TWI490811B - Support facilities for storage facilities, use of reservoir support facilities and support facilities for storage facilities - Google Patents

Description

Water storage facility operation support system, water storage facility operation support method and water storage facility operation support program

The present invention relates to a water storage facility operation support system, a water storage facility operation support method, and a program.

In order to efficiently perform hydroelectric power generation, an application support system caused by a computer is used. For example, in Patent Document 1, a generator operation plan and a water level plan that are automatically calculated based on the amount of water discharged on the first day, or a performance data of a past generator operation plan and a water level plan are displayed. And, in response to the need to accept the amendment, and to create a power generation application plan.

[Previous Technical Literature]

[Patent Literature]

[Patent Document 1] Japanese Patent Laid-Open No. 2006-39838

However, in the device described in Patent Document 1, although the water level is planned based on the amount of water discharged or the actual performance, the relationship between the water level plan and the generated power is not noticed, and it is impossible to A water level plan capable of achieving an increase in power generation is performed.

The present invention has been made in view of such a background, and an object of the present invention is to provide a system and method for supporting the operation of a water storage facility so that the amount of generated electric power due to hydroelectric power generation can be increased. And the program.

In order to solve the above-described problems, the main invention of the present invention is a system for supporting the operation of a water storage facility for hydroelectric power generation, characterized in that it is provided with an optimum water level gauge and a specific configuration. The water level at the start time of the unit period during the operation period is additionally associated with the inflow amount of the aforementioned water storage facility in the aforementioned unit period, and is optimal as the optimum water level at the end time point of the aforementioned unit period. The water level is obtained by the water level acquisition unit, and the water level of the water storage facility at the start time of the operation period is obtained; and the predicted inflow amount acquisition unit obtains a predicted value of the inflow amount in each of the unit periods; And the water level planing unit reads the optimum water level corresponding to the water level at the start time point and the predicted value of the inflow amount in the unit period for each of the unit periods, and reads the optimum water level from the optimum water level table. And, the aforementioned optimum water level read out is taken as the water level at the start time point of the unit period below one of the unit periods; The water level plan memory unit stores the water level of the water storage facility at the current time point at the end of each of the unit periods. In the current water level, the predicted inflow amount acquisition unit acquires a predicted value of the inflow amount after a unit period of the next unit period, and the water level plan unit uses the current water level as the next unit. The water level at the start time of the period, and for the aforementioned unit period after the next unit period, the above-mentioned optimum water level is read from the above-mentioned optimum water level table, and the optimum water level read out by the foregoing Update the aforementioned water level plan memory unit.

Further, in the water storage facility operation support system of the present invention, the predicted inflow amount acquisition unit may acquire a predicted value of the plurality of inflow amounts in each of the unit periods, and The probability of occurrence of the predicted value, the optimum water level table is based on the water level that the water storage facility is likely to be, the predicted value of the inflow amount in each of the unit periods, and the probability of occurrence of the predicted value of the inflow amount. It is created by the probability dynamic programming method.

Further, in the water storage facility operation support system of the present invention, the storage system may be configured to include a database for memorizing the actual value of the inflow amount for each of the unit periods. And the mode memory unit, which has a statistical mode in which the inflow amount at the second unit period immediately before a certain first unit period and the balanced inflow amount which is a specific constant are both The difference between the two is the explanatory variable, and the amount of increase in the inflow amount from the second unit period to the first unit period is the target variable; and the inflow amount pattern estimating unit is based on the memory in the aforementioned database. Regression analysis is performed on the actual value of the inflow amount and the statistical pattern, and the regression coefficient and the equalization inflow amount and the residual term frequency distribution generation unit are described for each unit period. And the first actual value of the inflow amount corresponding to the second unit period immediately before the first unit period of the unit period, and the first The second actual value corresponding to the inflow amount is read from the database, and the difference between the first and second actual performance values is subtracted from the balanced inflow amount estimated from the above. After the second actual value is multiplied by the value obtained by the regression coefficient, the residual term is calculated, and the frequency distribution of the residual term is generated; and the drift term frequency distribution generating unit is for each unit period described above. The first actual value of the inflow amount corresponding to the second unit period and the second actual value of the inflow amount corresponding to the first period are read from the database, and the calculation is performed from the foregoing a value obtained by subtracting the second actual value from the equalized inflow amount, multiplying the regression coefficient, calculating a drift term, and generating a frequency distribution of the drift term; and a probability distribution generating unit The frequency distribution of the residual term and the frequency distribution of the drift term are added to generate a frequency distribution of the inflow amount, and a probability distribution of the inflow amount is generated from the frequency distribution of the inflow amount, The predicted inflow amount acquisition unit calculates, as the predicted value of the inflow amount, each of the values of the specific range in which the inflow amount is generated, and calculates the occurrence of the predicted value of the inflow amount based on the occurrence probability distribution. Probability.

Further, another aspect of the present invention is a method for supporting the operation of a water storage facility for hydroelectric power generation, characterized in that it is provided with a start time point of a unit period during which a specific operation period is constituted. The water level and the computer that is the optimal water level table for the optimum water level at the most appropriate water level at the end time of the above-mentioned unit period are added in the same manner as the inflow amount of the aforementioned water storage facility in the aforementioned unit period, and the water level is performed. The process of obtaining the water level of the water storage facility at the start time of the operation period, and obtaining a predicted value of the inflow amount in each of the unit periods before the start of the operation period; And the water level at the start time point and the predicted water level corresponding to the inflow amount in the unit period are read from the optimum water level table, and the read optimal water level is taken as The water level at the start time of the unit period below the unit period; the aforementioned optimum water for each unit period mentioned above Memorizing in the memory, at the end of each of the aforementioned unit periods, obtaining the current water level of the water level of the water storage facility at the current time point; obtaining the aforementioned inflow after the unit period of the next unit period The predicted value of the quantity; the current water level is used as the water level at the start time point of the next unit period, and the aforementioned optimum water level is from the aforementioned optimum water level table for each of the aforementioned unit periods after the next unit period The reading is performed; the memory is updated by the optimum water level read as described above.

Further, another aspect of the present invention is a program for supporting the operation of a water storage facility for hydroelectric power generation, characterized in that it is provided with a start time point of a unit period in which a specific operation period is constituted. The water level and the computer that is the optimal water level table for the optimum water level at the most appropriate water level at the end time of the above-mentioned unit period are correspondingly added to the inflow amount of the aforementioned water storage facility in the aforementioned unit period, and the computer is implemented. a step of obtaining a water level of the water storage facility at a start time point of the operation period, and before the start of the operation period, performing a step of obtaining a predicted value of an inflow amount in each of the unit periods; For each of the unit periods described above, the water level corresponding to the water level at the start time point and the predicted value of the inflow amount in the unit period is read from the optimum water level table, and read out. The aforementioned optimum water level as a step of the water level at the start time point of the unit period below one of the unit periods; The step of storing the above-mentioned optimum water level in each unit period in the memory, at the end of each of the foregoing unit periods, is performed: obtaining the current water level of the water level of the aforementioned water storage facility at the current time point; a step of predicting the aforementioned inflow amount after the unit period of the unit period; the current water level is used as the water level at the start time point of the next unit period, and for the next unit period In the above-mentioned respective unit periods, the step of reading the optimum water level from the optimum water level table; and updating the memory by the read optimal water level.

The problems disclosed in the present application or the solutions thereof can be understood from the description of the embodiments of the invention and the description of the drawings.

According to the present invention, the amount of generated electric power due to hydroelectric power generation can be increased.

[System Components]

Hereinafter, an operation support system for a water storage facility according to one embodiment of the present invention will be described. The operation support system of the present embodiment performs support for appropriately applying the water level in the water storage facility such as a reservoir. In the operation support system of the present embodiment, the amount of precipitation is predicted using the form of the past weather, and the amount of water flowing into the water storage facility by the river or the precipitation is calculated using the statistical mode (below) The probability distribution of simply referred to as the inflow amount, and then plan the water level in such a way as to maximize the amount of generated electricity.

Fig. 1 is a view showing the overall configuration of the operation support system of the present embodiment. The operation support system of the present embodiment includes a precipitation amount prediction system 10, an inflow amount prediction system 20, and an operation planning system 30. The precipitation amount prediction system 10, the inflow amount prediction system 20, and the operation planning system 30 are respectively connected to the communication network 40 and are capable of communicating with each other. The communication network 40 is, for example, an Internet or a Local Area Network (LAN), and is constructed by an Ethernet (registered trademark) or public telephone line network, a wireless communication network, or the like.

The precipitation prediction system 10 is a computer that predicts the amount of precipitation. The inflow amount prediction system 20 is a computer that predicts the inflow amount flowing into the water storage facility. In the present embodiment, the inflow amount prediction system 20 calculates the probability distribution of the inflow amount and the inflow amount. The use of the planning system 30 is the most appropriate water level plan in the water storage facility. In the precipitation amount prediction system 10, the inflow amount prediction system 20, and the operation planning system 30, for example, a personal computer or a computer such as a workstation, a PDA (Personal Digital Assistant), or a mobile phone terminal can be used.

[precipitation prediction system 10]

The precipitation prediction system 10 predicts the amount of precipitation. In the present embodiment, the precipitation amount prediction system 10 morphs the profile of the past weather in advance, and shapes the future weather forecast, and uses the percentile of the precipitation of the past date of the same form. (percentile) value to predict precipitation.

2 is a diagram showing the hardware configuration of the precipitation prediction system 10. The precipitation amount prediction system 10 includes a CPU 101, a memory 102, a memory device 103, a communication interface 104, an input device 105, and an output device 106. The memory device 103 stores various programs or materials, such as a hard disk or a flash memory, a CD-ROM drive, and the like. The CPU 101 executes the reading of the program stored in the memory device 103 into the memory 102, thereby realizing various functions. The communication interface 104 is an interface for connecting to the communication network 40. The communication interface 104 is constructed, for example, by an Ethernet (registered trademark) or public telephone line network, a wireless communication network, or the like. The input device 105 receives input of data from a user, and is, for example, a keyboard or a mouse, a touch panel, a microphone, or the like. The output device 106 outputs the data, for example, a display or a printer, a speaker, and the like.

FIG. 3 is a diagram showing the software composition of the precipitation prediction system 10. The precipitation amount prediction system 10 includes a weather profile acquisition unit 111, a weather form registration unit 112, a predicted precipitation amount acquisition request receiving unit 113, a weather forecast acquisition unit 114, a precipitation amount prediction unit 115, and a predicted precipitation amount transmission unit. 116. A weather profile database 131 and a weather profile database 132. In addition, the weather profile acquisition unit 111, the weather form registration unit 112, the predicted precipitation amount acquisition request receiving unit 113, the weather forecast acquisition unit 114, the precipitation amount prediction unit 115, and the predicted precipitation amount transmission unit 116 are configured to reduce the amount of precipitation. The CPU 101 included in the prediction system 10 is realized by reading a program stored in the memory device 103 to the memory 102 and executing the same. Further, the weather profile database 131 and the weather form database 132 are realized as the memory 102 provided in the precipitation prediction system 10 or the memory area provided by the memory device 103.

The weather profile database 131 is a memory that records the weather of the past date and the performance of the precipitation (hereinafter referred to as weather profile information). FIG. 4 is a diagram showing a configuration example of weather profile information memorized in the weather profile database 131. As shown in the same figure, the weather profile information includes the date, the total value of the precipitation on the 1st, and the maximum amount of precipitation in the time zone (06:00 to 18:00) during the day. The maximum amount of precipitation in the night time zone (18:00 to 06:00 every other day), text information representing the weather in the time zone during the day (hereinafter referred to as weather profile), and at night The weather profile in the time zone. In addition, the weather profile information is information generally provided by the Meteorological Bureau or the weather company.

The weather profile acquisition unit 111 obtains weather profile information. The weather profile acquisition unit 111 can, for example, access the server operated by the weather bureau or the civil meteorological company and obtain weather profile information. Further, the weather profile acquisition unit 111 may be configured to receive an input of weather profile information from the user.

The weather form database 132 is information for morphing the weather profile for each date (hereinafter, referred to as weather form information). FIG. 5 is a diagram showing a configuration example of weather form information stored in the weather form database 132. In the weather form information, the date is attached to the date, and each term representing the weather (hereinafter referred to as weather term) and the term representing the time of the weather (hereinafter referred to as time term) are included. The flag value of the combination (hereinafter, referred to as weather form). In the present embodiment, the term "weather" is used to mean any of "clear", "yin", "snow", "rain", and "rain", and time words, It is set to any of "time", "one hour", and "after". In addition, in the time language, it is assumed that the blank text is also included (representing that the time is not specified). That is, as shown in FIG. 5, in the present embodiment, the weather patterns are "clear", "yin", "snow", "rain", "heavy rain", "timely heavy rain", " "Heavy rain", "post-heavy rain", "time rain", "seven rain", "post rain", "time snow", "snow snow", "back snow", "time yin", "time yin" , "Back Yin", "Timely Clear", "One Time Sunny" and "Post Sunny".

