KR102591679B1 - Method and apparatus for determining generated power for economic load dispatch - Google Patents

Abstract

A method of determining the amount of solar power performed by an apparatus for determining the amount of power for economic dispatch according to an embodiment includes the steps of estimating the amount of solar power generated by a solar power generator based on the amount of solar radiation predicted using meteorological data, and the estimated amount of solar power generated. A step of constructing a probability distribution model for each time reflecting the error in the actual solar power generation amount, and a predetermined number of solar power having the predicted power generation amount of the solar power generator as the average through the probability distribution model for each time. Generating power generation data, calculating an expected power generation cost for each time based on a predetermined number of net loads generated by the generated solar power generation data, and calculating an expected power generation cost for each time. It includes the step of determining the power generation amount of the non-renewable energy generator based on.

Description

Apparatus and method for determining power generation for economic dispatch {METHOD AND APPARATUS FOR DETERMINING GENERATED POWER FOR ECONOMIC LOAD DISPATCH}

The present invention relates to a device for determining the amount of power generation for economic dispatch and a method by which the device determines the amount of power generation.

The steady decline in the price of renewable energy sources confirms that it is a cost-effective climate decarbonization solution that allows the continued increase in the linkage of cost-competitive renewable energy sources, short-term economic needs, and mid- to long-term sustainable development goals. In addition, the expansion and introduction of renewable energy sources in the power system has positive effects such as reducing dependence on thermal power generation energy and minimizing power generation costs. However, the operation of the current power system is focused on how to meet the power supply with the volatility of power demand that changes in real time, and is only performed at thermal power plants, where cost function modeling is relatively easy.

The diversity of power sources complicates the formulation of equations for solving the problem of optimal operation of the system, and renewable energy sources that have intermittency due to the surrounding environment have difficulties in accurately modeling the output of power generation. Therefore, in order to satisfy the dissemination and expansion of microgrids using renewable energy sources, forecasting technology for the output of renewable energy sources and economic dispatch considering the output volatility of renewable energy sources are needed as a basic technology for operating/planning the system. .

Republic of Korea Patent Publication No. 10-2021-0047627, published on April 30, 2021.

According to one embodiment, the amount of solar power generation is estimated based on solar radiation, then used to construct a probability distribution model for each time, and the power generation amount of the non-renewable energy generator is determined based on the net load obtained based on this for economic dispatch. Provides a device and method for determining power generation.

The problems to be solved by the present invention are not limited to those mentioned above, and other problems to be solved that are not mentioned will be clearly understood by those skilled in the art to which the present invention pertains from the following description.

The method of determining the power generation amount performed by the power generation determination device for economic dispatch according to the first aspect includes the steps of estimating the solar power generation amount of the solar power generator based on the solar radiation amount predicted using meteorological data, and the estimated solar power generation amount A step of constructing a probability distribution model for each time reflecting the error in the actual solar power generation amount, and a predetermined number of solar power having the predicted power generation amount of the solar power generator as the average through the probability distribution model for each time. Generating power generation data, calculating an expected power generation cost for each time based on a predetermined number of net loads generated by the generated solar power generation data, and calculating an expected power generation cost for each time. It includes the step of determining the power generation amount of the non-renewable energy generator based on.

The power generation determination device according to the second aspect includes an input unit that receives weather data, and a processor unit that determines the power generation amount of a non-renewable energy generator for economic dispatch using the weather data, and the processor unit receives the weather data. Estimating the solar power generation amount of the solar power generator based on the solar radiation predicted using the solar power generator, constructing a probability distribution model for each time reflecting the error between the estimated solar power generation amount and the actually measured solar power generation amount, and constructing a probability distribution model for each time Through a probability distribution model, a predetermined number of solar power generation data having the predicted power generation amount of the solar power generator as an average is generated, and each time based on a predetermined number of net loads generated by the generated solar power generation data. Calculate the expected power generation cost for each time, and determine the power generation amount of the non-renewable energy generator based on the calculated expected power generation cost for each time.

