KR102153722B1 - PV Power Generation Management Apparatus and Method - Google Patents

Abstract

The present invention may provide a PV power generation management device. The PV power generation management device comprises: a PV generation unit producing electric energy from the light of the sun; a PVMS controlling and managing the electric energy produced from the PV generation unit; an inverter unit converting direct current power of the PV generation unit into alternating current power by control of the PVMS and outputting the alternating current power; a PCS receiving the alternating current power through a primary side of the inverter unit by control of the PVMS; an ESS receiving and storing the direct current power outputted from the PCS; a transformer unit converting the direct current power into the alternating current power and receiving the alternating current power through the primary side and a secondary side of the inverter unit; and a KEPCO system purchasing the alternating current power through the transformer unit. In addition, PV power generation management method comprises the steps of: identifying generation information of the PV power generation unit and storage information of the ESS; determining whether it is time for charging now; selling charging power of the ESS to the KEPCO system if it is it not the time for charging now; selling part of power generated by the PV power generation unit and charging the part of the power to the ESS if an amount of PV generation generated by the PV generation unit is greater than a predicted amount of generation reflecting the weather; and charging all the generation power generated by the PV generation unit to the ESS if the amount of PV generation is smaller than the predicted amount of generation.

Description

PV Power Generation Management Apparatus and Method {PV Power Generation Management Apparatus and Method}

The present invention relates to an apparatus and method for managing PV power generation power.

ESS (Energy Storage System) is a connection between a renewable energy generation system represented by solar power and a power storage system, storing renewable energy or surplus power from the power system in a battery capable of charging and discharging, and supplying power to the load when necessary. It is a supply system.

If there is no ESS, serious problems such as sudden power failure may occur in the power system due to unstable power supply that relies on wind or sunlight. Therefore, storage is emerging as a very important field in this environment.

When PV/ESS is installed and designed according to the maximum capacity, the cost is exceeded by selecting the capacity more than necessary, so the PV/ESS operator predicts the annual average generation amount and selects the ESS capacity based on the most frequent generation amount. Therefore, the ESS is charged as it is generated on the day when the amount of power generation is higher than the average, and after the ESS is fully charged, the electricity produced through PV (photovoltaic) is sold at a price without weight.

In this case, since the energy was operated without considering the capacity efficiency, not only the stored energy is small, but also the life of the battery is adversely affected.

The problem to be solved by the present invention is to provide a more efficient PV power generation power management apparatus and method.

The solution means of the present invention is a PV power generation unit for generating electrical energy from sunlight; PVMS for controlling and managing electric energy produced from the PV power generation unit; An inverter unit for converting and outputting the DC power of the PV power generation unit into AC power under the control of the PVMS; PCS that is supplied as DC power through the primary side of the inverter unit under the control of the PVMS; ESS that receives and stores DC power output from the PCS; AC power from DC power through the primary and secondary sides of the inverter unit A transforming unit that is converted into and received; A PV power generation power management apparatus including a KEPCO system for purchasing AC power through the transformer unit may be provided.

In addition, the present invention comprises the steps of identifying generation information of the PV power generation unit and storage information of the ESS; Determining whether or not the current charging time; Selling the charging power of the ESS to a KEPCO system if it is not the current charging time; When the PV generation amount generated by the PV power generation unit is greater than the predicted generation amount reflecting the weather, some of the power generated by the PV power generation unit is sold to the KEPCO system and charged to the ESS; When the PV generation amount is less than the predicted generation amount, a PV generation power management method comprising the step of charging all of the generation power produced by the PV generation unit in the ESS may be provided.

The present invention is an apparatus and method for managing generated power generated from a PV power generation unit, and the PV generation amount and the ESS charge amount can be set as efficiently as possible by predicting the total amount of generation per hour.

In other words, if the ESS can be recharged as much as possible, it can be recharged as much as possible, and the remaining power generation after the maximum storage of ESS can be sold to the KEPCO system to realize profit.

In the present invention, when the ESS is evenly charged, all power generation can be charged to the ESS until the power generation amount exceeds the equal charge amount.

The present invention compares the actual power generation amount of the PV power generation unit and the predicted power generation amount to be charged to the KEPCO system and charged to the ESS.

