JP2007094569A - Prediction method for billow in port, prediction device for billow in port, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve prediction accuracy of a billow in a port in connection with a prediction method for a billow in a port, a prediction device for the billow in the port and a program. <P>SOLUTION: The prediction method for the billow in the port predicts at a certain time of day about a specific port, the billow in the port inside the port at appointed time of day after the time of day. It contains a procedure to compute a statistical relational expression 230 for the billow out of the port at outside of the port and the billow in the port; a procedure to estimate a estimation value 330 for the billow out of the port at the appointed time of day by numerical simulation using a billow projection model 310; and a procedure to compute a prediction value 340 of the billow in the port at the appointed time of day based on the statistical relational expression 230 and the estimation value 330 of the billow out of the port. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a harbor wave prediction method, a harbor wave prediction apparatus, and a program. More specifically, when a specified time later than the current time is specified in a specific port, the significant wave height, significant wave period, wave direction, direction spectrum, etc. of the waves in the port at this specified time are predicted. The present invention relates to a harbor wave prediction method, a harbor wave prediction apparatus, and a program.

  For harbor management, harbor construction, etc., highly accurate harbor wave prediction is required. In other words, whether port entry / exit, port berthing, cargo handling work, etc. are determined based on the magnitude of the waves. If the judgment is wrong, damage to port facilities, worker safety, etc. will be seriously affected. Effect. On the other hand, however, when safety is emphasized excessively, the use efficiency of the port and the ship is lowered, and the advantage of the ship that it is suitable for mass transportation is lost.

  On the other hand, as a wave prediction method, various attempts have been made to calculate a prediction value based on a physical model called a wave prediction model, and a certain effect has been achieved with respect to wave prediction for a certain wide range. However, the desired prediction accuracy is not obtained for the wave prediction in a limited area called a harbor.

  The reason for this is that, within the harbor, the reflection of waves by quay walls, changes in seabed topography, narrow changes in the target area, and local changes are rapid. Even if you design a wave prediction model that incorporates these unique factors in the port, there are many variables, so it takes a lot of time to acquire and input parameters. In addition, since the calculation load is high, it is not suitable for practical use in which the predicted value must be repeatedly calculated in a short time.

  In the following Patent Document 1, “various amplitude gain data regarding bay water oscillations in the bay with respect to long-period waves outside the bay are prepared, and long-period waves in the bay are estimated from this amplitude gain data”. Is described. However, the calculation of “periodic characteristics of amplitude amplification factor” disclosed in this document still requires many variable inputs and calculation with a high processing load. Further, the method described in Patent Document 1 does not describe wave direction prediction. Here, the wave conditions in the bay greatly depend on the wave direction because the influence of reflection and diffraction by breakwaters and structures differs depending on the wave direction. Therefore, in the above method, there is no mention of the influence of the wave direction of the waves outside the port which is important in predicting the waves in the harbor, and it is difficult to expect a practical prediction accuracy of the waves in the harbor.

  Patent Document 2 below discloses a method for predicting the generation of long-period waves based on a small number of variables related to typhoons. However, this method is effective during a special period during which typhoons have been affected, and is not generally applicable to normal harbor management.

Further, Patent Document 3 below also describes a wave prediction device in which GPV data is applied to a wave prediction model based on multiple regression analysis. However, the wave prediction apparatus disclosed in this document makes no mention of the condition of being in a harbor, and does not disclose matters that can contribute to improvement of wave prediction in a harbor.
JP 2002-149767 A JP 2003-203300 A JP 2003-315468 A

  As described above, wave prediction using a physical model such as a wave prediction model is not suitable for wave prediction in a narrow port although the wave prediction in a port is an important technical problem. For this reason, the development of a technology that can effectively predict the waves in a harbor with practical accuracy and speed is awaited.