A computer-readable recording medium storing a computer program according to the third aspect, when the computer program is executed by a processor, causes the processor to perform a method of determining the amount of power generation performed by the device for determining the amount of power generation for economic dispatch. Includes commands for

The computer program stored in the computer-readable recording medium according to the fourth aspect, when the computer program is executed by a processor, causes the processor to perform the method of determining the amount of power generation performed by the device for determining the amount of power generation for economic dispatch. Includes commands for

According to an embodiment of the present invention, the amount of solar power generation is estimated based on the amount of solar radiation, then the probability distribution model for each time is constructed using this, and the amount of power generation of the non-renewable energy generator is determined based on the net load obtained based on this. Since this invention takes into account the intermittency and uncertainty of renewable energy sources, the expected value of reserve power and output limitation costs according to time during economic dispatch is calculated. In addition, it is possible to compare expected values for each scenario by substituting the intermittency and uncertainty of renewable energy sources into economic dispatch that does not include existing reserve capacity and output limitation costs. Through this, it is possible to calculate the expected profit in case of economic emergency funds and has the effect of enabling efficient economic emergency funds.

Figure 1 is a configuration diagram of a device for determining power generation for economic dispatch according to an embodiment of the present invention.
Figure 2 is a flowchart illustrating a method for determining the amount of power generation performed by the device for determining the amount of power generation for economic dispatch according to an embodiment of the present invention.
Figure 3 is an IEEE 13 busbar diagram used to analyze the effectiveness of economic dispatch according to an embodiment of the present invention.
Figure 4 is a graph showing solar radiation prediction results according to an embodiment of the present invention.
Figure 5 is a graph showing the results of predicting solar power generation according to an embodiment of the present invention.

The advantages and features of the present invention and methods for achieving them will become clear with reference to the embodiments described below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete and to be understood by those skilled in the art. It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims.

The terms used in this specification will be briefly explained, and the present invention will be described in detail.

The terms used in the present invention are general terms that are currently widely used as much as possible while considering the function in the present invention, but this may vary depending on the intention or precedent of a person working in the art, the emergence of new technology, etc. In addition, in certain cases, there are terms arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the relevant invention. Therefore, the terms used in the present invention should be defined based on the meaning of the term and the overall content of the present invention, rather than simply the name of the term.

When it is said that a part ‘includes’ a certain element throughout the specification, this does not mean excluding other elements, but may further include other elements, unless specifically stated to the contrary.

Additionally, the term ‘unit’ used in the specification refers to software or hardware components such as FPGA or ASIC, and the ‘unit’ performs certain roles. However, ‘wealth’ is not limited to software or hardware. The 'part' may be configured to reside on an addressable storage medium and may be configured to reproduce on one or more processors. Therefore, as an example, 'part' refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, procedures, Includes subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functionality provided within components and parts can be combined into a smaller number of components and parts or further separated into additional components and parts.

Below, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the present invention. In order to clearly explain the present invention in the drawings, parts unrelated to the description are omitted.

Figure 1 is a configuration diagram of a device for determining power generation for economic dispatch according to an embodiment of the present invention.

Referring to FIG. 1, an apparatus 100 for determining power generation for economic dispatch according to an embodiment includes an input unit 110 and a processor unit 120.

The input unit 110 receives weather data and provides the received weather data to the processor unit 120.

The processor unit 120 uses weather data to determine the power generation amount of the non-renewable energy generator for economic dispatch. This processor unit 120 estimates the solar power generation amount of the solar power generator based on the solar radiation amount predicted using weather data, and generates a probability distribution for each time reflecting the error between the estimated solar power generation amount and the actually measured solar power generation amount. Constructing a model, generating a predetermined number of solar power generation data with the predicted power generation amount of the solar power generator as the average through a probability distribution model for each time, and generating a predetermined number of orders generated by the generated solar power generation data Based on the load, the expected power generation cost for each time is calculated, and the power generation amount of the non-renewable energy generator is determined based on the calculated expected power generation cost for each time.