The weather form registration unit 112 generates weather form information based on the weather profile information, and registers it in the weather form database 132. Figure 6 is a diagram showing the flow of the registration process for weather form information. FIG. 7 is a view showing a configuration of a table (hereinafter, referred to as a form table 161) used in the registration process of the weather form information.

As shown in Fig. 7, in general, the form table 161 is a table in which weather is regarded as a line and time terms are used as columns. Further, in the rank of the form table 161, the items having a strong degree of precipitation are placed above and to the left. In the example of the form table 161 of FIG. 7, the line is arranged in the order of "heavy rain", "rain", "snow", "yin", and "clear", and the column is "one hour" and "time". And "after" are arranged in the order.

The weather profile acquisition unit 111 first acquires weather profile information (S1511).

The weather form registration unit 112 generates a form table 161 in which "0" is set in each cell (S1512). The weather form registration unit 112 is included in the weather profile information, and performs the following processes for each of the daytime weather profile and the nighttime weather profile.

The weather form registration unit 112 removes sentences other than the weather term and the time sentence from the weather profile (S1513), and divides the weather representative at the end of the term representing the weather, and divides the sentence as the weather language. Chunk (S1514). The weather form registration unit 112 sets the value of the form table 161 corresponding to the time phrase and the weather included in the weather block to "1" for each of the divided weather blocks (S1515).

The weather form registration unit 112 performs the above processing for each of the weather profile of the daytime time zone and the weather profile of the night time zone, and then sets the row of the form table 161 to "1". The leftmost one in the search is retrieved, and all the cells on the right side are set to "0" (S1516). The weather form registration unit 112 searches for the top one of the cells set to "1" for each column of the form table 161, and sets all the cells below the searched cell to "below". 0" (S1517).

The weather form registration unit 112 creates a weather form information including a combination of time words and weather terms and date, and registers it in the weather form database 132 (S1518). ).

Fig. 8 is a view for explaining a specific example of registration processing of weather form information based on weather profile information. In the example of Fig. 8, the weather profile 1711 of the time zone included in the weather profile information is "the weather is cloudy after the rain", and the weather profile of the night time zone is 1712, which is "clear after the weather" It’s a moment.

The weather form registration unit 112 divides the daytime weather profile 1711 and divides it after the end of the weather term, that is, "yin", "rain" and "clear", and creates a "yin" and "a rain" And the weather block of "Post" (S1514).

Next, the weather form registration unit 112 sets a cell corresponding to the weather term and the time phrase included in each weather block, that is, the cell 1721 and the "rain" corresponding to the "yin" and the empty text. In the case of the "one-time" corresponding cell 1722 and the cell 1723 corresponding to "clear" and "post", "1" is set (S1515).

The weather form registration unit 112 is also divided into "weather" and "after one time" weather blocks (S1514) for the weather profile 1712 of the night time zone, and corresponds to "clear" and empty text. In the case of the cell 1731 and the corresponding cells 1732 and 1733 corresponding to "yin", "one hour" and "after", "1" is set (S1515).

The weather form registration unit 112 retains only the leftmost "1" of each line of the form table 161, and sets "0" in the other cells (S1516). In the row 174, the leftmost cell 1741 among the cells set to "1" remains, and in the other cells 1742 and 1743, "0" is set.

The weather form registration unit 112 retains only the "1" at the top of each column of the form table 161, and sets "0" in the other cells (S1517). In the row 175, the top cell 1751 in the cell set to "1" remains, and in the other cell 1752, "0" is set.

In the case of performing the above-described processing, in FIG. 8, the form table 161 is only in the cell 1751 corresponding to the "fashion" and the blank, and the cell 1753 corresponding to the "rain" and the "one hour". It is set to "1". Then, according to the form table 161, the weather form information 177 having only "1" set in the "yin" and "one-time rain" is generated, and is registered in the weather form database 132.

The predicted precipitation acquisition request receiving unit 113 is an instruction for receiving a predicted value of precipitation (hereinafter also referred to as predicted precipitation) from another computer connected to the communication network 40 (hereinafter, also It is called the predicted precipitation acquisition requirement). In the predicted precipitation acquisition request, the set value of the percentile (hereinafter referred to as the percentile set value) is set. As the percentile set value, for example, it can be set to any value such as 5% or 10% (0<percentile set value <1).

The weather forecast acquisition unit 114 acquires information on the weather forecast (hereinafter referred to as weather forecast information). The weather forecast acquisition unit 114 may be, for example, a weather forecast information obtained from a weather bureau or a weather company, or may be input as a weather forecast information received from a user. Weather forecast information, including date, and weather forecast at that date. The weather forecast information is different from the weather profile information, and the weather is recorded in units of one day. On the other hand, the weather forecast information is the same as the weather profile, and describes the weather and the time terms that are described in response to the necessity.

The precipitation amount prediction unit 115 calculates the predicted precipitation amount, and the predicted precipitation amount transmission unit 116 transmits the predicted precipitation amount calculated by the precipitation amount prediction unit 115 to the source of the delivery requesting the predicted precipitation amount. The precipitation amount prediction unit 115 morphizes the weather forecast acquired by the weather forecast acquisition unit 114, obtains the amount of precipitation of the past date of the same form, sorts it, and obtains a hundred based on the percentile setting value. The quantile value is used as the predicted precipitation. Figure 9 is a diagram showing the flow of the prediction process for precipitation.

When the weather forecast acquisition unit 114 acquires the weather forecast information (S1531), the precipitation amount prediction unit 115 creates information (hereinafter referred to as forecast weather pattern information) that has been shaped by the weather forecast included in the weather forecast information (hereinafter referred to as forecast weather form information). S1532). Fig. 10 is a diagram showing the flow of the processing of the weather form information according to the weather forecast.

The precipitation amount predicting unit 115 creates the form table 161 of Fig. 7 and sets "0" in all the cells (S1551). The precipitation amount prediction unit 115 removes sentences other than the weather term and the time sentence from the weather forecast (S1552), divides the weather term at the end of each weather term, and divides the divided character string as a weather language. Block (S1553).

The precipitation amount predicting unit 115 sets "1" in the form of the form table 161 corresponding to the time word and the weather term included in the weather block for each weather block (S1554). The precipitation amount prediction unit 115 selects the leftmost one of the cells set to "1" for each row of the pattern table 161, and sets all the cells closer to the right side than the selected cells to " 0" (S1555). Further, the precipitation amount predicting unit 115 selects the uppermost one of the cells set to "1" for each column of the form table 161, and selects all of the lower ones than the selected ones. The grid is set to "0" (S1556).

The precipitation amount prediction unit 115 extracts the value of the cell corresponding to the combination for each combination between the time designation term and the weather term, and creates a weather forecast information set as the extracted value. (S1557).

Returning to Fig. 9, the precipitation amount predicting unit 115 selects the date of the weather forecast information that has been created, and the date in which the date included in the weather forecast information is the same month, from the weather form database 132. The search is performed (S1533), and the amount of precipitation corresponding to the retrieved date is obtained from the weather profile database 131 (S1534). The precipitation amount predicting unit 115 sorts the obtained precipitation amount from small to large (S1535), and multiplies the amount of the obtained precipitation amount by the predicted precipitation amount acquisition request. The value after the percentile set value (and the value after further conversion to an integer value) is set to n (S1536). The precipitation amount predicting unit 115 sets the amount of precipitation of the nth from the front of the sorted precipitation amount as the predicted precipitation amount (S1537).

As described above, according to the precipitation prediction system 10 of the present embodiment, the weather profile in the past is morphologically shaped, and the precipitation is added in association with the precipitation amount, and the future weather forecast is morphified. In addition, the amount of precipitation in the past in the same form can be used as a predicted value in the future. There is a correlation between the weather and the amount of precipitation, which is well known and can be predicted with high precision precipitation. Further, according to the precipitation amount prediction system 10 of the present embodiment, since the characters included in the weather profile can be morphologically and memorized, it is possible to easily set the past days of the same weather as the weather forecast. The amount of precipitation is drawn out.

Further, in the precipitation amount prediction system 10 of the present embodiment, it is assumed that the date is the same as the date included in the weather forecast information, but the date is the same as the date included in the weather forecast information, but It may also be set in the weather form information to include typhoon information indicating whether a typhoon has occurred in the area to be forecasted, or information on whether or not to enter the Meiyu season, and to report information on weather patterns. The matching and typhoon information or the entry into the plum information are consistent, and the same month as the forecast date is searched. In the case of a typhoon and the absence of a typhoon, the situation of entering the plum and the situation of not entering the plum, the form of precipitation is different. Therefore, it is possible to use the actual value of the amount of precipitation on the matching date as the predicted value of the precipitation amount only for the typhoon information or the information of the weather information that is consistent with the information, so that it can be expected to be the same amount of precipitation. The actual value of the form is used to predict the precipitation. Therefore, it is possible to predict the precipitation with higher accuracy.

[Inflow Forecasting System 20]

Next, the inflow amount prediction system 20 will be described. The inflow amount prediction system 20 is a forecast for the inflow into the water storage facility. In the present embodiment, the inflow amount prediction system 20 performs a regression analysis on the past temperature, the amount of precipitation, the amount of snowfall, the amount of snow, the amount of snow melt, and the amount of inflow by using a statistical mode described later, and estimates the regression coefficient. The predicted values of temperature and precipitation (for example, the forecast value caused by weather forecast) and the estimated regression coefficient are applied to the statistical model and the inflow is predicted. In addition, the following configuration is adopted; the prediction of the inflow amount is performed in units of a specific first unit period (in the present embodiment, one day), and the probability distribution of the inflow amount is based on the first The unit period is a unit of a longer second unit period (in the present embodiment, one month).

Figures 11-13 are graphs showing changes in influx. Even in the case of no precipitation or snow melting, there is a specific inflow amount, for example, by oozing from a mountain forest or the like. Therefore, if there is no precipitation or snowmelt, the inflow will be decremented to a specific equilibrium value (hereinafter referred to as the equilibrium inflow) (Fig. 11). On the other hand, if there is precipitation, the amount of water flowing temporarily increases, but it starts to decrease toward the equalization inflow amount at the same time as the stop of the precipitation ( FIG. 12 ). On the other hand, if the temperature rises, snowmelt occurs, and the amount of water flow increases. However, the change in the influence of the snowmelt on the inflow is gentle compared to precipitation (Fig. 13). In this case, for example, when the average temperature rises during the transition from winter to spring, the snow melt may continue to occur.

Therefore, in the present embodiment, attention is paid to the amount of precipitation, the amount of snow melt, and the amount of equalized inflow, and the regression model in which the inflow amount is calculated is used to predict the amount of inflow. In the inflow amount prediction system 20 of the present embodiment, the parameter is calculated based on the actual value of the meteorological data such as the past temperature or the amount of precipitation, and the regression mode described later, and the estimated parameter and the parameter are derived, for example, by The predicted value of the temperature taken out by the weather forecast or the like (hereinafter referred to as the predicted temperature) and the predicted precipitation amount obtained by the precipitation prediction system are applied to the regression mode, and the predicted value of the inflow amount is calculated. The details will be described below.

Fig. 14 is a view showing the hardware configuration of the inflow amount prediction system 20 of the present embodiment. The inflow amount prediction system 20 includes a CPU 201, a memory 202, a memory device 203, a communication interface 204, an input device 205, and an output device 206. The memory device 204 stores various programs or materials, such as a hard disk or a flash memory, a CD-ROM drive, and the like. The CPU 201 executes the reading of the program stored in the memory device 203 into the memory 202, thereby realizing various functions. The communication interface 204 is a connector for connecting to the communication network 40. The communication interface 204 is constructed, for example, by an Ethernet (registered trademark) or public telephone line network, a wireless communication network, or the like. The input device 205 receives input of data from a user, such as a keyboard or a mouse, a touch panel, a microphone, and the like. The output device 206 outputs the data, for example, a display or a printer, a speaker, and the like.

