hours; e3. Extract the amplitude value and compare it with the threshold. If the amplitude value is higher than the threshold, trigger an alarm.

2. The method for real-time analysis of ocean station tide level observation data according to claim 1, characterized in that, In step a: the tide data includes tide value, astronomical tide value and storm surge data; the storm surge data is the difference between tide value and astronomical tide value.

3. The method for real-time analysis of ocean station tide level observation data according to claim 1, characterized in that, In step b: a tide level extreme value test is performed every minute; the current tide level value is extracted every minute and compared with the historical extreme tide level values ​​for the same period. After the alarm is triggered, the on-duty personnel will be notified and prompted that the tide level is either higher than the historical maximum for the same month or lower than the historical minimum for the same month.

4. The method for real-time analysis of ocean station tide level observation data according to claim 1, characterized in that, In step c1: N is 10; In step c2: After the alarm is triggered, notify the on-duty personnel and indicate: The tide level is fluctuating abnormally; In step c3: After the alarm is triggered, notify the on-duty personnel and indicate that the tide level remains unchanged.

5. The method for real-time analysis of ocean station tide level observation data according to claim 1, characterized in that, In step c: the upper and lower thresholds are determined based on historical tide data statistics; the historical tide data is calculated every minute. To construct a cumulative probability curve, select the 0.5% and 99.5% quantiles as the upper and lower thresholds, respectively.

6. The method for real-time analysis of ocean station tide level observation data according to claim 1, characterized in that, In step d3: the set threshold is determined based on historical tide data statistics; the historical tide data is calculated hourly. To create a cumulative probability curve, the 99.5th percentile was selected as the threshold. After the alarm is triggered, the on-duty personnel will be notified and a message will be displayed: The hourly tide level is discontinuous.

7. The method for real-time analysis of ocean station tide level observation data according to claim 1, characterized in that, In step e3: extract the amplitude values ​​for 12 hours and 24 hours. and ; Will and The system is compared with a threshold. If one of the values ​​exceeds the threshold, an alarm is triggered, notifying the on-duty personnel and indicating that the tide gauge well is clogged or silted up.

Images