This processor unit 120 can predict solar radiation using LSTM (long short term memory), a deep learning technique.

Additionally, weather observation data may be used as the weather data input by the input unit 110, and temperature, wind direction, wind speed, precipitation, and humidity may be used as the initial independent variables. And, this weather data may have gone through a data pre-processing process. In this data preprocessing process, missing values can be explored and compensated. In addition, independent and dependent variables can be reclassified to reflect time reality and scale between variables can be adjusted. Afterwards, data such as trends in solar radiation, seasonality, hour, day, week, month, and year can be created as additional independent variables, and then some highly correlated independent variables can be removed to remove multicollinearity.

When constructing a probability distribution model, the processor unit 120 defines the average and variance of the power generation amount at each time through the error between the estimated solar power generation amount and the actually measured solar power generation amount and can use this to construct a probability distribution model. there is.

Additionally, when calculating the expected value of power generation cost, the processor unit 120 may calculate the expected value of power generation cost by adding the output limitation cost and reserve power cost to the net load.

Additionally, when calculating the expected value of power generation cost, the processor unit 120 may calculate the expected value of power generation cost through Monte Carlo simulation.

Additionally, when determining the power generation amount of a non-renewable energy generator, the processor unit 120 may determine the power generation amount using a particle swarm optimization (PSO) algorithm.

The power generation determination device 100 configured in this way can determine the power generation amount of a non-renewable energy generator (eg, diesel generator, etc.) for economic dispatch within the microgrid.

Hereinafter, with reference to FIGS. 1 and 2, we will take a closer look at the process by which the power generation determination device 100 according to an embodiment of the present invention determines the power generation amount of a non-renewable energy generator for economic dispatch within a microgrid. .

Microgrid refers to a power system in which renewable energy sources and loads are managed integratedly within a limited area and can be operated independently or connected to a small power grid connected to the commercial power grid at one connection point. While existing centralized control methods and large-scale power systems incur enormous costs, microgrids are efficient because they produce energy by predicting the power generation and consumption of renewable energy sources within the system and sell surplus power back to the central power plant. This can increase energy consumption. These microgrids can increase the efficiency of the power industry, effectively reduce energy use and greenhouse gas emissions, and increase the amount of power generation from distributed power sources.

These microgrids can include renewable energy generators and non-renewable energy generators. For example, a renewable energy generator may be a solar generator, and a non-renewable energy generator may be a diesel generator.

Diesel generators supply loads that cannot be met by the generation of solar generators within the microgrid. For modeling of diesel generators, existing fossil fuel power generation cost models such as Equation 1 can be used.

[Equation 1]

here, is a generator is the cost function for, is a generator is the output of represents the cost coefficient of the generator, and the constraints of the generator are as shown in Equation 2 and Equation 3 below.

[Equation 2]

[Equation 3]

here, is the total load of the current system, refers to the total transmission loss that occurs in the system during power generation. In this paper, an ideal form of power generation that does not incur losses Since it is assumed that the total power generation in a given time period is equal to the total load, maintaining supply and demand balance. class are the respective generators Indicates the minimum and maximum output of .

The power generated by a solar power generator is greatly influenced by the amount of solar radiation and also varies depending on the temperature of the module, ambient temperature of the solar light, and module characteristics. The relational expression that formalizes the output function for the solar power module considering the relevant conditions is as shown in Equation 4 to Equation 8.

[Equation 4]

[Equation 5]

[Equation 6]

[Equation 7]

[Equation 8]

here, Is Temperature of the cell at time [°C] is the ambient temperature [°C] and are the voltage [ /°C] and the temperature coefficient of current [mA/°C]. is the temperature [°C] for the temperature of the cell represents the charging rate and means the short-circuit current [A] and open-circuit voltage [V] of the module, respectively. and represents the maximum power point voltage and current, respectively. and Is Output and solar radiation of solar power generator in time [W/ ] means.