Fig. 15 is a view showing the software configuration of the inflow amount prediction system 20 of the present embodiment. The inflow amount prediction system 20 of the present embodiment includes a snowfall temperature estimation unit 211, a snowmelt amount estimation unit 212, an inflow amount pattern estimation unit 213, a predicted air temperature acquisition unit 214, and a predicted precipitation amount acquisition unit 215. And the predicted inflow amount acquisition request receiving unit 216, the inflow amount predicting unit 217, the predicted inflow amount transmitting unit 218, the inflow amount distribution obtaining request receiving unit 219, and the inflow amount generating probability estimating unit 220, and The inflow amount distribution generating unit 221, the inflow amount distribution transmitting unit 222, the mode storage unit 231, the parameter storage unit 232, and the weather performance database 233.

In addition, the snowfall temperature estimation unit 211, the snowmelt amount estimation unit 212, the inflow amount pattern estimation unit 213, the predicted air temperature acquisition unit 214, the predicted precipitation amount acquisition unit 215, and the predicted inflow amount acquisition request receiving unit 216, And the inflow amount prediction unit 217, the predicted inflow amount transmission unit 218, the inflow amount distribution acquisition request receiving unit 219, the inflow amount generation probability estimation unit 220, the inflow amount distribution generation unit 221, and the inflow amount distribution transmission. The portion 222 is realized by causing the CPU 11 of the inflow amount prediction system 20 to read out the program stored in the memory device 13 into the memory 12 and carry out the operation. Further, the mode storage unit 231, the parameter storage unit 232, and the weather performance database 233 are realized as the memory 202 provided in the inflow amount prediction system 20 or the memory area provided by the memory device 203.

In the meteorological performance database 233, the history of information (hereinafter referred to as meteorological performance information) including various performance values of meteorology is stored. FIG. 16 is a diagram showing a configuration example of weather performance information memorized in the weather performance database 233. As shown in the same figure, in the meteorological performance information, the temperature and precipitation, the amount of snowfall, the amount of snow, the amount of snow melt, and the amount of inflow are included in the corresponding weather. The temperature is the average temperature of the day. Precipitation, snowfall, and snowmelt are the cumulative values of precipitation, snowfall, and snowmelt for one day. The amount of snow is the amount of snow observed on that day. The inflow amount is an integrated value of the amount of water flowing into a water storage facility such as a reservoir during the day. The temperature, precipitation, snowfall, and amount of snow, for example, are provided by the Meteorological Bureau or a civil meteorological company. The inflow amount is a measured value measured in a water storage facility. The inflow amount may be, for example, a measured value of the river flow rate provided by an local government that manages the river. The amount of snow melting may be measured by a weather bureau, a civil meteorological company, a measurement company, or the like, or may be recorded as a real value by a value calculated by a mode described later.

In the mode storage unit 231, various statistical modes based on the meteorological performance information are stored, and the parameter storage unit 232 stores the regression coefficient applied to the mode stored in the mode storage unit 231 or A parameter such as a constant. In the mode memory unit 231, a statistical mode of precipitation amount (hereinafter referred to as precipitation amount mode A1), a statistical mode of snowfall amount (hereinafter referred to as snowfall amount mode A2), and statistics of snow accumulation amount are stored. The mode (hereinafter referred to as the snow amount mode A3), the statistical mode of the snowmelt amount (hereinafter referred to as the snowmelt amount mode A4), and the statistical mode regarding the increase amount of the inflow amount from the previous day (hereinafter, referred to as the inflow amount) The mode A5) and the statistical mode relating to the probability of occurrence of the inflow amount (hereinafter referred to as the inflow probability mode A6). In the parameter storage unit 132, the temperature in which the rain changes to snowfall (hereinafter referred to as snowfall temperature) δ, the equilibrium inflow amount μ _{1 of the} daily unit, and the equilibrium inflow amount μ _{3 of the} monthly unit, and The temperature of the snow melting μ ₂ , and the regression coefficients α ₁ to α ₄ , and β ₁ , and β _{2 will start} . In addition, the equilibrium inflow amount μ ₁ of the daily unit and the temperature μ _{2 at} which the snow melting is started are set as the specific constants and are stored in the parameter storage unit 132 in advance.

In addition, in the following description, the temperature, the amount of precipitation, the amount of snowfall, the amount of snow, the amount of snow melt, and the amount of inflow at the date t are T _t , P _t , S _t , D _t , M _{t ,} and F _t . The amount of increase in the inflow amount from the previous day is set to ΔF _t . The precipitation amount when the temperature T _t is higher than δ is P1 _t , and the precipitation amount when the temperature T _t is δ or less is P2 _t . Further, the inflow amount at the time of m month is set to F _m , and the average of the inflow amount with respect to the past of the m month is set to I _m . The inflow amount normalized by dividing F _m by I _m is F _m ^* , and the amount of increase in the inflow amount is normalized to ΔF _m ^* .

The precipitation amount pattern A1 is a pattern in which the precipitation amount P _t is expressed as P1 _t or P2 _t with the snowfall temperature δ as a boundary, and is expressed by the following formula.

Snowfall mode A2, when the line to the precipitation temperature T _t where δ is less than P2 _t variable of illustration, and snowfall S _t as the object variable regression model, and for performance by the following formula.

S _t = γP 2 _t ‧‧‧(A2)

Snowfall mode A3, the Department for the amount of snow for D _t in line with the plus until the day before the precipitation S _t M _t minus the amount of snowmelt were consistent on the issue of the relationship between _t-1 until the day before the snow amounts D The mode of performance and performance by the following formula.

D _t =( S _t + D _{t -1} - M _t ) ⁺ ‧‧‧(A3)

The snowmelt amount mode A4 is obtained by multiplying the value of the temperature μ ₂ at which the snow melting starts from the temperature T _t by the snow amount D _t , and P1 _t as the explanatory variable, and the snow melting The quantity M _{t is} used as a regression mode of the target variable and is expressed by the following formula.

M _t =α ₂ ( T _t -μ ₂ ) ⁺ D _{t -1} +α ₃ P 1 _t , if ( D _{t -1} >0) ‧‧‧(A4)

In the snowmelt amount mode A4, if the temperature T _t is lower than μ ₂ , the first term is 0, and if the amount of snow D _t-1 until the previous day is 0, the amount of snow melting M _t becomes 0.

The inflow amount mode A5 is a value obtained by subtracting the inflow amount F _t-1 of the previous day from the equilibrium inflow amount μ ₁ and a precipitation amount when the temperature T _t of the day is higher than δ. P1 _t and the amount of snow melted M _{t of the} day are used as explanatory variables, and the amount of increase ΔF _{t of the} inflow is taken as the regression mode of the target variable, and is expressed by the following formula.

Δ F _t =α ₁ (μ ₁ - F _{t -1} )+β ₁ P 1 _t +β ₂ M _t ‧‧‧(A5)

Inflow mode probability A6, based equalization units of months inflow μ _3, and the inflow amount is subtracted from the equalized μ ₃ month before the normalized inflow F _{m 1-*} after the value obtained, and becomes a residual The term u _{m is} used as a explanatory variable, and the amount of increase ΔF _m * of the normalized inflow amount is taken as a regression mode of the target variable, and is expressed by the following formula.

The snowfall temperature estimating unit 211 estimates the snowfall temperature δ based on the precipitation amount pattern A1 and the weather performance information, and registers the estimated snowfall temperature δ in the parameter storage unit 232. The details of the estimation process of δ due to the snowfall temperature estimating unit 211 will be described later.

In addition, the snowfall temperature estimation unit 211 reads the meteorological performance information corresponding to the date t from the meteorological performance database 233 for each date t, and calculates the precipitation amount and the temperature of the meteorological performance information. The snowfall temperature δ estimated above is applied to the aforementioned precipitation mode A1, and P2 _t is calculated. The snowfall temperature estimating unit 211 performs regression analysis based on the meteorological performance information, the P2 _t , and the snowfall amount mode A2, and estimates the regression variable γ. The snowfall temperature estimating unit 211 registers the estimated regression variable γ in the parameter storage unit 232.

The snowmelt amount mode estimating unit 212 is a formula for substituting the snow amount mode A3 into the snow melting amount mode A4.

M _t = α ₂ ( T _t - μ ₂ ) ⁺ ( S _{t -1} + D _{t -2} - M _{t -1} ) ⁺ + α ₃ P 1 _t

A regression analysis was performed and the regression coefficients α ₂ , α _{3 ,} and μ _{2 were} calculated. The snowmelt amount estimation unit 212 registers the estimated regression coefficients α ₂ , α _{3 ,} and μ ₂ in the parameter storage unit 232.

The inflow amount pattern estimating unit 213 estimates the regression coefficients α ₁ , β _{1 ,} and β ₂ based on the inflow amount pattern A5 and the weather performance information. Specifically, the inflow amount pattern estimating unit 213 calculates P1 _t based on the amount of precipitation of the precipitation amount patterns A1, δ and the weather performance information corresponding to the date t for each date _t , and uses each date. The meteorological performance information and the corresponding P1 _{t are} used for regression analysis of the inflow model A5, and the regression coefficients α ₁ , β ₁ , β _{2 ,} and μ ₁ are derived. When there is no snowmelt in the meteorological performance information, the weather information of the date t-1 and other weather information of the date t may be used, and the M _t calculated by the snow melting mode A4 may be used. The regression analysis is performed on the inflow mode A5, and the regression coefficients α ₁ , β ₁ , β _{2 ,} and μ ₁ are derived. The inflow amount pattern estimating unit 23 registers the estimated regression coefficients α ₁ , β ₁ , β _{2 ,} and μ ₁ in the parameter storage unit 232.

The predicted temperature acquisition unit 214 obtains a predicted value of the temperature (hereinafter referred to as a predicted temperature). The predicted temperature acquisition unit 214 can receive an input of a predicted temperature from the user, for example, and can access the computer of the weather bureau or the civil meteorological company to obtain the predicted temperature. Further, the predicted temperature acquisition unit 214 can also predict the temperature based on the weather performance information. In this case, for example, a statistical pattern used in the general prediction of the temperature is stored in the pattern storage unit 31 in advance, and the predicted temperature acquisition unit 214 performs regression analysis based on the statistical pattern and the weather performance information. The parameters are also calculated, and the estimated parameters and meteorological performance information are applied to the statistical model, and the predicted temperature can be calculated.

The predicted precipitation amount acquisition unit 215 obtains the predicted precipitation amount. In the present embodiment, the predicted precipitation amount acquisition unit 215 is configured to transmit a predicted precipitation amount request to the precipitation amount prediction system 10, and to receive a request for precipitation from the precipitation in response to the predicted precipitation amount acquisition request. The amount of predicted precipitation that the quantity prediction system 10 responds to, thereby obtaining predicted precipitation. The predicted precipitation amount acquisition unit 215 transmits a predicted precipitation amount acquisition request having a specific percentile setting value (for example, 10% or 90%, etc.) to the precipitation amount prediction system. Further, the predicted precipitation amount acquisition unit 215 can receive an input of the predicted precipitation amount from the user, for example, and can access the computer of the weather bureau or the civil meteorological company, and obtain the predicted precipitation amount. Further, the predicted precipitation amount acquisition unit 215 is the same as the predicted temperature acquisition unit 214, and may be configured to predict the amount of precipitation based on the weather performance information.

The predicted inflow amount acquisition request receiving unit 216 is a command for receiving a predicted value of the inflow amount (hereinafter referred to as a predicted inflow amount) from another computer connected to the communication network 40 (hereinafter referred to as Forecast inflows to obtain requirements).

The inflow amount prediction unit 217 uses the predicted air temperature obtained by the predicted air temperature acquisition unit 214, the predicted precipitation amount obtained by the predicted precipitation amount acquisition unit 215, the parameter estimated by the snowfall temperature estimation unit 211, the weather performance information, and the inflow amount. Mode A5, and the predicted inflow is calculated. The details of the calculation process of the predicted inflow amount will be described later.

The predicted inflow amount transmitting unit 218 transmits the predicted inflow amount calculated by the inflow amount predicting unit 217 to the source of the communication requesting the predicted inflow amount.

The inflow amount distribution acquisition request receiving unit 219 is a command for receiving the probability distribution of the inflow amount from the other computer connected to the communication network 40 (hereinafter referred to as an inflow amount distribution acquisition request).

The inflow probability pattern estimation unit 219 performs regression analysis on the inflow probability pattern A6, and estimates the parameters α ₄ and μ ₃ . In addition, the inflow amount probability pattern estimating unit 219 may be configured to register the estimated parameters α ₄ and μ ₃ in the parameter storage unit 232 in advance, but in the present embodiment, it is determined in response to the inflow amount. The estimation of the parameters of the inflow probability mode A6 is performed by the distribution acquisition request.

