KR102582302B1 - System Having a Function of Managing Power Demand - Google Patents

Abstract

According to the present invention, a storage unit for storing subscriber information including power usage for at least one power consumption device of the subscriber and a record of the subscriber's participation in a personal demand response event, and event information including the date of the demand response event; A receiving unit that receives a request signal for reducing power use; a selection unit that, upon receiving the request signal, selects a subscriber who wishes to request power usage reduction using the subscriber information and event information; It includes a transmitter that transmits guidance information on power usage reduction to the subscriber terminal of the selected subscriber, wherein the selection unit generates a credit scoring model, evaluates the subscriber's credit score according to the credit scoring model, and uses the power. A server having a power demand management function is provided that selects subscribers wishing to request savings according to their evaluated credit scores.

Description

System Having a Function of Managing Power Demand}

The present invention relates to a power demand management method, and more specifically, in the target selection process for selecting subscribers who are likely to participate in a demand response event, an explanation of the selection process that is easy to understand for general subscribers without energy-related knowledge. It relates to a server having a power demand management function that provides a method of selecting subscribers to be invited during a demand response event that can provide performance, and a power demand management method thereof.

Electricity consumption in the household sector is steadily increasing and is expected to account for 41% of total electricity consumption within the next 20 years. Moreover, the household sector can have a large impact because not only does it account for a large portion of electricity consumption, but it can also play an important role in improving social awareness of energy use through educational activities for the household sector. In this context, Residential Demand Response is considered an important tool to solve the problem of imbalance between electricity demand and supply.

There are two types of residential demand response: Price-Based Demand Response and Incentive-Based Demand Response. Among these, incentive-based demand response can provide subscribers with the flexibility to choose whether to participate in demand response, and does not require large initial investments in dedicated infrastructure. Therefore, incentive-based demand response is advantageous for introduction because it imposes less burden on subscribers, and can provide both efficiency and reliability when initially applying demand response.

This incentive-based demand response is generally operated in the form of an invitation-based event that encourages subscriber participation by sending scheduled invitations at the time when power reduction is most needed. In this case, incentive-based demand response cannot force subscribers to participate. Therefore, introducing a 'selective process' in incentive-based demand response, which sends invitations mainly to subscribers who are likely to participate, can be helpful in the following two aspects.

First, the target selection process may exclude demand response participants who only receive compensation without actually contributing to power reduction. Because people's electricity consumption patterns vary widely, some people may simply participate in all demand response events, make no effort to reduce electricity, and receive compensation even if they consume electricity according to their existing electricity consumption patterns. This problem occurs because the usage patterns of residential demand response participants are much more diverse than those of business demand response participants. Therefore, demand response program operators can improve operational efficiency by providing the option to detect and exclude such demand response participants by utilizing the target selection process.

Second, the target selection process can increase the satisfaction of demand response participants by appropriately controlling the frequency of sending invitations and notifications. In order for incentive-based demand response to work, an invitation must be sent, and a notification message to remind people of the time to reduce can help increase the efficiency of residential demand response. However, if there are many invitations and notifications, there is a disadvantage that it may be annoying to some demand response participants, which may eventually lead to complaints or refusal to receive notifications. For this problem, the audience selection process can determine and control the number of notifications for specific demand response participants.

However, most existing studies focused on optimizing the compensation amount, and there was not much research on the subject selection process. This means that although operational efficiency and demand response participant satisfaction can be improved through the target selection process, the participation opportunities of some consumers may be discriminatoryly limited due to selective invitations, which is likely to lead to fairness disputes and legal disputes.

In particular, when applying a target selection process, demand response program providers must ensure fairness to subscribers, make the process transparent, and explain the causal relationship between demand response compensation. In addition, residential demand response requires easy and concise explanations because it deals with a large number of general subscribers who have no knowledge of the energy industry. Therefore, 'Explainability' is a very important requirement when operating residential demand response.

However, the subject selection process is typically implemented through machine learning to maximize efficiency. Algorithms implemented through machine learning are fundamentally poor in explainability because they are accompanied by mathematical algorithms. In particular, most of the demand response subscribers are general energy cost experts, so it is very difficult to understand the machine learning algorithm.

Therefore, when operating a demand response program, there is an urgent need for a power demand management system that can guarantee explainability while utilizing an efficient target selection process through machine learning.