The present invention provides a first forecast value including weather information of the past, present and tomorrow and the amount of power generation in the past, satellite image data for grasping the amount of clouds in real time, a second forecast value including current weather information and power generation amount in the past, The power generation amount may be predicted by reflecting at least one of the third predicted values including a value calculated by including the power generation amount of the power plant in the first predicted value and/or the second predicted value.

delete

delete

1 is a configuration diagram of a PV power generation power management apparatus according to the present invention.
2 is a flow chart showing the management method of the present invention.

Hereinafter, specific details for carrying out the present invention will be described in detail based on the accompanying exemplary drawings.

1 is a configuration diagram of a PV power generation power management apparatus according to the present invention.

Referring to FIG. 1, a PV power generation unit 110 for producing electric energy from sunlight, a photovoltaic management system (PVMS) 120 for controlling and managing electric energy produced from the PV power generation unit 110, and the PVMS The inverter unit 130 converts and outputs the DC power (DC) of the PV power generation unit 110 into AC power (AC) under the control of 120, and the inverter unit 130 under the control of the PVMS 120. A PCS (Power Control System) 140 receiving DC power through the primary inverter unit of ), ESS 150 receiving and storing DC power output from the PCS 140, 1 of the inverter unit 130 It may include a transformer unit 160 that is transmitted from DC power to AC power through a primary inverter unit and a secondary inverter unit, and a KEPCO system 170 that buys AC power through the transformer unit 160.

The PVMS 120 receives power generation information of the PV power generation unit 110 and the storage information of the ESS 150 and controls the inverter unit 130 and the PCS 140 to control the charging amount of the ESS 150. In addition, it is possible to control the amount of power sold to the KEPCO system 170.

The inverter unit 130 may include a primary inverter unit and a secondary inverter unit, and the primary inverter unit may not convert the supplied DC power into AC power, and the secondary inverter unit may convert DC power to AC power. .

The PV power generation unit 110 may generate DC power by converting solar energy into electrical energy in a state consisting of a plurality of solar cells.

The PVMS 120 acquires power generation information (power generation, etc.) of the PV power generation unit 110, synchronizes the charging time of the ESS 150, and calculates the amount of current to be sold to the KEPCO system 170.

The storage information of the ESS 150 is information on the type and charging time of the battery whose capacity efficiency changes according to the amount of power during charging/discharging. More specifically, the storage information of the ESS 150 indicates the load capacity, and is information on how much current is stored or the current amount of charge.

In the present invention, the PV/ESS provider reflects the power generation information and storage information to be considered in the situation of producing and selling power, and following the predicted power generation information to be charged for the ESS 150 and sold to the KEPCO system 170. It is possible to propose an optimal PVMS (120).

[PV power generation and ESS installation capacity calculation]

In the case of connecting and installing the PV power generation unit 110 and the ESS 150, it may be important to calculate and install the ESS 150 of an appropriate capacity compared to the PV power generation unit 110 in order to increase profits. have.

In an embodiment of the present invention, it may be assumed that the PV power generation unit 110 has a power generation capacity of 100 kW.

When the PV power generation unit 110 and ESS 150 are installed, designing according to the maximum capacity will select the capacity more than necessary and the cost will be exceeded, so the annual average generation amount is estimated and the ESS capacity is selected based on the most frequent generation amount. can do.

In an embodiment of the present invention, the storage capacity of the ESS 150 may be assumed to be 300kW.

The generation power management method of the present invention can be made as follows.

Referring to FIG. 2, the generated power management method of the present invention may manage generated power by PVMS 120.

That is, the management method of the present invention may first include the step (S1) of determining power generation information of the PV power generation unit 110 and storage information of the ESS 150. Here, the PV power generation unit 110 may generate a difference between the actual power generation amount and the reference power generation amount generated according to the installed region, temperature, and climate. That is, since the power generation amount pattern generated by the PV power generation unit 110 is different every day, power generation information and storage information are transmitted to the PVMS 120 in real time, and the PVMS 120 recognizes the transmitted information.

Subsequently, the management method of the present invention may include a step (S2) of determining whether the current charging time is or not. Here, the charging time is calculated based on the generation information generated from the PV power generation unit 110 between 10 am and 16:00, for example, as described above, and the discharge time, that is, the ESS 150 The time to sell the power stored in) can be made from 16:00 to 10:00 a.m. the next day.

