JPH11323885A - River water level preestimating system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable accurate preestimate of a water level, corresponding to states of operation of a pump and a gate. SOLUTION: An upstream water level measuring means 100 and a downstream water level measuring means 200 which measure water levels on the upstream and downstream sides respectively are provided. A rainfall measuring means 300 is provided. A pump/gate state measuring means 600 which executes detection of states of a pump and a gate and measurement of the amount of discharge of the pump is provided. A model preparing means 430 which prepares an external water level model 454 in consideration of the water level and rainfall being measured and a pump model 455 in consideration of the water level, rainfall and amount of discharge of the pump being measured, on the basis of the data from the measuring means 100, 200, 300 and 600, is provided. A model switchover means 453 which selects the external water level model 454 or the pump model 455 corresponding to the states of the pump and the gate is provided. A model preestimate computing means 452 which computes the water level at a future time at a point of object of preestimate by inputting the measured data to a water level preestimate model selected is provided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a river water level prediction device in a river plant monitoring and control device applied to monitoring and control of facilities in a river basin.

[0002]

2. Description of the Related Art FIG. 6 shows a known river water level predicting device in a river plant monitoring and control device. In the figure, reference numeral 100 denotes an upstream water level measuring means for measuring a water level upstream of a river prediction target point. Reference numeral 200 denotes a downstream water level measuring unit that measures the water level at the prediction target point and the water level or tide level downstream thereof.

[0003] Reference numeral 300 denotes rainfall measuring means for measuring rainfall at a certain point in a river basin. Reference numeral 500 denotes an operator who monitors the river plant based on the water level prediction information and performs operation and control. Reference numeral 400 denotes a river water level prediction device that predicts a river water level. Reference numeral 410 denotes a water level / rainfall data storage unit for storing the water level / rainfall data measured by the upstream water level / downstream water level / rainfall measurement unit.

[0004] Reference numeral 420 denotes the following (1) and (2) for selecting water level / rainfall data to be adopted as input data of a river water level model from the water level / rainfall data storage means 410.
Data processing means for performing the above processing. (1) Normalization processing: the maximum and minimum values set in advance for each of the water level and rainfall data measured by the upstream water level measurement means 100, the downstream water level measurement means 200, and the rainfall measurement means 300 and stored in the storage means 410 Is converted to a value in the range of 0 to 1 on the basis of (2) Correlation calculation processing: A correlation value between output data and input data is obtained, and a variable having a high correlation value is adopted as input data.

[0005] Reference numeral 430 represents the water level / rainfall data selected by the data processing means 420 as input data.
This is a model creating means for creating a water level prediction model at a prediction target point by performing a calculation by the multiple regression analysis method described below and calculating a coefficient of each input data. Multiple regression analysis method: The relationship between input and output data of n inputs and 1 output is identified using a statistical method, and the coefficient of the input data is determined by the present analysis according to the magnitude of the influence of the input data on the change of the output data. .

[0006] Reference numeral 440 denotes a model storage unit for storing the water level prediction model created by the model creation unit 430. Reference numeral 450 denotes a model sequential updating unit 45 described later.
1 is a water level predicting means for calculating a water level predicted value at a future time designated by the operator 500 using the model predicting calculation means 452. Reference numeral 451 denotes a model sequential updating unit that adds the trend of the latest data to the model by using the sequential least squares method while maintaining the trend of the current model, and improves the prediction accuracy online.

Reference numeral 452 denotes a model successive updating unit 451.
The water level / rainfall data storage means 410
The latest and past (e.g., currently one hour ago,
The water level / rainfall data (2 hours before, 3 hours before 4 hours, etc.) are input after normalization by the data processing means 420, and the operator 5
Model prediction calculating means for calculating a predicted water level at a designated future time of 00 (for example, 30 minutes later or 1 hour later).

Reference numeral 460 denotes an input means serving as an interface for using the river water level prediction device 400, for example, when the operator 500 designates a water level prediction time. Reference numeral 470
Is a display unit serving as an interface for using the river water level prediction device 400 such that the operator 500 obtains a predicted water level. Reference numeral 480 is an average error between a measured value and a predicted value during a series of water level changes from a rise in water level after the start of rainfall to a decrease to a normal water level, and the value is used to determine an error determined by the operator 500. This is a prediction error determination unit that determines whether the value is larger than a threshold value.

Reference numeral 490 denotes a case where the prediction error determination means 480 determines that the error is larger than the threshold value, and adds the rainfall / water level time series data at the time of the series of water level changes to the model creation data up to that time. This is a model updating unit that creates new model creation data, creates and updates a model with this data, and stores it in the model storage unit 440.

