Self-Demand Response Platform Virtual Power Plant System

Abstract

The embodiment provides a virtual power plant system based on a self-demand response infrastructure. [Solution] The system according to the embodiment relates to a virtual power plant system that analyzes at least one of the following in real time and provides the analysis results, so that individual consumers can perform self-demand response (Self-DR) using a user terminal.

Description

Technical Field

[0001] The present invention relates to a virtual power plant system that optimizes AI and an energy storage system (ESS) in a self-demand response infrastructure to efficiently perform wholesale and retail power trading. This system includes AI-based power demand prediction, optimization of charging and discharging of an energy storage device using reinforcement learning, P2P power trading utilizing blockchain smart contracts, and an optimization engine for wholesale and retail power trading.

[0002] This patent is a research conducted with the support of the regional demand customized research and development project of Jeollanam-do and Jeonnam Technopark in 2024.

Background Art

[0003] Recently, along with the continuous increase in power consumption, the expansion of renewable energy resources such as solar and wind power has raised the main issue of ensuring the safety and efficiency of the power grid simultaneously. Renewable energy has the characteristic that its output changes rapidly depending on weather conditions, and flexible power grid operation based on real-time prediction of energy production and consumption is essential.

[0004] Existing power systems are based on a centralized control structure and cannot fully reflect the power usage patterns and demand changes of individual consumers. As a result, demand response (DR) technology, which can actively adjust power demand, has been introduced. However, this technology operates based on fixed plans or limited data and has the limitation that it is difficult to achieve autonomous responses optimized for individual users.

[0005] Meanwhile, the enforcement of the Special Act on the Activation of Distributed Energy has established an institutional framework that allows individual consumers to trade the electricity they produce. This enables direct (P2P) electricity trading between consumers and participation in the retail electricity market. In response to these institutional changes, there is a growing need for an autonomous and optimized electricity trading system that integrates artificial intelligence (AI), energy storage devices, and blockchain technology. [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] An embodiment of the present invention provides a virtual power plant system that optimizes AI and an Energy Storage System (ESS) on a Self-Demand Response platform to efficiently carry out wholesale and retail electricity trading.

[0007] The technical challenges addressed in these embodiments are not limited to those mentioned above. Furthermore, other technical challenges not mentioned can be considered by those with ordinary skill in the art from the various embodiments described below. [Means for solving the problem]

[0008] To solve the aforementioned problems, a system according to one embodiment of the present invention provides a virtual power plant system that enables individual consumers to perform Self-DR using a user terminal by analyzing at least one of the following in real time: power demand data, power generation data, power price data, and distributed power information, and providing the analysis results, wherein the system includes a transaction management unit that calculates power demand and supply based on the power price data, the power generation data received from the power plant, and weather data received from an external server, predicts an appropriate power price based on the calculation results, and generates a smart contract; and the power generation data is received from a safe power plant. The system may include: a power management unit that generates distributed power information based on power amount data, environmental power generation amount data received from environmental power plants, power status data received from a power storage device, and at least one of the power price data, power generation amount data, and weather data, and generates charge / discharge schedule information based on the distributed power information and transmits it to the power storage device; and a scheduling unit that analyzes time-of-day power consumption patterns and preferences based on power demand data and power consumption data received from consumer terminals, generates demand forecast data, and generates a consumer-ordered schedule based on the demand forecast data.

[0009] The transaction management unit can receive the electricity price data from the electricity exchange and transmit the smart contract to the electricity exchange. An emergency response unit that detects in real time whether or not an emergency condition is occurring; and a notification unit that transmits an emergency condition notification to the administrator and the consumer terminal when the emergency condition occurs, wherein the emergency condition includes at least one of the following: power plant failure, sudden increase in power demand, and worsening weather.

[0010] The scheduling unit generates consumer-customized schedule information based on the electricity demand data and the electricity price data, wherein the electricity demand data includes peak-time demand data and off-peak-time demand data, the electricity price data includes peak-time price data and off-peak-time price data, and the consumer-customized schedule information may include schedules for cost reduction during peak hours.

[0011] The scheduling unit can generate the consumer-customized schedule information using the power trading optimization algorithm of the artificial intelligence unit. [Brief explanation of the drawing]

[0012] [Figure 1] This diagram schematically shows a network environment in which virtual power plant systems according to one embodiment of the present invention are connected. [Figure 2] Figure 1 is a block diagram illustrating the configuration of the virtual power plant system. [Figure 3] Figure 2 is a block diagram illustrating the configuration of the artificial intelligence unit. [Figure 4] Figure 3 is a diagram illustrating a multilayered neural network. [Figure 5] Figure 1 is a block diagram illustrating the structure of the power storage device. [Figure 6] This flowchart illustrates a method for generating and providing schedules in a scheduling unit according to one embodiment of the present invention. [Figure 7] This is a flowchart illustrating a method for generating a charge / discharge schedule according to one embodiment of the present invention. [Modes for carrying out the invention]

[0013] The following embodiments are combinations of the components and features of the embodiments in a predetermined form. Each component or feature can be considered optional unless otherwise explicitly mentioned. Each component or feature can be implemented in a form that is not combined with other components or features. Furthermore, some components and / or features can be combined to form diverse embodiments. The order of operations described in the diverse embodiments may be changed. Some components or features of one embodiment may be included in other embodiments, or can be replaced with corresponding components or features of other embodiments.

[0014] In the descriptions of the drawings, we did not include any procedures or steps that could obscure the essence of the various embodiments, nor did we include any procedures or steps that could be understood by someone with ordinary skill in the art.

[0015] Throughout the specification, when a part “comprising or including” a component, this does not exclude other components unless specifically contradicted, but rather means that other components may be further included. Furthermore, terms such as “part,” “device,” and “module” as used in the specification mean a unit that processes at least one function or operation, which can be embodied in hardware, software, or a combination of hardware and software. Also, “a or an,” “one,” “the,” and similar related terms can be used in a singular and plural sense in the context of describing various embodiments (particularly in the context of the following claims), unless otherwise indicated herein or clearly refuted by the context.

[0016] Hereinafter, embodiments of various examples will be described in detail with reference to the accompanying drawings. The detailed description disclosed below, together with the accompanying drawings, is intended to describe exemplary embodiments of various examples and is not intended to represent only one embodiment.

[0017] Furthermore, the specific terminology used in the various embodiments is provided to aid in understanding those embodiments, and the use of such terminology can be modified to other forms as long as it does not deviate from the technical concept of the various embodiments.

[0018] Figure 1 is a schematic diagram showing a network environment in which a virtual power plant (VPP) system 100 according to one embodiment of the present invention is connected.

