JP2009167751A - Inflow forecasting system, inflow forecasting method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make possible easy forecast about the inflow of water into a water storage facility. <P>SOLUTION: The information on a weather track record is stored in a database 40 on the weather track record for every date, and the information on weather track record includes the track record value of the inflow, a track record value of a rainfall, and a track record value of snowfall. In the inflow model 1 stored in a model memory unit 31, a rainfall P1<SB>t</SB>and the difference between an inflow F<SB>t-1</SB>and an equilibrium inflow μ<SB>1</SB>are made description variables, and the amount of an increase ΔF<SB>t</SB>in the inflow is made an objective variable. An inflow model estimate part 23 conducts a regression analysis based on the track record value of the inflow model 1, the track record value of the inflow model 1, and the inflow model 1. Thus, regression coefficients α<SB>1</SB>, β<SB>1</SB>, and β<SB>2</SB>are estimated. An inflow forecast part 26 forecasts the inflow based on the information on the weather track record and the inflow model 1. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a system, method, and program for predicting the amount of water flowing into a water storage facility.

For the purpose of water level management in river dam storage facilities and disaster prevention measures for rivers, the amount of water flowing into water storage facilities and the flow rate of water in rivers (hereinafter referred to as inflow) are being predicted. Yes. For example, in Patent Document 1, the trend of forecast error of forecast rainfall is analyzed, the forecast rainfall is corrected using the result, and the future flow is predicted using the current flow rate, the current rainfall, and the corrected forecast rainfall. ing.
JP 2006-92058 A

  However, Patent Document 1 describes the amount of rainfall accumulated in the past by observing the amount of rainfall at many points that affect the inflow of rivers, etc., and taking into account the time during which the rain causes a change at the point where the inflow is measured. The inflow is estimated as a variable, and a complicated estimation is made.

  The present invention has been made in view of such a background, and provides an inflow amount prediction system, an inflow amount prediction method, and a program capable of easily predicting the inflow amount of water into a water storage facility. Objective.

  The invention according to claim 1 of the present invention for solving the above-mentioned problems is a system for predicting the amount of water flowing into a water storage facility, the actual value of the inflow and the amount of rainfall every predetermined period. Explain the difference between the amount of rainfall in a certain first period, the amount of rainfall in a certain first period, and the amount of inflow in the second period before the first period and the equilibrium inflow that is a predetermined constant And a model storage unit that stores a regression model having a target variable that is an increase amount of the inflow amount from the second period to the first period, the actual value of the inflow amount that is stored in the database, and the An inflow model estimation unit that performs regression analysis based on the actual value of rainfall and the regression model stored in the model storage unit, and estimates a regression coefficient for each of the explanatory variables; A predicted rainfall amount acquisition unit that acquires a predicted value of the rainfall amount in a prediction period that is between, the actual value of the inflow amount in the period before the predetermined period from the prediction period is acquired from the database, and Applying the predicted value, the actual value of the inflow amount, and each regression coefficient to the regression model, the increase amount prediction unit for calculating the predicted value of the increase amount in the prediction period, and the actual value of the inflow amount An inflow amount predicting unit that adds the predicted value of the increase amount to calculate the predicted value of the inflow amount in the prediction period.

  The predetermined period is a unit period such as one hour or one day. For example, the predetermined period may be one period when there is no series correlation. Further, when there is a series correlation, the predetermined period can be the number of periods subjected to variable conversion by, for example, Prais-Winstein conversion or the Cochran-Ocut method.

  According to the inflow amount prediction system of the present invention, since the inflow amount can be predicted based on the actual value of the inflow amount, it is easier and more accurate than the case where the inflow amount is predicted based only on the rainfall amount. The amount of inflow can be predicted.

  Moreover, invention of Claim 2 among this invention is an inflow amount prediction system of Claim 1, Comprising: The said database further memorize | stores the actual value of the amount of snowmelt for every said predetermined period, The regression model stored in the model storage unit includes, as an explanatory variable, the snowmelt amount in the first period, and further includes a predicted snowmelt amount acquisition unit that acquires a predicted value of the snowmelt amount in the prediction period. The inflow amount model estimation unit includes the actual value of the inflow amount stored in the database, the actual value of the rainfall amount, the actual value of the snowmelt amount, and the regression stored in the model storage unit. Based on the model, regression analysis is performed and the regression coefficient is estimated.

  In this case, the amount of water flowing into the water storage facility can be predicted in consideration of the predicted value of the amount of snowmelt, so there was no precipitation compared to the case where the amount of inflow was predicted based solely on the prediction of precipitation. Even in such a case, it is possible to predict that the inflow amount may increase due to melting snow. Therefore, more accurate prediction can be performed.

  Moreover, invention of Claim 3 among this invention is an inflow amount prediction system of Claim 2, Comprising: The said database further memorize | stores the actual value of temperature, and the actual value of snowfall, The said model In addition to the inflow model that is the regression model, the storage unit subtracts a constant indicating the temperature at which snow melting starts from the temperature in the first period, or sets a larger value of zero to the snow cover in the second period. A snowmelt amount model that is a regression model having the value multiplied by the amount and the rainfall amount in the first period as explanatory variables and the snowmelt amount in the first period as an objective variable is stored in the database. A regression analysis is performed based on the actual temperature value, the actual snow accumulation value, the actual rainfall value, and the snowmelt model stored in the model storage unit. A snowmelt amount model estimation unit that estimates a regression coefficient for each of the explanatory variables, and a predicted temperature acquisition unit that acquires a predicted value of the temperature in the prediction period, the predicted snowmelt amount acquisition unit, The actual value of the snow cover amount in the period before the predetermined period of the prediction period is acquired from the database, the predicted value of the temperature, the acquired actual value of the snow cover amount, the predicted value of the rainfall amount, and the snow melting amount Each explanatory variable related to the model is applied to the snow melting amount model to calculate a predicted value of the snow melting amount in the prediction period.