  In order to solve the above-mentioned problem, as a first mode of the present invention, for a specific port, at a certain time, an in-port wave prediction method for predicting an in-port wave inside the port at a specified time after that time A statistical relational expression of the measured values of the sea waves outside the port and the sea waves actually measured at the same time as the sea waves actually measured outside the port from the designated time of the sea waves outside the port. The procedure to calculate, the procedure to estimate the estimated value of the wave outside the port at the specified time by the numerical simulation using the wave prediction model, and the wave in the port at the specified time based on the statistical relational expression and the estimated wave outside the port There is provided a harbor wave prediction method including a procedure for calculating a prediction value. The predicted value preferably includes the wave height and the wave period. This makes it possible to perform highly accurate harbor wave predictions based on highly accurate wave prediction values outside the harbor, without performing numerical simulations using harbor wave prediction models that require numerous variables and complicated calculations. .

  Further, according to one embodiment, in the harbor wave prediction method, the wave prediction initial value used in the numerical simulation is corrected based on the ratio of the default wave prediction initial value and the significant wave height of the wave GPV data. In addition. As a result, the wave prediction initial value can be corrected in consideration of the ocean wave from which the actual measurement value cannot be obtained, so that a highly accurate harbor outside wave estimation value can be obtained. Therefore, the accuracy of the in-harbour wave prediction made based on the estimated wave value outside the harbor is also improved.

  According to another embodiment, in the harbor wave prediction method, the wave prediction initial value used in the numerical simulation is further corrected based on the direction spectrum calculated from the actually measured value of the wave outside the harbor. Thereby, since the wave prediction initial value can be corrected with reference to a large number of waves outside the harbor, a highly accurate harbor outside wave estimation value can be obtained. Therefore, the accuracy of the in-harbour wave prediction made based on the estimated wave value outside the harbor is also improved.

  Further, according to another embodiment, the above-mentioned harbor wave prediction method is a procedure for calculating a statistical value of an error by comparing an estimated value in the past rather than a time and an actually measured value of a wave outside the harbor at a specified time of the estimated value. And a procedure for correcting the estimated value using the statistical value. Thereby, it is possible to predict the wave in the harbor with reference to a more accurate estimated value.

  Further, according to another embodiment, the above-described harbor wave prediction method compares the calculated predicted value calculated in the past with respect to the time and the actually measured value of the harbor wave at the specified time of the calculated predicted value, to reduce the error. It further includes a procedure for calculating a statistical value and a procedure for correcting a predicted value newly calculated using the statistical value. As a result, it is possible to obtain a more accurate predicted wave value in the harbor.

  According to another embodiment, the harbor wave prediction method further includes a procedure of correcting the statistical relational expression by accumulating the actual values of the wave outside the harbor and the wave inside the harbor in the past rather than the time. This makes it possible to predict harbor waves with higher accuracy over time.

  In addition, according to another embodiment, the harbor wave prediction method further includes a procedure for determining whether or not a ship can enter or leave a port, come in a berth, or perform a cargo handling operation based on a predicted value. As a result, since harbor management can be performed based on highly accurate harbor wave prediction, not only the use efficiency of the ship is improved, but also the safety management of the harbor is improved.

  Furthermore, as a second aspect of the present invention, there is provided an in-port wave prediction device for predicting an in-port wave inside a port at a specified time later than that time for a specific port. Statistics that calculate and store the statistical relational expression of the actual measured values of the sea waves outside the port and the actual values of the sea waves measured at the same time as the actual measured values of the sea waves outside the port before the designated time of the outside sea waves Port-related wave simulator that estimates the estimated value of off-shore waves at a specified time by a numerical simulation using a wave prediction model, and a specified time based on statistical relational expressions and estimated off-shore waves There is provided a harbor wave prediction apparatus including a predicted value generation unit that calculates a predicted value of a harbor wave in Japan. Thereby, the apparatus which performs the method which concerns on a 1st form is supplied.

  Furthermore, as a third aspect of the present invention, there is an in-port wave prediction program for predicting an in-port wave inside a port at a specified time after that time for a specific port. Calculating a statistical relational expression of the measured values of the waves outside the harbor and the measured values of the waves inside the harbor measured at the same time as the measured values of the waves outside the harbor before the designated time of the outside waves of the harbor, A step of estimating the estimated value of the wave outside the port at the specified time by a numerical simulation using the wave prediction model, and calculating the predicted value of the wave inside the port at the specified time based on the statistical relational expression and the estimated wave outside the port A harbor wave prediction program including steps is provided. As a result, it is possible to predict a wave in a harbor with high accuracy even if a general-purpose information processing apparatus or the like is used as well as a dedicated machine.