According to the above river water level prediction device, the water level is measured by the upstream water level measurement means at the point to be predicted and the upstream point, and the water level or the tide level is measured by the downstream water level measurement means at the downstream side point. Also, the rainfall at the upstream point is measured by the rainfall measuring means, and these measured data are stored in the water level / rainfall data storage means 410. From these measured data, one hour ago, two hours ago, three hours ago Each data four hours before is input to the water level prediction model, and the water level at the prediction target point one hour later can be predicted.

[0011]

However, in the above-mentioned river water level prediction device, for example, the prediction target point is the river water level near the upstream of the pump or gate equipment, and the river is the water level of the pump or gate equipment. In the case of a small river whose water level changes greatly due to the operation, the water level is affected by the operation of the pump and gate, and the water level change tendency before and after the operation greatly changes, and the prediction error during operation and after the operation is large. (Hereinafter, the water level near the upstream of the pump and gate equipment is referred to as the internal water level, and the confluent mainstream downstream of the pump and gate equipment is referred to as the external water level: see FIG. 2).

The present invention has been made in view of the above circumstances, and has a river water level prediction that can significantly reduce prediction errors and improve the reliability of prediction even near the upstream of a pump or a gate facility. It is intended to provide a device.

[0013]

In order to achieve the above object, a river water level prediction apparatus according to claim 1 is a river water level prediction apparatus for predicting a water level in a facility in a river basin having a gate and a pump. Water level measuring means for measuring the water level at a plurality of locations such as upstream and downstream of the river, rainfall measuring means for measuring the amount of rainfall in the river basin, and detecting the state of the pump and the gate, A pump gate state measuring means for measuring the discharge amount, based on data from the water level measuring means, the rainfall measuring means and the pump gate state measuring means, a water level prediction model, taking into account the measured water level and rainfall amount. Model creation means for creating a water level model and a pump model in consideration of the measured water level, rainfall, and the discharge rate of the pump; The outer level model or pump model created by means, and the model change-over means for selecting in accordance with the operating state and closed state of the gate of said pump,
The measured data is input to either the external water level model or the pump model water level prediction model selected by the model switching means, and the water level at a future time at the prediction target point at which the facility in the river basin is installed is calculated. And a model prediction calculation means for obtaining the same.

That is, as the water level prediction model used when calculating the water level prediction by the model prediction calculation means, an external water level model in consideration of the measured water level and rainfall, and the measured water level, rainfall and pump discharge Model creation means for creating a pump model in consideration of the above, and model switching means for selecting an outside water level model or a pump model according to the operating state of the pump and the open / closed state of the gate are provided. The water level is predicted in consideration of the influence of the water level change due to the pump discharge amount, and the prediction error at the time of operating the pump is improved. In addition, the operation / stop state of the pump changes depending on the rainfall and the water level, but the outside water level model and the pump model are switched by the model switching means according to the operation state of the pump, so that the predicted value has the best accuracy and precision. Is done.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a river water level prediction apparatus according to the present invention will be described below with reference to the drawings. FIG. 1 shows a river water level prediction device 400 according to the present embodiment, which is applied to a river plant monitoring and control device. It should be noted that reference numerals 100, 200, 300, 420, and 43 in FIG.
0,440,451,452,460,470,48
Reference numerals 0, 490, and 500 denote the same means as in the conventional example shown in FIG.
0, 453, 454, 455, 600. The river water level prediction device 400 is provided with these components. First, the configuration of these characteristic portions will be described below.

Reference numeral 411 denotes a pump related to an upstream water level, a downstream water level, and water level / rainfall data measured by the rainfall measuring means, a pump discharge amount measured by a pump / gate state measuring means 600 described later, and a pump related to a gate open / closed state. Water level, rainfall, pump, and gate data storage means for storing gate data. Reference numeral 450 denotes a water level prediction unit that represents a feature of the present invention, and includes a model successive update unit 451, a model prediction calculation unit 452, and an outside water level model 4 described later.
Model switching means 4 for selecting pump model 455 and pump model 455
Using 53, a predicted water level at a future time specified by the operator 500 is calculated.