The economic dispatch method determines the active power output by each generator to minimize the overall power generation cost. In general, in the case of a microgrid composed of renewable energy sources, diesel, and an energy storage system (ESS), the diesel generator plays the role of controlling frequency and voltage, and the energy storage system plays the role of compensating for fluctuating voltage and frequency. It is common to perform . In this case, because the diesel generator has a large capacity and must be operated at all times, fuel consumption is high.

In order to minimize fuel consumption of the diesel generator, the economic power supply method according to an embodiment of the present invention predicts the power generation of the solar power generator and then operates the solar power generator at its maximum output at all times based on this prediction. In addition, the valve point effect is applied to consider the input and output characteristics of the diesel generator, and then economic dispatch is performed taking into account the output uncertainty of renewable energy sources through a probability distribution model.

The device 100 for determining the amount of power generation for economic dispatch according to an embodiment of the present invention predicts the amount of solar radiation to predict the amount of solar power generation and then applies it to a physical prediction model. For example, solar radiation prediction can use deep learning techniques. For example, LSTM can be used as a deep learning technique.

According to an embodiment of the present invention, synoptic weather observation data and disaster prevention weather observation data from 2015 to 2020 provided by the Korea Meteorological Administration were used as weather data for predicting solar radiation. The initial independent variables of the input data were temperature, wind direction, wind speed, precipitation, and humidity.

According to the solar radiation prediction algorithm according to an embodiment of the present invention, missing values are searched for and compensated for through a data preprocessing process for meteorological data. In addition, independent and dependent variables are reclassified, time reality is reflected, and scales between variables are adjusted. Afterwards, data such as trend and seasonality of solar radiation, hour, day, week, month, and year are created as additional independent variables, and highly correlated independent variables are removed to remove multicollinearity. Through this, overfitting can be prevented and optimal input variables can be selected. For example, through this process, 14 independent variables can be selected and the training set and test set can be separated as of January 1, 2020.

When data that has undergone this data preprocessing process is provided to the processor unit 120 through the input unit 110, the processor unit 120 can predict the amount of solar radiation by performing LSTM (S210). The relational expressions for MAE (mean absolute error) and MSE (mean square error), which can be used as a performance evaluation method for solar radiation predicted according to the solar radiation prediction algorithm, are as shown in Equation 9 and Equation 10.

[Equation 9]

[Equation 10]

here, is the measured value, is the predicted value is the number of data.

Next, the processor unit 120 estimates the solar power generation amount of the solar power generator based on the solar radiation amount predicted using weather data (S220), and the error between the estimated solar power generation amount and the actually measured solar power generation is reflected. Construct a probability distribution model for time (S230). Here, when constructing the probability distribution model, the processor unit 120 defines the average and variance of the power generation amount at each time through the error between the estimated solar power generation amount and the actually measured solar power generation amount and uses this to construct the probability distribution model. can do.

And, the processor unit 120 generates a predetermined number (e.g., 1,000) of solar power generation data having the average predicted power generation amount of the solar power generator through a probability distribution model for each time (S240), and the generated The expected power generation cost for each hour can be calculated based on a predetermined number (eg, 1,000) of net loads generated by solar power generation data. Here, when calculating the expected value of power generation cost, the processor unit 120 can add the output limitation cost and reserve cost to the net load and calculate the expected value of power generation cost through Monte Carlo simulation. Net load refers to the required power generation from diesel generators excluding solar power generation within the microgrid (S250).