Korean Patent Publication No. 10-2016-0143455 (published on December 14, 2016)

In order to solve the above problems, the present invention provides a server, communication device and system with a power demand management function that can guarantee explainability while utilizing an efficient target selection process when operating a demand response program, and power demand management thereof. We would like to provide a method.

In order to solve the above technical problem, a server having a power demand management function according to the present invention includes subscriber information including power usage for at least one power consumption device of a subscriber and a record of the subscriber's participation in a personal demand response event; a storage unit that stores event information including the date of the demand response event; A receiving unit that receives a request signal for reducing power use; a selection unit that, upon receiving the request signal, selects a subscriber who wishes to request power usage reduction using the subscriber information and event information; It includes a transmitter that transmits guidance information on power usage reduction to the subscriber terminal of the selected subscriber, wherein the selection unit generates a credit scoring model, evaluates the subscriber's credit score according to the credit scoring model, and uses the power. Subscribers wishing to request savings can be selected based on their assessed credit score.

Meanwhile, the power demand management method according to the present invention includes subscriber information including power usage for at least one power consumption device of the subscriber and the subscriber's personal demand response event participation record, and event information including the date of the demand response event. storing information; Receiving a request signal for reducing power use; Upon receiving the request signal, selecting a subscriber who wishes to request power usage reduction using the subscriber information and event information; A step of transmitting guidance information on power usage reduction to the communication device of the selected subscriber, wherein the selecting step includes generating a credit scoring model, and evaluating the subscriber's credit score according to the credit scoring model. Thus, it may include the step of selecting subscribers who wish to request reduction in power use according to the evaluated credit score.

In addition, the method of selecting a subscriber subject to a demand response event according to the present invention includes subscriber information including power usage for at least one power consumption device of the subscriber and the subscriber's personal demand response event participation record, and the date of the demand response event. A storage unit that stores event information including; A receiving unit that receives a request signal for reducing power use; a selection unit that, upon receiving the request signal, selects a subscriber who wishes to request power usage reduction using the subscriber information and event information; In a server having a power demand management function, including a transmitter that transmits guidance information on power usage reduction to the subscriber terminal of the selected subscriber, information values for a plurality of features generated based on the subscriber information selecting features based on; calculating logit values for the selected features and roughly classifying the features; and generating a credit scoring model by generating a score range for each range of the rough classification through machine learning.

Therefore, according to the present invention, a server, communication device and system having a power demand management function that can guarantee explainability while utilizing an efficient target selection process when operating a demand response program, and a power demand management method thereof are provided. .

Additionally, according to the present invention, subscribers who are likely to participate in a demand response event can be encouraged to participate in the demand response event, and thus power use can be reduced stably.

1 is a configuration diagram of a system with a power demand management function according to an embodiment of the present invention.
Figure 2 is a flowchart of a power demand management method according to an embodiment of the present invention.
Figure 3 is a configuration diagram of a server with a power demand management function according to an embodiment of the present invention.
Figure 4 is a configuration diagram of a subscriber terminal with a power demand management function according to an embodiment of the present invention.
Figure 5 is a flowchart of a method for selecting subscribers for a demand response event according to an embodiment of the present invention.
Figure 6 shows features for creating a credit scoring model according to an embodiment of the present invention.
Figure 7 shows features selected through a selection process among the features of Figure 6 according to an embodiment of the present invention.
Figure 8 is a diagram for explaining the concept of explainability according to the present invention.

The following merely illustrates the principles of the invention. Therefore, those skilled in the art will be able to invent various devices that embody the principles of the invention and are included in the concept and scope of the invention, although not clearly described or shown herein. In addition, all conditional terms and embodiments listed herein are, in principle, expressly intended solely for the purpose of enabling the inventive concept to be understood, and should be understood not as limiting to the embodiments and conditions specifically listed as such. do.

The above-mentioned purpose, features and advantages will become clearer through the following detailed description in relation to the attached drawings, and accordingly, those skilled in the art in the technical field to which the invention pertains will be able to easily implement the technical idea of the invention. .

Additionally, when describing the invention, if it is determined that a detailed description of known technology related to the invention may unnecessarily obscure the gist of the invention, the detailed description will be omitted. The following will be described with reference to the attached drawings.

Hereinafter, with reference to FIGS. 1 to 6, a server, communication device, and system having a power demand management function according to an embodiment of the present invention, and a power demand management method thereof will be described.