Reference numeral 453 denotes a model prediction calculating means 452 for selecting an appropriate water level prediction model in accordance with the equipment near the prediction target point, ie, the operating state of the pump and gate.
This is a model switching means for switching the water level prediction model used in. As an example of the prediction target, there is an internal water level in the rainwater drainage station as shown in FIG. 2, but when the general facilities are not operating, that is, when the pump P is stopped or the gate G is open, the internal water level = the external water level. Therefore, the external water level model 4 similar to the conventional method
54 is used. When the equipment is operating, that is, when the pump P is operating and the gate G is closed, the pump model 455 is used. The description of each model is shown below.

Reference numeral 454 denotes the latest and past water level / rainfall data stored in the water level / rainfall / pump / gate data storage means 411, similarly to the model used in the model prediction calculation means 452 in the conventional method of FIG. It is a water level prediction model that outputs a predicted water level at a future time specified by the operator 500 as an input, that is, an outside water level model. Code 4
Reference numeral 55 denotes a water level prediction model that inputs a pump discharge amount together with water level / rainfall data and outputs a predicted water level at a future time specified by the operator 500, that is, a pump model.

Reference numeral 600 denotes a pump / gate state measuring unit which is located near the downstream of the prediction target point A and measures the operation state and discharge amount of the pump P and the gate G which influence the change tendency of the predicted water level.

The reference numerals 100, 200, 300,
420, 430, 440, 451, 452, 460, 4
As described above, 70, 480, 490, and 500 are the same means as in the conventional example shown in FIG. 6, but will be described below.

Reference numeral 100 denotes an upstream water level measuring means for measuring the water level upstream of the river prediction point A. Reference numeral 200 denotes a downstream water level measuring unit that measures the water level of the prediction target point A and the water level or tide level downstream thereof. Reference numeral 300 denotes a rainfall measurement unit that measures rainfall at a certain point in a river basin. Reference numeral 500 denotes an operator who monitors the river plant based on the water level prediction information and performs operation and control.

Reference numeral 420 denotes the above-described normalization processing and correlation calculation processing for selecting water level / rainfall / pump / gate data to be adopted as input data of the river water level model from the water level / rainfall / pump / gate data storage means 411. Data processing means for performing Reference numeral 430 denotes a prediction target by taking in the water level, rainfall, pump, and gate data selected by the data processing means 420 as input data, performing a calculation by the multiple regression analysis method described above, and calculating a coefficient of each input data. This is a model creation means for creating a water level prediction model at a point.

Reference numeral 440 is a model storage unit for storing the water level prediction model created by the model creation unit 430. Reference numeral 450 denotes a water level prediction unit that calculates a predicted water level at a future time specified by the operator 500 using the model successive updating unit 451, the model switching unit, and the model prediction calculation unit 452.

Reference numeral 451 represents a recursive least squares method.
This is a model sequential updating means that adds the trend of the latest data to the model while maintaining the trend of the current model, and improves the prediction accuracy online. Reference numeral 452 denotes the latest and past (for example, one hour before, two hours before, three hours ago) stored in the water level / rainfall / pump / gate data storage unit 411 in the model obtained by the model successive updating unit 451. 4
The water level, rainfall, pump, and gate data (before time, etc.) are input after normalization by the data processing means 420, and the operator 500
Is a model prediction calculating means for calculating a predicted water level at a future time designated (for example, 30 minutes later or 1 hour later).

Reference numeral 460 denotes an input means serving as an interface for using the river water level prediction device 400, for example, when the operator 500 designates a water level prediction time. Reference numeral 470
Is a display unit serving as an interface for using the river water level prediction device 400 such that the operator 500 obtains a predicted water level.

Reference numeral 480 denotes an average error between a measured value and a predicted value during a series of water level changes from a rise in water level after the start of rainfall to a decrease to a normal water level, and the value is designated by the operator 500. Prediction error determination means for determining whether or not the difference is larger than the determined error determination threshold value.

Reference numeral 490 denotes a case where when the prediction error determination means 480 determines that the error is larger than the threshold value, the rainfall / water level time series data at the time of the water level change is added to the model creation data up to that time. This is a model updating unit that creates new model creation data, creates and updates a model with this data, and stores it in the model storage unit 440. Reference numeral 500 denotes an operator who monitors the river plant based on the water level prediction information and performs operation and control.

FIG. 2 shows a river basin of a river to be predicted in which the above-mentioned river water level prediction device 400 is installed. FIG. 2 is an explanatory diagram of a basin of a river to be predicted for which the water level is predicted by the river water level prediction device 400 shown in FIG. 1 described above, where A is a water level prediction target point near the upstream of the rainwater drainage plant (inland water level). , B is an upstream water level rainfall measurement point, C is a downstream water level measurement point, and D is an upstream water level rainfall measurement point in a small river connected to the rainwater drainage station. A 'indicates a point (outside water level) near the downstream of the rainwater drainage station.