The cost function of the fossil fuel power generation cost model can be expressed by approximating a two-dimensional function. However, the actual output curve of a diesel generator is highly non-linear and has many local solutions due to the sudden increase in losses that occur when the turbine's steam intake valve begins to open, which is called the valve point effect. Accordingly, in one embodiment of the present invention, a relational expression such as Equation 11 is used in consideration of the valve point effect, which is a practical limitation of a diesel generator.

[Equation 11]

here, and is a coefficient according to the generator characteristics.

Next, the processor unit 120 determines the power generation amount of the non-renewable energy generator based on the expected power generation cost for each time calculated in step S250. At this time, the processor unit 120 may determine the amount of power generation using the PSO algorithm (S260).

Operating the solar power generator and diesel generator in the microgrid based on the power generation amount of the non-renewable energy generator (e.g., diesel generator) determined according to the power generation amount determination method performed by the power generation amount determination device 100 for economic dispatch as described so far. If you do this, an economic emergency will occur. The PSO algorithm is an optimization technique that models the process of observing the characteristics of natural creatures when they move in groups or search for food, and then the objects within the group share their experiences and search for the optimal path. The PSO algorithm has the characteristic of converging more stably than other statistical methods and uses the cost function of each generator as the objective function for economic dispatch. Due to the non-linearity and discontinuity of input-output characteristics such as the valve point effect depending on the load, economic dispatch is performed using the PSO algorithm based on empirical and probabilistic search.

Figure 3 is an IEEE 13 busbar diagram used to analyze the effectiveness of economic dispatch according to an embodiment of the present invention, Figure 4 is a graph showing the solar radiation prediction results according to an embodiment of the present invention, and Figure 5 is a graph showing the results of solar radiation prediction according to an embodiment of the present invention. This is a graph showing the results of predicting solar power generation according to an embodiment of .

Hereinafter, the results of examining the effectiveness and effect of economic emergency payments according to an embodiment of the present invention through simulation will be described.

Two scenarios are performed to analyze the effectiveness of economic emergency funds according to an embodiment of the present invention. Scenario 1 is an economic dispatch method that considers the intermittency and uncertainty of solar power generation through a probabilistic method, and scenario 2 is an existing economic dispatch method that does not consider the intermittency and uncertainty of solar power generation. To perform the scenario, a solar generator 1400 is installed at node 632 in Figure 3 and the IEEE 13 busbar schematic diagram. was installed, and the data of the four diesel generators are shown in Table 1.

No. a b c 561 7.92 0.00156 800 1700 310 7.85 0.00194 600 1300 78 7.97 0.00482 300 700 280 7.78 0.00323 400 1000

Here, a, b, and c are the coefficients of the generator fuel cost function.

Scenario 1 predicted solar radiation using LSTM, one of the deep learning techniques, and the results are shown in Figure 4.

The MAE and MSE for Figure 4 have high prediction performance of 0.08 and 0.02, respectively. The predicted insolation amount was applied to the physical model to calculate solar power generation, and a day-ahead probability distribution model was constructed through the error of the actual measured data and predicted data. The results are shown in Figure 5.

To confirm the expected value of hourly operating power generation costs, a probability distribution model was used and hourly solar power generation data was generated. The net load is determined according to the generated data, and according to the net load, the power generation cost of the diesel generator, output limitation cost, and reserve power cost are considered. The output limitation cost was calculated as 180 [KRW/kWh], and the reserve power cost was calculated as 125.81 [KRW/kWh].

Monte Carlo simulation was used to calculate the expected power generation cost considering the intermittency and uncertainty of solar power generation. Through Monte Carlo simulation, the power generation cost plus the output limitation cost and reserve cost were calculated, and the results are shown in Table 2 below.