[0019] Referring to Figure 1, the virtual power plant system 100 can exchange signals or data with the power plant 200, the power exchange 300, the consumer terminal 400, and the power storage device 500 via a network.

[0020] The virtual power plant system 100 can be configured to allow individual consumers to perform Self-Demand Response (Self-DR) without the need for a power exchange. This system can analyze power demand, price, and the charging / discharging status of power storage devices in real time through coordination with the artificial intelligence unit 120 and the power management unit 140, and based on this, generate and execute an optimal power trading strategy.

[0021] The virtual power plant system 100 can collect electricity usage data from consumer terminals 400 and electricity price data from the electricity exchange 300. The virtual power plant system 100 can predict peak pricing periods using an artificial intelligence model and perform self-demand response scheduling based on this.

[0022] The virtual power plant system 100 can collect weather data and use an artificial intelligence model to predict the power output of environmentally friendly power plants 220. The virtual power plant system 100 can compare power production forecast data with power demand forecast data to provide power storage (charge / discharge) scheduling and generate power price forecast data.

[0023] The power plant 200 may include a safety-type power plant 210 and an environmentally friendly power plant 220. The safety-type power plant 210 may mean a power plant that can stably produce electricity regardless of the weather. For example, a thermal power plant or a nuclear power plant may fall into this category. The safety-type power plant 210 may include a fuel supply device, a generator, and a control unit. The fuel supply device can stably supply fossil fuels or nuclear fuels. The generator can convert thermal energy generated from the combustion of fuel or nuclear reactions into electrical energy. The control unit can monitor the status of the power plant in real time and adjust the output power.

[0024] The environmentally friendly power plant 220 may include environmentally friendly power source infrastructure power plants that produce electricity in accordance with climate change. For example, solar power plants, wind power plants, and hydroelectric power plants may fall into this category. The environmentally friendly power plant 220 can convert the kinetic energy of sunlight, wind, or water into electricity. The environmentally friendly power plant 220 can store the electricity it produces and supply it when needed. The environmentally friendly power plant 220 can send the stored electricity to the electricity exchange 300 or the virtual power plant system 100.

[0025] The power exchange 300 may include a variety of power trading markets, such as a capacity market where power is traded on a monthly or yearly basis, and a power market where power is traded on an hourly or work basis. The power exchange 300 can manage power trading between power plants 200 and consumers. The power exchange 300 may include a trading management unit 310, a trading database 320, and an exchange communication unit 330. The trading management unit 310 can record power received from safe power plants 210 and environmentally friendly power plants 220 in real time, provide it for trading, and distribute it to consumer terminals 400. The trading database 320 can analyze power consumption, production, and price information and process power trading data in real time. The exchange communication unit 330 can exchange data with power plants 200, the virtual power plant system 100, and consumer terminals 400.

[0026] The consumer terminal 400 may be a device connected to the virtual power plant system 100 that monitors the consumer's power usage status and receives and executes a consumer-ordered schedule. The consumer terminal 400 may include a processor, memory, communication module, and display.

[0027] The processor of the consumer terminal 400 can process data received from an external source and execute instructions. The processor of the consumer terminal 400 can analyze a customized schedule sent from the virtual power plant system 100 and generate an execution plan suitable for the consumer. The processor of the consumer terminal 400 can analyze power consumption data in real time and generate warning messages when abnormal conditions occur.

[0028] The memory of the consumer terminal 400 can store power usage data, schedule information, user settings, etc. The memory of the consumer terminal 400 may include volatile and non-volatile memory. The memory of the consumer terminal 400 can store past power consumption records and schedule data, learn recurring patterns, or be used for user-specific recommendations.

[0029] The communication module of the consumer terminal 400 can support network connectivity that enables it to send and receive data with the virtual power plant system 100. The communication module of the consumer terminal 400 can connect in real time with the power management unit 140 and scheduling unit 150 (described later) via Wi-Fi, LTE, or other wireless communication technologies. The communication module of the consumer terminal 400 can also receive notifications transmitted from the notification unit 170 (described later) and send user input data. The display of the consumer terminal 400 provides a user interface and can visually display schedule information, power consumption status, estimated costs, etc. The user can input commands such as setting schedule priorities and activating energy saving mode via the touchscreen display of the consumer terminal 400. The display of the consumer terminal 400 can display warning and suggestion messages received from the notification unit 170 (described later). The touch display can be an LCD (liquid crystal display) panel, an OLED (organic light emitting diode) panel, etc. The touch display may also include a touch screen panel (TSP) that includes touch electrodes for sensing touches to the display panel. In one embodiment, the touch display may be of the in-cell type, where the display panel and the touch panel are integrated into a single unit. However, this is illustrative, and embodiments of the present invention are not limited thereto. The touch display can electrically recognize contact from a finger, pen, etc., generate an electrical signal, and transmit it to a processor.

[0030] The power storage device 500, under the control of the virtual power plant system 100, can store power, supply it when needed, monitor the battery status, and manage an optimized energy flow. The power storage device 500 can check temperature, voltage, SOC (State of Charge), SOH (State of Health), etc. in real time to detect emergency conditions and can send and receive data with the virtual power plant system 100 and external systems. The configuration and operation of the power storage device 500 will be described in detail later.

[0031] Figure 2 is a block diagram illustrating the configuration of the virtual power plant system 100 in Figure 1. Figure 3 is a block diagram illustrating the configuration of the artificial intelligence unit 120 in Figure 2. Figure 4 is an illustrative diagram showing the multilayer neural network 121 in Figure 3.

[0032] In one embodiment, the virtual power plant system 100 can be designed to perform at least one of the following functions: power demand forecasting, charging and discharging scheduling of power storage devices 500, P2P power trading, and establishing wholesale and retail market trading strategies.

[0033] Referring to Figure 2, the data acquisition unit 110 of the virtual power plant system 100 can collect power consumption data, power generation data, weather data, and power price data from external devices (e.g., power plant 200, power exchange 300, etc.).

[0034] The power consumption data may include information on the consumer's real-time power consumption and time-of-day consumption patterns, obtained through the consumer terminal 400 and smart meter.

[0035] The power generation data may include power data received at power plant 200. The power generation data may also include safe power data received from safe power plant 210 and environmentally friendly power data received from environmentally friendly power plant 220.

[0036] Weather data can be collected through weather stations (not shown) or the Japan Meteorological Agency API (not shown), and may include data such as weather forecasts, wind direction, and solar radiation.

[0037] Electricity price data can be collected from electricity exchanges 300 and may include time-of-day electricity price data.