Subsequently, if it is not the current charging time, a step (S3) of selling the charging power of the ESS 150 to the KEPCO system 170 may be included. That is, the power stored in the ESS 150 is sold to the KEPCO system 170 at the time when the PV power generation unit 110 does not generate electricity, and supplies power to the necessary loads, and the PV/ESS business operator is provided to the KEPCO system 170. You can generate revenue through the electricity sold.

Subsequently, the management method of the present invention may include a step S4 of analyzing and optimizing the power generation information of the PV power generation unit 110 and the charging information of the ESS 150 if the current charging time is the current charging time.

The step of analyzing and optimizing the power generation information of the PV power generation unit 110 and the charging information of the ESS 150 (S4) may be performed through the calculation unit 122 of the PVMS 120.

Here, some of the power generated by the PV power generation unit 110 through the PVMS 120 may be sold to the KEPCO system 170 and charged to the ESS 150.

The amount of power sold may be calculated as a difference between the actual amount of power generated by the PV power generation unit 110 and the amount of power charged in the ESS 150.

Since the amount of PV power generation can vary depending on the weather, temperature, etc., it is important to predict such power generation and to establish and operate a power sales and charging plan by comparing the predicted power generation and actual power generation.

Here, in Figure 2, for convenience, it is shown as if it is sold only to the KEPCO system 170, but if the PV generation amount is greater than the predicted generation amount, some of the power generated by the PV power generation unit 110 is sold to the KEPCO system 170, and It may include a step (S5) of charging the ESS (150).

And, if the PV generation amount is less than the predicted generation amount, the step of charging all of the generated power generated by the PV power generation unit 110 in the ESS 150 (S6) may be included.

The PV power generation unit 110 may have an actual power generation amount greater than a predicted power generation amount depending on temperature and weather. At this time, the PVMS 120 receiving the power generation information of the PV power generation unit 110, while viewing the power generation information from 10:00 to 16:00, rapidly charge the ESS 150 according to the power generation amount of the PV power generation unit 110. By adjusting the amount of charge so that it is slowly charged for each period of time without proceeding, not only the usable capacity of the ESS 150 can be increased, but the lifespan can be further improved.

Here, the method of predicting the amount of power generated by the PV power generation unit 110 may be predicted through the determination unit 124 of the PVMS 120. Such a power generation prediction method is a deep learning algorithm using artificial intelligence, and may be a time-series prediction method computed using a recurrent neural network (RNN) and a long-short-term memory (LSTM).

The generation amount prediction method reflects at least one or more of the following three prediction values (first, second, and third prediction values), and through such reflection, the accuracy score of each prediction method is managed and applied, thereby increasing the prediction accuracy.

As a method of predicting the amount of power generation, the first predicted value may reflect at least one of the past, present, and tomorrow's weather and past power generation amount based on the weather forecast as a factor.

The first predicted value is historical numeric data and may include past and present temperature, humidity, solar radiation, wind direction, air volume, and rainfall.

Numerical data in the past can be reflected by setting a specific period of the past period.

The first predicted value may include tomorrow's temperature, humidity, solar radiation, wind direction, air volume, and rainfall through weather forecasting.

The first prediction value may input data (numeric data) on weather factors of the past, present, and tomorrow and the amount of generation in the past for a specific period, and the amount of generation may be predicted using RNN and LSTM techniques.

The first predicted value may be calculated in at least one of an hour, a half day, and a day.

The second prediction value may reflect a change (image) in the amount of clouds generated over the power plant through real-time satellite photographs.

The second prediction value is obtained by inputting satellite image data, current weather data (wind direction, air volume, temperature, humidity), and past power generation, and using a convolutional neural network (CNN) and a recurrent neural network (RNN) technique. It is to predict future generation.

Real-time satellite photos can be calculated for an hour or half a day, using, for example, Cheonlian real-time satellite image data.

In other words, the second predicted value supplements the power generation forecast by predicting the amount of clouds above the power plant.

The third forecast can reflect the estimated generation of other power plants in the region, and can be calculated for an hour or a half.

The third predicted value reflects the power generation amount of other power plants in the vicinity to the first and/or second forecast values, and calculates the predicted power generation amount using RNN and LSTM techniques, thereby reflecting the internal issues of the relevant power plant. It can be reflected in the prediction.