At the points A, B, C, and D shown in FIG. 2, the variables of the prediction model in the case where the water level at the prediction target point A after one hour is predicted by the river water level prediction device 400 are compared with those of the conventional example. The results are shown in Table 1 below.

[0030]

[Table 1]

In the embodiment of the present invention, at the prediction target point A, the upstream water level rainfall measurement point B and the upstream water level rainfall measurement point D, the water level and rainfall are measured by the upstream water level measurement means 100, and the downstream water level measurement point is measured. In C, the water level or the tide level is measured by the downstream water level measurement means 200, and these measurement data are stored in the water level / rainfall / pump / gate data storage means 411.

Also, pump / gate state measuring means 600
Then, the operating state of the pump P and the open / closed state of the gate G are detected, and when the pump P is operating, the discharge amount of the pump P is measured. These data are, as shown in Table 1, when the pump P is stopped and the gate G is open (outside water level model),
Operation and when the gate G is closed (pump model).
From these data, as shown in Table 1,
The data before time, two hours before, three hours before, and four hours before are input to the outside water level model or the pump model, and the water level at the prediction target point A one hour later is predicted.

On the other hand, in the conventional example, as shown in the left column of Table 1, water level data at the prediction target point A, the upstream water level rain measurement point B and the downstream water level rain measurement point C, and the water level data at the upstream point B Only the rainfall data is input to one prediction model to predict the water level at the prediction target point A one hour later.
That is, unlike the present embodiment, the water level, the rainfall amount, the operation state of the pump P, and the open / close state of the gate G at the upstream point D of the small river are not taken into consideration.

Here, such points A, B, C, D
Based on the input data and pump / gate data at
The model structure of the water level prediction model for calculating the water level at point A one hour later (a prediction formula based on multiple regression analysis) is shown in the following (1),
Equation (2) shows. Equation (1) shows the outside water level model, and equation (2) shows the pump model.

[0035]

## EQU1 ## A outside water level (+1) = a0 * A outside water level (0) + a1 * A outside water level (-1) + b0 * B water level (0) + b1 * B water level (-1) + b2 * B water level (-2) + b3 * B water level (-3) + b4 * B water level (-4) + c0 * C water level (0) + c1 * C water level (-1) + c2 * C water level (-2) + c3 * C water level (-3) + br0 * B rainfall (0) + br1 * B rainfall (-1) + br2 * B rainfall (-2) + br3 * B rainfall (-3) + br4 * B rainfall ( -4) …… (1) Water level in A (+1) = a0 * Water level in A (0) + a1 * Water level in A (-1) + a2 * Water level in A (-2) + d0 * D water level ( 0) + d1 * D water level (-1) + d2 * D water level (-2) + d3 * D water level (-3) + d4 * D water level (-4) + dr0 * D rainfall (0) + dr1 * D Rainfall (-1) + dr2 * D rainfall (-2) + dr3 * D rainfall (-3) + dr4 * D rainfall (-4) + p0 * P discharge rate (0) + p1 * P discharge rate (-1 ) + p2 * P discharge amount (-2) + p3 * P discharge amount (-3) …… (2)

In the above equations (1) and (2), a0, a
1, b0 to b4, c0 to c3, br0 to br4, d0 to d4, dr0 to dr4, p0
Pp3 are constants, and A, B, C, and D represent respective points. The numbers in parentheses indicate time, (+1) is one hour later, (0) is present, (-1) is one hour ago, (-2) is two hours ago, (-3) ) Is 3 hours ago and (-4) is 4 hours ago.

That is, A water level (n) is the water level at point A at time n, B water level (n) is the water level at point B at time n, and C water level (n) is the water level at point C at time n. Water level, B rain
(n) is the amount of rainfall at point B at time n, A internal water level (n) is the internal water level at point A at time n, D water level (n) is the water level at point D at time n, D rainfall (n ) Indicates the rainfall amount at the point D at time n, and the P discharge amount (n) indicates the discharge amount of the pump P at time n.

Here, the model structure of the conventional example is shown as equation (3) in order to compare it with the model structure of the water level prediction model at one hour later at the prediction target point A in the apparatus of the present embodiment.