Time [MW] [MW] [MW] [MW] Cost [1,000 won] 0 999.4 749.6 399.7 490.2 29,604.7 One 800 899.1 349.8 573.5 29,490.2 2 899.7 749.6 349.8 598.9 29,157 3 899.7 674.7 349.8 662.9 29,118 4 899.7 824.3 49.8 519.5 29,094.1 5 999.4 749.6 349.8 500.2 29,098.7 6 899.7 974.0 399.7 495.2 31,279 7 1498 899.1 300 558.7 37,788 8 1697 1273 549.3 940.4 55,374.8 9 1697 1273 549.3 954.3 60,545.5 10 1597 1273 599.1 696.3 58,635.1 11 1398 1198 399.7 862.8 55,731.2 12 999.4 824.3 349.8 590.6 39,536.6 13 1298 973.9 449.5 681.2 47,922 14 1498 1123 399.7 584.7 51,834.2 15 1099 1048 499.4 690.0 45,102.4 16 1298 899.1 300 609.7 41,864.9 17 899.7 749.5 399.7 502.3 32,465.7 18 800 899.1 300 504.3 31,231.1 19 800 899.2 399.7 409.5 28,109.4 20 899.7 600 349.8 477.7 25,851.6 21 899.7 674.8 300 455.2 25,877.7 22 800 600 349.8 577.5 25,894.9 23 899.7 674.7 349.7 400 25,737.6

Scenario 2 maintains supply and demand balance while minimizing the power generation costs of diesel generators. The PSO algorithm was used to obtain the optimal power generation of the diesel generator for the predicted net load, and the results are shown in Table 3.

Time [MW] [MW] [MW] [MW] Cost [1,000 won] 0 899.7 749.5 399.7 590 29,659.5 One 800 973.9 349.8 498.7 29,492.3 2 999.8 824.6 300 474.9 29,129.6 3 999.4 674.7 349.8 563.2 29,044.9 4 899.7 749.5 349.8 594.3 29,080 5 899.7 899.1 300 500.2 29,113.7 6 1099 969.5 300 400 31,312.8 7 1198 973.9 399.7 683.4 37,532.3 8 1697 1273 599.1 890.3 54,608 9 1697 1273 549.3 950.1 54,698.9 10 1498. 1273 549.3 838.1 50,143.8 11 1498 1198 449.5 709.6 45,803 12 1198 749.5 300 504.6 31,099.3 13 1398 973.9 399.7 627.9 39,565.6 14 1398 1123 349.8 731.9 42,349.7 15 1398 1048 399.7 488.4 38,669.5 16 1198 824.3 449.5 629.7 35,658.1 17 899.7 899.1 300 448.3 28,561.1 18 999.4 749.5 349.8 402.2 27,893.7 19 800 824.3 399.7 484.3 28,059.1 20 800.0 674.7 349.8 502.6 25,825.6 21 800 600 349.8 579.8 25,916.9 22 800 600 349.8 577.4 25,894.8 23 800 674.7 300 549.4 25,857.5

To verify the effectiveness of Scenario 1, the power generation amount of each generator determined in Scenario 2 is substituted into Monte Carlo simulation. Through this, it is possible to obtain probabilistic expected values for the results of existing economic dispatches. Its effectiveness can be verified through comparison with the expected value of Scenario 1. The results of the expected values for scenarios 1 and 2 are shown in Table 4, and the expected profits were calculated between 9:00 and 19:00 when the solar power generator is in operation.

Time Scenario 1 Cost
[One thousand won] Scenario 2 Cost
[One thousand won] Comparison of expected profits
[One thousand won] 9 60,545.5 60,589.6 44.1 10 58,635.1 58,687.5 52.4 11 55,731.1 55,768.8 37.7 12 39,536.6 39,769.8 233.2 13 47,921.9 48,052.1 130.2 14 51,834.2 51,927.2 93 15 45,102.4 45,081.8 -20.6 16 41,864.9 41,952.6 87.7 17 32,465.7 32,564.2 98.5 18 31,231.1 31,186.8 -44.3 19 28,109.3 28,059.1 -50.2

Through the results in Table 4, it can be seen that when using the proposed economic dispatch algorithm, an economic benefit of approximately 661.7 [thousand won/day] can be obtained compared to the existing economic dispatch method. The expected economic profit that can be obtained in one year through repetition of the above-described simulation was calculated to be approximately 230 million won. This result can be said to be highly effective because it is the result of increased operational efficiency through the system operation algorithm without installing or reinforcing additional facilities in the line or system.

