JP6046530B2 - Maximum power generation prediction system - Google Patents

Description

  The present invention relates to a maximum power generation amount prediction system that predicts the maximum power generation amount of a power generation facility such as a gas turbine plant that generates power by burning fuel such as natural gas.

  In recent years, the tight supply and demand of electric power has become a serious problem due to the review of electric power measures. Under such circumstances, not only the power supplied by general electric utilities (electric power companies) but also the power supplied by other electric power companies (electric power companies other than power companies, such as specified scale electric power companies) This is expected to ease the tight supply and demand situation.

  When power supply and demand becomes tight, general electric utilities and other electric utilities have the maximum power generation capacity of power generation facilities in order to avoid a situation where the amount of power supplied is insufficient even if the power generation facilities are operated at the maximum output. It is necessary to predict the amount.

  At this time, if a general electric utility mistakenly estimates the maximum power generation capacity of the power generation facility larger than the actual amount, the amount of power supplied will be short of demand, resulting in a serious situation such as a large-scale power outage. This can be a problem.

  Also, at this time, if another electric power company mistakenly estimates the maximum power generation capacity of the power generation facility larger than the actual power generation, it is impossible to supply the amount of power tendered to the power exchange or the amount of power as contracted. This may be a problem because it may not be possible to supply to customers. Further, in this case, other electric power companies may have a problem because it may be necessary to bear the responsibility for compensation for the shortage of the amount of power supplied relative to the amount of power promised to be supplied. In addition, thermal power generation with high followability to demand fluctuation is useful as a regulated power source, and it is particularly required to predict the maximum power generation amount.

  However, in a power generation facility that performs thermal power generation by burning fuel using oxygen in the atmosphere, such as a gas turbine plant, the amount of power generation fluctuates with changes in atmospheric conditions (Patent Document 1 and non-patent documents). Reference 1). Therefore, it is difficult to accurately predict the maximum power generation amount of such a power generation facility.

Japanese Patent Laid-Open No. 5-321610

“Prospects for New Energy Gas Turbine Technology”, Institute of Energy Engineering, March 2007, pp. 6-9

  Even if it is difficult to accurately predict the maximum power generation amount of the power generation facility, the above situation can be avoided if the maximum power generation amount of the power generation facility is estimated as small as possible. However, if the maximum power generation amount is estimated in such a manner, there is a problem because the tightness of power supply and demand becomes serious and the amount of wasted power that is not supplied to consumers increases.

  Accordingly, the present invention provides the maximum power generation for predicting the maximum power generation amount of the power generation facility so that the power amount supplied by the power generation facility can be increased as much as possible while avoiding the shortage of the power amount supplied by the power generation facility. An object is to provide a quantity prediction system.

  In order to achieve the above object, the present invention provides a maximum power generation amount prediction system that predicts a predicted maximum power generation amount that is the maximum power generation amount in a prediction target period of a power generation facility that generates power by burning fuel using oxygen in the atmosphere. A predicted atmospheric state data generation unit that generates predicted atmospheric state data indicating a predicted atmospheric state that is a prediction result of the atmospheric state in the prediction target period, and high reliability of prediction of the predicted atmospheric state A reliability data generation unit that generates reliability data indicating reliability, and a maximum power generation amount prediction unit that calculates the predicted maximum power generation amount based on the predicted atmospheric state data and the reliability data, The maximum power generation amount prediction unit calculates the predicted maximum power generation amount that increases as the reliability increases, if the predicted atmospheric state is the same. To provide.

  According to this maximum power generation amount prediction system, the predicted maximum power generation amount increases as the reliability, which is the reliability of the prediction of the predicted atmospheric state, is higher. Therefore, the predicted maximum power generation amount is the actual maximum power generation amount. It can prevent becoming larger than the amount, and can prevent the predicted maximum power generation amount from becoming excessively smaller than the actual maximum power generation amount.

Furthermore, the maximum power generation amount prediction system of the above features, the maximum power generation amount prediction unit calculates a provisional predicted maximum power generation amount corresponding to the predicted atmospheric conditions, the maximum the provisional prediction small decrease higher the reliability It is preferable to calculate the predicted maximum power generation amount by reducing the power generation amount.