  Above a certain temperature, the amount of snow melt increases as the temperature rises, and the snow melts due to rain. Therefore, the amount of snow melt can be regressed using the temperature, the amount of snow and the amount of rainfall as explanatory variables. In general, temperature and rainfall can be easily predicted using a weather forecasting method. Based on these easily obtainable predicted values of temperature and rainfall, and actual values of snow cover. Since the snow melting amount can be predicted, it is possible to cope with the case where it is difficult to predict the snow melting amount itself.

Moreover, invention of Claim 4 among this invention is an inflow amount prediction system of Claim 3, Comprising: The said database further memorize | stores the actual value of snowfall, The said snow melting model estimation part is the following. The explanatory variable of the snow cover amount included in the snow cover model is represented by the formula: snow cover amount in the first period = max (the snow fall amount in the first period + the snow cover amount in the second period−the snow cover amount in the first period). Snow melting, 0)
The regression coefficient for each of the explanatory variables related to the snowmelt amount model is estimated by performing regression analysis on the model in which is substituted.

  Moreover, invention of Claim 5 among this invention is an inflow amount prediction system of Claim 1, Comprising: It replaces with the actual value of the said rainfall, and the actual value of the precipitation is stored in the said database. A predicted precipitation amount acquisition unit that is stored and that acquires a predicted value of the precipitation amount in the prediction period; and a predicted temperature acquisition unit that acquires a predicted value of the temperature in the prediction period, the predicted rainfall amount The acquisition unit sets the precipitation as the rainfall when the predicted value of the temperature exceeds a predetermined temperature, and sets the rainfall as 0 when the predicted value of the temperature is equal to or lower than the predetermined temperature. I decided to.

  Further, the invention according to claim 6 of the present invention is the inflow amount prediction system according to claim 5, wherein the database stores the actual value of precipitation instead of the actual value of the rainfall. The database further stores an actual value of temperature and an actual value of snowfall, a snowfall model that is a regression model using the temperature and precipitation as explanatory variables, and the snowfall as an objective variable, and the database A snowfall model estimation unit that performs regression analysis based on the actual temperature value, the precipitation actual value, and the snowfall actual value stored in the memory, and estimates a regression coefficient for each of the explanatory variables; When the temperature is less than or equal to δ for each of the actual temperature values stored in the database, the estimated regression coefficient, the actual temperature value, and the previous The value calculated by applying the actual precipitation value to the snowfall model is the estimated snowfall amount, and when the temperature is higher than δ, the estimated snowfall amount is 0. A snow temperature estimation unit that determines the δ as the predetermined temperature that minimizes an error from the estimated snow amount.

  Further, the invention according to claim 7 of the present invention is a system for predicting the inflow amount of water into the water storage facility, and for each predetermined period, the actual value of the inflow amount, the actual value of the rainfall amount, the temperature A database that stores actual values of snow and actual values of snow cover, the amount of rainfall in a certain first period, the amount of snow melt in the first period, and the amount of inflow in a second period before the first period from the first period And an inflow model, which is a regression model with the increase in the inflow from the second period to the first period as an objective variable, A value obtained by subtracting a constant indicating the temperature at which snow melting starts from the temperature in the first period or a larger value of zero and the amount of snow in the second period and the amount of rainfall in the first period As an explanatory variable, the first period A model storage unit for storing a snowmelt amount model, which is a regression model having the snowmelt amount as an objective variable, an actual value of the temperature, an actual value of the snow cover, and an actual value of the rainfall stored in the database. And a snowmelt amount model estimating unit that performs regression analysis based on the snowmelt amount model stored in the model storage unit and estimates a regression coefficient for each of the explanatory variables related to the snowmelt amount model, and the database Applying the actual temperature value, the actual snow accumulation value, the actual rainfall value, and the estimated regression coefficient stored in the snowmelt amount model, The theoretical value is calculated, the actual value of the inflow amount stored in the database, the actual value of the rainfall amount, the theoretical value of the calculated snow melting amount, and the model description An inflow model estimation unit that performs regression analysis based on the inflow model stored in the unit and estimates a regression coefficient for each of the explanatory variables related to the inflow model, and a prediction that is a prediction target period A predicted rainfall amount acquisition unit that acquires a predicted value of the rainfall amount in a period, a predicted temperature acquisition unit that acquires a predicted value of the temperature in the prediction period, and the snow cover amount in the period before the predetermined period of the prediction period Obtained from the database, the predicted value of the temperature, the obtained actual value of the snow cover, the predicted value of the rainfall, and the respective explanatory variables related to the snow melt model are stored in the snow melt model. And applying a predicted snowmelt amount acquisition unit for calculating a predicted value of the snowmelt amount in the prediction period, and an actual value of the inflow amount in a period before the predetermined period from the prediction period. A predicted value of the rainfall amount, a predicted value of the snowmelt amount, an actual value of the inflow amount, and the regression coefficient are obtained from the database, and the prediction of the increase amount in the prediction period is applied to the inflow amount model. An increase amount prediction unit that calculates a value; and an inflow amount prediction unit that calculates a predicted value of the inflow amount in the prediction period by adding the predicted value of the increase amount to the actual value of the inflow amount. And