[0038] The artificial intelligence unit 120 can learn from multiple data sets, including power consumption data, power generation data, weather data, and electricity price data, through machine learning and deep learning algorithms, find patterns, and generate a model.

[0039] Referring to Figure 3, the artificial intelligence unit 120 can include a multilayer neural network 121, a learning engine 122, and a memory 123.

[0040] The learning engine 122 can pre-train the multilayer neural network 121 using multiple training data sets. A multilayer neural network is a predictive model embodied in software or hardware that mimics the computational power of a biological system by utilizing a large number of artificial neurons (or nodes).

[0041] The learning engine 122 can guide and train the multilayer neural network 121 using training data that includes input data and ground truth data, so that the multilayer neural network 121 can generate accurate predicted values ​​based on multiple training data.

[0042] In this context, "instructional learning" refers to learning that uses data containing input values ​​and corresponding output values ​​as training data to find the output value for a given input value, and it means learning that is done with the correct answer already known. The set of input and output values ​​given in instructional learning is called training data.

[0043] Memory 123 can store a variety of data used by at least one component of the artificial intelligence unit 120. This data may include, for example, software and input or output data for associated instructions. Memory 123 may include volatile or non-volatile memory.

[0044] The demand forecasting unit 125 can be implemented as a component of the artificial intelligence unit 120. Furthermore, the demand forecasting unit 125 can analyze individual consumer electricity consumption data through LSTM (Long Short-Term Memory) and Transformer-based AI models to forecast short-term and long-term demand. The demand forecasting unit 125 can generate consumer-specific demand forecast information by utilizing weather data, consumer electricity usage history, and time-of-day patterns as input values. This demand forecast information is linked with the scheduling unit 150 and the power management unit 140 (described later) and provided as basic data for establishing a self-demand response strategy.

[0045] Referring to Figure 4, the multilayer neural network 121 can include an input layer, one or more hidden layers, and an output layer.

[0046] In one embodiment, the multilayer neural network 121 may include an input layer that receives input values ​​and has nodes corresponding to the number of components in a first feature vector, a first hidden layer that multiplies each output value of the input layer by a weight and outputs a bias, a second hidden layer that multiplies each output value of the first hidden layer by a weight and outputs a bias, and an output layer that multiplies each output value of the second hidden layer by a weight and outputs the result using an activation function. Although only two hidden layers are shown in Figure 4, one or more hidden layers may include more than one hidden layer in addition to the first and second hidden layers.

[0047] For example, the activation function may be the Softmax function, but the embodiments of this invention are not limited to this, and the activation function may be a variety of other functions such as the ReLU function. The weights and biases can be continuously updated through instructional learning.

[0048] Specifically, the output vector can be input to a loss function layer connected to the output layer. The loss function layer can output a loss value using a loss function that compares the output vector with the correct vector for each training data. The parameters of the multilayer neural network 121 can be trained to minimize the loss value.

[0049] Referring again to Figure 2, the virtual power plant system 100 can include a transaction management unit 130. The transaction management unit 130 can include a power trading unit 135 and a wholesale / retail power trading optimization engine 137.

[0050] The power trading unit 135 can be implemented on a smart contract platform and can automatically execute P2P (peer-to-peer) power trading, allowing consumers to buy and sell electricity directly without going through the power exchange 300. Based on electricity price data and supply / demand data, the power trading unit 135 can calculate the optimal trading timing and price and automatically execute buy or sell orders through smart contracts. For example, if a sharp rise in electricity prices is predicted, the system can be set to automatically sell electricity, and during times when prices are cheap, it can execute storage or purchase strategies.

[0051] The wholesale and retail electricity trading optimization engine 137 can perform the role of determining the optimal price when consumers purchase electricity in the retail electricity market or when electricity generated or stored in the wholesale market is supplied. This engine can establish individualized consumer-tailored strategies by considering consumer preferences (e.g., carbon reduction, cost savings), price sensitivity, the status of the electricity storage device 500, and the amount of electricity that can be supplied. In the retail market, it can propose an automated purchasing strategy based on user settings, and in the wholesale market, it can implement an optimized sales strategy utilizing distributed energy sources (DERs) and electricity storage devices 500.

[0052] The trading management unit 130 can analyze electricity price data received from the electricity exchange 300, power generation data received from the safety-type power plants 210 and environmentally friendly power plants 220 of the power plants 200, and integrate weather data provided by an external server to calculate supply and demand. Based on this data, the trading management unit 130 can predict the appropriate electricity price, generate a smart contract, and transmit it to the electricity exchange 300.

[0053] In one embodiment, the transaction management unit 130 can receive time-of-day electricity price data from the electricity exchange 300 and collect safe power generation data from safe power plants 210 and environmentally friendly power generation data from environmentally friendly power plants 220 in real time.

[0054] The transaction management unit 130 then analyzes weather data (solar radiation, wind speed, etc.) to predict the expected power generation of the environmentally friendly power plant 220, and can calculate the time-of-day supply-demand imbalance based on the demand data provided by the scheduling unit 150.

[0055] The trading management unit 130 can adjust electricity prices in a way that minimizes the supply-demand difference. The trading management unit 130 can generate smart contracts to send to the electricity exchange 300. For example, it can set a lower price during a surplus of electricity to induce demand, and set a higher price during a shortage to induce production.

[0056] A smart contract is a computerized agreement based on the code principle that "code is law," where the elements necessary for a contract are automatically executed through code. It is a technology that records the terms of the agreement between contracting parties as program code, and when the contract conditions are met, the program code is executed to automatically carry out the contract terms. By using a programming language and pre-coding the contract period, amount, conditions, etc., any type of contract, such as real estate transactions, used car transactions, and trade transactions, can be automated.

[0057] In this specification, a smart contract may include a smart buy contract for purchasing electricity from the electricity exchange 300, and a smart sell contract for selling electricity to the electricity exchange 300.

[0058] In another embodiment, the transaction management unit 130 can receive weather data in real time via an external server. The transaction management unit 130 can analyze the weather data using the model of the learning unit (artificial intelligence unit) 120 and predict fluctuations in power generation due to weather changes. When the expected power generation decreases, the transaction management unit 130 can generate a smart contract that reflects a plan for utilizing stored resources, including discharge from power storage devices. It can also secure additional power in the electricity market by setting a high price in a supply shortage situation.

[0059] In another embodiment, the trading management unit 130 can receive real-time electricity price data from the electricity exchange 300 and analyze the current electricity trading situation based on this data. The trading management unit 130 can calculate the appropriate electricity price in real time based on the power generation data and electricity price data. The trading management unit 130 can generate the optimal smart contract to sell or store electricity in response to real-time market price changes and transmit it to the electricity exchange 300.