The inflow amount distribution generation unit 221 generates a probability distribution of the inflow amount (hereinafter referred to as an inflow amount distribution). In the present embodiment, the inflow amount distribution is a distribution of the conditional probability Pr(F _m ∣F _m-1 ) with the occurrence of the inflow amount in the previous period as a condition, and occurs with a conditional probability. The inflow amount is set as the average value of the inflow amount in the month unit. In addition, the details of the process of generating the inflow amount distribution will be described later.

The inflow amount distribution transmitting unit 222 transmits the inflow amount distribution generated by the inflow amount distribution generating unit 221 .

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

[Throwing treatment of snowfall temperature δ]

First, the estimation process of the snowfall temperature δ by the snowfall temperature estimating unit 211 will be described. FIG. 17 is a diagram showing the flow of the estimation process of the snowfall temperature δ by the snowfall temperature estimating unit 211.

First, the snowfall temperature estimating unit 211 obtains weather performance information having a snowfall amount larger than 0 from the weather performance information database 233 (S2501). As a result, as shown in FIG. 18, in the meteorological performance information memorized in the weather performance information database 233, only those who have a greater snowfall than 0 will be drawn.

Next, the snowfall temperature estimating unit 211 is based on the extracted meteorological performance information, and the mode is

Snowfall = a × temperature + b × precipitation

Perform a regression analysis and calculate the regression coefficients a and b (S2502).

When the temperature is δ or less, the snowfall temperature estimation unit 211 calculates "a temperature of the meteorological performance information + b × precipitation of meteorological performance information", and when the temperature is higher than δ, "0" is used as the estimated amount of snowfall, and δ which minimizes the square value of the difference between the amount of snowfall of the estimated snowfall and the amount of snowfall of the weather performance information is calculated (S2503).

The snowfall temperature estimating unit 211 can be, for example, δ which increases the temperature of a specific range in each specific step, and as shown in FIG. 19, for each meteorological performance information, if the temperature is higher than δ Then, the temperature is multiplied by the value of the above regression coefficient a and the value of the above regression coefficient b is multiplied by the precipitation amount, and is calculated as the amount of snowfall, and the snowfall and weather performance are calculated. The difference between the snowfall of the information is squared and calculated as the square of the error, and the square of the error is determined as the minimum δ. Further, in the above-described processing for determining the above δ so that the square of the error is minimized, a general statistical technique can be utilized.

The snowfall temperature estimating unit 211 registers the δ determined as described above in the parameter storage unit 32 (S2504).

As mentioned above, the temperature of the snow melting is determined to be δ.

[Predictive processing of inflows]

Next, the prediction of the inflow amount will be described. Fig. 20 is a view showing the flow of the inflow amount prediction processing in the inflow amount prediction system 20 of the present embodiment. Further, in the processing of FIG. 20, it is assumed that the date t is the date to be predicted.

The predicted temperature acquisition unit 214 obtains the predicted temperature T _t (S2521). As described above, the predicted temperature obtaining unit 214 can obtain T _t by, for example, receiving input from a user or by accessing a computer of a weather bureau or a civil meteorological company and obtaining data. . The predicted precipitation amount acquisition unit 215 obtains the predicted precipitation amount P _t (S2522). As described above, the predicted precipitation amount acquisition unit 215 sets the predicted precipitation amount acquisition request with the specific percentile set value for the precipitation amount prediction system 10, and receives the response from the precipitation amount prediction system 10 The predicted precipitation is used to obtain the predicted precipitation.

The inflow amount prediction unit 217 reads δ from the parameter storage unit 132 (S2523), and applies δ, T _{t ,} and P _t in the precipitation amount pattern A1, and calculates P1 _t (S2524). That is, if the predicted temperature T _t is larger than δ, it is P1 _t = P _t , and if the T _t is δ or less, it is P1 _t =0.

The inflow amount prediction unit 217 reads α ₁ to α3, β ₁ , β ₂ , μ ₁ , and μ ₂ from the parameter storage unit 232 (S2525). The inflow amount prediction unit 217 reads out the weather performance information corresponding to the previous day t-1 from the weather performance database 233, and sets the snow amount of the read weather performance information to D _t-1 (S2526). ), and the inflow amount of the read weather performance information is set to F _t-1 (S2527).

The inflow amount prediction unit 217 substitutes α ₂ , T _t , μ ₂ , D _t-1 , α ₃ , and P1 _t in the snow melting amount mode A4 to calculate the predicted value M _t of the snow melting amount (S2528). In the inflow amount mode A5, α ₁ , μ ₁ , F _t-1 , β ₁ , P1 _t , β ₂ , and M _{t are} substituted, and the predicted value ΔF _t of the inflow increase amount is calculated (S2529). The inflow amount prediction unit 217 calculates the predicted inflow amount F _t by adding ΔF _{t to} F _t-1 (S2530).

As described above, according to the inflow amount prediction system 20 of the present embodiment, it is possible to predict the amount of inflow of the precipitation amount and the amount of snow melt based on the predicted temperature and the predicted precipitation amount and the meteorological performance information. Even in the case where there is no precipitation, the amount of inflow is increased by the snow melting. Therefore, by considering the amount of snowmelt and predicting the amount of inflow, the accuracy of the prediction can be improved.

Further, according to the inflow amount prediction system 20 of the present embodiment, the amount of snow melting can be calculated from the amount of precipitation and the temperature. Precipitation and temperature predictions, as a means of weather forecasting, are available in a variety of ways and can be easily obtained. Therefore, even when the prediction of the amount of snowmelt is difficult, it is possible to consider the amount of snowmelt inflow by predicting the amount of snowmelt based on the amount of precipitation that can be easily obtained or the predicted value of the temperature. Forecasting is easy to implement.

Further, in the above-described inflow amount mode A5, since the first term is set as the difference between the equalized inflow amount and the inflow amount, the equalized inflow amount is also taken into consideration, and therefore, the inflow amount is simply compared with Explain the case of variables and enable more accurate inflow forecasting.

In addition, in the inflow amount prediction system of the present embodiment, the amount of water flowing into the water storage facility such as a reservoir from the river is predicted, but the system for predicting the amount of water flowing in the river is predicted. Medium can also be easily applied. In this case, it is assumed that the predicted value and the actual value of the temperature or precipitation in the upper stream region of the river are acquired and recorded.

Further, in the present embodiment, the equalizing inflow amount μ ₁ and the temperature μ _{2 at} which the snow melting is started are stored in the parameter storage unit 32 in advance, but the present invention is not limited thereto. It is assumed that the value of the compliance is high based on the past weather performance information.

Further, when the water storage facility is present in a region where snowfall or snow is small, the inflow amount mode A5 may be set so as not to consider the amount of snow melt M. In this case, the inflow amount mode A5 is expressed by the following formula.

Δ F _t =α ₁ (μ ₁ - F _{t -1} )+β ₁ P 1 _t ‧‧‧(A5')

Further, in each regression mode of the present embodiment, when there is a case where there is a general order correlation in the error term, the snowfall temperature estimating unit 211 may be estimated by the Prais-Winstein transform or the Cochrane-Orcutt method. Out of parameters. In this case, for example, when the parameter of the inflow amount mode A5 is estimated by the Cochrane-Orcutt method, the residual is taken as ε _t and is set as the explanatory variable X _t* =X _t -ρX _t-1 (that is, set to F _t* =F _t -ρF _t-1 , P1 _t* =P1 _t -ρP1 _t-1 , M _t* =M _t -ρM _{t-1 ,} etc.), and the inflow can be Mode A5 is set to be as follows:

Δ F _{t *} =α ₁ (μ ₁ - F _{t -1*} )+β ₁ P _{1 t *} +β2 M _{t *} +σε _t ‧‧‧(A5")

If the above formula (A5") is deformed, it becomes the following formula:

Δ( F _t -ρ F _{t -1} )=α ₁ {μ ₁ -( F _{t -1} -ρ F _{t -2} ))}+β ₁ ( P 1 _t -ρ P 1 _{t -1} )+β2( M _t -ρ M _{t -1} )+σε _t

As described above, it is possible to extract the inflow amount pattern A5 which is deformed back to the n-stage before the residual ε _t is not present in the sequence correlation, and the same can be applied to other modes. The parameter is derived in such a way that there is no sequence correlation in the error term. As described above, the sequence of the residual εt is generally correlated with the mode in which the n-phase is delayed, and is stored in the pattern storage unit 131 in advance, and the inflow amount is predicted using the mode delayed by the n-phase. Even when there is a sequence correlation in the residual term, it becomes a measure that can perform appropriate parameters, and an appropriate parameter can be used to predict the inflow with good accuracy.

[Processing of the inflow amount distribution]

Next, the creation processing of the inflow amount distribution will be described. The inflow distribution is a distribution of the conditional probability of taking the inflow of the previous month as a condition. In the present embodiment, the inflow amount F _m is set so as to be the maximum value (hereinafter referred to as the minimum inflow amount and expressed as F _min ) from the specific minimum value (hereinafter referred to as the maximum inflow amount). Between the ranges up to F _max ), a discrete value of one unit amount (for example, 1 cubic meter or 10 cubic meters, 1000 cubic meters, etc., hereinafter referred to as unit inflow amount) is added at a time, and inflow distribution, and system setting from the minimum _min to the inflow until the maximum amount F since each unit of the inflow amount of F _max until the inflow of additional inflow amount corresponding to the probability for the description of the table.

If the inflow probability mode A6 is applied in the conditional probability Pr(F _m ∣F _m-1 ), the system can be deformed into the following general formula:

Here, in the residual u _m , since there is no sequence correlation, the A6′ can be further transformed into the following general formula:

That is, the conditional probability Pr(F _m ∣F _m-1 ) is the drift term α ₄ (μ ₃ -F _m ) which varies according to the inflow of the previous period (m-1 month). _-1 *) I _m and the residual term u _m I _m .

In addition, the probability of attaching the condition in the present embodiment is a condition that the inflow amount of the previous month is used as a condition, but the inflow amount of one month ago and two months ago may be used as a condition. The probability, that is, it can also be set to Pr (F _m ∣F _m-1 , F _m-2 ).

Fig. 21 is a view for explaining a flow of a process of creating an inflow amount distribution.

The inflow amount probability pattern estimating unit 219 sets the following month (January to December) as m, and performs the following processing.

The inflow probability pattern estimating unit 219 calculates the inflow amount corresponding to the date of the m month of all the years from the meteorological performance database 233, and calculates I _m (S2541), and will be m with all the years. - The date corresponding to the January date is averaged, and I _m-1 (S2542) is calculated.

The inflow probability pattern estimating unit 219 averages the inflow amount of the date corresponding to the date of the m month of the year for each of the years to which the date of the meteorological performance information stored in the meteorological performance database 233 belongs. And F _m (S2543) is calculated and normalized by dividing F _m by I _m to calculate F _m * (S2544). Further, the inflow probability pattern estimating unit 219 calculates the F _m-1 from the weather performance database 233 by averaging the inflow amount on the date corresponding to the date of the _m-1 month of the year (S2545). And normalized by dividing F _m-1 by I _m-1 to calculate F _m-1 * (S2546).

The inflow probability pattern estimating unit 219 performs the above-described processing for each year, and then performs regression analysis on the inflow probability pattern A6", and estimates α ₄ and μ ₃ (S2547), and the inflow probability mode estimating unit In 219, the drift term α ₄ (μ ₃ -F _m-1 *)I _m (S2548) is calculated.

Further, the inflow probability pattern estimating unit 219 calculates the residual u _{m of the} inflow probability pattern A6" by the following equation (S2549).

The inflow probability pattern estimating unit 219 calculates the residual term u _m I _m (S2550).

The inflow probability pattern estimating unit 219 repeatedly performs the above processing for each month, calculates the drift term and the residual term for each month, and then generates the drift term α ₄ (μ ₃ -F _m-1 *). The frequency distribution of I _m (S2551), and the frequency distribution (S2552) of the residual term u _m I _m is generated, and these are added to generate a frequency distribution of the inflow amount F _m (S2553). Inflow probability models pushing meter unit 219, based inflows The F _m of the frequency distribution calculated occurrence conditional inflow amount of a probability _{Pr (F m |F m-1} ) (S2554), and made to calculate the The probability of occurrence is described as the inflow distribution. Further, the inflow distribution may be described as a calculation formula of the probability distribution.