  According to this maximum power generation amount prediction system, the predicted maximum power generation amount is calculated with the provisional predicted maximum power generation amount corresponding to the predicted atmospheric state as the upper limit. Therefore, it is possible to effectively prevent the predicted maximum power generation amount from becoming larger than the actual maximum power generation amount.

  Furthermore, in the maximum power generation amount prediction system having the above characteristics, the maximum power generation amount prediction unit is a measured atmospheric state that is the state of the atmosphere measured in the past and a maximum power generation amount of the power generation facility that has been measured in the past. It is preferable to calculate the provisional predicted maximum power generation amount corresponding to the predicted atmospheric state using the correspondence relationship with the actually measured maximum power generation amount.

  According to this maximum power generation amount prediction system, the provisional predicted maximum power generation amount is calculated based on past measurement results and reliability. Therefore, the predicted maximum power generation amount can be effectively brought close to the actual maximum power generation amount.

  Furthermore, in the maximum power generation amount prediction system having the above characteristics, the maximum power generation amount prediction unit includes a measured atmospheric state that is the state of the air actually measured in the past, and a change in the maximum power generation amount of the power generation facility that has been measured in the past. Using the correspondence relationship with the measured fluctuation amount that is the magnitude of the tentative amount, the provisional predicted fluctuation amount corresponding to the predicted atmospheric state is calculated, and the temporary predicted fluctuation amount is adjusted so as to decrease as the reliability increases. It is preferable to calculate the predicted maximum power generation amount by calculating the adjusted fluctuation amount and subtracting the adjusted fluctuation amount from the provisional predicted maximum power generation amount.

  According to the maximum power generation amount prediction system, the predicted maximum power generation amount is calculated by subtracting the adjusted fluctuation amount calculated based on the past actual measurement result and the reliability from the provisional predicted maximum power generation amount. Therefore, it is possible to effectively bring the predicted maximum power generation amount close to the actual maximum power generation amount and to effectively prevent the predicted maximum power generation amount from becoming excessively smaller than the actual maximum power generation amount.

  Furthermore, in the maximum power generation amount prediction system having the above characteristics, the system further includes a warning notification unit that notifies a user of a warning, and the maximum power generation amount prediction unit determines the predicted maximum power generation amount until the prediction target period. When the predicted maximum power generation amount changes beyond a predetermined threshold, it is preferable that the warning notification unit notifies the user of a warning.

  According to this maximum power generation amount prediction system, a warning is notified to the user when the predicted maximum power generation amount changes greatly exceeding the threshold value. As a result, the user is given an opportunity to review the predicted maximum power generation amount calculated previously, so that it is possible to prevent the amount of power supplied from becoming insufficient or the amount of wasted power from increasing. It becomes.

  Furthermore, in the maximum power generation amount prediction system having the above characteristics, it is preferable that the predicted atmospheric state data indicates a prediction result of at least one of atmospheric pressure and temperature.

  This maximum power generation prediction system is based on atmospheric pressure and temperature, which are factors that are strongly related to fuel combustion in power generation facilities such as gas turbine plants that generate power by burning fuel using oxygen in the atmosphere. Thus, the predicted maximum power generation amount can be calculated.

  According to the maximum power generation amount prediction system having the above characteristics, the higher the reliability, which is the reliability of the prediction of the predicted atmospheric state, the larger the predicted maximum power generation amount calculated. While being able to prevent becoming larger than the maximum power generation amount, it is possible to prevent the predicted maximum power generation amount from becoming excessively smaller than the actual maximum power generation amount. Therefore, according to the maximum power generation amount prediction system having the above characteristics, it is possible to increase the amount of power supplied by the power generation facility as much as possible while avoiding a situation where the amount of power supplied by the power generation facility is insufficient. It is possible to predict the maximum power generation amount.

The block diagram shown about an example of the structure of the maximum electric power generation amount prediction system which concerns on embodiment of this invention. The figure explaining an example of the calculation method of reliability. The figure explaining an example of the calculation method of prediction maximum electric power generation amount. The figure explaining an example of the alerting | reporting method of a warning.