  It should be noted that the above summary of the invention does not enumerate all the necessary features of the present invention. Sub-combinations of these feature groups can also be an invention.

  Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the invention according to the claims. In addition, not all the combinations of features described in the embodiments are essential for the solving means of the invention.

  FIG. 1 is a diagram schematically showing the structure of a harbor wave prediction apparatus 10 capable of executing the harbor wave prediction method according to the above. As shown in the figure, the harbor wave prediction apparatus 10 is connected to an information processing apparatus 100 having an internal element for executing each calculation procedure described later so that data can be input to the information processing apparatus 100. And an input device such as a mouse 106 and a keyboard 107, and a display 108 capable of displaying information output from the information processing apparatus 100. In addition, the information processing apparatus 100 can read data from a recording medium 102 such as a disk or a magnetic disk with light. Furthermore, the information processing apparatus 100 is also connected to a communication line 104 that can exchange data with the outside.

  The information processing apparatus 100 includes a statistical relational expression generation unit 110 that generates a statistical relational expression 230 in accordance with a procedure described later, and an estimated value 330 of the offshore harbor by a numerical simulation using a physical model. The wave simulator 120, the storage unit 130 that stores the statistical relational expression 230 generated by the statistical relational expression generation unit 110, the estimated value 330 of the offshore wave output from the offshore wave simulator 120, and the statistics stored in the storage unit 130 The predicted value generation unit 140 that outputs the predicted value 340 of the wave in the harbor with reference to the general relational expression 230, and the predicted value that the predicted value generation unit 140 outputs refers to the entry / exit of the ship in the port, Each is provided with a determination unit 150 that determines whether or not berthing, cargo handling work, civil engineering work and the like are possible.

  Each internal element of the information processing apparatus 100 can be formed as hardware, but can also be formed by software read through the recording medium 102 or the communication line 104. Further, the read software may be retained only when processing is executed, or may be stored in the information processing apparatus 100 so that it can be reused.

  FIG. 2 is a diagram showing a data structure 20 of data handled here in order to explain the processing in the statistical relational expression generating unit 110 of the harbor wave prediction apparatus 10 described above. As shown in the figure, the statistical relational expression generating unit 110 includes an actual measurement value 210 of the off-harbour wave measured in the past and an actual measurement value 220 of the in-harbour wave measured at the same time as the actual measurement value 210 of the off-shore wave. Are input over a period of time. These measured values 210 of the off-harbour waves and the measured values of the off-harbour waves are input via an input device such as the communication line 104 or the mouse 106 and the keyboard 107.

  The actually measured value 210 of the harbor outside wave and the actually measured value 220 of the harbor wave inputted at this time are combined with the actually measured values inside and outside the harbor at the same time. Based on these input values, the statistical relational expression generating unit 110 generates a statistical relational expression 230 that is established between the actually measured value 210 of the sea wave outside the harbor and the actually measured value 220 of the sea wave within the harbor. As shown in the figure, the statistical relational expression 230 is generated for each wave direction defined by dividing the azimuth by 16, and is stored in the storage unit 130 as a regression equation for each wave direction. Further, not only the wave direction but also individual regression equations are generated not only for the wave period. Thus, the generated statistical relational expression 230 is held in the storage unit 130 as a statistical model. It should be noted that the statistical relational expression 230 used here is not limited to the linear regression equation as described above, and it goes without saying that another method known as a statistical model can be applied.

  A method of using the statistical relational expression 230 will be described later with reference to FIG. Further, the actually measured value 210 of the sea wave outside the harbor is the height, period, wave direction, direction spectrum, etc. of the sea wave measured at an observation point outside the harbor for a specific harbor. Further, the actually measured value 220 of the wave in the harbor is the height, period, wave direction, direction spectrum, etc. of the wave measured at the observation point inside the breakwater in the same harbor. In addition to the method of collecting data directly by installing a measuring instrument, such actual measurement values include wave data provided by public facilities such as the Japan Meteorological Agency, wave data provided externally by bayside power stations, etc. Alternatively, it may be acquired via the communication line 104 or the like. Alternatively, the data may be read from the recording medium 102 at once.