[Usage Project System 30]

In the plan system 30, the amount of generated electric power in the period of the long-term operation plan (hereinafter referred to as "long-term operation period, in the present embodiment, one year" is maximized). The simulation of the most appropriate water level in each of the periods (hereinafter, referred to as the medium-term operation plan, which is one month in the present embodiment), which is the target of the medium-term operation plan, is followed by In the period of the most suitable water level in the medium-term operation period, and in the period of the short-term use (hereinafter referred to as the short-term operation period, in the present embodiment, the sales amount is set to 6 days), the sales amount is the largest. In order to extract the water level of each unit period (in the present embodiment, it is set to 1 day) in the short-term operation period. In addition, in the following description, the water level at the water storage facility is set to be in a specific unit amount (for example, 1 meter or 5 meters, etc., hereinafter referred to as unit water level). Discrete value. The water level during the long-term operation period will be the maximum water level during the normal mid-term use period, which is obtained by the Stochastic Dynamic Programming (SDP). The water level during each unit in the short-term operation period is the maximum value of the water level at the water storage facility or the amount of water used in hydropower generation (hereinafter referred to as the water withdrawal amount, and is denoted as Q). Under the limit of the minimum value and the like, the predicted inflow amount predicted by the inflow amount prediction system 20 is used to change the water level of each day by one unit water level, and the simulation is performed, and the calculated sales amount becomes The largest water level.

Figure 22 is a diagram showing the hardware configuration of the application planning system 30. As shown in the same figure, the application planning system 30 includes a CPU 301, a memory 302, a memory device 303, a communication interface 304, an input device 305, and an output device 306. The memory device 303 stores various data or programs, such as a hard disk or a flash memory, a CD-ROM drive, and the like. The CPU 301 executes the reading of the program stored in the memory device 303 into the memory 302, thereby realizing various functions. The communication interface 304 is an interface for connecting to the communication network 40, and is, for example, a connector for connecting to an Ethernet (registered trademark) or for connecting to a public telephone line network. A data machine, a communicator for wireless communication, and the like. The input device 305 receives input of data, such as a keyboard or a mouse, a touch panel, a microphone, and the like. The output device 306 outputs the data, for example, a display or a printer, a speaker, and the like.

FIG. 23 is a diagram showing the configuration of the software using the planning system 30. As shown in the same figure, the plan system 30 includes data input unit 311, water storage amount setting value input unit 312, inflow amount distribution acquisition unit 313, medium-term plan unit 314, and predicted inflow amount acquisition unit. 315. The short-term plan unit 316, the data storage unit 331, the mode storage unit 332, the power price database 333, and the optimal water level database 334. Further, the data input unit 311, the water storage amount setting value input unit 312, the inflow amount distribution acquisition unit 313, the medium-term plan unit 314, the predicted inflow amount acquisition unit 315, and the short-term plan unit 316 are used to make the operation plan system The CPU 201 provided by the 30 is realized by reading the program stored in the memory device 203 to the memory 202 and performing the same. Further, the data storage unit 331, the power price database 333, and the optimal water level database 334 are realized as the memory area provided by the memory system 202 or the memory device 203 provided in the planning system 30. In addition, the data storage unit 331, the mode storage unit 332, the power price database 333, and the optimal water level database 334 may be managed by a database server different from the application planning system 30, and The application planning system 30 is made accessible to the database server.

The data storage unit 331 stores information (hereinafter referred to as data information) including setting values of various data such as a water storage facility, a river, and a power generation facility. Fig. 24 is a diagram showing a configuration example of data information memorized in the data storage sections 331. As shown in the same figure, the data information includes data names, units, and set values.

The data input unit 311 receives the input of the data information and registers the received data information in the data storage unit 331. For example, the data input unit 311 can be configured to receive input of each item of data information from an input device 205 such as a keyboard or a mouse, or to access the host computer of the power company, for example. And get the data information.

Further, the data input unit 331 is in advance: as the data relating to the water storage facility, the highest value of the received water level (the highest applied water level, hereinafter, denoted as H _max , the unit is m) and the lowest value of the water level ( The lowest water level, hereinafter, denoted as H _min , the unit is the input of m), and the data information including the highest used water level received and the data information including the lowest used water level received are registered. In the data storage unit 331, as the data relating to the river, the reception flow rate (hereinafter, referred to as S0, the unit is m ³ /s) is received, and data information including the received maintenance flow rate is created. Registered in the data storage unit 331; as the data relating to the power generation device, the maximum amount of water available for power generation (maximum water withdrawal amount, hereinafter referred to as Q _max , unit is m ³ /s), The lowest amount of water in power generation (minimum water withdrawal, hereinafter, denoted as Q _min , unit is m ³ /s), and the height of water released after power generation (water level, below, denoted as H _out , unit system) For m) And a loss of gap (hereinafter denoted as H _los, unit system to m) is input, and creating comprises receiving the the various data of the maximum quantity of water of information comprising various data minimum quantity of water of information, comprising discharge all the water level The data information and the data information including the loss are recorded in the data storage unit 331.

The electricity price database 333 is a memory price for each date. FIG. 25 is a diagram showing a configuration example of the power price database 333. As shown in the same figure, in the power price database 333, the power price (unit is yuan/kWh) is stored in association with the date. In the present embodiment, the electric power price is set to be changeable on each date, and the electric power price of each day is input by the user in advance and registered in the electric power price database 333.

The water storage amount setting value input unit 312 receives an input of a predetermined value (hereinafter referred to as initial water storage amount) of the water storage amount at the start time point (0 point) of the first day of the short-term operation period. Further, for example, the water storage amount setting value input unit 312 may be configured to acquire the water storage amount of the current water storage facility as the initial water storage amount. Further, the water storage amount setting value input unit 312 may be configured to predict the initial water storage amount from the past water level.

The optimum water level database 334 is to be remembered for the water level after the most appropriate intermediate period of use corresponding to the month and the water level at the beginning of the month and the inflow during the month. Fig. 26 is a view showing a configuration example of the optimum water level database 334. As shown in the same figure, in the optimal water level database 334, the water level inflow target target water level table 351 is stored, which is in each of the months m in the long-term use period, and The optimum water level (the optimum water level H _m+1 after one month) is set correspondingly to the water level H _m and the inflow amount F _m . As will be described later, by reading the optimum water level one month after the month and the water level at the start of the month and the predicted inflow amount from the optimum water level database 334, it is possible to obtain the month. The target water level at the final point in time.

The inflow amount distribution acquisition unit 313 acquires the inflow amount distribution. In the present embodiment, the inflow amount distribution acquisition unit 313 is configured to transmit a flow inflow amount distribution request to the inflow amount prediction system 20, and to receive an inflow due to the inflow amount distribution acquisition request. The inflow amount distribution sent by the quantity prediction system 20 is obtained by which the inflow amount distribution is obtained. Further, the inflow amount distribution acquisition unit 313 may be configured to receive an input of the inflow amount distribution from the user.

The Medium Term Planning Department 314 is the most suitable water level plan for each medium-term use period. The medium-term planning unit 314 uses the inflow amount distribution and the statistical pattern stored in the pattern storage unit 332, which will be described later, to create an optimum water level plan by the probability dynamic programming method, and registers it in the optimal water level database 334. in. The details of the processing for creating the optimum water level plan by the medium-term planning unit 314 will be described later.

The predicted inflow amount acquisition unit 313 acquires the predicted value (predicted inflow amount) of the inflow amount R for each day in the short-term operation period predicted by the inflow amount prediction system 20 by accessing the inflow amount prediction system 20. Further, the predicted inflow amount acquisition unit 313 may be configured to receive an input of a predicted inflow amount, for example, from an input device 205 such as a keyboard or a mouse.

The short-term plan department 316 (optimum water storage amount determination unit) calculates the most appropriate water level at each of the short-term use periods by simulation (hereinafter, referred to as short-term use water level, and is denoted as H). The calculated short-term water level of each day is output to the output device. The short-term plan unit 316 performs simulation using a statistical mode stored in the mode storage unit 332 which will be described later.

In the mode memory unit 332, the following modes B1 to B11 are stored.

The mode B1 is used to calculate the water level H (water level calculation mode) based on the water storage amount V, and is expressed by the following equation. In addition, a is a constant inherent to the water storage facility.

The mode B2 is an upper limit for calculating the water storage amount based on the highest operating water level H _max (hereinafter referred to as an upper limit water storage amount, and is expressed as V _max ), and is expressed by the following formula.

V _max = a × H _max ² ÷24÷3600‧‧‧(B2)

The mode B3 is used to calculate the lower limit of the water storage amount (hereinafter referred to as the lower limit water storage amount and is expressed as V _min ) based on the lowest operating water level H _min , and is expressed by the following formula.

V _min = a × H _min ² ÷24÷3600‧‧‧(B3)

In the mode B4, the difference in the amount of water stored from the start time point of the unit period to the end time point is calculated based on the water storage amount from 0:00 to 24:00 (i.e., 0 o'clock of the next day). Hereinafter, it is referred to as a difference in water storage amount and is expressed as ΔV), and the water storage amount at the zero point of a certain date t is set to V _t and expressed by the following formula.

Δ V _t = V _{t +1} - V _t ‧‧‧(B4)

The mode B5 is used to calculate the amount of water (R0) obtained by subtracting the maintenance flow rate S0 and the water storage amount difference ΔV from the inflow amount R, and is expressed by the following equation.

R 0= R - S 0-Δ V ‧‧‧(B5)

The mode B6 is used to determine the water intake amount Q (the water intake amount calculation mode) based on the minimum water withdrawal amount Q _min , the maximum water withdrawal amount Q _max , and the minimum operation output Q0 _min and R0, and is expressed by the following equation.

That is, when R0 is the minimum water withdrawal amount Q _min or more and is the maximum water withdrawal amount Q _max or less, R0 is the water withdrawal amount Q; when R0 is smaller than the minimum water withdrawal amount Q _min , the minimum is the quantity of water-based Q _min Q be the quantity of water; R0 line when more than the maximum quantity of water for the case where Q _max, Q _max the maximum quantity of water-based abstraction becomes Q. However, when R0 is smaller than the minimum operation output Q0 _min , the water intake amount Q becomes 0.

The mode B7 is used to calculate the amount of water discharged in the water storage facility in the day (hereinafter referred to as the ordinary discharge amount, and is denoted as S) based on R0 and the water intake amount Q, and is expressed by the following formula.

S = R 0- Q ‧‧‧(B7)

The mode B8 is used to calculate the effective drop hn according to the applied water level, the discharge level H _out and the loss drop H _los at the end of the 1 day, and sets the operating water level at the 0 point of the date t to Ht, which is used to convert the water level to a high altitude, is set to b, and is expressed by the following equation.

The mode B9 is used to calculate the generated electric power Pn generated in the first day based on the water intake amount Q and the effective drop hn, and the coefficient for the change efficiency of the power generation is c, and the gravitational acceleration is g. It is expressed by the following formula.

Pn = Q × hn × c × g ‧‧‧(B9)

The mode B10 is a method for calculating the amount of generated electric power En (power amount calculation mode) generated in one day from the generated electric power Pn, and is expressed by the following equation.

En = Pn ×24‧‧‧(B10)

The mode B11 is used to calculate the amount of electricity sold based on the amount of generated electric power En on a certain date and the price of electricity corresponding to the date, and is expressed by the following formula.

Sales amount = En × electricity price ... (B11)

[interim plan]

The medium-term planning department 314 calculates the most appropriate water level by the probability dynamic programming method using the inflow distribution and the above-mentioned statistical mode. Specifically, the medium-term planning unit 314 seeks to take out a combination of water levels satisfying the following formula.

In formula B12, GE(H _m , H _m+1 , F _m ) represents: when the water level at the beginning of the month m is H _m , at the end of the month m (month m+1 At the start time point, the water level at _m+1 is H _m+1 , and the amount of generated electric power En when the inflow amount of the month m is F _m . Here, the amount of generated electric power at the month m is calculated by multiplying the number of generated electric power En of one day calculated in the above-described mode B10 by the number of days of the month m.

27 to 29 are diagrams for explaining a method for extracting the most appropriate water level by the probability dynamic programming method. In the examples of FIGS. 27 to 29, in order to simplify the description, it is assumed that the long-term operation period is 4 months, and the water level of the water storage facility is only a value of any of h1 to h3. In addition, the medium-term planning unit 314 performs the following processing for the case where the water level at the end time point m5 is h1 to h3.

As shown in FIG. 27, in general, the medium-term planning unit 314 sets the total power generation amount E (1, H ₁ , F ₁ ) of the previous period (phase) to 0, and for the start time of the fourth month. Each of the water levels at point m4, and the amount of generated power is calculated for each of the inflows in the fourth month, and at the calculated amount of generated power, the first month is added. The expected value of each water level calculated by the probability of occurrence of the inflow amount (here, since the total amount of generated electric power is 0, the expected value is also 0), and this is calculated as the expected value of the fourth month. come out. The medium-term plan unit 314 selects the expected value of the fourth month (shown by a thick line in the figure) (step 1).