[0060] In another embodiment, the trading management unit 130 can collect status data of the power storage device to confirm the currently stored amount of electricity and the available discharge capacity. The trading management unit 130 can predict the optimal time for power storage device discharge and establish a price adjustment strategy to maximize the use of stored electricity. The trading management unit 130 can generate a smart contract that reflects the power storage device discharge plan and transmit it to the power exchange 300 to maximize the utilization of the power storage device.

[0061] The power management unit 140 can comprehensively control the operation of the power storage device 500, including charging and discharging. The power management unit 140 can generate distributed power information based on data provided by the power storage device 500. It can generate charge / discharge schedule information and transmit it to the power storage device 500, thereby controlling the power storage device 500.

[0062] The trading management unit 130 can receive electricity demand data from the electricity exchange 300 and, based on this, request distributed electricity information from the electricity management unit 140. The trading management unit 130 can generate smart contracts based on the distributed electricity information provided by the electricity management unit 140.

[0063] The power management unit 140 can receive data from the power plant 200 and monitor the power production status of the environmentally friendly power plant 220 and the safety-type power plant 210 in real time. Furthermore, the power management unit 140 can generate charge / discharge schedule information based on the power production status and transmit it to the power storage device 500.

[0064] The power management unit 140 can generate distributed power information, including information such as the power generation capacity and power generation speed of the power plant 200, and the maximum charge amount and charging speed of the power storage device 500. Here, distributed power information can mean data indicating the power production, storage, and consumption status associated with distributed energy sources (parent environment power plant 220, safety type power plant 210, and power storage device 500) in the virtual power plant system 100. The distributed power information may include information on the time-of-day power generation amount of the safety type power plant 210, power production efficiency relative to fuel use, the operating status of the power plant and whether maintenance and repair are required, and the maximum amount of power that can be supplied during a specific time period.

[0065] Furthermore, the distributed power information may include information on the predicted power generation of the environmentally friendly power plant 220, based on the amount of solar and wind power generated by the environmentally friendly power plant 220 at different times of the day, as well as weather data such as solar radiation and wind speed.

[0066] The distributed power information may include power state data for the power storage device 500. The power state data may include information for at least one of the following: SOC (State of Charge), current charge percentage (0-100%), maximum chargeable capacity, charging speed (kW or MW), maximum dischargeable capacity, currently dischargeable capacity, percentage of energy lost during discharge, and SOH (State of Health).

[0067] The distributed power information may also include at least one of the following collected by the consumer terminal 400: time-of-day power consumption, peak and off-peak time-of-day consumption data, data by consumption device, consumption pattern data, and consumer preference and setting data (e.g., cost reduction priority, carbon emission reduction).

[0068] The power management unit 140 may include a charge / discharge optimization unit 145. The charge / discharge optimization unit 145 can monitor the status of the power storage device 500 in real time and optimize the charge / discharge strategy using an AI-based reinforcement learning algorithm. Specifically, the charge / discharge optimization unit 145 can use reinforcement learning algorithms including DQN (Deep Q-Network), PPO (Proximal Policy Optimization), and Linear Programming to generate a charge / discharge policy that charges during periods of low electricity prices and discharges during periods of high prices. At this time, the charge / discharge optimization unit 145 can determine the available charge / discharge capacity and time periods based on data such as SOC (State of Charge) and SOH (State of Health) provided by the status monitoring unit 540.

[0069] In one embodiment, the transaction management unit 130 and the power management unit 140 can work complementaryly to optimize power trading and storage strategies. The transaction management unit 130 analyzes power price data and supply / demand data to derive the optimal power trading strategy and can automatically execute buy and sell transactions using smart contracts. This allows for the prediction of power price fluctuations during specific time periods and the execution of a strategy to purchase or store power when prices are cheap and sell it when prices rise.

[0070] The power management unit 140 optimizes the charging and discharging schedule of the power storage device 500 based on the trading strategy provided by the trading management unit 130, monitors the status of the power storage device 500 in real time, and uses a reinforcement learning algorithm to determine the optimal charging and discharging policy. For example, if the trading management unit 130 predicts a future rise in electricity prices and decides to store electricity in the power storage device 500, the power management unit 140 adjusts the optimal charging time and speed considering the State of Charge (SOC) and State of Health (SOH) status of the power storage device 500. Conversely, if a decline in electricity prices is expected, the power management unit 140 adjusts the discharge of the power storage device 500 to generate optimal new revenue. Through this cooperative structure, the virtual power plant system 100 can maximize economic efficiency and effectively manage the volatility of the energy market.

[0071] The power management unit 140 can generate a charge / discharge schedule for the power storage device 500 based on distributed power information (e.g., power demand data and power production status).

[0072] The scheduling unit 150 can predict at least one of the following: consumer energy demand, power plant power output, and power status data of power storage devices, using the model of the artificial intelligence unit 120. The scheduling unit 150 can predict demand by reflecting consumers' power consumption patterns and preferences by time of day. The scheduling unit 150 can also predict power supply by predicting environmentally friendly power generation data such as solar and wind power based on weather data. The scheduling unit 150 can simulate various power sales or power storage strategies based on the calculated prediction data. The scheduling unit 150 can set schedule priorities according to user preferences. The scheduling unit 150 can generate consumer-tailored schedule information based on the calculated prediction data.

[0073] The scheduling unit 150 can collect electricity demand data from consumer terminals 400 and receive electricity price data from the electricity exchange 300. Here, the electricity demand data may include peak-time demand data and off-peak-time demand data, and the electricity price data may include peak-time price data and off-peak-time price data.

[0074] The scheduling unit 150 can analyze electricity demand data and electricity price data to generate consumer-tailored schedule information. The schedule information can include schedules for various purposes, such as schedules for reducing peak-hour costs, schedules for maximizing the use of environmentally friendly power generation, schedules for energy conservation, schedules for reducing carbon emissions, and schedules that reflect consumer preferences.

[0075] The scheduling unit 150 can generate schedules that, based on time-of-day consumption data, restrict the use of non-essential devices during peak hours or suggest operating home appliances during off-peak hours. For example, an electric vehicle charging schedule can be set to avoid peak hours and charge during daytime hours when solar power generation is high or during off-peak hours.

[0076] Furthermore, the scheduling unit 150 can utilize power generation data from the environmentally friendly power source (environmentally friendly power plant 220) to generate a schedule that guides the system to consume or store energy during periods when renewable energy is abundant. For example, it can control the system to release energy stored in the power storage device 500 during periods when renewable energy is scarce, and provide the power management unit 140 with a charging schedule to store energy during periods when renewable energy is abundant.