According to one embodiment of the present invention, LSTM, one of the deep learning techniques, was used to increase prediction accuracy for solar power output. In order to consider the intermittency and uncertainty of solar power generation during economic dispatch, Monte Carlo simulation and probability distribution model, which are one of the probabilistic methods, were applied. When the proposed economic dispatch is performed, approximately 660,000 won per day is generated compared to the existing economic dispatch method. There was a difference in operating costs. Finally, by determining the optimal power generation amount through the PSO algorithm, economic operating benefits according to the expected value can be realized.

Meanwhile, each step included in the method for determining the amount of power performed by the device for determining the amount of power for economic dispatch according to the above-described embodiment is a computer-readable record that records a computer program including instructions for performing these steps. It can be implemented in any medium.

Combinations of each step in each flowchart attached to the present invention may be performed by computer program instructions. Since these computer program instructions can be mounted on the processor of a general-purpose computer, special-purpose computer, or other programmable data processing equipment, the instructions performed through the processor of the computer or other programmable data processing equipment perform the functions described in each step of the flowchart. It creates the means to carry out these tasks. These computer program instructions may also be stored on a computer-usable or computer-readable recording medium that can be directed to a computer or other programmable data processing equipment to implement a function in a particular way, so that the computer program instructions are computer-usable or computer-readable. The instructions stored in the recording medium can also produce manufactured items containing instruction means that perform the functions described in each step of the flowchart. Computer program instructions can also be mounted on a computer or other programmable data processing equipment, so that a series of operational steps are performed on the computer or other programmable data processing equipment to create a process that is executed by the computer, thereby generating a process that is executed by the computer or other programmable data processing equipment. Instructions that perform processing equipment may also provide steps for executing the functions described in each step of the flowchart.

Additionally, each step may represent a module, segment, or portion of code containing one or more executable instructions for executing specified logical function(s). Additionally, it should be noted that in some alternative embodiments it is possible for the functions mentioned in the steps to occur out of order. For example, two steps shown in succession may in fact be performed substantially simultaneously, or the steps may sometimes be performed in reverse order depending on the corresponding function.

The above description is merely an illustrative explanation of the technical idea of the present invention, and those skilled in the art will be able to make various modifications and variations without departing from the essential quality of the present invention. Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention, but are for illustrative purposes, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention shall be interpreted in accordance with the claims below, and all technical ideas within the scope equivalent thereto shall be construed as being included in the scope of rights of the present invention.

100: Power generation determination device
110: input unit
120: Processor unit

Claims (16)