  In this case, the theoretical value of the amount of snowmelt is calculated using a snowmelt amount model with temperature, rainfall, and snow cover as explanatory variables, and the amount of water flowing into the water storage facility is predicted in consideration of the amount of snowmelt. be able to. Therefore, even if measured values of snowmelt cannot be obtained from the Japan Meteorological Agency or private weather companies, etc., the amount of snowmelt is generally based on the predicted values of temperature and rainfall that are easy to obtain and the actual value of snow cover. The predicted value can be calculated. This makes it possible to predict the inflow of water into the water storage facility, taking into account that the inflow can increase due to snow melting, even if the predicted value of the amount of snow melt cannot be easily obtained or when there is no precipitation. Therefore, it is possible to predict the inflow amount with high accuracy.

  Other problems and solutions to be disclosed by the present application will be made clear by the embodiments of the invention and the drawings.

  According to the present invention, it is possible to easily predict the amount of water flowing into a water storage facility.

Hereinafter, an inflow amount prediction system 10 according to an embodiment of the present invention will be described.
The inflow amount prediction system 10 of the present embodiment predicts the amount of water flowing into a water storage facility such as a dam from a river (hereinafter referred to as the inflow amount). The inflow forecasting system 10 estimates a regression coefficient by performing regression analysis on past precipitation, snowfall, snow cover, snowmelt, and inflow using a statistical model to be described later, and predicts predicted temperatures and precipitations. (For example, a forecast value from a weather forecast can be adopted.) And the estimated regression coefficient are applied to a statistical model to predict the inflow. In addition, the inflow amount is predicted on a daily basis.

  1 to 3 are graphs showing changes in the inflow amount. Even when there is no precipitation or snowmelt, there is a predetermined inflow due to, for example, oozing from a forest. Therefore, if there is no precipitation or snowmelt, the inflow will gradually decrease to a predetermined equilibrium value (hereinafter referred to as equilibrium inflow) (Fig. 1). On the other hand, when there is precipitation, the amount of water flowing temporarily increases accordingly, but the precipitation stops and begins to decrease again toward the equilibrium inflow (Fig. 2). On the other hand, when the temperature rises, snow melts and the amount of water flowing increases accordingly, but the effect of snow melt on the inflow is more gradual than that of precipitation (Fig. 3). This is a case where, for example, snow melting continuously occurs when the average temperature rises during the period from winter to spring.

  Therefore, in this embodiment, paying attention to precipitation, snowmelt, and balanced inflow, the inflow is predicted using a regression model that calculates the inflow. The inflow prediction system 10 of the present embodiment estimates parameters based on actual values of weather data such as past temperature and precipitation and a regression model described later, and is obtained by the estimated parameters and, for example, weather forecasts. A predicted value of inflow is calculated by applying a predicted value of temperature (hereinafter referred to as predicted temperature) and a predicted value of precipitation (hereinafter referred to as predicted rainfall) to the regression model. Details will be described below.

== System configuration ==
In the present embodiment, it is assumed that the inflow rate prediction system 10 is configured by a single computer. FIG. 4 is a diagram illustrating a hardware configuration of the inflow amount prediction system 10 of the present embodiment. The inflow amount prediction system 10 includes a CPU 11, a memory 12, a storage device 13, an input device 14, and an output device 15. The storage device 13 stores various programs and data, for example, a hard disk drive, a flash memory, a CD-ROM drive, or the like. The CPU 11 implements various functions by reading the program stored in the storage device 13 into the memory 12 and executing it. The input device 14 is a keyboard, mouse, touch panel, microphone, or the like that accepts data input from the user. The output device 15 is, for example, a display, a printer, or a speaker that outputs data.

  FIG. 5 is a diagram illustrating a software configuration of the inflow amount prediction system 10 of the present embodiment. The inflow prediction system 10 according to the present embodiment includes a snowfall temperature estimation unit 21, a snowmelt model estimation unit 22, an inflow model estimation unit 23, a predicted temperature acquisition unit 24, a predicted precipitation acquisition unit 25, an inflow prediction unit 26, A model storage unit 31, a parameter storage unit 32, and a weather record database 40 are provided.

  The snowfall temperature estimation unit 21, the snowmelt amount model estimation unit 22, the inflow amount model estimation unit 23, the predicted temperature acquisition unit 24, the predicted precipitation amount acquisition unit 25, and the inflow amount prediction unit 26 are executed by the CPU 11 of the inflow amount prediction system 10. This is realized by reading the program stored in the storage device 13 into the memory 12 and executing it.