  Hereinafter, a maximum power generation amount prediction system according to an embodiment of the present invention will be described with reference to the drawings. Hereinafter, the maximum power generation amount of the power generation facility in the prediction target period predicted by the maximum power generation amount prediction system is referred to as “predicted maximum power generation amount”. Moreover, below, although the case where the length of a prediction object period is 1 day is demonstrated as an example of embodiment of this invention, even if a prediction object period is a half day (12 hours), several hours, and several days, Well, not necessarily limited to one day.

<Configuration of maximum power generation prediction system>
First, an example of the configuration of the maximum power generation amount prediction system according to the embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram showing an example of the configuration of the maximum power generation amount prediction system according to the embodiment of the present invention.

  As shown in FIG. 1, the maximum power generation amount prediction system 1 according to the embodiment of the present invention includes a predicted atmospheric state data generation unit 10, a reliability data generation unit 11, a database 12, and a maximum power generation amount prediction unit 13. The output part 14 and the warning alerting | reporting part 15 are provided.

  The predicted atmospheric state data generation unit 10 generates predicted atmospheric state data based on GPV (Grid Point Value) data distributed from the outside (for example, a server of the Japan Meteorological Agency) via a network or the like. It comprises a communication device to be connected and a calculation device such as a CPU (Central Processing Unit) that calculates a predicted atmospheric state using the acquired GPV data.

  The predicted atmospheric state data indicates a predicted atmospheric state that is a prediction result of the atmospheric state in the prediction target period. The predicted atmospheric state may include any element that indicates the atmospheric state (for example, atmospheric pressure, temperature, humidity, etc.), but power is generated by burning fuel using atmospheric oxygen. It is preferable that an element having a strong relationship with the combustion of fuel in a power generation facility such as a gas turbine plant is included. Specifically, for example, it is preferable that the predicted atmospheric state includes a prediction result of at least one of atmospheric pressure and temperature.

  The GPV data indicates weather prediction results at each grid point of the earth divided in a grid, and is distributed periodically (for example, four times a day, once a day, etc.). The distributed GPV data is recorded in the database 12 or the like.

  The interval between lattice points in GPV data is as large as several kilometers or several tens of kilometers, and may exceed 100 kilometers in long-term prediction. Therefore, there may be a case where the position of the power generation facility and the grid point in the GPV data do not match. In addition, the time of the weather prediction result indicated by the GPV data may not match the time of the weather prediction result desired by the user. In such cases, the predicted atmospheric condition data generation unit 10 corrects or interpolates (or corrects and interpolates) the GPV data spatially and temporally (or spatially or temporally) as necessary. It is preferable to calculate the predicted atmospheric state.

  For example, the predicted atmospheric state data generation unit 10 may use the GPV data in the past fixed period (for example, the last one month or the same month in the previous year) and the actual measurement result of the atmospheric state measured at the power generation facility. A predicted atmospheric state is calculated by setting a correction method based on the correlation and correcting the newly acquired GPV data using the correction method. This correction method can be constructed by using various estimation methods such as a MOS (Model Output Statistics), a Kalman filter, and a neural network.

  Also, for example, the predicted atmospheric state data generation unit 10 may determine the grid point closest to the power generation facility in the GPV data in the past certain period (for example, the most recent month or the same month of the previous year). A primary (or several-order) relational expression that minimizes the difference between the data and the actual measurement result of the atmospheric state measured at the power generation facility may be set as the correction method.

  Further, for example, the predicted atmospheric state data generation unit 10 may calculate the predicted atmospheric state at the position of the power generation facility by interpolating data of grid points around the power generation facility in the GPV data.

  The reliability data generation unit 11 is based on ensemble data distributed from the outside (for example, a server of the Japan Meteorological Agency) via a network or the like, and indicates a reliability indicating the reliability of the prediction of the predicted atmospheric state. Data is generated, and includes a communication device connected to a network or the like, and an arithmetic device such as a CPU that calculates reliability using the acquired ensemble data.

  Ensemble data is data used for ensemble prediction performed when the above-described GPV data is obtained. Ensemble data is also periodically distributed and recorded in the database 12 and the like, similar to GPV data.