Before explaining the present invention, terms mainly mentioned in this specification will be defined.

First, the request signal for power usage reduction may include information on at least one of the amount of power required to be reduced, the time period for which reduction is requested, and the area where reduction is required. For example, the request signal may include information to save 10,000 KWh (required power savings) from 2 PM to 5 PM (savings required time period) on May 1, 2015. Furthermore, the request signal may further include information designating a specific area (eg, Suwon-si, Gangnam-gu, Seoul, etc.) as the reduction request area.

Next, the subscriber information is raw data for creating features, including personal event records of subscribers who signed up for the demand response program (e.g., participation by past events, highest reduction goal success, reduction goal and use), and It includes electricity usage data collected through a power measuring device (eg, smart meter) per unit time for each demand response program subscriber over a certain period of time.

Next, the guidance information on power usage reduction includes at least one of the following: the amount of power saved required for subscribers selected to be invited to a demand response event, the subscriber's power use status, the time period for which savings are required, and expected compensation for savings. It may be included, and may be provided for all power consumption devices or individual power consumption devices used by the subscriber.

Lastly, the information regarding compensation for power usage reduction may include information about power usage discounts granted to the subscriber in proportion to the actual amount of power saved or points available when paying power usage fees to correspond to the guidance information. .

FIG. 1 shows a configuration diagram of a system with a power demand management function according to an embodiment of the present invention, and FIG. 2 shows a flowchart of a power demand management method according to an embodiment of the present invention.

As shown in FIG. 1, the system 700 with a power demand management function displays an invitation to reduce power use received from the demand response operation server 300, the power measurement device 200, and the server 100. It may include a subscriber terminal 400 that outputs or outputs a power supply server 600 that transmits a request signal for reducing power use to the demand response operation server 300.

Additionally, according to an embodiment of the present invention, the system 700 with a power demand management function may be additionally connected to the sponsor server 610.

In this case, the demand response operation server 300, the power measurement device 200, and the subscriber terminal 400 are connected through a network 100, such as the Internet.

The demand response operation server 300 can receive power usage of power consumption devices 500 inside the house from the power measurement device 200 through the network 100 in real time. Additionally, the demand response operation server 300 may select participants from among subscribers to participate in the demand response program based on the power consumption.

Additionally, the demand response operation server 300 may receive a request signal for reducing power use from the power supplier server 600. When receiving a request signal, the demand response operation server 300 may select a participant in the demand response program based on the power consumption and the subscriber's profile. Afterwards, the demand response operation server 300 sends an invitation requesting participation in the demand response event to the subscriber terminal 400 for the selected participation target and, for example, a reminder for participation in the demand response event, etc. Notifications can be sent.

Additionally, the demand response operation server 300 may provide compensation to subscribers who participated in the demand response event. In this case, the demand response operation server 300 is connected to the sponsor server 610 and can provide coupons, cash, etc. provided by the sponsor server 610 as compensation to subscribers who participated in the demand response event.

The power measurement device 200 supplies the power usage of power consumption devices 500 inside the house to the demand response operation server 300 in real time.

Meanwhile, as described above, the subscriber terminal 400 may receive an invitation and notification from the demand response operation server 300 and guide the user to participate in the demand response event. Additionally, various information provided by the demand response operation server 300 can be displayed. The subscriber terminal 400 may be various types of communication devices, such as a mobile phone, tablet, or personal computer (PC).

Below, the operation of each component for implementing the present invention will be described with reference to FIG. 2.

The demand response operation server 300 receives a request signal for reducing power use from the power supply company server 600 (S210).

Here, the power supply company server 600 is a server managed by a power supply source (eg, Korea Electric Power Corporation) or a power supply agency, and can perform overall operations related to power supply.

Upon receiving the request signal, the demand response operation server 300 uses subscriber information to select a subscriber who wishes to request reduction in power use through a target selection process (S220).

Here, the demand response operation server 300 can store subscriber information and update the subscriber information by reflecting the subscriber's power usage status/pattern or update the subscriber information according to the subscriber's request.