An internal issue of the plant could be, for example, an inverter failure. In this case, if there is a difference of more than the average value compared to the predicted power generation amount of other nearby power plants, and compared with the estimated power generation amount of the corresponding power plant, identify the device failure of the power plant and reflect this as a third predictive method factor to predict the power generation amount. You can, and by doing this, you can predict more accurate power generation.

When setting the amount to be charged in the ESS 150, at least one or more of the following weights may be reflected.

The first weight may reflect weather information, the second weight may reflect temperature information, and the third weight may reflect air quality information.

Weather information, temperature information, and air quality information may reflect the analysis of weather forecasts provided by the Meteorological Agency or other weather forecast publishing organizations (kweather or openweathermap sites).

Since weather information can affect the current, the first weight is weather information, and the first weight is weather information, such as past and present humidity, solar radiation, wind direction, air volume, rainfall, and tomorrow's humidity, solar radiation, wind direction, wind volume, Can reflect rainfall.

The second weight reflects temperature information. If the temperature of the solar panel of the PV power generation unit 110 is too high or too low, the voltage may be affected and power generation may not be performed well, and thus the temperature may be reflected.

The temperature information reflected as the second weight may reflect the temperature of the past, present, and tomorrow in the weather forecast.

The third weight reflects air quality, and the amount of power generation may vary depending on the fine dust condition is good, normal, bad, or very bad. Therefore, the air quality state can be reflected as the third weight.

When the actual power generation amount of the PV power generation unit 110 is higher than the predicted power generation amount, the PVMS 120 controls the charging to the ESS 150 and saves the power sold to the KEPCO system 170 as a sales time and an amount of power. By arbitrarily adjusting and selling electricity after 16:00, you can increase sales profit.

Meanwhile, the PV power generation unit 110 may generate a difference between the actual power generation amount and the predicted power generation amount according to changes in the location, season, and climate of the installed area, as well as temperature and weather. That is, since the actual power generation pattern of the PV power generation unit 110 is different every day, the present invention analyzes power generation information through the PVMS 120, and calculates the charge amount of the ESS 150 and the sales amount of the KEPCO system 170. It can be adjusted arbitrarily to improve the life of the battery and effectively generate profits from power sales.

In the present invention, when the amount of power generation of the PV power generation unit 110 is greater than the predicted value, 100% charge is not allocated to the ESS 150 and can be sold through the KEPCO system 170.

The amount of power generated by the PV power generation unit 110 to be charged to the ESS 150 is allocated in a predetermined ratio (%) to the amount of PV power generation by power control by the PCS 120, or a constant capacity regardless of the amount of PV power generation. You can set the amount of charge by assigning it with.

At this time, the amount of power generation in the PV power generation unit 110 may be predicted by reflecting the weather.

That is, the weather of the day can be predicted at intervals of 30 minutes or 1 hour by setting each weather condition such as sunny, cloudy, rain, or snow.

The predicted PV generation amount can be reflected in the determination of the amount of electricity sold and the amount to be charged in the ESS 150.

For example, if the actual PV power generation is less than the predicted PV power generation, the entire amount is charged to the ESS 150, and if the actual PV power generation amount is greater than the predicted PV power generation amount, some of them are sold, and a part is charged to the ESS 150. can do.

In addition, when the ESS 150 is charged, the total PV revenue according to Equation 1 below may be predicted, and the charging amount and the charging time may be considered.

Here, Pr is total revenue, Ppv is PV sales, PB is ESS discharge, SMP is System Marginal Price, and REC is Renewable Energy Certificate.

SMP is the system marginal price, which is the unit price of electricity purchased from solar power plants by Korea Electric Power Corporation. In other words, the amount purely sold to Korea Electric Power Corporation according to the amount of electricity generated by the solar power plant can be said to be SMP.

REC is a Renewable Energy Certificate, which is issued by the Energy Management Corporation according to the amount of electricity produced by the solar power plant, and the certificate is sold only to power generation companies (18 companies) that are obligated to supply renewable energy. Is to do.

Equation 1 can be simply expressed as follows.