  FIG. 3 is a diagram schematically showing a calculation procedure 30 of the harbor wave predicted value by the harbor wave prediction method executed in the harbor wave prediction apparatus 10 as described above. As shown in the figure, the calculation procedure 30 is broadly divided into an estimation procedure 301 for calculating an estimated value 330 of the wave outside the port and a calculation procedure 302 for calculating a predicted value 340 of the wave inside the port.

  It should be noted that the wave prediction model 310 held by the sea wave simulator 120 outside the harbor of the harbor wave prediction apparatus 10 estimates the estimated value by time integrating the direction spectrum, which is the energy of the wave, Information such as the mean wave period and direction is calculated from the direction spectrum.

  In addition, the accuracy of estimation using such a numerical model largely depends on the accuracy of the initial value. Therefore, in this embodiment, an initial value is obtained from the wave GPV data 312 for an ocean wave that cannot be measured, and the initial wave value 314 is corrected using the ratio with the initial value of the wave prediction model 310. is doing. For coastal waves where measured values can be obtained, the measured value is obtained from the data of the Japan Meteorological Agency via the communication line 104, etc., and the directional spectrum is calculated, and the wave prediction initial value 314 is directly corrected based on it. is doing. The accuracy of the wave prediction initial value 314 corrected in this way becomes very high.

  Further, in this embodiment, the harbor wave simulator 120 corrects the estimated value based on the error statistical model 320 calculated from the error of the past estimated value. That is, the error statistical model 320 is calculated by accumulating the error between the past estimated value 322 of the offshore wave and the actually measured value 324 at the same time. By applying such a statistical model 320 of the error to the estimated value 330 of the wave outside the harbor, it is possible to estimate an estimated value with higher accuracy.

  In addition, it is already known that the estimated value based on the physical model such as the wave prediction model 310 is reliable for the estimation of the wave outside the harbor. Another method known as the Kalman filter method is that the error of the estimated value is reduced by introducing the statistical model 320 of the error.

  As described above, the estimated value 330 of the wave outside the harbor at the specified time is estimated. In the subsequent calculation procedure 302 for calculating the predicted value 340 of the harbor wave, the estimated value 330 of the harbor wave is used.

  In the calculation procedure 302 of the predicted value 340 of the harbor wave, the estimated value 330 of the harbor wave at the specified time in the predicted value generation unit 140 of the harbor wave prediction apparatus 10 has the statistical relationship described with reference to FIG. Introduced in Equation 230. Here, when the estimated value 330 of the wave outside the harbor input to the predicted value generation unit 140 is, for example, related to an eastward wave, among a plurality of statistical relational expressions 230, the one for the eastward wind Is used. As described above, the statistical relational expression 230 is a statistical model, and the wave height and period prediction value 340 in the harbor corresponding to the wave height, period, and wave direction estimation value 330 input as a variable. Is output.

  According to the calculation procedure 30 as described above, since the predicted value 340 of the in-harbour wave is calculated using the highly accurate estimated value 330 of the off-harbour wave, a highly reliable predicted value 340 of the in-harbour wave can be obtained. . Further, in the calculation procedure 302 for calculating the predicted value 340 of the wave in the harbor, the predicted value is calculated by a statistical model using a single variable called the estimated value 330 of the wave outside the harbor, so the calculation load is very low. Therefore, for example, it is possible to calculate repeatedly for the same designated time, or to calculate predicted values for a plurality of designated times in a short time.

  As illustrated in FIG. 1, the harbor wave prediction apparatus 10 according to the present embodiment includes a determination unit 150. The determination unit 150 is input with the predicted value 340 of the in-harbour wave, and determines whether or not entry / exit from the port, berthing, cargo handling work, civil engineering work, and the like with respect to the port at a specified time is performed based on a predetermined threshold. Specifically, for example, the standard for canceling cargo handling work is that the significant wave height of the wave in the harbor is 0.5 m and the significant wave period in the harbor is 10 seconds or more. In addition, the port departure standard is that the significant wave height of the harbor wave is 0.8 m, and the significant wave period of the harbor wave is 10 seconds or more. The determination result is displayed, for example, on the screen of the display 108 and transmitted to the user. In this way, whether or not the port can be used is determined according to the predicted value with high accuracy, so that efficient port management can be performed. Therefore, not only the use efficiency of the ship is improved, but also the safety management in the harbor is improved.