[0077] The scheduling unit 150 can provide a schedule that reflects event-based consumption patterns, based on the consumer's power usage history. For example, it can generate a schedule that adjusts the operating hours of heating and cooling systems on public holidays, or restricts the use of home appliances during specific time periods on weekends.

[0078] In other words, the scheduling unit 150 transmits optimized schedule information to the consumer terminal 400 based on electricity price data and electricity demand data, thereby supporting consumers in achieving cost savings, energy savings, and carbon emission reductions.

[0079] The emergency response unit 160 works in cooperation with the power management unit 140, the scheduling unit 150, and the transaction management unit 130 to collect power demand data, power price data, and external weather data in order to maintain the safety of the virtual power plant system 100 and to detect emergency conditions in real time.

[0080] Here, 'emergency conditions' may include at least one of the following: power plant failure, sudden surge in power demand, and adverse weather conditions. The emergency response unit 160 can classify the emergency conditions and, depending on the situation, adjust the charging and discharging state of the power storage device (energy storage device) 500, or send a command to the consumer terminal 400 to restrict the use of non-essential devices. The emergency response unit 160 can determine whether or not an emergency condition has occurred based on at least one of the following: power demand data, power generation data, and weather data.

[0081] For example, the emergency response unit 160 can activate the discharge of the power storage device 500 in the event of a surge in power demand and execute a strategy to purchase additional power through the transaction management unit 130. Furthermore, if environmentally friendly power generation decreases sharply due to adverse weather conditions, the output of the safety-type power plant 210 can be increased to compensate for the power shortage.

[0082] The emergency response unit 160 can provide the results of the emergency situation response to the artificial intelligence unit 120 to improve the automatic response capability to the emergency situation and maximize the reliability of the virtual power plant system 100.

[0083] The notification unit 170 can inform users and administrators of the system status and abnormal conditions in real time.

[0084] The notification unit 170 can generate and transmit status-specific notifications based on real-time data generated and collected by various components of the virtual power plant system 100. Here, the notifications may include emergency status notifications, schedule execution status notifications, transaction status notifications, and forecast notifications.

[0085] The notification unit 170 can send a notification to the administrator and users when an emergency is detected, including the cause of the emergency, the components affected, and the proposed measures. For example, if the power generation of the safety-type power plant 210 suddenly decreases, the notification unit 170 can send a notification to the administrator terminal stating, "Safety-type power plant output decrease: Output reduced by 50%, inspection required immediately."

[0086] Furthermore, the notification unit 170 can transmit the execution status of the consumer-ordered schedule generated by the scheduling unit 150 to the consumer terminal 400. For example, the notification may include a message such as, "Electric vehicle charging is scheduled for tonight at 10:00. Estimated charging cost: 3,000 won."

[0087] The notification unit 170 can send notifications to various targets such as administrator terminals, consumer terminals 400, and power exchanges 300, thereby providing both user convenience and system security simultaneously.

[0088] The notification unit 170 is responsible for real-time information transmission for the virtual power plant system 100, and can enhance the ability of users and administrators to take action through abnormal situation detection, schedule execution status notifications, transaction status reports, etc.

[0089] Database 180 can classify and store data generated or collected by various components of the virtual power plant system 100, such as power consumption data, power generation data, weather data, power price data, power demand data, and distributed power information. Database 180 can have a general data structure implemented in the storage space (hard disk or memory) of a computer system using a database management program (DBMS). Database 180 can have a data storage format that allows for free retrieval (extraction), deletion, editing, and addition of data. Database 180 can be implemented to suit the purposes of one embodiment of this disclosure using relational database management systems (RDBMS) such as Oracle, Infomix, Sybase, and DB2, object-oriented database management systems (OODBMS) such as Gemstone, Orion, and O2, and XML native databases such as Excelon, Tamino, and Sekaiju, and can have appropriate fields or elements to achieve its functions.

[0090] The virtual power plant system 100 can have the same hardware configuration as a typical web server or WAP server. However, in terms of software, it can include program modules that perform various functions and are implemented through languages ​​such as C, C++, Java®, Visual Basic, and Visual C. Furthermore, the virtual power plant system 100 generally refers to a computer system and the computer software (server program) installed for it that is connected to an unspecified number of clients and / or other servers via an open computer network such as the internet, receives work requests from clients or other servers, derives and provides the results of those work. In addition to the server program mentioned above, the virtual power plant system 100 should be understood as a broader concept that includes a series of application programs that run on the virtual power plant system 100 and, in some cases, various databases (DBs, hereinafter referred to as "DBs") built internally or externally.

[0091] Here, "network" refers to a connected structure that enables information exchange between nodes such as terminals and servers, or a network connecting the virtual power plant system 100 and smart factory control devices (including, for example, a power exchange 300). "Network" includes, but is not limited to, the Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), 3G, 4G, LTE, 5G, and Wi-Fi. While a network can be a closed network such as a LAN or WAN, it is preferable for it to be an open network like the Internet. "Internet" refers to a global, open computer network structure that provides the TCP / IP protocol and many services at its higher layers, namely HTTP (Hyper Text Transfer Protocol), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), SNMP (Simple Network Management Protocol), NFS (Network File Service), and NIS (Network Information Service).

[0092] Figure 5 is a block diagram illustrating the structure of the power storage device 500 shown in Figure 1.

[0093] Referring to Figure 5, the power storage device 500 may include a charging unit 510, a discharging unit 520, a power storage control unit 530, a state monitoring unit 540, a communication unit 550, and a control unit 560 for storing and supplying energy when needed. In one embodiment, some of these components may be omitted from the power storage device 500, or one or more other components may be added.

[0094] The charging unit 510 can perform the function of storing energy in the power storage device 500. The charging unit 510 can charge the battery using electricity generated at the power plant 200 or surplus electricity from the power grid. The charging unit 510 can analyze the input power to optimize the charging speed and efficiency, and can adjust the charging voltage and current according to the battery state.

[0095] The discharge unit 520 can supply energy stored in the power storage device 500 to the outside as needed. The discharge unit 520 can maintain the safety of the power grid by releasing stored energy when the demand for the power grid increases. The discharge unit 520 can minimize energy loss by adjusting the output voltage and current during discharge in real time.

[0096] The power storage control unit 530 can manage the energy storage state of the power storage device 500. The power storage control unit 530 can monitor the battery status (charge amount, temperature, capacity reduction rate, etc.) in real time and save the data. The power storage control unit 530 can manage and tune the overall operation of the power storage device 500. The power storage control unit 530 can optimize the energy flow by controlling the charging unit 510 and the discharging unit 520. The power storage control unit 530 can receive data from the status monitoring unit 540 and switch the system to safety mode when an emergency occurs.