  The weather record database 40 stores a history of information (hereinafter referred to as weather record information) including various kinds of record values of weather. FIG. 6 is a diagram illustrating a configuration example of weather performance information stored in the weather performance database 40. As shown in the figure, the weather result information includes temperature, precipitation, snowfall, snow cover, snowmelt, and inflow in association with the date. The temperature is the average daily temperature. Precipitation, snowfall and snowmelt are cumulative values of daily precipitation, snowfall and snowmelt. The snow cover is the snow cover observed on that day. The inflow is the cumulative amount of water that has flowed into a water storage facility such as a dam per day. The temperature, precipitation, snowfall, and snow cover are data provided by, for example, the Japan Meteorological Agency or a private weather company. The inflow is a measured value measured at the water storage facility. The inflow amount may be, for example, a measurement value of a river flow rate provided by a local government that manages the river. The amount of snowmelt may be measured by the Japan Meteorological Agency, a private weather company, a surveying company, or the like, or a value calculated by a model described later may be recorded as an actual value.

Various statistical models based on weather performance information are stored in the model storage unit 31, and parameters such as regression coefficients and constants applied to the models stored in the model storage unit 31 are stored in the parameter storage unit 32. Is done. The model storage unit 31 includes a precipitation statistical model (hereinafter referred to as precipitation model 1), a snowfall statistical model (hereinafter referred to as snowfall model 2), and a snowfall statistical model (hereinafter referred to as snowfall model). 3), a statistical model of snowmelt (hereinafter referred to as snowmelt model 4), and a statistical model (hereinafter referred to as inflow model 5) relating to the increase in inflow from the previous day. In the parameter storage unit 32, the temperature (hereinafter referred to as the snow temperature) δ at which the rain changes to snow, the equilibrium inflow amount μ ₁ , the temperature μ _{2 at which} snow melting starts, and the regression coefficients α _{1 to} α ₃ , β ₁ , β ₂ are stored. It is stored in the parameter storage unit 32. It is assumed that the equilibrium inflow amount μ ₁ and the temperature μ _{2 at} which snow melting starts are stored in the parameter storage unit 32 in advance as predetermined constants.

In the following description, the temperature at the date t, rainfall, snowfall, snowfall, snowmelt and inflow _{respectively,} and _{T t, P t, S t} , D t, M t , and _{F t.} Let ΔFt be the increase in the inflow from the previous day. The precipitation when the temperature Tt is higher than δ is P1 _t , and the precipitation when the temperature T _t is δ or less is P2 _t .

降雪量モデル２は、気温Ｔ_ｔがδ以下である場合の降水量Ｐ２ｔを説明変数とし、降雪量Ｓｔを目的変数とした回帰モデルであり、次式で表される。

積雪量モデル３は、積雪量Ｄ_ｔが、前日までの積雪量Ｄ_ｔ−１に当日の降雪量Ｓ_ｔを加え、そこから融雪量Ｍ_ｔを引いたものに一致するという関係を示すモデルであり、次式で表される。

融雪量モデル４は、気温Ｔ_ｔから融雪が始まる気温μ_２を減じた値に積雪量Ｄ_ｔを乗じた値と、Ｐ１_ｔとを説明変数とし、融雪量Ｍ_ｔを目的変数とした回帰モデルであり、次式で表される。

融雪量モデル４において、気温Ｔ_ｔがμ_２よりも低ければ第１項は０になり、前日までの積雪量Ｄ_ｔ−１が０であれば融雪量Ｍ_ｔは０になる。
流入量モデル５は、均衡流入量μ_１から前日の流入量Ｆ_ｔ−１を減じた値と、当日の気温Ｔ_ｔがδより高い場合の降水量Ｐ１_ｔと、当日の融雪量Ｍ_ｔとを説明変数とし、流入量の増加量ΔＦ_ｔを目的変数とした回帰モデルであり、次式で表される。

Precipitation Model 1, precipitation P _t is, the boundary of snow temperature [delta], is a model that indicates that the P1 _t or P2 _t, is expressed by the following equation.

The snowfall model 2 is a regression model in which the precipitation P2t when the temperature T _t is equal to or less than δ is an explanatory variable and the snowfall St is an objective variable, and is represented by the following equation.

Snowfall model 3, snow accumulation D _t is a model showing the relationship of addition of the day of snowfall S _t in snowfall D _t-1 of the previous day, matches therefrom minus snowmelt M _t Yes, it is expressed by the following formula.

The snow melting amount model 4 is a regression model in which a value obtained by subtracting the temperature μ _{2 at} which the snow melting starts from the temperature T _t is multiplied by the snow accumulation amount D _t , P1 _t is an explanatory variable, and the snow melting amount M _t is an objective variable. And is represented by the following equation.

In the snowmelt amount model 4, the first term is 0 if the temperature T _t is lower than μ ₂ , and the snowmelt amount M _t is 0 if the snow accumulation amount D _t−1 up to the previous day is 0.
Inflow model 5, a value obtained by subtracting the inflow F _t-1 of the previous day from equilibrium inflow mu _1, and precipitation P1 _t where temperature T _t of the day is higher than [delta], and the day of snowmelt M _t Is an regression variable, and an inflow increase ΔF _t is a target variable, which is expressed by the following equation.

  The snowfall temperature estimating unit 21 estimates the snowfall temperature δ based on the precipitation model 1 and the weather record information, and registers the estimated snowfall temperature δ in the parameter storage unit 32. The details of the process of estimating δ by the snowfall temperature estimating unit 21 will be described later.

Moreover, the snowfall temperature estimation part 21 reads the weather performance information corresponding to the date t from the weather performance database 40 for each date t, the precipitation amount and temperature of the read weather performance information, and the estimated snowfall temperature δ are described above. P2 _t is calculated by applying it to precipitation model 1. The snowfall temperature estimation unit 21 performs a regression analysis based on the weather performance information, P2 _t , and the snowfall amount model 2 to estimate the regression variable γ. The snowfall temperature estimation unit 21 registers the estimated regression variable γ in the parameter storage unit 32.