  Ensemble prediction is a weather forecast simulation that gives different variability to the initial value that is an actual measurement value, and obtains a plurality of forecast simulation results (members), and averages the members to obtain the forecast result. The prediction method to get. Note that the data obtained by averaging the members is the above-described GPV data, and the data indicating each of the plurality of members is the ensemble data.

  A method of calculating reliability using ensemble data will be described with reference to the drawings. FIG. 2 is a diagram illustrating an example of a reliability calculation method. FIG. 2 is a graph showing the standard deviation of ensemble data, where the horizontal axis is the prediction time and the vertical axis is the standard deviation (error).

  The average standard deviation indicated by the broken line in FIG. 2 is the average standard deviation for each member's predicted time indicated by the ensemble data in the past fixed period (for example, the most recent month or the same month of the previous year). It has become. On the other hand, the target standard deviation indicated by the solid line in FIG. 2 is the standard deviation for each predicted time of the member indicated by the ensemble data newly acquired by the reliability data generation unit 11. In addition, since each member (namely, prediction simulation result) is near, so that a standard deviation is small, reliability becomes high.

  In general, the member indicated by the ensemble data increases in variation every time the predicted time passes, so that the standard deviation increases as the predicted time elapses. In particular, the average standard deviation changes so as to increase approximately monotonically with the passage of the prediction time because the influence of weather fluctuations (for example, the passing of the rainy season front and the approach of a typhoon) is weakened by averaging. On the other hand, the target standard deviation is not necessarily a monotonically increasing change because the influence of weather fluctuations appears directly.

  Therefore, the reliability data generation unit 11 calculates the reliability based on the magnitude relationship between the average standard deviation and the target standard deviation. Specifically, for example, since the target standard deviation is larger than the average standard deviation at the predicted time T1 in FIG. 2, the reliability data generation unit 11 calculates the reliability low. Further, for example, since the target standard deviation is smaller than the average standard deviation at the predicted time T2 in FIG. 2, the reliability data generation unit 11 calculates the reliability high.

  At this time, the reliability data generation unit 11 has a quotient obtained by dividing the average standard deviation by the target standard deviation or a value corresponding to a difference obtained by subtracting the target standard deviation from the average standard deviation (specifically, for example, the difference is 0 or more). The reliability may be expressed by a numerical value, such as reliability = 2 if there is a difference, reliability = 0.5 if the difference is −50 or more and less than 0). In this case, the higher the reliability, the larger the reliability value. However, the reliability is a number of 0 or more.

  In addition, the reliability data generation unit 11 determines the rank corresponding to the quotient or the difference of the average standard deviation and the target standard deviation (specifically, for example, A: high reliability, B: somewhat high reliability, C: reliability) Slightly low, D: low reliability), the reliability may be expressed.

  The database 12 includes a recording device that records various data, and records a maximum power generation amount table, a fluctuation amount table, and predicted maximum power generation amount data. Details of the maximum power generation amount table, the fluctuation amount table, and the predicted maximum power generation amount data will be described later.

  The maximum power generation amount prediction unit 13 is composed of an arithmetic device such as a CPU, for example. The predicted atmospheric state data generated by the predicted atmospheric state data generation unit 10, the reliability data generated by the reliability data generation unit 11, and the database 12 The predicted maximum power generation amount is calculated based on the maximum power generation amount table and the fluctuation amount table to be recorded. Then, the maximum power generation amount prediction unit 13 generates predicted maximum power generation amount data indicating the predicted maximum power generation amount, and inputs the generated data to the output unit 14 and the database 12. Note that details of a specific calculation method of the predicted maximum power generation amount by the maximum power generation amount prediction unit 13 will be described later. Further, for example, the maximum power generation amount prediction unit 13 calculates the predicted maximum power generation amount every predetermined time (for example, every 30 minutes or every hour), and is for one day or a plurality of days (for example, seven days). Min) is calculated.

  The output unit 14 includes an output device (for example, a display) that outputs data in a format that can be perceived by the user, such as an image or sound, and outputs the predicted maximum power generation amount data generated by the maximum power generation amount prediction unit 13 to the user. To output the predicted maximum power generation amount calculated by the maximum power generation amount prediction unit 13 to the user.