The medium-term planning unit 314 calculates the amount of generated electric power at the third month for each combination of the water level at the time point m4 and the water level at the time point m3, and adds the fourth calculated value to the calculated value. The expected value of each water level calculated by the probability of occurrence of the inflow of the month is calculated as the expected value after the third month. The medium-term plan unit 314 selects the expected value after the third month (the thick line is shown in the figure) (step 2).

In the same manner, the medium-term planning unit 314 calculates the amount of generated electric power at the second month for each combination of the water level at the time point m3 and the water level at the time point m2, and adds it to the calculated value. The expected value of each water level calculated by the probability of the inflow of the third month is calculated as the expected value of the amount of generated electric power after the second month, and the expected value after the second month becomes The largest combination (shown in bold in the figure) is chosen (step 3).

In the first month, the medium-term planning unit 314 calculates the amount of generated electric power in the first month for each combination of the water level at the time point m2 and the water level at the time point m1, and adds it to the calculation. The value is multiplied by the expected value of each water level calculated by the probability of the inflow of the second month, and this is calculated as the expected value after the first month, and then the expected value becomes the largest after the first month. The combination (shown by bold lines in the figure) is chosen (step 4).

As above, the most appropriate water level is determined.

In the above description, it is calculated by combining the water level (the target water level) at the end of the long-term operation period and the water level of each month in which the expected value of the generated electric power is maximized. The processing procedure is described. However, depending on the water level at the end of the long-term operation period, it is also conceivable that the performance of the next long-term operation period will deteriorate. In order to solve this problem, the processing is repeated until the expected value of the total amount of generated electric power in the long-term operation period is maximized.

The medium-term plan unit 314 creates the total value of the amount of generated electric power from the first month to the fourth month and the water level H _{1 of the} first month, at the end of each of the above-described processes (stages). The inflow amount F ₁ of the first month is added to the water level inflow amount amount table 341 which is managed and managed. In Fig. 28, a configuration example of the water level inflow amount electric power meter 341 is shown. In the second and subsequent stages, the medium-term planning unit 314 reads the total generated electric power amount of the previous stage from the water level inflow amount electric power meter 341, and performs the above-described processing (refer to FIG. 29). The medium-term planning unit 314 repeats the above-described processing until the difference between the total electric power amount at the previous stage and the total electric power amount at the current stage is equal to or lower than a specific value.

Figure 30 is a diagram showing the flow of the process of planning the optimum water level during the long-term use period.

The medium-term plan unit 314 sets 1 in the ST at the stage of the representative process (S3601), and performs each month from the last month (in the present embodiment, the twelfth month) to the first month. The processing of the water level caused by the probability dynamic programming method (S3602). Fig. 31 is a view showing the configuration of the water level inflow amount generation electric power amount table 352 which is used in the water level plan processing by the probability dynamic programming method. Do not scale power generation 352 in FIG. 31 inflows of water, the water level of the power generation system 12 H ₁₂ months and 12 months inflow F ₁₂ is attached correspondence is registered in the first 12 months of The amount of electricity.

32 and 33 are diagrams showing the flow of the water level plan processing by the probability dynamic programming method.

Figure 32 is a diagram showing the flow of the water level plan processing for the final month (December). The mid-term planning unit 314 will process the month until the month of the first one (because it is processed for each month from the 12th month to the first month), therefore, before the 12th month In the month of the first month, the cumulative power generation E(1, H ₁ , F ₁ ) is 0 (S3621), and the minimum water level H _{min is} set at the water level H ₁₂ of the final month. (S3622). The medium-term planning unit 314 sets the cumulative power generation amount E (12, H ₁₂ , F ₁₂ ) up to _December (0362), and sets the minimum inflow amount F at the inflow amount F _{12 in December} . _Min (S3624). The medium-term planning unit 314 sets the minimum water level H _min (S3625) at the water level H _{1 in January} , and applies the data stored in the data storage units 331 to the above-mentioned mode to calculate the power generation. Power En (S3626). The medium-term planning unit 314 multiplies 31 (the number of days in December) at En, and calculates an expected value GE (S3627) of the amount of generated electric power for one month. The medium-term planning unit 314 obtains the probability of occurrence of the inflow amount F _{1 in January} (S3628), and multiplies the cumulative generated electric power amount E (12, H ₁₂ , F ₁₂ ) for each F ₁ . F ₁₂ is the total value of the conditional probability of F ₁ as a condition, and is added to GE, and E1 (12, H ₁₂ , F ₁₂ ) is calculated (S3629).

The medium-term planning unit 314, when E1 (12, H ₁₂ , F ₁₂ ) is equal to or greater than E (12, H ₁₂ , F ₁₂ ) (S3630: YES), is in E (12, H ₁₂ , F ₁₂ ) E1 (12, H ₁₂ , F ₁₂ ) is set (S3631), and the month 12, H ₁₂ , H ₁ , and F _{12 are} registered in the water level inflow target water level table 351 and the water level inflow power generation electric power meter 352 ( S3632).

The medium-term planning unit 314 adds the unit water level at H ₁ (S3633), and if the added H ₁ does not become larger than the highest water level H _max (S3634: NO), the process proceeds from step S3626. deal with. When the added water level H ₁ is greater than the highest water level H _max (S3634: YES), the medium-term planning unit 314 adds the unit inflow amount to the inflow amount F ₁₂ (S3635). When the inflow amount F ₁₂ does not exceed the maximum inflow amount F _max (S3636: NO), the processing from step S3625 is repeatedly performed. When the inflow amount F ₁₂ exceeds the maximum inflow amount F _max (S3636: YES), the medium-term planning unit 314 adds the unit water level to the water level H ₁₂ (S3637), and if it is added, the H ₁₂ does not become When the maximum water level H _{max is} larger (S3638: NO), the processing from step S3623 is repeated. The medium-term planning unit 314 terminates the processing if the water level H ₁₂ is greater than the maximum water level H _max (S3638).

Fig. 33 is a diagram showing the flow of the water level plan processing for the month m (m = 11, 10, ..., 1) other than the final month (December). First, the medium-term planning unit 314 calculates the cumulative power generation amount E(m+1, H _m+1 , F _m+1 ) from the previous one to the month m+1, from the water level inflow amount. The generated electric power meter 352 is calculated and calculated (S3641). In the above-described process of the water level plan for the final month of FIG. 32, the cumulative generated electric power amount E(1, H ₁ , F ₁ ) is calculated as 0 (FIG. 32, S3621). In the meantime, when the processing of FIG. 32 is first performed, the water level inflow amount generation electric power amount table 352 is not registered with a value.

Next, the mid-section plan 314, FIG. 32 of the processing system and the same, the water level in the month m H _m, the lowest set water level H _min (S3642), and m until the month cumulative generating electric power E (m, H _m , F _m ) is set to 0 (S3643), and at the inflow amount F _m of the month m, the minimum inflow amount F _{min is set} (S3644). The medium-term planning unit 314 sets the lowest water level H _min (S3645) at the water level H _m+1 of the previous month m+1, and applies the data stored in the data storage units 331 to the above mode. In the middle, the amount of generated electric power En is calculated (S3646). The medium-term planning unit 314 multiplies the number of days of the month by the month of En, and calculates the expected value GE of the power generation amount for one month (S3647). The medium-term plan department 314 obtains the probability of occurrence of the inflow amount F _m+1 of the previous month m+1 (S3648), and for each F _m+1 , the accumulated power generation amount E (m+1, H) _m+1 , F _m+1 ) is multiplied by the conditional probability of F _m with F _m+1 as the condition, and added to GE, and E1(m,H is calculated. _m , F _m ) (S3649).

When the E1 (m, H _m , F _m ) is equal to or greater than E (m, H _m , F _m ) (S3650: YES), the medium-term planning unit 314 is in E(m, H _m , F _m ). E1(m, H _m , F _m ) is set (S3651), and the months m, H _m , H _m+1 , and F _{m are} registered in the water level inflow target water level table 351 and the water level inflow power generation electric power meter 352. At (S3652).

The medium-term plan department 314 adds the unit water level (S3653) at H _m+1 , and if the added H _m+1 does not become larger than the highest water level H _max (S3654: NO), the process proceeds repeatedly. Processing from step S3646. When the water level H _m+1 after the addition is larger than the highest water level H _max (S3654: YES), the medium-term planning unit 314 adds the unit inflow amount to the inflow amount F _m (S3655). When inflow amount does not exceed the maximum F _m F _max inflows case when (S3656: NO), the system from step S3645 repeated processing of starting. When the inflow amount F _m exceeds the maximum inflow amount F _max (S3656: YES), the medium-term planning unit 314 adds the unit water level to the water level H _m (S3657), and if the addition, the H _m does not become When the maximum water level H _{max is} larger (S3658: NO), the processing from step S3643 is repeated. The medium-term plan unit 314 ends the process if the water level H _m becomes larger than the maximum water level H _max (S3658).

After the above-described water level plan processing by the probability dynamic programming method is performed for each month from the twelfth month to the first month, the medium-term plan unit 314 performs the expected total amount shown in FIG. Calculation processing of the amount of generated electric power (S3603). The medium-term planning unit 314 sets H _min (S3661) at H ₁ and acquires the probability of occurrence of each F ₁ per unit inflow amount (S3662). The medium-term plan unit 314 calculates the cumulative generated electric power amount E(1, H ₁ , F ₁ ) up to the first month from the water level inflow amount electric power generation amount table 352 (S3663), and for each F1, The E value is multiplied by the value of the conditional probability of F ₁ with F ₁₂ as the condition, and the expected total power generation electric power EE(1, H ₁ ) is calculated (S3664). In the medium-term plan, the unit water level is added to H ₁ (S3665), and if the added H ₁ does not exceed H _max (S3666: NO), the process from step S3663 is repeated.

Referring back to FIG. 30, when the ST system is 1 (S3604: 1), the medium-term planning unit 314 sets the expected total generated electric power amount EE(1, H ₁ ) as E02 (1, H ₁ ). (S3605) When the ST system is 2 (S3604: 2), the total generated power amount EE (1, H ₁ ) is set as E01 (1, H ₁ ) (S3606). The medium-term planning unit 314 increments the ST (S3607), and repeats the process of starting the water level planning process by the probability planning method from the final month to the first month of step S3602.

On the other hand, when the ST system is 3 or more (S3604: 3 or more), the medium-term planning unit 314 expects the total generated electric power amount EE(1, H ₁ ) to be E00(1, H ₁ ). Set (S3609), and subtract the first value after E01(1,H ₁ ) from E02(1,H ₁ ), and subtract E00(1 from E01(1,H ₁ ). The second value after H ₁ ), and whether the quotient exceeds a certain threshold value, thereby determining the convergence. When the above quotient exceeds the threshold (S3610: YES), the medium-term planning unit 314 sets E01(1, H ₁ ) (S3611) at E02 (1, H ₁ ), and at E01 (1). E1 (1, H ₁ ) is set at H ₁ ) (S3612), and proceeds to step S3607 described above. If the quotient does not exceed the threshold (S3610: NO), the medium-term planning unit 314 ends the processing.

As described above, the water level inflow target water level table 351 for each month stored in the optimum water level database 334 is updated.

According to the operation plan system 30 of the present embodiment, it is possible to use the predicted value of the inflow amount in the future, and to extract the expected value of the amount of generated electric power during the long-term operation period, and to increase the start time point and end of the mid-term operation period. The combination of water levels at the time. Therefore, the user of the water storage facility will use the optimum water level database 334 as a reference at the beginning of the medium-term operation period, and the water level corresponding to the water level at that time point, As the target water level, and the use of water storage facilities, it is possible to increase the amount of electricity generated during the year. Therefore, even when the experience of the user of the water storage facility is shallow, it can be easily adjusted to the most appropriate water level.

Further, according to the operation planning system 30 of the present embodiment, the water level can be calculated based on the objective data as compared with the intuition and experience of the user who has passed through the water storage facility. Planning. Further, in the operation planning system 30 of the present embodiment, even when the predicted value of the future inflow amount is obtained as the probability distribution by using the probability dynamic programming method, it can be easily derived. An optimum water level plan to increase the expected value of the amount of generated electricity at the water storage facility.

[short-term plan]

As described above, the medium-term planning unit 314 seeks to extract the combination of the optimum water level at the start time point and the end time point of the mid-term operation period (month), whereas the short-term plan unit 334 is used for short-term use. During the period (6th), the amount of electricity sold increases, and the water level in the short-term use period is planned.