を回帰分析して、回帰係数α_２及びα_３を推計する。融雪量モデル推計部２２は、推計した回帰係数α_２及びα_３をパラメータ記憶部３２に登録する。 The snow melting amount model estimation unit 22 is an equation in which the snow accumulation amount model 3 is substituted into the snow melting amount model 4.

And regression coefficients α ₂ and α ₃ are estimated. The snowmelt amount model estimation unit 22 registers the estimated regression coefficients α ₂ and α ₃ in the parameter storage unit 32.

The inflow amount model estimation unit 23 estimates the regression coefficients α ₁ , β ₁ , and β ₂ based on the inflow amount model 5 and the weather performance information. Specifically, the inflow amount model estimation unit 23 calculates P1 _t for each date t based on the precipitation amount of the precipitation model 1, δ, and the weather performance information corresponding to the date t, and the weather on each date t. The inflow model 5 is subjected to regression analysis using the actual information and the corresponding P1 _t , and regression coefficients α ₁ , β ₁ , and β ₂ are estimated. If there is no snowmelt amount in the actual weather information, the inflow amount model 5 is calculated using M _t calculated using the snowmelt amount model 4 using the snow cover amount on the date t-1 and other weather information on the date t. It is also possible to estimate regression coefficients α ₁ , β ₁ , and β ₂ by performing regression analysis. The inflow model estimation unit 23 registers the estimated regression coefficients α ₁ , β ₁ , and β ₂ in the parameter storage unit 32.

  The predicted temperature acquisition unit 24 acquires a predicted temperature value (hereinafter referred to as a predicted temperature). For example, the predicted temperature acquisition unit 24 may receive an input of the predicted temperature from the user, or may access the computer of the Japan Meteorological Agency or a private weather company to acquire the predicted temperature. Moreover, you may make it the estimated temperature acquisition part 24 estimate temperature based on weather performance information. In this case, for example, a statistical model used for general temperature prediction is stored in the model storage unit 31, and the predicted temperature acquisition unit 24 performs regression analysis based on the statistical model and the actual weather information to set the parameter The estimated temperature can be calculated by applying the estimated parameters and the weather performance information to the statistical model.

  The predicted precipitation acquisition unit 25 acquires a predicted value of precipitation (hereinafter referred to as predicted precipitation). For example, the predicted precipitation acquisition unit 25 may receive an input of predicted precipitation from a user, or may access a computer of the Japan Meteorological Agency or a private weather company to acquire the predicted precipitation. In addition, the predicted precipitation amount acquisition unit 25 may predict the precipitation amount based on the weather performance information, similarly to the predicted temperature acquisition unit 24.

  The inflow amount prediction unit 26 includes the predicted temperature acquired by the predicted temperature acquisition unit 24, the predicted precipitation acquired by the predicted precipitation acquisition unit 25, the parameters estimated by the snowfall temperature estimation unit 21, weather performance information, and the inflow amount model 5. Is used to calculate the predicted value of the inflow amount (hereinafter referred to as the predicted inflow amount). Details of the predicted inflow calculation process will be described later.

  Hereinafter, details of processing in the inflow amount prediction system of the present embodiment will be described.

== Estimation processing ==
First, the process for estimating the snowfall temperature δ by the snowfall temperature estimation unit 21 will be described. FIG. 7 is a diagram showing the flow of the process for estimating the snowfall temperature δ by the snowfall temperature estimation unit 21.

First, the snowfall temperature estimation part 21 acquires the weather performance information whose snowfall amount is larger than 0 from the weather performance database 40 (S501). Thereby, as shown in FIG. 8, only the snowfall amount larger than 0 is extracted among the weather performance information stored in the weather performance database 40.
Next, the snowfall temperature estimation unit 21 performs regression analysis of the model snowfall amount = a × temperature + b × precipitation amount based on the extracted weather record information, and estimates regression coefficients a and b (S502).
The snowfall temperature estimation unit 21 sets “a × weather performance information temperature + b × weather performance information precipitation” when the temperature is δ or less, and “0” as the estimated snowfall when the temperature is higher than δ. Δ that minimizes the value obtained by squaring the difference between the estimated amount of snowfall and the amount of snowfall in the weather information is calculated (S503).

  For example, as shown in FIG. 9, the snowfall temperature estimation unit 21 sets the regression coefficient a to the temperature if the temperature is higher than δ for each meteorological record information as shown in FIG. The sum of the value multiplied by the above and the value obtained by multiplying the above-mentioned area coefficient b by the amount of precipitation is calculated as the estimated snowfall, and the value obtained by squaring the difference between the estimated snowfall and the snowfall in the weather information is an error. Can be determined as δ, and the one with the smallest error square can be determined as δ. A general statistical method can be used for the process of determining δ so that the square of the error is minimized.

The snowfall temperature estimation unit 21 registers δ determined as described above in the parameter storage unit 32 (S504).
The snow melting temperature δ is determined as described above.

== Inflow prediction process ==
Next, prediction of the inflow amount will be described. FIG. 10 is a diagram illustrating a flow of the inflow amount prediction process in the inflow amount prediction system 10 of the present embodiment. In the process of FIG. 10, it is assumed that the date t is the date to be predicted.