  The warning notification unit 15 can be perceived by a calculation device such as a CPU that determines whether or not a warning is to be notified to the user based on the predicted maximum power generation amount data recorded in the database 12, and a user such as an image or sound. And an output device (for example, a display) for notifying a user of a warning in a simple format. The details of the warning notification method by the warning notification unit 15 will be described later.

  Each process of the maximum power generation amount prediction system 1 is performed by arithmetic processing using computer hardware resources (CPU, various storage devices, etc.) and software resources (OS: Operating System, various drivers, etc.). Further, such arithmetic processing is realized by software by executing a program whose execution is controlled by the CPU.

<Calculation method of predicted maximum power generation>
Next, an example of a method for calculating the predicted maximum power generation amount by the maximum power generation amount prediction unit 13 will be described with reference to the drawings. FIG. 3 is a diagram illustrating an example of a method for calculating the predicted maximum power generation amount. 3A is a diagram showing a maximum power generation amount table, and FIG. 3B is a diagram showing a fluctuation amount table.

  Each of the maximum power generation amount table and the fluctuation amount table shown in FIGS. 3A and 3B is a power generation facility in a past fixed period (for example, the most recent one month or the same month in the previous year). It is a table created using the actual measurement result of the measured atmospheric state and the actual measurement result of the maximum power generation amount actually measured at the power generation facility.

  FIGS. 3A and 3B illustrate the maximum power generation amount table and the variation amount table in the case where the actual measurement result of the atmospheric state is the actual measurement result of the atmospheric pressure and the air temperature, respectively, for concrete description. However, the measurement results of other elements may be adopted as the measurement results of the atmospheric state. However, as described above, the atmospheric pressure and the temperature are factors that are strongly related to the combustion of fuel in a power generation facility such as a gas turbine plant that generates power by burning fuel using oxygen in the atmosphere. It is preferable to create a maximum power generation amount table and a fluctuation amount table that employ at least one of the actual measurement results.

  The maximum power generation amount table shown in FIG. 3A shows a correspondence relationship between the actual measurement results of atmospheric pressure and temperature (measured atmospheric state) and the actual measurement results of maximum power generation (measured maximum power generation amount). In the maximum power generation amount table illustrated in FIG. 3A, the actually measured maximum power generation amount at a certain atmospheric pressure and a certain air temperature is an average value of the maximum power generation amount actually measured at the certain atmospheric pressure and the certain air temperature. In the maximum power generation amount table of FIG. 3A, the actually measured maximum power generation amount at a certain atmospheric pressure and a certain air temperature may be the maximum value of the maximum power generation amount actually measured at the certain atmospheric pressure and the certain air temperature.

  The fluctuation amount table shown in FIG. 3 (b) shows the correspondence between the actual measurement results of atmospheric pressure and temperature (measured atmospheric conditions) and the magnitude of fluctuation of the actually measured maximum power generation amount (actual fluctuation amount). It is. In the fluctuation amount table illustrated in FIG. 3B, the actually measured fluctuation amount at a certain atmospheric pressure and a certain air temperature is a standard deviation of the maximum power generation actually measured at the certain atmospheric pressure and the certain air temperature. In addition, in the fluctuation amount table of 3 (b), the measured fluctuation amount at a certain atmospheric pressure and a certain air temperature is the absolute value of the difference between the maximum value and the minimum value of the maximum power generation actually measured at the certain atmospheric pressure and the certain air temperature. It is good.

  The maximum power generation amount prediction unit 13 obtains the maximum power generation amount (provisional predicted maximum power generation amount) corresponding to the predicted atmospheric state, using the maximum power generation amount table shown in FIG. For example, when the predicted atmospheric state is atmospheric pressure: 996 hPa and temperature: 15.1 ° C., the provisional predicted maximum power generation amount obtained by the maximum power generation amount prediction unit 13 is 50 MWh.