The demand response operation server 300 calculates the participation rate, maximum number of consecutive participation, cumulative rebate, maximum number of consecutive non-participation, and past At least one subscriber who will participate in the demand response event can be selected according to predetermined criteria considering success in the last event, participation in the last event, or failure to reduce power. Accordingly, the demand response operation server 300 may select subscribers who are likely to reduce power use by participating in a demand response event in response to a request to reduce power use through a selective process. Alternatively, the demand response operation server 300 may select subscribers who are predicted to have a high response to the request to reduce power use.

The subject selection process for subscriber selection will be described in detail later with reference to FIG. 3.

The demand response operation server 300 transmits an invitation containing guidance information on reducing power use to the subscriber terminal device 400 selected in the selection step (S220) (S230).

Here, the subscriber terminal 400 is a subscriber's communication device registered with the demand response operation server 300 for data communication with the demand response operation server 300, and includes mobile devices such as smart phones, laptops, tablet PCs, etc. It may include all fixed home appliances equipped with communication functions (for example, TVs, refrigerators, air conditioners, etc.).

The demand response operation server 300 monitors the power usage by at least one power consumption device 500 of the subscriber selected in the selection step (S220) (S240).

Accordingly, the demand response operation server 300 may check whether the actual power usage has decreased and the amount of power saved in the power consumption device of the selected subscriber to correspond to the guidance information transmitted in the transmission step (S230) (S250). .

For example, if the demand response operation server 300 determines that the power usage has decreased to correspond to the guidance information in the confirmation step (S250), it may grant compensation corresponding to the reduction in power usage (S260).

The demand response operation server 300 may determine the compensation to be granted to the corresponding subscriber by considering the actual amount of power saved to correspond to the guidance information.

For example, the demand response operation server 300 may grant compensation in proportion to the amount of power saved. Additionally, the demand response operation server 300 may grant compensation related to power usage fees, and may provide compensation in the form of a discount on power usage fees or payment of points that can be used to pay power usage fees.

Alternatively, compensation according to an embodiment of the present invention may include cash, gifts, coupons, etc. provided by a sponsor server 610 other than a server related to a power supply company. In this case, the demand response operation server 300 can receive cash, gifts, coupons, etc. from the sponsor server 610 in electronic form. Alternatively, the demand response operation server 300 may accumulate conversion points in electronic form to the sponsor's account to correspond to cash, gifts, or coupons provided by the sponsor. The demand response operation server 300 calculates the amount of power usage reduction corresponding to cash, gifts, coupons, or conversion points provided by each sponsor. Additionally, the sponsor server 610 may obtain carbon dioxide emissions credits from the carbon dioxide emissions trading server 620 based on the calculated power usage reduction amount. In this case, “based on the amount of power usage reduction” means all cases where the amount of power usage reduction is converted into carbon dioxide reduction, and a person skilled in the art can convert the power usage reduction into carbon dioxide reduction using existing methods. In addition, the carbon dioxide emissions trading server 620 confirms that a power reduction has occurred corresponding to the cash, gifts, coupons, etc. provided by the sponsor server 610 from the power supply company server 600 or the demand response operation server 300. You can.

The demand response operation server 300 may transmit information about the compensation granted in the granting step (S260) (hereinafter referred to as compensation information) to the corresponding subscriber terminal 400 (S270).

Accordingly, the subscriber of the subscriber terminal 400 can check whether the actual power usage fee is discounted and the degree of discount due to the subscriber's power saving behavior. This can have the effect of encouraging subscribers to actively participate in power saving when power saving is required.

FIG. 3 shows a detailed configuration diagram of the demand response operation server 300 having the power demand management function shown in FIG. 1.

According to FIG. 3, the demand response operation server 300 includes a storage unit 310 for storing subscriber information, a reception unit 321 for receiving a request signal for power use reduction, and the subscriber information upon receiving the request signal. A selection unit 331 that selects a subscriber who wishes to request power usage reduction using It may include a monitoring unit 332 that monitors the power consumption by the power consumption device and a compensation management unit 333 that grants a predetermined compensation to the subscriber when the power consumption by the power consumption device decreases according to the guide information. there is.

Here, the receiving unit 321 and the transmitting unit 322 may be implemented as separate communication modules or may be implemented as a single communication module 320. Additionally, the selection unit 331, the monitoring unit 332, and the compensation management unit 333 may be implemented as separate modules or may be implemented as one control module 330.