[0097] The condition monitoring unit 540 can check the overall condition of the power storage device 500 in real time. The condition monitoring unit 540 can measure and record temperature, voltage, current, SOC (State of Charge), SOH (State of Health), etc. The condition monitoring unit 540 can detect emergency conditions (e.g., overcharging, over-discharging, high temperature conditions, etc.) and generate notifications.

[0098] The communications unit 550 can be responsible for sending and receiving data between the power storage device 500 and the virtual power plant system 100, the power exchange 300, or other external systems.

[0099] The communication unit 550 can send charge and discharge status and storage status data to the virtual power plant system 100. The communication unit 550 can also receive and execute charge / discharge commands from an external system.

[0100] In one embodiment, the virtual power plant system 100 determines the charge amount (C) of the power storage device 500 based on predicted values ​​of power demand and power price. t ), discharge amount (D t ), purchase price (P t ^buy), selling price (P t The process of determining ^sell) will be explained with an example.

[0101] In this specification, the power trading optimization algorithm is comprised of an AI-based demand forecasting system, a reinforcement learning-based power storage device charging and discharging strategy, and a blockchain-based power trading engine. Furthermore, an agent is defined as a current state (s t ) observe and respond to this behavior (a t After selecting and applying to the environment, compensate (r t This can mean a learning subject that receives (this information).

[0102] An agent according to an embodiment of the present invention may act on a unit resource or node of an individual consumer within the virtual power plant system 100. The agent can be configured as a control arithmetic module or an information processing device that can operate independently within the virtual power plant system 100. For example, the agent includes a processor, a memory, and a communication interface, and can be embodied as a device or a software module that can perform functions such as demand prediction, state determination, action selection, and compensation calculation in real time. In other embodiments, it can be embodied in a cooperative structure that does not directly belong to the virtual power plant system 100 as an independent module for optimizing the operation of the distributed energy source (DER) of individual consumers. The agent can be embodied as an individual module physically separated from the virtual power plant system 100 similar to the consumer terminal 400, and can exchange information or signals with the consumer terminal 400 and the virtual power plant system 100 to analyze at least one of the power consumption pattern, the state of the power storage device 500, and the real-time power price, and perform scheduling based on the analysis result.

[0103] The reinforcement learning-based power trading optimization algorithm can be defined in a structure including the following state (s t ), action (a t ), and compensation (r t ).

[0104] [Mathematical formula 1] JPEG2026111488000002.jpg3089

[0105] In [Mathematical formula 1], s t can mean the state at time t. a t can mean the action at time t, and r t can mean the reward at time t.

[0106] [Mathematical formula 2] JPEG2026111488000003.jpg1989

[0107] In [Mathematical Formula 2], a is an action, A is the set of possible actions, and s t is the state at time t, r t E[·] can represent the outcome obtained when a specific action a is taken at time t. argmax can represent an operator that searches for the value that maximizes a particular expression, and E[·] represents the expected value, which can show the average result of probabilistically possible values. That is, [Mathematical Equation 2] is E[r t |s t Maximize a] t This can be interpreted as follows: [Mathematical formula 2] can be used in reinforcement learning to guide an agent to select an action that maximizes the compensation expected in the current state, thereby learning a strategy that maximizes compensation in the long term through optimal action selection.

[0108] In [Mathematical Formula 2], s t s can represent the state at time t. t SOC t , L t , P t ^(buy), P t ^(sell), forecast(L (t+1:t+H) This can mean a state vector that contains at least one of the following: SOC t This can mean the charge state of the energy storage device at time t, L t This can be interpreted as the load demand at time t. t ^(buy) is the electricity purchase price, P t ^(sell) can mean the electricity selling price, and forecast(L (t+1:t+H) This can be interpreted as the predicted power demand value for the upcoming H-hour interval.

[0109] [Mathematical formula 3] JPEG2026111488000004.jpg21102

[0110] In [Mathematical Equation 3], P t ^(buy) is the electricity purchase price per unit at time t, Pt ^(sell) is the electricity selling price per unit at time t, C t D is the amount of charge at time t. t This can represent the amount of discharge at time t. According to [Mathematical Equation 3], a t is the electricity purchase price (P t ^(buy)), electricity sales price (P t ^(sell)), charge amount (C t ), discharge amount (D t It can be defined as the optimal action selected to minimize transaction costs, with ) as a variable. That is, if we follow [Equation 3], a t Since it induces cost minimization through the combination of each variable, it can contribute to establishing scheduling that takes economic efficiency into account in a real-time electricity market environment.

[0111] Referring again to [Equation 1] and [Equation 2], r t This can mean compensation (reward) at time t. t This can generally be calculated based on the net profit or expense loss arising from the transaction.

[0112] [Mathematical formula 4] JPEG2026111488000005.jpg2089

[0113] In one embodiment, r t P can be defined as shown in [Mathematical Equation 4]. In [Mathematical Equation 4], P t ^(buy) is the electricity purchase price per unit at time t, P t ^(sell) is the electricity selling price per unit at time t, C t D is the amount of charge at time t. t This can represent the amount of discharge at time t.

[0114] As mentioned above, r t Since this operates through numerical feedback provided by the agent's actions, it can substantially contribute to the reinforcement learning agent learning even better action policies in the future.

[0115] Furthermore, as mentioned above, when using [Mathematical Formula 1] to [Mathematical Formula 4], s t This provides the agent with the basic information to recognize the current environmental state and formulate a strategy, and can therefore effectively contribute to setting up reinforcement learning-based charging / discharging and power trading optimization policies. Here, state (s t ) represents the current remaining capacity (SOC) of the 500 units of power storage equipment. t ), load demand (L t ), electricity trading price (P t ^buy, P t ^sell), future demand forecast (forecast(L (t+1:t+H) This can include actions such as )) and actions (a t ) is the charge amount (C t ), discharge amount (D t ), and can be comprised of setting electricity purchase and sales prices.

[0116] In one embodiment, the state of the power storage device 500 can be updated by the following [Mathematical Formula 5].

[0117] [Mathematical formula 5] JPEG2026111488000006.jpg2489

[0118] [Mathematical formula 5] is an equation for calculating the State of Charge (SOC) of a power storage device 500 according to one embodiment of the present invention. t (η) is the state of charge of the energy storage device at time t. c )C t This can mean charging efficiency, C t D is the amount of energy charged in time t. t η is the amount of electrical energy discharged in time t. D This can be interpreted as a coefficient that corrects the amount of electricity that can actually be supplied to the outside from the internal energy of the energy storage device, reflecting the losses during discharge.