Total profit = PV sales profit + ESS sales profit

Here, PV sales profit = PV sales amount × (SMP + REC × 1.0)

Profit from ESS sales = ESS discharge amount (discharged in the rest of the time excluding 6 hours of peak time) × (SMP + REC × 5.0)

ESS discharge amount = ESS charge amount × power conversion efficiency (approximately 80% during charging and discharging)

Total power generation = PV power supply + ESS charge

PV generation power management method according to the present invention can be summarized as follows.

[PV generation and ESS storage and discharge optimal schedule]

[Prediction of power generation through PVMS]

The first forecast including weather information of the past, present and tomorrow and the amount of power generation in the past, satellite photo data to determine the amount of clouds in real time, the current weather information and the second forecast including the amount of power generation in the past, the amount of power generation of power plants in the region The amount of power generation may be predicted by reflecting at least one or more of third predicted values including values calculated by including in the first predicted value and/or the second predicted value.

[Calculation of PV power generation and ESS storage]

① Incentives (REC weights) can be provided only for electricity that is charged from 10 o'clock to 16 o'clock in the PV power generation unit and then discharged through the ESS 150.

It can alleviate the shortage of power grid access capacity caused by maximum power generation during the daytime and respond to the high power demand that occurs at night in spring, autumn and winter.

② When the weather is predicted to be sunny all day (maximum PV power generation)

Since the daily PV generation amount will be more than the ESS 150's maximum storage capacity (300kW), the ESS 150 can be charged to the maximum capacity between 10 o'clock and 16:00, and evenly charged as slowly as possible. When the ESS 150 is slowly discharged and charged, the state of health (SOH) can be maintained for the longest period.

When the ESS 150 is charged evenly and slowly, for example, when the total charging capacity of the ESS 150 is 300kW, the unit charging amount is 30kW, and it can be evenly charged for 10 hours. On the other hand, it is possible to equally charge for 6 hours at 50kW by varying the unit charge amount, but in this case, the facility capacity may increase proportionally as the unit charge amount increases, and thus the cost may increase.

Therefore, when charging the ESS 150, it may be desirable to lower the unit charge amount to evenly charge it slowly over several hours.

PV generation, excluding ESS charging, can be sold at a price without weight.

③ When the weather is predicted to be cloudy throughout the day (PV generation is the lowest)

Daily PV generation may not reach the maximum ESS storage. Therefore, it is possible to charge all of the power generation amount between 10 o'clock and 16 o'clock in the ESS 150.

④ If you predict that it will be cloudy in the morning and sunny in the afternoon

If the predicted value of the sum of the morning power generation and the afternoon power generation is more than the ESS maximum charging amount (300kW), the ESS 150 is equally charged to the maximum capacity (300kW), and the remaining generation excluding the ESS charging amount can be sold at a price without weight. .

If the predicted value of the sum of the morning power generation and the afternoon power generation does not reach the ESS maximum charging amount (300kW), both morning and afternoon power generation may be charged to the ESS 150.

⑤ When it is predicted to be sunny in the morning and cloudy in the afternoon

If the predicted value of the sum of the morning power generation and the afternoon power generation is more than the ESS maximum charging amount (300kW), the ESS (150) is charged equally with the maximum capacity (300kW), and the remaining generation excluding the ESS (150) charging amount is sold at a price without weight. can do.

If the predicted value of the sum of the morning power generation and the afternoon power generation does not reach the ESS maximum charging amount (300kW), both morning and afternoon power generation may be charged to the ESS 150.

⑥ If the predicted value of the sum of the morning power generation and the afternoon power generation does not reach the ESS maximum charging amount (300kW), both morning and afternoon power generation can be charged to the ESS 150.

[Charging plan according to predicted generation amount and actual PV generation amount]

For example, the estimated power generation amount from 13:00 to 14:00 was calculated as 80 kWh, and according to the actual PV generation amount and ESS (150) storage calculation method, the ESS charge amount from 13:00 to 14:00 was determined as 50 kWh and the KEPCO grid sales volume was determined as 30 kWh. Assuming you can plan and execute something like this:

① When PV generation amount is larger than predicted generation amount

If the predicted power generation does not reach the ESS maximum charging amount, the additional PV generation amount exceeding the predicted power generation amount is used until the ESS is fully charged, and even after the ESS is fully charged, the additional PV generation amount can be sold to the KEPCO system.