  Moreover, the harbor wave prediction according to the above-described series of procedures is a prediction regarding waves at observation points determined inside and outside the harbor. Therefore, by providing a plurality of observation points inside and outside the harbor, it is possible to predict the waves in the whole harbor.

  Further, in the harbor wave prediction apparatus 10, in parallel with the operation of calculating the harbor wave prediction value 340, the actually measured value 210 of the harbor outside wave newly obtained by touching the event and the harbor wave are obtained. , And the accuracy of the statistical relational expression 230, which is a statistical model, can be sequentially improved. Thereby, with the progress of the operating time of the harbor wave prediction apparatus 10, the accuracy of the harbor wave prediction is further improved.

  FIG. 4 is a diagram showing another calculation procedure 40 of the predicted value 340 of the harbor wave, which is executed in the harbor wave prediction apparatus 10. The calculation procedure 40 corresponds to the calculation procedure 302 shown in FIG. 3, and the estimation procedure 301 shown in FIG. 3 is executed immediately before the calculation procedure 40.

  As shown in the figure, the Kalman filter method is applied to improve the accuracy of the predicted value 340 in the harbor by applying the Kalman filter method similarly to the estimation of the estimated value 330 of the sea wave outside the port even when calculating the predicted value 340 of the harbor wave. be able to. That is, first, after calculating the predicted value 340 of the harbor wave by the statistical relational expression 230, the actual value 414 at the same time as the predicted value is obtained. As a result, when the actual measurement value 414 at the same time as the predicted value is acquired, it is possible to compare the predicted value that has already become the past predicted value 412 of the wave in the harbor with the actual measurement value. By forming the error statistical model 410 and introducing it into the calculation of the predicted value 340 of the harbor wave, it is possible to calculate the predicted value 340 of the harbor wave with higher accuracy.

  FIG. 5 shows the predicted value of the wave outside the port for 3 days in a certain port according to the estimation procedure 301 shown in FIG. 3 as the wave height, and also plots the wave height actually measured for the same 3 days outside the port. It is a graph. As shown in the figure, the predicted value of the wave height of the sea wave outside the port estimated by correcting the initial value has a high correlation with the actually measured value.

  Moreover, FIG. 6 shows the predicted value of the wave in the harbor for the same three days calculated according to the calculation procedure 302 of FIG. 3 as the wave height of the significant wave, and plotted the measured value together using the estimated value. It is a graph. As shown in the figure, the wave height of the wave in the port calculated based on the highly accurate predicted value of the wave outside the port has a high correlation with the measured value, and the prediction of the wave in the port using the statistical model is effective. It turns out that it is.

  As described in detail above, the harbor wave prediction method includes an epoch-making method of obtaining, using a statistical model, an intra-harbour predicted value that is difficult to match with a predicted value based on a physical model because there are many factors involved. Thereby, although there are few variables which should be input and calculation load is also low, the complicated wave prediction in a harbor can be performed with high precision. In addition, this predicted value includes the wave height and wave period of the wave inside the port calculated by introducing the wave direction prediction of the wave outside the port. Can be predicted with high accuracy. Therefore, it can be executed with relatively simple hardware resources, and highly accurate port management can be performed even in a small port that is not equipped with facilities.

  Although the present invention has been described above using the embodiment, the technical scope of the present invention is not limited to the scope described in the embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

The figure which shows typically the structure of the harbor wave prediction apparatus 10 which concerns on one embodiment. The figure which shows typically the data structure 20 of the data which concern on the statistical relational expression 230. The figure which shows the procedure of the calculation procedure 30 of the harbor wave prediction value. The figure which shows the procedure of the calculation procedure 40 of the harbor wave prediction value which concerns on other embodiment. It is a graph which compares and shows a predicted value and a measured value about the wave height of a sea wave outside a harbor. It is a graph which compares and shows a predicted value and a measured value about the wave height of the wave in a harbor.