[0119] Here, η cη is a coefficient that reflects the amount of energy effectively stored, taking into account losses during charging. D This is a coefficient that corrects the amount of electricity that can be supplied to the outside from the internal energy of the power storage device 500, reflecting the loss during discharge. Since the state of the power storage device 500 fluctuates over time due to repeated charging and discharging, as mentioned above, when using [Mathematical Equation 5], it becomes possible to predict the state of affairs (SOC) at the next point in time, which can help the reinforcement learning agent select a strategic action that takes the remaining energy into consideration.

[0120] In one embodiment, the equilibrium condition for a P2P electricity trading market can be defined as follows [Mathematical Equation 6].

[0121] [Mathematical formula 6] JPEG2026111488000007.jpg2589

[0122] [Equation 6] is an equation that shows the equilibrium condition for electricity purchase and sale over the entire time interval. In [Equation 6], P t ^(sell) can represent the amount of electricity sold by the seller at time t. t ^(buy) can represent the amount of electricity purchased by the buyer at time t. Using [Mathematical Formula 6], it can be guaranteed that the sales volume and purchase volume must always be kept equal within P2P electricity trading, thus contributing to ensuring real-time equilibrium of electricity supply and demand within the market.

[0123] In one embodiment, a reinforcement learning agent can be trained by the following [Mathematical Equation 7].

[0124] [Mathematical formula 7] JPEG2026111488000008.jpg25102

[0125] [Equation 7] is the objective function for minimizing the total cost. In [Equation 7], P t ^(buy) can represent the electricity purchase price per unit at time t. tThis can represent the amount of electricity discharged during the relevant time period. t ^(sell) can mean the electricity selling price per unit at time t, C t This can represent the amount of electricity charged during the relevant time. According to [Mathematical Formula 7], the total power trading cost can be minimized by reflecting the costs incurred during charging and the revenue obtained during discharging, thus effectively contributing to the reinforcement learning agent learning the optimal charging, discharging, and trading strategy.

[0126] In this case, the condition for satisfying electricity demand can be defined as follows [Mathematical Equation 8].

[0127] [Mathematical formula 8] JPEG2026111488000009.jpg2689

[0128] [Equation 8] is an equation that shows the constraints for satisfying the time-dependent load demand. In [Equation 8], D t This can represent the amount of discharge at time t. t L can mean the amount of charge at time t, t This can represent the actual load power demand at time t.

[0129] In one embodiment, the charge-discharge limit condition to reflect the technical limitations of the power storage device 500 can be defined as follows [Mathematical Equation 9].

[0130] [Mathematical formula 9] JPEG2026111488000010.jpg25102

[0131] [Equation 9] is an equation that shows the charge and discharge limit conditions for the power storage device 500. In [Equation 9], C t This can represent the amount of charge at time t. (max) This can mean the maximum allowable charge amount that cannot be exceeded by the amount of charge at time t. t This can mean the amount of discharge at time t, and D (max)This can be interpreted as the maximum allowable discharge amount that cannot be exceeded by the discharge amount at time t. As mentioned above, when using [Mathematical Equation 9], it is possible to ensure that the charging and discharging operations are performed within the limits of the physical limitations of the power storage device 500, thereby contributing to the stable operation of the power system.

[0132] In one embodiment, the power trading optimization system forms an optimization problem that integrates the aforementioned objective function and constraints, and by iteratively learning this problem through a reinforcement learning-based agent, it can help individual consumers perform self-demand response (Self-DR) without going through a power exchange. Through this, the agent can establish optimal charging, discharging, and trading strategies in real time based on power demand and price forecasts, thereby simultaneously ensuring the efficiency and safety of the overall power system.

[0133] In one embodiment, the power trading optimization system forms an optimization problem that integrates the aforementioned objective function and constraints, and learns iteratively through a reinforcement learning-based agent, enabling individual consumers to perform self-demand response (Self-DR) without going through a power exchange. Through this, the agent can establish optimal charging, discharging, and trading strategies in real time based on power demand and price forecasts, thereby simultaneously ensuring the efficiency and safety of the overall power system.

[0134] The amount of charge (C) of the power storage device 500 as described above. t ), discharge amount (Dt), purchase price (P t ^buy), selling price (P t The method for determining ^sell can be carried out by the artificial intelligence unit 120 and the power management unit 140 of the virtual power plant system 100.

[0135] Figure 6 is a flowchart illustrating a method for generating and providing schedules by a scheduling unit 150 according to one embodiment of the present invention.

[0136] Referring to Figure 6, at step S610, the scheduling unit 150 can collect consumer data from the consumer terminal 400. Here, consumer data may include electricity usage by time of day, usage data by consumer device, and consumer preferences. The scheduling unit 150 can also receive electricity price data from the electricity exchange 300 and renewable energy generation data and status data of the electricity storage device 500 from the electricity management unit 140. Weather data may include data such as solar radiation and wind speed necessary for solar and wind power generation forecasting.

[0137] At stage S620, the scheduling unit 150 can generate consumer-specific forecast information based on the collected data. The forecast information may include time-of-day electricity consumption, consumption changes due to specific events, time-of-day estimated costs, and renewable energy production. For example, the scheduling unit 150 can use weather data to predict time periods with high solar power generation, or use electricity price data to identify time periods with low costs.

[0138] At stage S630, the scheduling unit 150 can generate consumer-customized schedule information based on the generated forecast information. Here, the schedule information may include at least one of the following: a cost-saving schedule, a renewable energy utilization schedule, and an energy-saving schedule.

[0139] For example, the scheduling unit 150 can generate schedule information to guide power consumption to concentrate during off-peak hours, avoiding peak hours, or to charge the energy storage device 500 during periods of high solar power generation. The scheduling unit 150 can also generate customized schedule information that reflects consumer preferences as needed.

[0140] At step S640, the scheduling unit 150 can transmit the generated schedule information to the consumer terminal 400. Here, the schedule information can be provided visually through the display of the consumer terminal 400. The scheduling unit 150 can monitor the schedule execution status in real time and update the schedule information and retransmit it to the consumer terminal 400 if changes occur in power demand data, power price data, or weather data. For example, the scheduling unit 150 can send a message to the consumer terminal 400 saying, "Electric vehicle charging will start tonight at 10pm. Estimated cost: 3,000 won."

[0141] As described above, the schedule generation and provision method according to one embodiment of the present invention can optimize consumer electricity consumption, analyze consumption time periods, and increase the utilization of environmentally friendly power generation, thereby stabilizing the power grid and providing the technical effect of utilizing user demand response infrastructure resources. Through this, the virtual power plant system 100 can maximize energy efficiency and realize sustainable energy management.