The short-term planning unit 316 sets the initial water storage amount received by the water storage amount setting value input unit 312 as V _{1 of the} first day of the short-term operation period, and applies V ₁ to the mode B1 to calculate the water level H ₁ of the first day. . The short-term plan department 316 reads out the optimum water level after the month in which the short-term operation period belongs and H ₁ corresponds to the end of the short-term use period (also the start time of the 7th day). ) the water level H ₂ . The short-term planning unit 316 calculates the target value of the water storage amount (hereinafter referred to as the final destination water storage amount) V ₇ at the end time of the short-term operation period using the water level H ₂ and the mode B2 or B3. The short-term planning unit 316 sets the water storage amount V _t at 0 o'clock from V _min to V _max to a specific stage (for example, 0.1 or 0.5, for each date t (t=2 to 6). 1)) to change and simulate, and let En be the biggest way to determine V _t . The short-term planning unit 316, in this simulation, is set to use the dynamic programming method. By using the dynamic programming method, the most appropriate combination of V _t can be quickly calculated.

[Screen example]

Fig. 35 is a diagram showing an example of a screen 60 used in the simulation of the water level application. The screen 60 is provided with an input field 611 during operation, an initial water storage amount input field 612, a final destination water storage amount input field 613, and various data display fields 621 to 627.

The short-term planning unit 316 reads the set values (H _min and H _max ) corresponding to the minimum operating water level and the maximum operating water level from the data storage units 331 and displays the read H _min and H _max in The display column 621 and the display column 622 are displayed. The short-term planning unit 316 reads out the set value (S0) corresponding to the maintenance flow rate from the data storage unit 331, and displays the read S0 in the display field 623. The short-term planning unit 316 reads the set values (Q _min and Q _max ) corresponding to the minimum water intake amount and the maximum water withdrawal amount from the data storage units 331 , and displays the read Q _min and Q _max , respectively. At display column 624 and display column 625. The short-term planning unit 316 reads out the set value (H _out ) corresponding to the water discharge position from the data storage unit 331 and displays the read H _out in the display field 626. The short-term planning unit 316 reads out the set value (H _los ) corresponding to the loss difference from the data storage unit 331 and displays the read H _los in the display field 627.

Further, the short-term planning unit 316 calculates V _max and V _min using the above-described mode B2 and mode B3, and displays the calculated V _max and V _min in the display column 631 and the display column 632, respectively.

The predicted inflow amount acquisition unit 313 acquires the predicted value of the inflow amount R for each day in the operation period predicted by the inflow amount prediction system 20 by accessing the inflow amount prediction system 20, and acquires the obtained value. R is displayed at display column 633.

In the input fields 611, 612, and 613, the operation period, the initial water storage amount, and the final destination water storage amount are input. If the key 641 is depressed, the water storage amount setting value input unit 312 receives the input fields 611 and 612. And the input of the 613 input period, the initial water storage amount, and the final destination water storage amount, the short-term planning unit 316 simulates the water storage amount V for each day in the operation period. The short-term planning unit 316 performs simulation of the water storage amount V as follows, for example. Further, in the example of Fig. 35, the coefficient inherent to the water storage facility is a = 30000, and the constant for converting the water level to the altitude is b = 500.

The short-term planning unit 316 sets the initial water storage amount to be input into the input field 612 as the water storage amount V _{1 on the} first day, and sets the final destination water storage amount input to the input field 613 as the seventh day. The water storage amount V ₇ is simulated, and the water storage amount V _t from the first day to the sixth day is simulated, and the water storage amount V is displayed on the display column 651.

The short-term plan unit 316 applies the water storage amount V for each date to the mode B1, calculates the operation water level H, and displays the calculated operation water level H in the display field 652. The short-term planning unit 316 applies V _t and V _t+1 to the mode B4 for the date t of the first to sixth days, and calculates the water storage difference ΔV _t for each date _t , and calculates the The water storage difference ΔV _{t is} displayed at the display column 653.

The short-term planning unit 316 applies the predicted value R of the inflow amount, the maintenance flow rate S0, and the water storage amount difference ΔV to the mode B5, and calculates R0, and then determines the water intake amount Q corresponding to R0 by the mode B6. The determined water withdrawal amount Q is displayed on the display column 654. The short-term planning unit 316 applies R0 and the water intake amount Q to the mode B7, calculates the normal discharge flow rate S, and displays the calculated S in the display column 655.

The short-term planning unit 316 applies the calculated operating water level H _t+1 , the water discharge level H _{out ,} and the loss drop H _los to the mode B8 for each day t of the first to sixth days, and calculates the effective The difference hn is displayed, and the calculated hn is displayed in the display field 656.

The short-term planning unit 316 applies the water intake amount Q and the effective drop hn to the mode B9, calculates the generated electric power Pn, and displays it in the display column 657, and applies the calculated Pn to the mode B10, and The generated power amount En is calculated, and the calculated En is displayed on the display column 660.

The short-term planning unit 316 reads out the electric power price corresponding to the date of each day in the operation period from the electric power price database 333, and displays it in the display column 659, and reads the above En and reads out The electric power price is applied in the mode B11, and the electric power is calculated, and the calculated electric power amount is displayed on the display column 658.

Short-term plan unit 316, water storage system so that Vt (t = 2 ~ 6) from the maximum V _max V _min until up to make changes, and the above-described process repeatedly, with this, the simulated and decided ambassador The total amount of electricity sold during the period of use becomes the combination of the largest water storage amount V. Further, the short-term planning unit 316 determines the combination of the water storage amount V by the dynamic programming method, whereby the optimum water storage amount can be efficiently determined. The short-term plan unit 316 displays the display columns 651 to 660 in a manner corresponding to the combination of the determined water storage amounts V.

Figure 36 is a graph showing the results of a simulation of past performance levels in a water storage facility.

Fig. 36 (a) is a graph showing the water withdrawal amount and the change in the water level at the time of July 2003 when the comparatively high water period is known as the wet season. At the water storage facility, the water level was determined based on the experience of the user. The actual value of the electricity sales in the July 2003 period was approximately 258 (million). On the other hand, the amount of electric power sold corresponding to the combination of the water storage amount V as a result of the above simulation is about 269 (million). That is to say, an increase of about 4% of the sales amount of electricity can be obtained.

Fig. 36(b) is a graph showing the water withdrawal amount and the change in the water level at the time of April 2007 when the comparatively known dry season is known. The actual sales value of the electricity sales in the April 2007 period is 107 (million). In contrast, the electricity sales price corresponding to the combination of the water storage amount V as a result of the above simulation is 114 (hundreds). Ten thousand yuan), and can get an increase of about 6%.

In this way, it can be confirmed that by using the above simulation to determine the combination of the amount of electricity to be sold, which is the largest amount of water, and the use of the water level H, the application is based on the experience of the user. It is able to increase the amount of electricity sold.

As described above, according to the water level operation support system of the present embodiment, it is possible to indicate the operating water level at which the sales amount of electricity is maximized in accordance with the initial water storage amount and the final destination water storage amount. Therefore, the user of the water storage facility will use the tips provided by the water level operation support system as a reference, and use the water level of the water storage facility to thereby perform the hydraulic power more efficiently and effectively. Power generation.

Further, in the present embodiment, the precipitation amount prediction system 10, the inflow amount prediction system 20, and the operation planning system 30 are respectively different computers, but they may be realized by one computer. . Further, any of the precipitation amount prediction system 10, the inflow amount prediction system 20, and the operation planning system 30 may be realized by a plurality of computers.

Further, in the precipitation amount prediction system 10 of the present embodiment, the weather is set to be any one of "clear", "yin", "snow", "rain", and "rain", but for example, The system may also include "霙" or "雹". Further, in the present embodiment, the time phrase is defined as any of "time", "one hour", and "after", but for example, it may include "gradually" or "Begin" and so on.

Further, in the precipitation amount prediction system 10 of the present embodiment, the weather profile information is a weather profile including a time zone during the day and a time zone at night, but is not limited thereto. It can be set to include only one full-day weather profile, or it can be set to include weather profiles of more than three time zones, such as morning, afternoon, and night.

Further, in the precipitation amount prediction system 10 of the present embodiment, the weather term and the time word are arranged side by side in the order of the degree of precipitation in the same manner as the row of the form table 161, but for example, It is also possible to provide a priority memory unit that stores the intensity (priority) of the precipitation amount in association with the weather term and the time term. In this case, the weather form registration unit 112 does not use the form table 161, but generates a list of weather blocks in step S1514 of FIG. 6, and for the weather blocks included in the list. Each of them obtains the intensity corresponding to the weather term included in the block and the intensity corresponding to the time phrase included in the block from the priority memory unit, and sets the total value of the acquired strengths as The priority of the weather block. The weather form registration unit 112 specifies the weather block included in the weather term for each of the weather terms, and leaves the highest priority among the specified weather blocks, and the other is The person is removed from the list. The weather form registration unit 112 can generate weather form information based on the weather blocks remaining in the list, and register in the weather form database 132.

Further, in the precipitation amount prediction system 10 of the present embodiment, the weather form database 132 is set to have a flag value corresponding to the date and weather block, but is not limited thereto. For example, it is also possible to register a list of weather blocks in association with the date. Further, in the weather form database 132, the amount of precipitation included in the weather profile information may be memorized in association with the date.

Further, in the inflow amount prediction system 20 of the present embodiment, the predicted precipitation amount is determined using the percentile value, but the present invention is not limited thereto, and the average value or the center may be used. Value, etc.

Further, in the present embodiment, the inflow amount and the water level are set to be discrete values per unit inflow amount and unit water level, but may be continuous values of real numbers. In this case, the inflow amount distribution created in the inflow amount prediction system 20 is described as a function for extracting the probability of the inflow amount, and the optimum water level database 334 of the planning system 30 is used. It is described as a function of taking the month, the water level, and the inflow amount as an argument and taking out the optimum water level one month later.

In addition, in the operation plan system 30 of the present embodiment, the long-term operation period is one year, the medium-term operation period is one month, and the short-term operation period is six days, but the system is not It is limited to this and can be set to any length. For example, the long-term use period can be half a year or three months, and the medium-term use period can also be three months or ten days (10 days) or one week. In addition, the short-term operation period can be set to 2 weeks or 1 week or 4 days.

Further, in the operation planning system 30 of the present embodiment, the combination of the most appropriate water level at the start time point and the end time point of the medium-term operation period is obtained by the probability dynamic programming method. The water level is not limited thereto, and various water level estimation methods capable of maximizing the expected value of the amount of generated electric power during the long-term operation period can be used to calculate the water level.

Further, in the operation planning system 30 of the present embodiment, the combination of the water storage amount V is determined so that the amount of electric power is maximized. However, it is also possible to determine that the generated electric power amount En is maximized. General combination. In this case, there is no change in price, and it is possible to generate more efficient power generation.

Further, the plan system 30 can be modeled by setting the upper limit of the normal discharge flow rate S. In this case, the short-term planning unit 316 receives the input of the upper limit value from the user, and determines that the amount of power sold is the largest from the combination of the water storage amount V in which the normal discharge flow rate S does not exceed the upper limit value. By. 37 is a theoretical value of the discharge flow S when the actual value of the discharge flow S at the July 2003 period and the sales amount when the upper limit is not set is the maximum, and When the upper limit value is set to 15 m ³ /s, the theoretical value of the discharge flow rate S when the power sale amount becomes the maximum is compared. As shown generally in Fig. 37, for example, on July 24, the discharge flow rate S is increased to approximately 25 (m ³ /s). In this case, the amount of electricity sold is also about 269 (million), and it can be increased by about 4%. Therefore, in the actual use of the water storage facility, for example, considering the influence on the downstream when there is a fisherman or a camper, etc., the upper limit is set at the discharge rate. However, even if the upper limit value is set, it is possible to perform more efficient power generation by using the water level of the water storage facility based on the result of the above simulation.

In addition, in the inflow amount prediction system 20 of the present embodiment, the amount of water flowing into a water storage facility such as a reservoir from a river is predicted, but the amount of water flowing in the river is predicted. In the system, it can also be easily applied. In this case, it is assumed that the predicted value and the actual value of the temperature or precipitation in the upper stream region of the river are acquired and recorded.

Further, in the operation planning system 30 of the present embodiment, the inflow amount R is obtained from the inflow amount prediction system 20, but the present invention is not limited thereto. For example, the user may be Receive input of the inflow R. Further, it is also possible to store the actual value of the past inflow amount in the database in advance, and read the past actual value as R, and perform simulation.