  Further, the maximum power generation amount prediction unit 13 obtains a fluctuation amount (temporary prediction fluctuation amount) corresponding to the predicted atmospheric state using the fluctuation amount table shown in FIG. For example, when the predicted atmospheric state is atmospheric pressure: 996 hPa and temperature: 15.1 ° C., the provisional predicted fluctuation amount obtained by the maximum power generation amount prediction unit 13 is 2 MWh.

  Then, the maximum power generation amount prediction unit 13 calculates the predicted maximum power generation amount based on the following formula (1). When [Reliability] in the following formula (1) is expressed by rank as described above, a numerical value greater than 0 corresponding to the rank is substituted for [Reliability]. However, it is assumed that the numerical value assigned to [Reliability] in the following equation (1) increases as the rank increases in reliability.

  [Adjustment coefficient] in the following formula (1) may be a constant larger than 0, or may be a variable of 0 or more according to the reliability. Note that the value or calculation method of the “adjustment coefficient” is obtained empirically by, for example, comparing the prediction result of the maximum power generation amount prediction unit 13 with the actually measured maximum power generation amount.

[Predicted maximum power generation] = [Temporary predicted maximum power generation]-[Adjusted fluctuation amount]
[Adjusted fluctuation amount] = [Provisional prediction fluctuation amount] × [Adjustment coefficient] / [Reliability] (1)

  In the above formula (1), as [Reliability] increases (Reliability increases), [Adjusted fluctuation amount] decreases. Therefore, [Predicted maximum power generation amount] calculated by subtracting [Adjusted fluctuation amount] from [Temporary predicted maximum power generation amount] increases as [Reliability] increases (the reliability increases). Nearest tentative predicted maximum power generation].

  When the predicted maximum power generation amount is calculated as described above, the predicted maximum power generation amount is prevented from becoming larger than the actual maximum power generation amount, and the predicted maximum power generation amount is excessively smaller than the actual maximum power generation amount. Is prevented. Therefore, the maximum power generation amount prediction unit 13 calculates the predicted maximum power generation amount so that the power amount supplied by the power generation facility can be increased as much as possible while avoiding a situation where the power amount supplied by the power generation facility is insufficient. It becomes possible.

  Moreover, when the maximum predicted power generation amount is calculated by the maximum power generation amount prediction system 1 as described above, the maximum power generation amount of the power generation facility can be predicted in advance and automatically. This eliminates the need for labor (personnel) to predict the maximum power generation capacity of the power generation facility, and makes it possible to determine the operation plan of the power generation facility and the amount of power sold to the power exchange at an early stage. Can be increased.

  In the above-described maximum power generation amount prediction system 1, the case where the maximum power generation amount prediction unit 13 calculates the predicted maximum power generation amount using the maximum power generation amount table and the fluctuation amount table which are actual measurement results is illustrated. The method for calculating the maximum power generation amount is not limited to this example. For example, a relational expression may be constructed by performing regression analysis on the actual measurement result, and the predicted maximum power generation amount may be calculated using the relational expression. Thus, it is preferable to calculate the predicted maximum power generation amount using the relational expression because the predicted maximum power generation amount corresponding to any predicted atmospheric state can be calculated.

  The parameters of the relational expression are elements (for example, atmospheric pressure, temperature, humidity, etc.) that form the predicted atmospheric state described above. In this case, the value of the element may be used as it is as the parameter of the relational expression, or the power value of the element may be used as the parameter of the relational expression. Moreover, it is preferable that the relational expression is updated as needed every time the actual measurement result is obtained, because the prediction accuracy can be maintained high.

  Thus, when the predicted maximum power generation amount is calculated using a relational expression that is updated as needed based on the actual measurement result, for example, in a power generation facility that combines a plurality of powers such as a gas turbine combined cycle, it contributes to fluctuations in the power generation amount. Since various factors (for example, deterioration of the gas turbine over time, characteristics of air taken in by the gas turbine, etc.) can be taken into account, the predicted maximum power generation amount can be calculated with high accuracy. Even when the table data described above is used, the maximum power generation amount can be calculated in consideration of various factors contributing to the above-described fluctuations in the power generation amount if the table data is updated as needed.

  Further, as in the case of using the above table data, it is preferable to use the above relational expression when the reliability is taken into consideration. In this case, for example, the predicted maximum power generation amount can be calculated based on the following formula (2).