The storage unit 310 may store subscriber information and event information, and in particular, may store information on power use for each subscriber. For example, as subscriber information, information on power consumption by all power consumption devices or individual power consumption devices for each subscriber, power use profile by time, and power saving history according to guidance information may be stored. More specifically, the power saving history according to the guidance information corresponds to the previous guidance information including the compensation method and amount, whether the actual power usage was reduced, the amount of power saved, information about the power consumption device that performed the power reduction, etc. may include.

Event information includes event day information (date, start time, weather, temperature, and basic incentive amount) common to all demand response participants.

The receiving unit 321 may receive a request signal for reducing power use from the power supply-related server 600 before the start of the period/time zone in which power use reduction is required.

For example, the power supply-related server 600 may transmit a request signal one week, one day, or one hour before the start of the period/time zone in which power usage reduction is required. This is to encourage active power usage reduction behavior by providing subscribers with guidance information on power usage reduction before the period/time during which power usage reduction is required.

The selection unit 331 determines the Participation Rate, Maximum Number of Consecutive Participation, Cumulative Rebate, Maximum Number of Consecutive Non-Participation, and past events. At least one subscriber who wishes to request a reduction in power use can be selected according to predetermined criteria that take into account success in the last event, participation in the last event, or failure to reduce power use. Accordingly, the demand response operation server 300 can select subscribers who are expected to have a high response to the request to reduce power use.

Hereinafter, the subject selection process, which is a detailed operation of the selection unit 331, will be described with reference to FIG. 5.

The selection unit 331 according to an embodiment of the present invention is basically designed to maximize efficiency by utilizing machine learning and increase explainability by incorporating a credit scoring algorithm.

The selection unit 331 has two modes: a credit scoring learning mode and a credit score evaluation mode. In this case, the credit scoring learning mode is a mode that creates a credit scoring model to select demand response invitees based on existing subscriber information, and the credit score evaluation mode is a mode that generates credit scoring models for individual subscribers based on the created credit scoring model. This is a mode that evaluates and selects invitees to demand response events.

Before explaining the two modes of the selection unit 331, the concepts of explainability and credit scoring, which are important concepts for the present invention, will be explained in detail.

demand response in the program explanatory importance

The rise of automated decision-making using big data and machine learning has raised privacy concerns among many demand response participants. To resolve these concerns, the EU's 'GDPR (General Data Protection Regulation)' and legal regulations called 'Communication on Delivering a New Deal for energy consumers (hereinafter referred to as New Deal)' have been established. It is being implemented.

GDPR was passed in May 2016 and will come into effect in 2018. It is intended to provide transparency and fairness when personal data is used for automated decision-making. According to the GDPR, data subjects have the right to monitor all processes and receive understandable explanations. That is, the models used for automated decision-making must be designed in an explainable way so that they can provide answers when demand response participants raise questions or communicate complaints.

According to the EU's New Deal (2015), explainable modeling is needed in energy markets for the same reasons. In the energy market, all decision-making processes related to energy services must be transparently disclosed to demand response participants, and demand response participants have the right to freely monitor and change program options at any time. Therefore, demand response and target selection processes should also follow the same regulations, and in particular, since residential demand response programs are intended for the general public, simpler explanations should be available.

explanatory Justice

The efficiency and accuracy of demand response can be measured using various measurement techniques, such as Area Under Receiver Operation Characteristics (AUROC) and Area Under Precision-Recall Curve (AUPRC).

However, explainability is not easy to define because it depends on a person's background and knowledge base. Because of the subjective nature of explainability, most existing studies define and evaluate explainability indirectly. For example, explainability is shown in Figure 8 according to existing research (e.g., Learning Interpretable Models, S. Ruping, Dortmund Univ. (PhD Thesis) 2006, Towards a Rigorous Science of Interpretable Machine Learning, Finale Doshi-Velez, arXiv, 2017). It is defined in a qualitative way as follows.

According to the definition of explainability in previous studies in Figure 8, explainability can be defined using empirical methods, but cannot be evaluated using quantitative metrics. In particular, the existing studies above point out that the only way to ensure explanatory power is to use models that are already known to be explainable. Therefore, the present invention adopts credit scoring, known as an explainable approach, to enhance explainability in demand response.

credit scoring

Credit scoring, invented by Durand in 1941, sets the score for each range through logistic regression or linear programming of features generated from historical data of demand response subscribers. Credit scoring uses a score card, a simple and familiar format that can be understood by demand response subscribers without statistical expertise. Credit scoring primarily helps banks minimize and account for financial risk. For example, when you visit a bank to get a mortgage loan, the banker uses a credit scoring model to predict the person's likelihood of repaying the loan and determines the amount, interest rate, and term that can be borrowed. To evaluate the score, -10 points are given for delinquency in the past three months, +50 points for an average salary of more than $5,000, and +20 points for owning a car. In this case, if a demand response participant is denied a loan, the demand response participant can file a complaint about the bank's fairness.