Further, in the operation planning system 30 of the present embodiment, the electric power price is stored in the electric power price database 333 in advance, but for example, it may be set as the unloading power exchange in Japan. The electricity price (JEPX price) is automatically obtained. In this case, the power price database 333 may be omitted, and the JEPX price may be obtained in the simulation performed by each of the short-term planning units 316.

The above description of the embodiments has been made, but the above-described embodiments are intended to facilitate the understanding of the present invention and are not intended to limit the present invention. The present invention can be variously modified and improved without departing from the spirit and scope of the invention, and the equivalents are also included in the present invention.

10. . . Precipitation prediction system

20. . . Inflow forecasting system

30. . . Application planning system

40. . . Communication network

101. . . CPU

102. . . Memory

103. . . Memory device

104. . . Communication interface

105. . . Input device

106. . . Output device

111. . . Weather Profile Acquisition Department

112. . . Weather form registration department

113. . . Predicting the amount of precipitation obtained by the Ministry of Communications

114. . . Weather forecast acquisition department

115. . . Precipitation Forecasting Department

116. . . Predicting precipitation information

131. . . Weather profile database

132. . . Weather form database

201. . . CPU

202. . . Memory

203. . . Memory device

204. . . Communication interface

205. . . Input device

206. . . Output device

211. . . Snowfall temperature estimation department

212. . . Snowmelt mode estimation department

213. . . Inflow mode

214. . . Forecast temperature acquisition department

215. . . Predicted precipitation acquisition department

216. . . Predicted inflows are required to be requested by the Ministry of Communications

217. . . Inflow forecasting department

218. . . Predicted inflow amount

219. . . Inflow distribution obtained by the request receiving department

220. . . Inflow rate probability pattern estimation unit

221. . . Inflow distribution generation unit

222. . . Inflow distribution department

231. . . Mode memory

232. . . Parameter memory

233. . . Meteorological performance database

211. . . Snowfall temperature estimation department

212. . . Snowmelt mode estimation department

213. . . Inflow mode

214. . . Forecast temperature acquisition department

215. . . Predicted precipitation acquisition department

216. . . Predicted inflows are required to be requested by the Ministry of Communications

217. . . Inflow forecasting department

218. . . Predicted inflow amount

219. . . Inflow distribution obtained by the request receiving department

220. . . Inflow rate probability pattern estimation unit

221. . . Inflow distribution generation unit

222. . . Inflow distribution department

231. . . Mode memory

232‧‧‧Parameter Memory

233‧‧‧Meteorological Performance Database

311‧‧‧ Data Input Department

312‧‧‧Water storage setting value input unit

313‧‧‧Inflow distribution acquisition department

314‧‧ Mid-term Planning Department

315‧‧‧ Forecast Inflows Acquisition Department

316‧‧‧Short-term Planning Department

331‧‧‧ Data Memory Department

332‧‧‧Mode Memory

333‧‧‧Power Price Database

334‧‧‧Optimal water level database

351‧‧‧Water level inflow target target water level table

352‧‧‧Water level inflow power generation electricity meter

Fig. 1 is a view showing the overall configuration of the operation support system of the present embodiment.

FIG. 2 is a diagram showing the hardware configuration of the precipitation amount prediction system 10.

[Fig. 3] A diagram showing the software composition of the precipitation amount prediction system 10.

FIG. 4 is a diagram showing a configuration example of weather profile information memorized in the weather profile database 131.

FIG. 5 is a diagram showing a configuration example of weather form information stored in the weather form database 132.

[Fig. 6] A diagram showing a flow of registration processing of weather form information.

FIG. 7 is a diagram showing the configuration of the form table 161 used in the registration process of the weather form information.

[Fig. 8] A diagram for explaining a specific example of registration processing of weather form information based on weather profile information.

[Fig. 9] A diagram showing the flow of the prediction processing of the precipitation amount.

[Fig. 10] A diagram showing a flow of processing of weather form information according to a weather forecast.

[Fig. 11] A graph showing changes in the amount of inflow when there is no precipitation.

[Fig. 12] A graph showing changes in the amount of inflow when there is precipitation.

[Fig. 13] A graph showing changes in the amount of inflow when there is snow melting.

Fig. 14 is a view showing a hardware configuration of the inflow amount prediction system 20 of the present embodiment.

Fig. 15 is a view showing a soft body configuration of the inflow amount prediction system 20 of the present embodiment.

FIG. 16 is a diagram showing a configuration example of weather performance information memorized in the weather performance database 233.

[Fig. 17] A flow chart showing the flow of the estimation process of the snowfall temperature δ by the snowfall temperature estimating unit 211.

[Fig. 18] A diagram for explaining the extraction of weather performance information having a snowfall amount larger than zero.

[Fig. 19] A graph showing the results of calculations for the amount of snowfall and the square of the error for a certain snowfall temperature δ.

[Fig. 20] A diagram showing the flow of the inflow amount prediction processing.

FIG. 21 is a diagram for explaining a flow of a process of creating an inflow amount distribution. FIG.

FIG. 22 is a diagram showing the hardware configuration of the application planning system 30.

FIG. 23 is a diagram showing the configuration of the software using the planning system 30.

[Fig. 24] A diagram showing a configuration example of data information memorized in the data storage sections 331.

FIG. 25 is a diagram showing a configuration example of the power price database 333.

FIG. 26 is a diagram showing a configuration example of the optimum water level database 334.

[Fig. 27] A diagram for explaining a method of extracting an optimum water level by a probability dynamic programming method.

[Fig. 28] A diagram showing the configuration of a table used in the method of extracting the most appropriate water level by the probability dynamic programming method.

[Fig. 29] A diagram for explaining a method of extracting an optimum water level by a probability dynamic programming method.

[Fig. 30] A diagram showing the flow of processing for calculating the most appropriate water level.

[Fig. 31] A diagram showing a configuration example of the water level inflow amount electric power amount table 352 of the work table used in the plan processing of the optimum water level.

[Fig. 32] A diagram showing the flow of the processing of the water level caused by the probability dynamic programming method for the final month.

[Fig. 33] A diagram showing the flow of the processing of the water level caused by the probability dynamic programming method for the months other than the final month.

[Fig. 34] A diagram showing a flow of a calculation process for expecting a total amount of generated electric power.

[Fig. 35] A diagram showing an example of a screen 60 used in the simulation of the water level application.

[Fig. 36] A graph showing the results of simulations performed on a past performance water level in a certain water storage facility.

[Fig. 37] The actual value of the discharge flow rate, the theoretical value of the discharge amount in the result of the simulation in which the upper limit value exists, and the discharge amount in the result of the simulation in which the upper limit value is not present are performed. A chart of comparison.

Claims (5)

A water storage facility operation support system is a system for supporting the operation of a water storage facility for hydroelectric power generation, and is characterized in that: an optimum water level table is provided for a unit period during a specific operation period. The water level at the start time point is additionally associated with the inflow amount of the aforementioned water storage facility in the aforementioned unit period, and the optimum water level which is the most appropriate water level at the end time point of the aforementioned unit period is memorized; and the water level is obtained. The water level of the water storage facility at the start time of the operation period is obtained, and the predicted inflow amount acquisition unit obtains a predicted value of the inflow amount in each of the unit periods; and the water level plan unit For each of the unit periods described above, the water level corresponding to the water level at the start time point and the predicted value of the inflow amount in the unit period is read from the optimum water level table, and read out. The aforementioned optimum water level is used as the water level at the start time point of the unit period below the unit period; and the water level plan memory unit The water level obtaining unit acquires the current water level of the water level of the water storage facility at the current time point, the predicted inflow amount, at the end of each of the unit periods. The acquisition unit acquires a predicted value of the inflow amount after a unit period of one of the unit periods, and the water level plan unit sets the current water level as the unit period of the next unit period. The water level at the start time point, and for the aforementioned unit period after the next unit period, the above-mentioned optimum water level is read from the above-mentioned optimum water level table, and the aforementioned optimum water level is read by the aforementioned The water level plan memory is updated. The water storage facility operation support system according to the first aspect of the invention, wherein the predicted inflow amount acquisition unit acquires a predicted value of the inflow amount in a plurality of the unit periods and occurrence of the predicted value. The probability that the optimum water level table is based on the water level that the water storage facility is likely to be, the predicted value of the inflow amount in each of the unit periods, and the probability of occurrence of the predicted value of the inflow amount, Dynamic planning method is used to create. The water storage facility operation support system described in claim 2, wherein the database includes a database for memorizing the actual value of the inflow amount for each of the unit periods; and mode memory In the department, there is a statistical mode in which the difference between the inflow amount at the second unit period immediately before a certain first unit period and the equilibrium inflow amount which is a specific constant is used as The variable is described as the target variable from the second unit period until the first unit period, and the inflow amount pattern estimating unit is based on the inflow amount stored in the database. The actual value of the performance, as well as the aforementioned statistical model, and the regression score The regression coefficient and the residual inflow amount: and the residual term frequency distribution generation unit for the unit time period described above are compared with the first unit immediately after the unit period. The first actual value of the inflow amount corresponding to the second unit period before the period and the second actual value of the inflow amount corresponding to the first period are read from the database, and the first And the difference between the second actual value, the value obtained by subtracting the second actual value from the estimated second in-kind value and multiplying the regression coefficient, and calculating the residual term, and generating the aforementioned The frequency distribution of the residual term; and the drift term frequency distribution generating unit, for each of the unit periods, the first actual value of the inflow amount corresponding to the second unit period, and the first period corresponding to the first period The second actual value of the inflow amount is read from the database, and the value obtained by subtracting the second actual performance value from the estimated inflow amount is multiplied by the regression coefficient. Calculating a drift term and generating a frequency distribution of the drift item; and a probability distribution generating unit adding the frequency distribution of the residual item and the frequency distribution of the drift item to generate a frequency distribution of the inflow amount And generating a probability distribution of the inflow amount from the frequency distribution of the inflow amount, wherein the predicted inflow amount acquiring unit is each of the values of the specific range in which the inflow amount is likely to be the inflow amount Predicting the value and calculating the aforementioned inflow amount based on the aforementioned probability distribution The probability of occurrence of the measurement. A water storage facility operation support method is a method for supporting the operation of a water storage facility for hydroelectric power generation, and is characterized in that it is provided with a water level at a start time point of a unit period during which a specific operation period is constituted. And the computer which is the optimum water level table for storing the optimum water level of the most suitable water level at the end time of the above-mentioned unit period in addition to the inflow amount of the aforementioned water storage facility in the aforementioned unit period, Processing: obtaining a water level of the water storage facility at a start time point of the operation period, and obtaining a predicted value of an inflow amount in each of the unit periods before the start of the operation period; and for each unit period The water level at the start time point and the predicted water level corresponding to the inflow amount in the unit period are read from the optimum water level table, and the read optimal water level is used as the The water level at the start time of the unit period below the unit period; the aforementioned optimum water for each unit period mentioned above Memorizing in the memory, at the end of each of the aforementioned unit periods, obtaining the current water level of the water level of the water storage facility at the current time point; obtaining the aforementioned inflow after the unit period of the next unit period The predicted value of the quantity; The current water level is used as the water level at the start time point of the next unit period, and the optimum water level is read from the optimal water level table for each of the unit periods after the next unit period; The memory is updated by the optimum water level read as described above. A water storage facility operation support program is a program for supporting the operation of a water storage facility for hydroelectric power generation, and is characterized in that it has a water level with a start time point of a unit period during which a specific operation period is constituted. And the computer which is the optimum water level table for storing the optimum water level of the most suitable water level at the end time of the above-mentioned unit period in addition to the inflow amount of the aforementioned water storage facility in the aforementioned unit period, the following is performed a step of obtaining a water level of the water storage facility at a start time point of the operation period, and before the start of the operation period, performing the step of: obtaining a predicted value of the inflow amount in each of the unit periods; In the above-described respective unit periods, the water level corresponding to the water level at the start time point and the predicted value of the inflow amount in the unit period are read from the optimum water level table, and the read-out is read. The aforementioned optimum water level as a step of the water level at the start time point of the unit period below one of the unit periods; During each unit of the optimum level, the memory in the memory in step, At the end of each of the foregoing unit periods, a step of obtaining the current water level of the water level of the aforementioned water storage facility at the current time point is obtained; and obtaining the prediction of the aforementioned inflow amount after the unit period of the next one of the unit periods a step of value; the current water level is used as the water level at the start time point of the next unit period, and the aforementioned optimum water level is obtained from the aforementioned optimum water level table for each of the aforementioned unit periods after the next unit period The step of reading; the step of updating the memory by the optimum water level read as described above.