The predicted temperature acquisition unit 24 acquires the predicted temperature _Tt (S601). Predicted rainfall obtaining unit 25 obtains the predicted rainfall _{P t} (S602). As described above, the prediction temperature acquisition unit 24 and the predicted rainfall obtaining unit 25 is, for example, and receives the input data from a user, by acquiring the data by accessing the JMA and private meteorological office computer, T _t And P _t can be acquired.

The inflow amount prediction unit 26 reads δ from the parameter storage unit 32 (S603), and calculates P1 _t by applying δ, T _t and P _t to the precipitation model 1 (S604). That is, P1 _t = P _t if the predicted temperature T _t is greater than δ, and P1 _t = 0 if T _t is less than or equal to δ.

The inflow amount prediction unit 26 reads α _{1 to} α ₃ , β ₁ , β ₂ , μ ₁ , and μ ₂ from the parameter storage unit 32 (S605). The inflow prediction unit 26 reads the weather record information corresponding to t-1 on the previous day from the weather record database 40, sets the snow cover amount of the read weather record information as D _t-1 (S606), The inflow amount is set to F _t-1 (S607).

The inflow amount prediction unit 26 substitutes α ₂ , T _t , μ ₂ , D _t−1 , α ₃ , and P1 _t into the snow melting amount model 4 to calculate a predicted value M _t of the snow melting amount (S608). Then, α ₁ , μ ₁ , F _t−1 , β ₁ , P 1 _t , β ₂ , and M _t are substituted into the inflow amount model 5 to calculate a predicted value ΔF _t of the inflow increase amount (S609). The inflow amount prediction unit 26 adds ΔF _t to F _t−1 to calculate a predicted inflow amount F _t (S610).

  As described above, according to the inflow amount prediction system 10 of the present embodiment, the inflow amount is predicted in consideration of the precipitation amount and the snowmelt amount based on the predicted temperature, the predicted precipitation amount, and the weather performance information. Can do. Even when there is no precipitation, the amount of inflow increases due to snow melting, so the prediction accuracy can be improved by predicting the amount of inflow taking into account the amount of snow melting.

  Moreover, according to the inflow amount prediction system 10 of the present embodiment, the snowmelt amount can be calculated from the precipitation amount and the temperature. Prediction of precipitation and temperature has various methods as weather forecasting methods and is easily available. Therefore, even if it is difficult to predict the amount of snowmelt, it is possible to easily predict the inflow amount in consideration of the amount of snowmelt by predicting the amount of snowmelt based on the easily obtained precipitation and temperature predicted values. can do.

  Also, in the inflow model 5 described above, the first term is the difference between the balanced inflow and the inflow, and the balanced inflow is taken into account. Compared to the case where the inflow is simply used as an explanatory variable, More accurate inflow prediction can be performed.

  In the inflow amount prediction system of the present embodiment, the amount of water flowing into the water storage facility such as a dam is predicted from the river. However, the inflow amount prediction system can be easily applied to a system that predicts the amount of water flowing through the river. it can. In this case, the forecast value and the actual value of the temperature and precipitation in the upstream area of the river shall be acquired and recorded.

In the present embodiment, the equilibrium inflow amount μ ₁ and the temperature μ _{2 at} which snow melting starts are stored in the parameter storage unit 32 in advance. However, the present invention is not limited to this, and based on past weather record information. A good value may be estimated.

Further, when the water storage facility exists in an area where there is little snowfall or snowfall, the inflow amount model 5 may not consider the snowmelt amount M. In this case, the inflow model 5 is expressed by the following equation.

上記式（５”）を変形すると次式となる。

In addition, for each regression model of the present embodiment, when there is a series correlation in the error term, the snow temperature estimation unit 21 may estimate the parameters by the Prais-Winstein transform or the Cochrane / Aucut method. In this case, for example, when the parameters of the inflow model 5 are estimated by the Cochrane-Aucut method, the residual is ε _t and the explanatory variable X _{t *} = X _t −ρX _t−1 (that is, F _{t *} = F _t −ρF _t−1 , P1 _{t *} = P1 _t −ρP1 _t−1 , M _{t *} = M _t −ρM _{t−1, and the} like), and the inflow amount model 5 can be expressed as an equation.

When the above equation (5 ″) is modified, the following equation is obtained.

As described above, it is possible to obtain the inflow amount model 5 in which the equation modification is performed retroactively until the n period before the series correlation disappears in the residual ε _t. Parameters can be estimated so that they disappear. As described above, an n-phase delay model in which the series correlation of the residual ε _t disappears is stored in the model storage unit 31, and the inflow amount is predicted using the n-phase delay model, whereby a residual term is obtained. Even when there is series correlation, it is possible to estimate an appropriate parameter, and it is possible to accurately predict the inflow amount using the appropriate parameter.

  Although the present embodiment has been described above, the above embodiment is intended to facilitate understanding of the present invention and is not intended to limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and the present invention includes equivalents thereof.