[0039]

## EQU2 ## A outside water level (+1) = a0 * A outside water level (0) + a1 * A outside water level (-1) + b0 * B water level (0) + b1 * B water level (-1) + b2 * B water level (-2) + b3 * B water level (-3) + b4 * B water level (-4) + c0 * C water level (0) + c1 * C water level (-1) + c2 * C water level (-2) + c3 * C water level (-3) + br0 * B rainfall (0) + br1 * B rainfall (-1) + br2 * B rainfall (-2) + br3 * B rainfall (-3) + br4 * B rainfall ( -4) …… (3)

As described above, conventionally, the water level one hour later at the prediction target point A is predicted only by the water level time changes at the points A, B, and C and the rain time change at the point B. That is, the water level of the prediction target point A is predicted using only the expression (3) similar to the expression (1) in the present embodiment.

FIGS. 3 and 4 are flowcharts showing the flow of processing in the river water level prediction device 400 according to the embodiment of the present invention. The outline of the processing will be described below with reference to the block diagram of FIG. I do.

In the figure, when the rainfall starts and the prediction starts, first, in step S1, the upstream water level measuring means 100, the downstream water level measuring means 200, and the rainfall measuring means 3
At 00, water levels at points A, B, C, and D, and rainfall at points B and D are measured. Further, the operating state of the pump P and the open / closed state of the gate G are detected by the pump / gate state measuring means 600. These data are stored in the water level / rainfall / pump / gate data storage unit 411.

Next, in step S2, data processing such as normalization and correlation analysis is performed by the data processing means 420, and in step S3, the coefficients of the input data are determined by the model creation means 430 using a statistical method. Model creation processing is performed, and the coefficients of the created model are stored and stored in the model storage unit 440 in step S4.

Next, in step S5, a water level prediction process is performed by the water level prediction means 450. The processing in step S5 will be described. First, the coefficients of the water level prediction model stored in the model storage unit 440 are read (step S5-1), and the model successive updating unit 451 is executed.
The latest data is added to the coefficient read in step (1), and the data is sequentially processed and corrected to the latest data (step S).
5-2), and save the latest updated model coefficient data (step S5-3).

Here, the water level prediction model for which the coefficient is processed includes the outside water level model 454 and the pump model 455 as described above, and therefore, the model switching means 453 is used.
Selects an appropriate water level prediction model from the outside water level model 454 or the pump model 455 in accordance with the operation state of the facilities near the prediction target point A downstream, that is, the operation state of the pump P and the open / closed state of the gate G, Model prediction calculation means 45
Switch the water level prediction model used in step 2 (step S5)
-4). That is, the model switching means 453 reads the states of the pump P and the gate G (step S5-4-).
1) Switching to outside water level model 454 (step S5)
-4-2) or switching to the pump model 455 (step S5-4-3).

Thereafter, a model prediction operation is performed by the model prediction operation means 452 using the updated model coefficient data (step S5-5).
70 is displayed (step S5-6).

As shown in Table 1 above, this calculation processing is as follows.
The current water level of the prediction target point A, the current at points B, C, and D, one hour before, two hours, three hours, and four hours ago, the water level data at points B and D, and one hour before 2 hours ago, 3
Using the rainfall data four hours ago and four hours ago, (1),
The water level at the prediction target point A one hour later is obtained by the equation (2).

Next, in step S6, step S
The prediction value calculated in step 5 is compared with the current value to calculate a prediction error. Then, in step S7, based on the value predicted by the calculation in step S5, the end of rainfall is predicted, and it is determined whether the change in the predicted water level has ended (stable). If not, the process returns to step S1. Is repeated, and if it is completed, the process proceeds to step S8.

Thereafter, the new data file is stored in step S8, and in step S9, the prediction error determination means 480 performs a prediction error average value calculation process of calculating the average of the error between the actually measured value and the prediction value.

In step S10, the operator 500
Determines whether the calculated average error is smaller than a target value that is a specified threshold, and ends the process if smaller. If it is larger, in step S11, the model updating means 490 uses the model
A model update process for creating new model creation data from the water level time series data is performed, and the process returns to step S4, and the model coefficient is stored in the model storage unit 440.

In the figure, steps S3, S4, S8,
Steps S9, S10, and S11 are offline processes performed outside the online water level prediction process by the water level prediction unit 450 in step S5.

FIG. 5 is a graph showing the predicted results of the actual water level at the prediction target point A by the river water level predicting apparatus 400 of the present embodiment and the predicted results by the conventional river water level predicting apparatus. The value and the dashed line are the predicted values by the apparatus of the present embodiment, and the thin solid lines are the predicted values by the conventional apparatus. As shown in this graph, it can be seen that the predicted value by the river water level prediction device 400 of the present embodiment is closer to the actually measured value than the predicted value by the conventional device. That is, the river water level prediction device 400 of the present embodiment
According to this, the prediction error due to the influence of the operation of the pump P and the gate G can be suppressed to an extremely small value.