As a method of determining the amount of power generation performed by a device for determining the amount of power generation for economic dispatch,
A step of estimating the solar power generation amount of the solar power generator based on the amount of solar radiation predicted using meteorological data;
Constructing a probability distribution model for each time reflecting the error between the estimated solar power generation amount and the actually measured solar power generation amount;
Generating a predetermined number of solar power generation data having the predicted power generation amount of the solar power generator as an average through the probability distribution model for each time;
calculating an expected power generation cost for each hour based on a predetermined number of net loads generated by the generated solar power generation data;
Including the step of determining the power generation amount of the non-renewable energy generator based on the expected power generation cost for each time calculated above.
How to determine power generation. According to claim 1,
In the step of estimating the amount of solar power generation, the amount of solar radiation is predicted using LSTM (long short term memory) among deep learning techniques.
How to determine power generation. According to claim 2,
The weather data used in the step of estimating the amount of solar power generation has gone through a data preprocessing process,
In the data preprocessing process, in addition to the initial independent variable, at least one of the trend and seasonality of solar radiation, hour, day, week, month, and year is created as an additional independent variable, and according to the correlation between the initial independent variable and the additional independent variable, Some independent variables have been removed
How to determine power generation. According to claim 1,
The step of configuring the probability distribution model for each time is to define the average and variance of the power generation amount for each time through the error and use this to construct the probability distribution model.
How to determine power generation. According to claim 1,
The step of calculating the expected value of power generation cost is to calculate the expected value of power generation cost by adding the output limitation cost and reserve cost to the net load.
How to determine power generation. According to claim 1,
The step of calculating the expected value of power generation cost is calculating the expected value of power generation cost through Monte Carlo simulation.
How to determine power generation. According to claim 1,
The step of determining the power generation amount involves determining the power generation amount using a PSO (particle swarm optimization) algorithm.
How to determine power generation. An input unit that receives weather data,
A processor unit that determines the amount of power generated by a non-renewable energy generator for economic dispatch using the weather data,
The processor unit,
Estimating the solar power generation of the solar power generator based on the solar radiation predicted using the weather data, constructing a probability distribution model for each time reflecting the error between the estimated solar power generation and the actually measured solar power generation, Through the probability distribution model for each time, a predetermined number of solar power generation data having the average predicted power generation amount of the solar power generator is generated, and a predetermined number of net loads generated by the generated solar power generation data are generated. Based on this, the expected power generation cost for each time is calculated, and the power generation amount of the non-renewable energy generator is determined based on the calculated expected power generation cost for each time.
Power generation determination device. According to claim 8,
The processor unit predicts the amount of solar radiation using LSTM, a deep learning technique.
Power generation determination device. According to clause 9,
The input weather data has gone through a data preprocessing process,
In the data preprocessing process, in addition to the initial independent variable, at least one of the trend and seasonality of solar radiation, hour, day, week, month, and year is created as an additional independent variable, and according to the correlation between the initial independent variable and the additional independent variable, Some independent variables have been removed
Power generation determination device. According to claim 8,
The processor unit defines the average and variance of the power generation amount at each time through the error and uses this to construct the probability distribution model.
Power generation determination device. According to claim 8,
The processor unit calculates the expected power generation cost by adding an output limitation cost and a reserve cost to the net load.
Power generation determination device. According to claim 8,
The processor unit calculates the expected power generation cost through Monte Carlo simulation.
Power generation determination device. According to claim 8,
The processor unit determines the power generation amount using a particle swarm optimization (PSO) algorithm.
Power generation determination device. A computer-readable recording medium storing a computer program,
When the computer program is executed by a processor,
A step of estimating the solar power generation amount of the solar power generator based on the solar radiation amount predicted using meteorological data, and constructing a probability distribution model for each time reflecting the error between the estimated solar power generation amount and the actually measured solar power generation amount. A step of generating a predetermined number of solar power generation data having the average predicted power generation amount of the solar power generator through the probability distribution model for each time, and a predetermined number of solar power generation data generated by the generated solar power generation data. The processor comprising the steps of calculating an expected generation cost for each hour based on the number of net loads, and determining the generation amount of the non-renewable energy generator based on the calculated expected generation cost for each hour. A computer-readable recording medium containing instructions for executing a computer-readable recording medium. A computer program stored on a computer-readable recording medium,
When the computer program is executed by a processor,
A step of estimating the solar power generation amount of the solar power generator based on the solar radiation amount predicted using meteorological data, and constructing a probability distribution model for each time reflecting the error between the estimated solar power generation amount and the actually measured solar power generation amount. A step of generating a predetermined number of solar power generation data having the predicted power generation amount of the solar power generator as an average through the probability distribution model for the time, and a predetermined number generated by the generated solar power generation data. The processor performs a method comprising calculating an expected power generation cost for each hour based on the net load, and determining the power generation amount of the non-renewable energy generator based on the calculated expected generation cost for each hour. A computer program that contains instructions to execute.

Images