It is a graph which shows the change of inflow when there is no precipitation. It is a graph which shows the change of inflow when there is precipitation. It is a graph which shows the change of inflow when there is snow melting. It is a figure which shows the hardware constitutions of the inflow amount prediction system 10 of this embodiment. It is a figure which shows the software structure of the inflow amount prediction system 10 of this embodiment. It is a figure which shows the structural example of weather performance information. It is a figure which shows the flow of the estimation process of the snowfall temperature (delta) by the snowfall temperature estimation part. It is a figure explaining extracting the weather performance information larger than 0 snowfall. It is a figure which shows the result of having calculated the estimated snowfall amount and the square of an error about a certain snowfall temperature (delta). It is a figure which shows the flow of an inflow amount prediction process.

Explanation of symbols

10 Inflow prediction system 11 CPU
DESCRIPTION OF SYMBOLS 12 Memory 13 Storage device 14 Input device 15 Output device 21 Snowfall temperature estimation part 22 Snow melting amount model estimation part 23 Inflow amount model estimation part 31 Model storage part 32 Parameter storage part 40 Weather results database

Claims (9)

A system for predicting the amount of water flowing into a water storage facility,
A database for storing the actual value of the inflow and the actual value of the rainfall for each predetermined period;
The difference between the rainfall amount in a certain first period and the inflow amount in the second period before the first period from the first period and the equilibrium inflow amount that is a predetermined constant is used as an explanatory variable, and the second period from the second period to the second period. A model storage unit for storing a regression model whose objective variable is an increase amount of the inflow until one period;
Based on the actual value of the inflow and the actual value of the rainfall stored in the database, and the regression model stored in the model storage unit, regression analysis is performed for each of the explanatory variables. An inflow model estimation unit for estimating a coefficient;
A predicted rainfall amount acquisition unit that acquires a predicted value of the rainfall amount in a prediction period that is a prediction target period;
The actual value of the inflow from the prediction period to the period before the predetermined period is acquired from the database, and the predicted value of the rainfall, the actual value of the inflow, and the regression coefficients are applied to the regression model. An increase amount prediction unit for calculating a predicted value of the increase amount in the prediction period;
An inflow amount prediction unit that calculates the predicted value of the inflow amount in the prediction period by adding the predicted value of the increase amount to the actual value of the inflow amount;
An inflow rate prediction system comprising: The inflow amount prediction system according to claim 1,
The database further stores the actual value of the amount of snow melt for each predetermined period,
The regression model stored in the model storage unit further includes the amount of snow melt in the first period as an explanatory variable,
A predicted snowmelt amount acquisition unit for acquiring a predicted value of the snowmelt amount in the prediction period;
The inflow amount model estimation unit includes the actual value of the inflow amount stored in the database, the actual value of the rainfall amount, the actual value of the snowmelt amount, and the regression model stored in the model storage unit. Performing a regression analysis based on the above and estimating the regression coefficient,
An inflow prediction system characterized by The inflow amount prediction system according to claim 2,
The database further stores an actual temperature value and an actual snow amount value,
In addition to the inflow model that is the regression model, the model storage unit subtracts a constant indicating the temperature at which snow melting starts from the temperature in the first period or a larger value of zero in the second period. Storing a snowmelt amount model which is a regression model having the value obtained by multiplying the amount of snow and the amount of rainfall in the first period as explanatory variables, and the amount of snowmelt in the first period as an objective variable;
Performing regression analysis based on the actual temperature value stored in the database, the actual snow accumulation value, the actual rainfall value, and the snowmelt model stored in the model storage unit, A snowmelt amount model estimation unit for estimating a regression coefficient for each of the explanatory variables related to the snowmelt amount model;
A predicted temperature acquisition unit that acquires a predicted value of the temperature in the prediction period;
With
The predicted snowmelt amount acquisition unit acquires the actual value of the snow cover amount in the period before the predetermined period of the prediction period from the database, the predicted value of the temperature, the actual value of the acquired snow cover amount, the rainfall amount Applying the predicted value of each and the explanatory variables related to the snowmelt amount model to the snowmelt amount model to calculate the predicted value of the snowmelt amount in the prediction period;
An inflow prediction system characterized by The inflow amount prediction system according to claim 3,
The database further stores the actual amount of snowfall,
The snow melting model estimation unit uses an expression of the snow accumulation amount included in the snow accumulation model as an expression variable: snow accumulation amount in the first period = max (the snowfall amount in the first period + the snow accumulation amount in the second period) -The amount of snow melt in the first period, 0)
Estimating the regression coefficient for each of the explanatory variables related to the snowmelt amount model by performing regression analysis of the model into which
An inflow prediction system characterized by The inflow amount prediction system according to claim 1,
In the database, instead of the actual rainfall value, the actual rainfall value is stored,
A predicted precipitation acquisition unit for acquiring a predicted value of the precipitation in the prediction period;
A predicted temperature acquisition unit that acquires a predicted value of the temperature in the prediction period;
With
The predicted rainfall amount acquisition unit sets the precipitation amount as the rainfall when the predicted value of the temperature exceeds a predetermined temperature, and the rainfall when the predicted value of the temperature is equal to or lower than the predetermined temperature. Set the amount to 0,
An inflow prediction system characterized by The inflow amount prediction system according to claim 5,
The database further stores an actual temperature value and an actual snowfall value,
A snowfall model that is a regression model with the temperature and the precipitation as explanatory variables, and the snowfall as an objective variable, and the actual value of the temperature stored in the database, the actual value of the precipitation, and the A snowfall model estimation unit that performs regression analysis based on the actual amount of snowfall, and estimates a regression coefficient for each of the explanatory variables;