In addition, if the predicted generation amount is more than the ESS maximum charging amount, all of the additional PV generation amount can be sold to the KEPCO system.

An example of this situation is as follows.

If the actual PV power generation amount is 90 kWh and the predicted power generation amount is 80 kWh from 13:00 to 14:00, the ESS charge amount is maintained at 50 kWh, and an additional PV generation amount exceeding the predicted power generation amount, 10 kWh is added to the KEPCO system to sell, a total of 40 kWh. Can be sold to the KEPCO system.

② When PV generation amount is less than predicted generation amount

If the predicted generation amount is more than the ESS maximum charging amount, it can be used to reduce the sales plan of KEPCO system and fully charge the ESS.

In addition, if the predicted generation amount is less than the ESS maximum charging amount, the entire PV generation amount may be charged to the ESS 150.

An example of this situation is as follows.

From 13:00 to 14:00, if the actual PV power generation amount is 60 kWh and the estimated power generation amount is 80 kWh, only 10 kWh is sold to the KEPCO system by subtracting 20 kWh from the planned KEPCO sales volume of 30 kWh, and the ESS charge is maintained at 50 kWh.

On the other hand, if the actual PV power generation amount is 40 kWh and the predicted power generation amount is 80 kWh from 13:00 to 14:00, the ESS 150 is charged with all 40 kWh of actual PV power generation without selling 30 kWh, the planned KEPCO grid sales volume.

③ When PV generation amount is the same as predicted generation amount

The planned ESS charging amount of 50 kWh and KEPCO system sales amount of 30 kWh can be carried out.

110: PV power generation unit 120: PVMS
122: operation unit 124: determination unit
130: inverter unit 140: PCS
150: ESS 160: transformer
170: KEPCO system

Claims (12)

delete delete delete delete delete Determining the power generation information of the PV power generation unit and the storage information of the ESS;
Determining whether or not the current charging time;
Selling the charging power of the ESS to a KEPCO system if it is not the current charging time;
If the PV generation amount generated by the PV power generation unit is greater than the predicted generation amount, selling some of the power generated by the PV power generation unit to the KEPCO system and charging the ESS;
If the PV generation amount is less than the predicted generation amount, charging all of the generated power generated by the PV power generation unit in the ESS;
Including,
When the PV generation amount is greater than the predicted generation amount,
If the predicted generation amount is less than the ESS maximum charging amount, the additional PV generation amount exceeding the predicted generation amount is used until the ESS maximum charging, and the additional PV generation amount remaining after the ESS is fully charged is sold to the KEPCO system.
If the predicted generation amount is more than the ESS maximum charging amount, all of the additional PV generation is sold to the KEPCO system,
The storage information of the ESS is current charging information, and the storage information and the generation information and predicted generation amount of the PV power generation unit are identified by the PVMS that controls and manages the electric energy produced by the PV power generation unit,
When predicting the amount of power generation in the PVMS,
The first forecast including weather information of the past, present, and tomorrow and the amount of power generation in the past, satellite image data to determine the amount of clouds in real time, and the second forecast value including the amount of power generation in the past, the first forecast value of the power generation amount of power plants in the region And/or reflecting at least one of a third predicted value including a value calculated by including in the second predicted value.
The method of claim 6,
The third predicted value,
Reflects the issues inside the power plant and reflects it in the power generation forecast
If a power plant abnormality occurs as an internal issue of the relevant power plant, determine the predicted power generation amount of other nearby power plants, and if there is a difference of more than the average value compared to the predicted power generation amount of the corresponding power plant, identify and reflect the internal abnormality of the power plant PV generation power management method, characterized in that.
The method of claim 6,
When the PV generation amount is less than the predicted generation amount,
If the predicted generation amount is more than the ESS maximum charging amount, it is used to reduce the KEPCO system sales plan and fully charge the ESS,
PV generation power management method, characterized in that when the predicted generation amount does not reach the maximum charge amount of ESS, the entire PV generation amount is charged to the ESS.
The method of claim 6,
When charging the ESS, PV power generation power management method, characterized in that the charging amount is adjusted to be charged at a predetermined time period without rapidly proceeding according to the power generation amount of the PV power generation unit in order to increase the usable capacity of the ESS and extend the lifespan. delete delete delete

Images