To address this issue, the financial sector has introduced the National Credit Act of 2005 states in article 62, which mandates banks to disclose their decision-making processes and clearly explain the causal relationships in each process. ') and other legal regulations were implemented. In other words, an explainable model should be used in situations where demand response participants can ask questions about fairness.

Credit scoring has been actively used in the industry for over 60 years due to its high predictive accuracy rate in addition to meeting explainability requirements. There are two reasons for the high prediction performance. First, overfitting is controlled by extracting only the features necessary for prediction. Second, non-linearity is handled by adaptively selecting different score ranges.

Because credit scoring performs classification based on probability, prediction performance improves compared to the simple case of using raw data as is. To improve the predictive performance of credit scoring, machine learning can be applied to the feature selection or feature generation steps. However, even if machine learning is adopted, the simple scorecard format must be maintained to ensure explainability. Although machine learning is applied in several steps, such as feature selection and credit score range assignment, the final model is designed to remain a score card that can be understood by the average demand response participant.

Hereinafter, the credit scoring learning mode of the target selection process of the selection unit 331 will be described in detail with reference to FIG. 5.

Credit scoring model for selecting demand response participants

1. Feature Selection (Step S710)

Feature selection in credit scoring is the step of selecting several important criteria features necessary to maximize efficiency while maintaining explainability. Credit scoring for demand response selects a standard feature based on the information value defined by Equation 1 below.

(Equation 1)

Here, IV is the information value, and Pr() is the probability function. Equation 1 above is a value that measures how well two groups (Y=0 or Y=1) are separated by the feature (x). Here, the features that are the target of feature selection are illustrated in FIG. 6. Typically, features with an information value of 0.1 or higher can be considered important features.

In this case, the selection unit 331 may select a set number of features, for example, five features, in order of high information value. Or, as shown above, you can select all features with an information value of 0.1 or more. In the embodiment of the present invention, 5 features were selected among 56 features shown in FIG. 6. Here, information on the 56 features shown in FIG. 6 can be calculated from individual subscriber information and event information by the selection unit 331.

2. Fine Classing Step (Step S720)

Detailed classification is the step of dividing features with continuous values into classification values in a range of 20 or less. For example, if the standard feature is power usage, and 95% of the power usage data of all subscribers is between 250kWh and 450kWh, the range of one detailed classification is 10kWh ((450kWh - 250kWh)/20 ) has a gap. In total, if a feature has less than 20 values (for example, the number of engagements of a particular user over the last 10 demand response events), the engagement count itself as an integer is subject to detailed classification.

3. Coarse Classing Step (Step S730)

The selection unit 331 merges some classifications from the detailed classification and adjusts the classification into a rough classification. While detailed classification provides a simple, linear way to define a classification, maintaining a good balance between the number of values (population) corresponding to the classification is important for explainability. Rough classification merges the classification steps into a more stable and statistically significant classification.

In this case, credit scoring approximates the probabilities of the two groups by logistic regression. At this time, the logit value of logistic regression is in a linear relationship with the SCORE variable. (See Equation 4 below) The SCORE variable is the score value of each classification for each feature. Equation 2 and Equation 3 are steps for calculating the logit value of Equation 4.

(Equation 2)

(Equation 3)

(Equation 4)

_In this case _, Additionally, the left side of Equation 4 is the logit value. In order to increase explainability by using a smaller number of classification steps, the selection unit 331 rounds the logit value calculated using the detailed classification step and then integrates the classification steps with the same value.

4. Score adjustment step through machine learning (S740)

The selection unit 331 finally creates a credit scoring model through score adjustment through machine learning.

According to the present invention, the score is calculated by dividing a value between 0 and 100 into steps of 0.05, which is a scaling factor. In other words, the value obtained by dividing 0 to 100 into 2000 steps is set as the SCORE value, and then a meaningful score range is set. In an embodiment of the present invention, for example, the score range is divided based on an increase of 0.001 in AUROC (Area Under Receiver Operation Characteristics).