  [Predicted maximum power generation amount] = [Calculated maximum power generation amount] − [Calculated fluctuation amount] (2)

  [Calculated maximum power generation amount] in the above formula (2) is the maximum power generation amount calculated by substituting elements (atmospheric pressure, temperature, humidity, etc.) forming the predicted atmospheric state into the above-described relational expression. [Calculated maximum power generation amount] corresponds to [Temporary predicted maximum power generation amount] in the above formula (1).

  [Calculated fluctuation amount] in the above formula (2) is a fluctuation range calculated based on the reliability, and is calculated using the above-described relational expression when it is assumed that the elements forming the predicted atmospheric state fluctuate. It is the fluctuation range of the maximum power generation. Specifically, for example, it is assumed that the parameter of the relational expression is only air temperature, the predicted air temperature is 20 ° C., and the temperature fluctuation range calculated based on the reliability is ± 2 ° C. In this case, by substituting 22 ° C. and 18 ° C. into the above-described relational expression, the absolute value of the difference between the obtained maximum power generation amounts becomes the calculated fluctuation amount. [Calculated power generation amount] corresponds to [Adjusted fluctuation amount] in the above equation (1). In addition, when there are a plurality of parameters in the relational expression, it is assumed that each parameter fluctuates with a fluctuation range based on the reliability, and the maximum value of the maximum power generation amount calculated by substituting the parameter into the relational expression described above The calculated fluctuation amount may be calculated by subtracting the minimum value from.

  In addition, the [calculated fluctuation amount] in the above equation (2) is obtained by empirically obtaining the relationship (relational expression) between the reliability and the fluctuation range of the maximum power generation amount, and then calculating the relationship (relational expression). It may be calculated by applying (substituting) the obtained reliability.

<Warning notification method>
Next, a warning notification method by the warning notification unit 15 will be described with reference to the drawings. FIG. 4 is a diagram illustrating an example of a warning notification method. In the following, for concrete explanation, the maximum power generation amount prediction system 1 determines the predicted maximum power generation amount on January 5, which is the prediction target period, twice on January 1 and January 2. An example of the case of calculation will be described.

  In the example illustrated in FIG. 4, the predicted maximum power generation amount in the prediction target period (January 5) calculated on January 1 is 55 MWh. On the other hand, the predicted maximum power generation amount in the prediction target period (January 5) calculated on January 2 is 50 MWh.

  For example, every time a new predicted maximum power generation amount is calculated, the warning notification unit 15 acquires the predicted maximum power generation amount data from the database 12 and calculates the amount of change in the predicted maximum power generation amount. This change amount may be an absolute value of a difference between the maximum value and the minimum value of the predicted maximum power generation amount in the prediction target period (January 5) calculated up to now.

  Next, the warning notification unit 15 confirms whether or not the amount of change in the predicted maximum power generation amount (5 MWh in the example of FIG. 4) exceeds the threshold (3 MWh in the example of FIG. 4). When the amount of change in the predicted maximum power generation amount exceeds the threshold, the warning notification unit 15 notifies the user of a warning. On the other hand, if the amount of change in the predicted maximum power generation amount does not exceed the threshold value, the warning notification unit 15 does not notify the user of the warning.

  When the possibility of a sudden change in atmospheric conditions (for example, a sudden typhoon hits directly) or the expected change in atmospheric conditions (for example, a significant typhoon shift) increases For example, a predicted maximum power generation amount that is significantly different from the previously calculated predicted maximum power generation amount can be calculated.

  In such a case, if the user makes an operation plan for the power generation facility based on the previously calculated maximum power generation amount or determines the amount of power sold for bidding on the power exchange, the predicted maximum power generation amount The actual power generation facility's maximum power generation amount deviates, so that the amount of power to be supplied is insufficient or the amount of power that is wasted increases.

  Therefore, as described above, the warning notification unit 15 notifies the user of a warning when the predicted maximum power generation amount greatly changes beyond the threshold value. As a result, the user is given an opportunity to review the predicted maximum power generation amount calculated previously, so that it is possible to prevent the amount of power supplied from becoming insufficient or the amount of wasted power from increasing. It becomes.