Explanation of symbols

10 Port Wave Prediction Device, 20 Data Structure, 30, 40 Calculation Procedure, 100 Information Processing Device, 102 Recording Medium, 104 Communication Line, etc. 106 Mouse, 107 Keyboard, 108 Display, 110 Statistical Relational Expression Generation Unit, 120 Port External Wave Simulator, 130 Storage Unit, 140 Predicted Value Generation Unit, 150 Judgment Unit, 210 Measured Value of Waves Outside Port, 220 Measured Value of Waves in Harbor, 230 Statistical Relational Expression, 301 Estimation Procedure, 302 Calculation Procedure, 310 Wave Prediction model, 312 wave GPV data, 314 wave prediction initial value, 320, 410 error statistical model, 322 past estimated value of offshore wave, 324 estimated value at the same time as estimated value, 330 estimated value of offshore wave, 340 Predicted value of waves in harbor 412 Predicted value of waves in harbor 414 Forecast Measured value at the same time as the measured value

Claims (10)

For a specific port, at a certain time, a harbor wave prediction method for predicting the harbor wave inside the port at a specified time after that time,
Calculate the statistical relational expression of the measured value of the sea wave outside the port outside the port and the actually measured value of the sea wave outside the port before the designated time and the actually measured value of the sea wave outside the port at the same time as the actually measured value of the sea wave outside the port. And the steps to
A procedure for estimating the estimated value of the wave outside the port at the specified time by a numerical simulation using a wave prediction model;
A harbor wave prediction method comprising: calculating a predicted value of the harbor wave at the specified time based on the statistical relational expression and the estimated value.   The harbor prediction method according to claim 1, wherein the predicted value includes a wave height and a wave period.   2. The harbor wave prediction method according to claim 1, further comprising a step of correcting a wave prediction initial value used in the numerical simulation based on a ratio between a default wave prediction initial value and a significant wave height of the wave GPV data.   The harbor wave prediction method according to claim 1, further comprising a step of correcting a wave prediction initial value used in the numerical simulation based on a direction spectrum calculated from an actually measured value of the wave outside the harbor. A procedure for calculating the statistical value of the error by comparing the estimated value in the past than the time and the measured value of the wave outside the harbor at the specified time of the estimated value,
The harbor wave prediction method according to claim 1, further comprising a step of correcting the estimated value using the statistical value. A procedure for calculating a statistical value of an error by comparing a calculated predicted value calculated in the past from the time and an actual value of the wave in the harbor at a specified time of the calculated predicted value;
The harbor wave prediction method according to claim 1, further comprising a step of correcting the predicted value newly calculated using the statistical value.   2. The harbor wave prediction method according to claim 1, further comprising a step of sequentially accumulating measured values of the sea wave outside the harbor and the sea wave in the harbor before the time and correcting the statistical relational expression.   The harbor wave prediction method according to claim 1, further comprising a step of determining whether or not a ship can enter and leave the port, berthing or cargo handling work based on the predicted value. For a specific port, at a certain time, an in-harbour wave prediction device for predicting a wave in a port inside the port at a specified time later than that time,
Calculate the statistical relational expression of the measured value of the sea wave outside the port outside the port and the actually measured value of the sea wave outside the port before the specified time and the actually measured value of the sea wave outside the port at the same time as the actually measured value of the sea wave outside the port. A statistical relational expression generating unit,
An offshore harbor wave simulator that estimates an estimated value of the offshore harbor wave at the specified time by a numerical simulation using a wave prediction model;
A harbor wave prediction device including a predicted value generation unit that calculates a predicted value of the harbor wave at the specified time based on the statistical relational expression and the estimated harbor wave. About a specific harbor, at a certain time, a harbor wave prediction program for predicting the harbor waves inside the harbor at a specified time after that time,
Calculate the statistical relational expression of the measured value of the sea wave outside the port outside the port and the actually measured value of the sea wave outside the port before the specified time and the actually measured value of the sea wave outside the port at the same time as the actually measured value of the sea wave outside the port. And steps to
A step of estimating an estimated value of the wave outside the port at the specified time by a numerical simulation using a wave prediction model;
A harbor wave prediction program including a step of calculating a predicted value of the harbor wave at the specified time based on the statistical relational expression and the estimated harbor wave wave.

Images