Additionally, in an embodiment of the present invention, machine learning was performed based on GBM (Gradient Boosting Method). There may be various machine learning methods such as LR (Logistic Regression), Neural Network, and NB (Naive Bayes), but as a result of review, machine learning based on GBM showed the best performance.

The final features selected according to the embodiment of the present invention and the classification step (Range) and score classification according to the features are shown in Figure 7.

Referring to Figure 7, the five features selected among the 56 demand response-related features shown in Figure 6 are Participation Rate, Maximum Number of Consecutive Participation, Cumulative Rebate, and Maximum Maximum Number of Consecutive Non-Participation and Success in the Last Event (failure to participate in the last event or reduce power).

of the selection department (331) Credit score evaluation mode

The selection unit 331 selects subscribers to be invited to the demand response event among subscribers based on the features and scores selected in the above credit evaluation scoring learning mode.

For example, if the participation rate in the demand response event to date is 0.6, the maximum number of consecutive participation is 3, the cumulative reward is 500, the maximum number of consecutive non-participation is 6, and the user participated in the last event, refer to Figure 7, You will obtain a credit score of 75+28.35+17.75+9.1+40=95.95 points.

Meanwhile, referring again to FIG. 3, the transmitter 322 may transmit the guidance information (①) regarding power usage reduction to the subscriber terminal 400 selected by the selection unit 331.

For example, the guidance information includes the time period when power saving is required, the amount of power required to be saved, the amount of power that can be saved for all power consumption devices or individual power consumption devices, and a guide for reducing power use (e.g., how many degrees to set the air conditioner temperature to (lower, increase refrigerator refrigeration/freezing temperature, unplug unused electronic appliances, etc.), compensation information for reducing power use, etc. may be included.

The monitoring unit 332 may monitor power usage by the power consumption device of the selected subscriber after transmitting the guidance information. This is to determine whether the power use by the subscriber's power consumption devices has actually decreased or to what extent the amount of power has been reduced in accordance with the guidance information, and to grant compensation later based on this.

If the power usage by the selected subscriber's power consumption device decreases according to the transmitted guide information, the compensation management unit 333 may grant a predetermined compensation to the subscriber in proportion to the reduced power usage. More specifically, the compensation management unit 333 may grant compensation for power usage fees.

For example, the compensation management unit 333 may grant compensation in the form of a discount on power usage fees or payment of points that can be used to pay power usage fees. Alternatively, the compensation management unit 333 may grant cash, gifts, coupons, etc. received from the sponsor server 610 as compensation.

Furthermore, the transmitter 322 may transmit information on power savings (hereinafter referred to as savings information) according to guide information for each subscriber to the power supply-related server 600, and the power supply-related server 600 may send the power saving information. It is possible to determine whether or not power is saved and compensation corresponding to the amount of power saved, and transmit related information to the demand response operation server 300. Accordingly, the compensation management unit 333 may grant compensation to the corresponding subscriber based on information about compensation received from the power supply-related server 600.

Meanwhile, the compensation management unit 333 predetermines compensation grant criteria based on whether or not power is saved according to the guidance information and the amount of power saved, and may grant compensation to the relevant subscriber based on the compensation grant criteria.

The transmitter 322 may transmit information regarding the granted compensation (hereinafter referred to as compensation information) (②) to the communication device 400 of the subscriber.

FIG. 4 shows a configuration diagram of the subscriber terminal 400 having the power demand management function shown in FIG. 1.

According to FIG. 4, the subscriber terminal 400 includes a communication unit 410 that receives guidance information on reducing power use from the demand response operation server 300, and a control unit that executes a power-related application upon receiving the guidance information ( 420) and an output unit 430 that outputs the guide information by executing the power-related application. Additionally, the communication device 400 may further include a storage unit 450 that stores power-related applications, guidance information, compensation information, etc.

Here, the control unit 420 may control the operation of at least one of the communication unit 410, output unit 430, input unit 440, and storage unit 440.

Here, the power-related application refers to an application for managing the power usage of power consumption devices owned by the subscriber of the subscriber terminal 400, and is installed at the time of production of the subscriber terminal 400 or downloaded from an external server at the subscriber's selection. and can be installed.