[0142] Figure 7 is a flowchart illustrating a method for generating a charge / discharge schedule according to one embodiment of the present invention.

[0143] Referring to Figure 7, at step S710, the power management unit 140 can collect at least one of the following: status data, power price data, power generation data, and weather data, in order to generate a charge / discharge schedule for the power storage device 500.

[0144] Here, the state data may include the State of Charge (SOC), State of Health (SOH), charging rate, discharging rate, and maximum charge / discharge capacity of the power storage device 500.

[0145] The power management unit 140 can receive time-of-day electricity price data from the electricity exchange 300. Here, the electricity price data may include peak time-of-day price data and off-peak time-of-day price data.

[0146] The power management unit 140 can also collect environmental power generation data and stable power generation data from environmental power plants 220 and safety-type power plants 210, and collect weather data (solar radiation, wind speed, etc.) from an external server.

[0147] At stage S720, the power management unit 140 can generate predictive information necessary for charging and discharging based on the collected data. The charging prediction information can generate predictive data so that charging is possible during times when renewable energy production is high. The discharging prediction information can generate predictive data so that stored energy is released due to increased electricity demand during peak hours. The cost prediction information can predict the costs of charging and discharging by analyzing electricity price data by time of day.

[0148] In step S730, the power management unit 140 can generate a charge / discharge schedule based on step S720. The charging schedule can be generated by selecting a time period when renewable energy production is high or when electricity prices are low. For example, the power storage unit 500 can be planned to be charged during the daytime when solar power generation reaches its maximum value.

[0149] The discharge schedule can be generated to release electricity during peak hours when electricity demand is high or when electricity prices are high. For example, the schedule can be planned to supply electricity released from the power storage device 500 to the power grid during peak hours in the evening.

[0150] The power management unit 140 can generate a charge / discharge schedule that maximizes charge / discharge efficiency by utilizing a variety of optimization algorithms.

[0151] At step S740, the power management unit 140 can transmit the generated charge / discharge schedule to the power storage device 500. The power storage device 500 can receive the schedule information so that the charging unit 510 and the discharging unit 520 can perform charging and discharging according to the schedule.

[0152] As described above, the charge / discharge schedule generation method according to one embodiment of the present invention has the effect of enhancing the safety of the power storage device 500 and increasing the efficiency of electrical energy. Furthermore, such effects can contribute to maximizing the overall performance and longevity of the virtual power plant system 100, which includes the power storage device 500.

[0153] The embodiments described above can be embodied in hardware components, software components, and / or combinations of hardware and software components. For example, the apparatus, methods, and components described in the embodiments can be embodied using one or more general-purpose or special-purpose computers, such as a processor, controller, ALU (arithmetic logic unit), digital signal processor, microcomputer, FPGA (field programmable gate array), PLU (programmable logic unit), microprocessor, or any other device capable of giving and receiving instructions. The processing device can perform an operating system (OS) and one or more software applications performed on the OS. The processing device can also access, store, manipulate, process, and generate data in response to software execution. For convenience of understanding, the processing device has sometimes been described as being used by only one, but a person with ordinary skill in the art will know that the processing device can include multiple processing elements and / or multiple types of processing elements. For example, the processing device can include multiple processors or one processor and one controller. Furthermore, other processing configurations, such as parallel processors, are also possible.

[0154] Software can include computer programs, code, instructions, or combinations of these, which can configure a processing unit to operate as desired, or independently or collectively, instruct the processing unit. Software and / or data can be permanently or temporarily embodied in a type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave, in order to be interpreted by a processing unit or to provide instructions or data to a processing unit. Software can also be distributed, stored in a distributed manner, or executed on a network of connected computer systems. Software and data can be stored on one or more computer-readable recording media.

[0155] The methods according to the embodiment can be embodied in a program instruction form that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., individually or in combination. The program instructions recorded on the medium may be specifically designed and configured for the embodiment, or they may be publicly known and available to those skilled in the computer software art. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include not only machine code, such as that produced by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above can be configured to operate as one or more software modules to perform the operations of the embodiment, and vice versa.

[0156] Although the embodiments have been described above with limited drawings, a person with ordinary skill in the relevant art can apply a variety of technical modifications and variations based on the foregoing. For example, the described technique may be carried out in a procedure different from the described method, and / or the components of the described system, structure, apparatus, circuit, etc. may be combined or assembled in a manner different from the described method, or juxtaposed or substituted by other components or equivalents, and the appropriate results may be achieved.

[0157] Therefore, other embodiments, other embodiments, and those equivalent to the claims described below also fall within the scope of the claims.

Claims

1. In a virtual power plant system that enables individual consumers to perform self-demand response (Self-DR) using user terminals, the system analyzes at least one of the following in real time: electricity demand data, generation data, electricity price data, and distributed power information, and provides the analysis results: A transaction management unit that calculates the demand and supply of electricity based on the aforementioned electricity price data, the amount of electricity generated data received from power plants, and weather data received from an external server, and predicts an appropriate electricity price based on the calculation results to generate a smart contract; the amount of electricity generated data includes safe-type power generation data received from safe-type power plants and environmentally friendly power generation data received from environmentally friendly power plants. A power management unit that generates distributed power information based on power status data received from a power storage device and at least one of the following: the power price data, the power generation amount data, and the weather data; generates charge / discharge schedule information based on the distributed power information; and transmits it to the power storage device. Includes a scheduling unit that analyzes time-of-day power consumption patterns and preferences based on power demand data and power consumption data received from consumer terminals, generates demand forecast data, and generates a consumer-customized schedule based on the said demand forecast data; A virtual power plant system.

2. The transaction management unit receives the electricity price data from the electricity exchange and transmits the smart contract to the electricity exchange. The virtual power plant system according to claim 1.

3. An emergency response unit that detects in real time whether or not an emergency situation has occurred; and The emergency condition includes at least one of a power plant failure, a surge in electricity demand, and a severe weather condition, and further includes a notification unit that transmits an emergency condition notification to the administrator and the consumer terminal when the emergency condition occurs; The virtual power plant system according to claim 1.

4. The aforementioned scheduling unit, Based on the aforementioned electricity demand data and electricity price data, consumer-customized schedule information is generated. The aforementioned electricity demand data includes peak-time demand data and off-peak-time demand data. The aforementioned electricity price data includes price data for peak hours and price data for off-peak hours. The aforementioned consumer-customized schedule information includes a schedule for cost reduction during peak hours. The virtual power plant system according to claim 1.

5. The aforementioned scheduling unit, The system is characterized by generating consumer-ordered schedule information using the power trading optimization algorithm of the artificial intelligence unit. The virtual power plant system according to claim 4.

Images