<Deformation, etc.>
In the maximum power generation amount prediction system 1, the predicted atmospheric state data generation unit 10 generates predicted atmospheric state data based on GPV data, and the reliability data generation unit 11 generates reliability data based on ensemble data. However, as long as the predicted atmospheric state data generation unit 10 calculates the predicted atmospheric state and the reliability data generation unit 11 can calculate the reliability, the GPV data and the ensemble data may not necessarily be used.

  The maximum power generation amount table and the fluctuation amount table illustrated in FIG. 3 may be used properly according to the operation method of the power generation facility and the season. For example, a maximum power generation table and fluctuation table in summer when a cooler for cooling the air introduced into the gas turbine is used, and a maximum power generation table and fluctuation table in winter when a cooler is not used, and May be created separately, and these tables may be used properly according to the time when the predicted maximum power generation amount is calculated. Further, as described above, when using a relational expression instead of the table data, the relational expression may be properly used according to the operation method and the season of the power generation facility.

  The maximum power generation amount prediction system of the present invention can be suitably used for a maximum power generation amount prediction system that predicts the maximum power generation amount of a power generation facility such as a gas turbine plant that generates power by burning fuel such as natural gas.

1: Maximum power generation amount prediction system 10: Predicted atmospheric state data generation unit 11: Reliability data generation unit 12: Database 13: Maximum power generation amount prediction unit 14: Output unit 15: Warning notification unit

Claims (6)

A maximum power generation amount prediction system that predicts a predicted maximum power generation amount that is a maximum power generation amount in a prediction target period of a power generation facility that generates power by burning fuel using oxygen in the atmosphere,
A predicted atmospheric state data generation unit that generates predicted atmospheric state data indicating a predicted atmospheric state that is a prediction result of the atmospheric state in the prediction target period;
A reliability data generation unit that generates reliability data indicating reliability that is a high reliability of prediction of the predicted atmospheric state;
A maximum power generation amount predicting unit that calculates the predicted maximum power generation amount based on the predicted atmospheric state data and the reliability data,
The maximum power generation amount prediction unit is configured to calculate the predicted maximum power generation amount that increases as the reliability increases, if the predicted atmospheric state is the same. The maximum power generation amount prediction unit, the prediction corresponding to the atmospheric conditions to calculate a provisional predicted maximum power generation amount, the reliability that reduces the tentative maximum predicted power generation amount higher small reduction, the predicted maximum power The maximum power generation amount prediction system according to claim 1, wherein the amount is calculated.
  The maximum power generation amount prediction unit uses the correspondence relationship between the measured atmospheric state, which is the state of the atmosphere actually measured in the past, and the actually measured maximum power generation amount, which is the maximum power generation amount measured in the past. 3. The maximum power generation amount prediction system according to claim 2, wherein the provisional predicted maximum power generation amount corresponding to an atmospheric state is calculated. The maximum power generation amount prediction unit
Corresponding to the predicted atmospheric state by using the correspondence relationship between the measured atmospheric state, which is the atmospheric state actually measured in the past, and the measured fluctuation amount, which is the magnitude of the fluctuation of the maximum power generation amount measured in the past. Calculated the tentative forecast fluctuation amount,
By adjusting the provisional prediction fluctuation amount so as to be smaller as the reliability is higher, an adjusted fluctuation amount is calculated,
4. The maximum power generation amount prediction system according to claim 2, wherein the predicted maximum power generation amount is calculated by subtracting the adjusted fluctuation amount from the provisional predicted maximum power generation amount. 5. A warning notification unit for notifying the user of a warning;
The maximum power generation amount prediction unit calculates the predicted maximum power generation amount a plurality of times at different timings until reaching the prediction target period,
5. The warning according to claim 1, wherein when the predicted maximum power generation amount changes beyond a predetermined threshold, the warning notification unit notifies the user of a warning. Maximum power generation prediction system.   6. The maximum power generation amount prediction system according to claim 1, wherein the predicted atmospheric state data indicates a prediction result of at least one of atmospheric pressure and temperature.

Images