When receiving guidance information, the control unit 420 may automatically execute a power-related application, or provide a notification of receipt of guidance information to the subscriber and then execute the power-related application only when an execution command is input from the subscriber.

For example, the guidance information reception notification may display brief contents of the guidance information, or may output a vibration/lamp/notification sound simply to indicate whether or not the information has been received.

In addition, the communication unit 410 receives information about compensation granted (hereinafter referred to as compensation information) corresponding to the reduction in power use according to the guidance information from the demand response operation server 300, and the output unit 430 receives power The received compensation information can be output by executing the relevant application.

Regarding the notification of receipt of compensation information, the description of the notification of receipt of the above-described guidance information can be cited, so detailed description thereof will be omitted.

Therefore, according to the present invention, a server, communication device and system having a power demand management function that can guarantee explainability while utilizing an efficient target selection process when operating a demand response program, and a power demand management method thereof are provided. .

Additionally, according to the present invention, subscribers who are likely to participate in a demand response event can be encouraged to participate in the demand response event, and thus power use can be reduced stably.

The above description is merely an illustrative explanation of the technical idea of the present invention, and various modifications, changes, and substitutions can be made by those skilled in the art without departing from the essential characteristics of the present invention. will be.

Accordingly, the embodiments disclosed in the present invention and the attached drawings are not intended to limit the technical idea of the present invention, but are for illustrative purposes, and the scope of the technical idea of the present invention is not limited by these embodiments and the attached drawings. . The scope of protection of the present invention should be interpreted in accordance with the claims below, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of rights of the present invention.

Claims (21)

A power measuring device that measures and supplies power usage for at least one power consumption device of a subscriber in real time;
A power supplier server that transmits a request signal for power use reduction; and
When receiving the request signal, a demand response operation server that selects a subscriber who wishes to request power usage reduction;
The demand response operation server,
a storage unit that stores subscriber information including power usage for at least one power consumption device of the subscriber and a record of the subscriber's participation in a demand response event, and event information including the date of the demand response event;
A receiving unit that receives a request signal for reducing power use;
a selection unit that, upon receiving the request signal, selects a subscriber who wishes to request power usage reduction using the subscriber information and event information;
It includes a transmitter that transmits guidance information about the requested power usage reduction to the subscriber terminal of the selected subscriber,
The selection unit generates a credit scoring model, evaluates the subscriber's credit score according to the credit scoring model, and selects a subscriber who wishes to request reduction in power use according to the evaluated credit score,
The credit scoring model selects features based on information values for a plurality of features generated based on subscriber information, divides the selected features into a plurality of classification steps, and then Rough classification into fewer classification steps by merging at least some of them,
A system with power demand management functions. According to claim 1,
The credit scoring model is created by calculating the logit value for the selected features, roughly classifying the features, and then generating a score range for each range of the rough classification through machine learning.
A system with power demand management functions. According to claim 2,
The rough classification is generated by classifying the selected features in detail, rounding the logit value for the detailed classification range, and integrating the detailed classification range when they have the same logit value.
A system with power demand management functions. According to claim 3,
The detailed classification range is 20 or less for one feature,
A system with power demand management functions. According to claim 2,
The above selected features include Participation Rate, Maximum Number of Consecutive Participation, Cumulative Rebate, Maximum Number of Consecutive Non-Participation and success in past events. Including whether (Success in the Last Event, participation in the last event or failure to reduce power),
A system with power demand management functions. According to claim 1,
The selection unit generates a plurality of features based on the subscriber information and the event information.
A system with power demand management functions. According to claim 1,
a monitoring unit that monitors power usage by the selected at least one power consumption device; and
Further comprising a compensation management unit that grants a predetermined compensation to the subscriber when the power usage by the power consumption device decreases according to the guide information,
A system with power demand management functions. According to claim 7,
The transmitting unit transmits information about the granted compensation to the subscriber terminal of the corresponding subscriber,
A system with power demand management functions. According to claim 1,
The information value is defined by Equation 1 below,
(Equation 1)

Here, IV is the information value, Pr() is the probability function, x is the feature, Y=1 and Y=0 are the groups that must be distinguished by feature x,
A system with power demand management functions. According to claim 2,
The logit value is calculated by Equation 4 below,
(Equation 4)

(Here,
A system with power demand management functions.



delete delete delete delete delete delete delete delete delete delete delete

Images