For each of the actual temperature values stored in the database, when the temperature is δ or less, the estimated regression coefficient, the actual temperature value, and the actual precipitation value are stored in the snowfall model. When the estimated snowfall amount is a value calculated by application, and 0 is the estimated snowfall amount when the temperature is higher than δ, the error between the actual value of the snowfall amount and the estimated snowfall amount is the smallest. A snow temperature estimating unit for determining δ as the predetermined temperature;
An inflow rate prediction system comprising: A system for predicting the amount of water flowing into a water storage facility,
For each predetermined period, a database that stores the actual value of the inflow, the actual value of the rainfall, the actual value of the temperature, and the actual value of the amount of snow;
The difference between the rainfall amount in a certain first period, the snowmelt amount in the first period, and the inflow amount in the second period before the predetermined period from the first period and the equilibrium inflow amount that is a predetermined constant is used as an explanatory variable. , An inflow model that is a regression model with an increase in the inflow from the second period to the first period as an objective variable, and a constant indicating the temperature at which snow melting starts is subtracted from the temperature in the first period. A value obtained by multiplying the larger value or zero by the amount of snow in the second period and the amount of rainfall in the first period are explanatory variables, and the amount of snow melt in the first period is an objective variable. A model storage unit that stores a snowmelt amount model that is a regression model
Performing regression analysis based on the actual temperature value stored in the database, the actual snow accumulation value, the actual rainfall value, and the snowmelt model stored in the model storage unit, A snowmelt amount model estimation unit for estimating a regression coefficient for each of the explanatory variables related to the snowmelt amount model;
Applying the actual temperature value, the actual snow accumulation value, the actual rainfall value, and the estimated regression coefficient stored in the database to the snow melting model, the snow melting for each predetermined period. The theoretical value of the amount is calculated, the actual value of the inflow amount stored in the database, the actual value of the rainfall amount, the theoretical value of the calculated snow melting amount, and the model storage unit An inflow model estimation unit that performs a regression analysis based on the inflow model, and estimates a regression coefficient for each of the explanatory variables related to the inflow model;
A predicted rainfall amount acquisition unit that acquires a predicted value of the rainfall amount in a prediction period that is a prediction target period;
A predicted temperature acquisition unit that acquires a predicted value of the temperature in the prediction period;
The actual value of the snow cover amount in the period before the predetermined period of the prediction period is acquired from the database, the predicted value of the temperature, the acquired actual value of the snow cover amount, the predicted value of the rainfall amount, and the snow melting amount Applying each of the explanatory variables related to the model to the snowmelt amount model, and calculating a predicted value of the snowmelt amount in the prediction period;
The actual value of the inflow amount in the period from the prediction period to the predetermined period is acquired from the database, the predicted value of the rainfall amount, the predicted value of the snowmelt amount, the actual value of the inflow amount, and each regression coefficient Applying to the inflow amount model, an increase amount prediction unit that calculates a predicted value of the increase amount in the prediction period;
An inflow amount prediction unit that calculates the predicted value of the inflow amount in the prediction period by adding the predicted value of the increase amount to the actual value of the inflow amount;
An inflow rate prediction system comprising: A method for predicting the amount of water flowing into a water storage facility,
Computer
For each predetermined period, the actual value of the inflow and the actual value of the rainfall are stored in a memory,
The difference between the rainfall amount in a certain first period and the inflow amount in the second period before the first period from the first period and the equilibrium inflow amount that is a predetermined constant is used as an explanatory variable, and the second period from the second period to the second period. Storing in the memory a regression model whose objective variable is the amount of increase in inflow up to one period;
The regression analysis is performed based on the actual value of the inflow amount and the actual value of the rainfall amount stored in the memory, and the regression model stored in the model storage unit, and the regression for each of the explanatory variables. Estimate the coefficient,
Get the predicted value of the rainfall in the forecast period, which is the forecast target period,
The actual value of the inflow from the prediction period to the period before the predetermined period is acquired from the memory, and the predicted value of the rainfall, the actual value of the inflow, and the regression coefficients are applied to the regression model. Calculating a predicted value of the increase amount in the prediction period,
Adding the predicted value of the increase amount to the actual value of the inflow amount to calculate the predicted value of the inflow amount in the prediction period;
An inflow rate prediction method characterized by A program for predicting the amount of water flowing into a water storage facility,
On the computer,
Storing the actual value of the inflow and the actual value of the rainfall in a memory for each predetermined period;
The difference between the rainfall amount in a certain first period and the inflow amount in the second period before the predetermined period from the first period and the equilibrium inflow amount that is a predetermined constant is used as an explanatory variable, and the first period to the first period Storing, in the memory, a regression model whose objective variable is an increase in the amount of inflow until a period;
The regression analysis is performed based on the actual value of the inflow amount stored in the memory, the actual value of the rainfall amount, and the regression model stored in the model storage unit. Estimating a regression coefficient;
Obtaining a predicted value of the rainfall amount in a prediction period that is a period to be predicted;
The actual value of the inflow amount in the period before the predetermined period from the prediction period is acquired from the memory, the predicted value of the rainfall amount, the predicted value of the snowmelt amount, the actual value of the inflow amount, and each regression coefficient Applying to the regression model to calculate a predicted value of the increase in the prediction period;
Adding the predicted value of the increase amount to the actual value of the inflow amount to calculate the predicted value of the inflow amount in the prediction period;
A program for running

Images