As described above, in the example of predicting the internal water level of the rainwater drainage station, when the pump P is stopped (when the gate G is open), the internal water level is equal to the external water level, and the water level change tendency is between the present and the past. Determined by water level and rainfall data. By the way, when the pump P is operated (when the gate G is closed), the inner water level and the outer water level are shut off by the gate G, and only the inner water level is lowered due to the pump drainage. If is used as it is, the prediction error will be large (the outside water level does not change in the water level change tendency due to the operation of the pump P because the river scale is larger than the inside water level).

However, according to the river water level prediction device 400 of the above embodiment, when the pump P is operated, the discharge amount of the pump P also affects the water level change. Since the pump model 455 added as input data is used, a prediction in consideration of a tendency of a water level drop due to pump drainage is realized, and a prediction error when the pump P is operated can be improved. Moreover, the operation / stop state of the pump P changes depending on the rainfall and water level, but in the apparatus according to the present embodiment, the two models described above, that is, the two models described above, By switching between the outside water level model 454 and the pump model 455, the prediction value can be set to the best accuracy.

[0055]

As described above, according to the river water level prediction apparatus of the present invention, the following effects can be obtained. According to the river water level prediction apparatus of the first aspect, in a facility in a river basin, a water level prediction model used when calculating a water level prediction by a model prediction calculation means, in consideration of a measured water level and a rainfall amount. A model for preparing a water level model and a pump model in consideration of a measured water level, a rainfall amount, and a discharge amount of a pump is provided, and a model for selecting an outside water level model or a pump model according to a pump operating state and a gate opening / closing state. Since the switching means is provided, even when the pump is operating, the water level can be predicted in consideration of the effect of the water level change due to the pump discharge amount, and the prediction error at the time of operating the pump can be improved. In addition, although the operation / stop state of the pump changes depending on the rainfall and water level, the above-mentioned two models, ie, the external water level model and the pump model, are switched by the model switching means according to the operation state of the pump. , The predicted value can be of the best accuracy.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a river water level prediction device illustrating a configuration and functions of a river water level prediction device according to an embodiment of the present invention.

FIG. 2 is a schematic layout diagram of a river explaining a river basin where a river water level prediction device according to an embodiment of the present invention is installed.

FIG. 3 is a flowchart illustrating a flow of a water level prediction process performed by the river water level prediction apparatus according to the embodiment of this invention.

FIG. 4 is a flowchart illustrating a flow of a water level prediction process performed by the river water level prediction device according to the embodiment of this invention.

FIG. 5 is a graph showing a comparison between an actual measured water level, a predicted water level by the river water level prediction device of the present embodiment, and a predicted water level by the conventional river water level prediction device.

FIG. 6 is a functional block diagram of a river water level prediction device illustrating the configuration and functions of a conventional river water level prediction device.

[Explanation of symbols]

 Reference Signs List 100 upstream water level measurement means (water level measurement means) 200 downstream water level measurement means (water level measurement means) 300 rainfall measurement means 400 river water level prediction device 430 model creation means 452 model prediction calculation means 453 model switching means 454 outside water level model (water level prediction model ) 455 pump model (water level prediction model) 600 pump gate state measuring means A prediction target point G gate P pump

Claims (1)

[Claims] 1. A river water level prediction device for predicting a water level in a facility in a river basin having a gate and a pump, comprising: a water level measuring means for measuring water levels at a plurality of locations such as upstream and downstream of the river; Rainfall measuring means for measuring the amount of rainfall in the basin, and detecting the state of the pump and the gate,
Pump gate state measuring means for measuring the discharge amount of the pump, Based on data from the water level measuring means, the rainfall measuring means and the pump gate state measuring means, the measured water level and rainfall as a water level prediction model Model creation means for creating a pump model in consideration of the outside water level model considered and the measured water level, rainfall, and the discharge rate of the pump; and the outside water level model or the pump model created by the model creation means, Model switching means for selecting according to the operating state of the gate and the opening / closing state of the gate; and inputting measurement data to any of the water level prediction models of the outside water level model or the pump model selected by the model switching means, Model prediction calculation means for calculating and calculating a water level at a future time at a prediction target point where facilities in a river basin are installed. River level prediction device, characterized in that to become to.