KR102358532B1 - Apparatus and method for predicting energy use and generation through self-enhancement learning - Google Patents

Abstract

Disclosed are an apparatus and method for predicting energy usage and power generation through self-enhancement learning. The energy usage and generation amount prediction method includes the steps of: determining whether there is missing data in input data, and processing the missing data if there is missing data; retrieving data related to a target for estimating energy usage and energy generation amount from input data in which missing data has been processed; synchronizing the retrieved data according to a time reference point or a spatial reference point; labeling the synchronized data; filtering the labeled data to extract data corresponding to the target; and analyzing the extracted data to predict the energy usage and energy generation of the target.

Description

Apparatus and method for predicting energy use and power generation through self-enhancement learning

The present invention relates to an apparatus and method for predicting energy usage and solar power generation using self-enhanced learning.

Renewable energy, including solar energy, has a fatal disadvantage that energy is generated intermittently and the amount of usable energy is limited. This is a fatal disadvantage when compared to power generation using fossil fuels, which can be used whenever necessary.

For example, the use of solar energy is limited because only the remaining energy of the electrical energy obtained from solar cells during the day can be used at night. Also, in cloudy weather, the energy is intermittent, so its use is bound to be more limited. To compensate for these shortcomings, it is necessary to predict the future solar power generation energy as accurately as possible and use the solar power generation energy efficiently according to the prediction result.

Conventional solar power generation prediction methods use Regressive, ANN, and SVR. For example, a hybrid-statistical method among conventional methods of predicting solar power generation is a method of predicting power generation by using a plurality of different machine learning techniques together. In addition, Hybrid-physical predicts the amount of solar power generation by using both the physical method using the physical configuration of the solar system and the prediction using machine learning.

The more accurate the prediction of the amount of solar power generation and energy consumption, the more efficiently the solar power generation energy can be used. Therefore, a method of increasing the prediction accuracy of the amount of solar power generation and energy usage has been requested.

The present invention may provide an apparatus and method for estimating energy usage and power generation.

In addition, the present invention can provide an apparatus and method capable of increasing the accuracy of the analysis result by making the prediction result into knowledge.

A method for estimating energy usage and power generation according to an embodiment of the present invention includes the steps of: checking whether there is missing data in input data, and if there is missing data, processing the missing data; retrieving data related to a target for estimating energy usage and energy generation amount from input data in which missing data has been processed; synchronizing the retrieved data according to a time reference point or a spatial reference point; labeling the synchronized data; filtering the labeled data to extract data corresponding to the target; and analyzing the extracted data to predict the energy usage and energy generation of the target.

According to an embodiment of the present invention, it is possible to predict energy usage and power generation by pre-processing and analyzing measurement data collected from energy devices and sensors and knowledge data collected from a big data platform.

In addition, according to an embodiment of the present invention, by storing and managing the prediction result into knowledge, when the energy consumption and power generation amount are predicted in the future, the accuracy of the analysis result can be increased.

1 is a diagram illustrating an apparatus for predicting energy usage and power generation according to an embodiment of the present invention.
2 is an example of database servers according to an embodiment of the present invention.
3 is a flowchart illustrating a method for predicting energy usage and generation amount according to an embodiment of the present invention.
4 is a flowchart illustrating a data pre-processing process of a method for predicting energy usage and power generation according to an embodiment of the present invention.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes may be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all modifications, equivalents and substitutes for the embodiments are included in the scope of the rights.

The terms used in the examples are used for the purpose of description only, and should not be construed as limiting. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present specification, terms such as "comprise" or "have" are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It is to be understood that this does not preclude the possibility of addition or existence of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

In addition, in the description with reference to the accompanying drawings, the same components are given the same reference numerals regardless of the reference numerals, and the overlapping description thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description thereof will be omitted.

The energy usage and generation amount prediction method according to an embodiment of the present invention may be performed by an energy usage and generation amount prediction apparatus.

1 is a diagram illustrating an apparatus for predicting energy usage and power generation according to an embodiment of the present invention.

The energy usage and generation amount prediction apparatus 100 may include a data collection unit 110 , a data preprocessor 120 , and a data analysis unit 130 as shown in FIG. 1 . In this case, the data collection unit 110 is a communicator, and the data preprocessor 120 and the data analysis unit 130 may be different processes or each module included in a program performed by one process.

The data collection unit 110 may collect input data by searching for input data from at least one of the sensor 101 , the Korea Meteorological Administration server 102 , the Internet 103 , and the database server 140 . At this time, the input data collected by the data collection unit 110 includes current metering data and state data collected from the device or sensor 101 , weather-related climate data collected from the Meteorological Administration server 102 , and social It may include at least one of context data collected through the media and the Internet 103 , policy data collected from the database server 140 , and knowledge data.

In this case, the measured data and the state data may be data measured at a current time by the sensor 101 capable of measuring data related to the amount of solar power generation. Also, the context data may include local data related to a local location. And, the policy data may include power unit price, equipment unit cost, depreciation cost, and the like. In addition, the knowledge data may include analysis results accumulated from the point in time when the energy usage and generation amount prediction apparatus 100 starts to operate to the present.

In addition, the data collection unit 110 may independently collect data from the sensor 101 , the Korea Meteorological Administration server 102 , the Internet 103 , and the database server 140 . In addition, the data collection unit 110 operates independently of the operations of the data preprocessor 120 and the data analysis unit 130 , and may collect data in real time or at regular time intervals.

The data preprocessing unit 120 may preprocess the data collected by the data collecting unit 110 so that the data analyzing unit 130 can analyze it. In this case, the data preprocessor 120 may check whether there is missing data in the input data. And, when there is missing data, the data preprocessor 120 may process the missing data.

Also, the data preprocessor 120 may search for data related to a target for predicting energy usage and energy generation from input data in which missing data is processed. Next, the data preprocessor 120 may synchronize the searched data according to a time reference point or a spatial reference point. Next, the data preprocessor 120 may label the synchronized data. Finally, the data preprocessor 120 may extract data corresponding to the target by filtering the labeled data.

The data analyzer 130 may analyze the data preprocessed by the data preprocessor 120 to predict the energy usage and energy generation of the target. In this case, the data analyzed by the data analysis unit 130 may be data corresponding to a target extracted by filtering the labeled data.

In addition, the data analysis unit 130 may provide a visualization service by visualizing the analysis result of the data. In addition, the data analysis unit 130 may store and manage the analysis result and the data visualized by the analysis result in the database server 140 through the knowledgeization technique.

The database server 140 may be a big data platform that receives, stores, and manages analysis results from the energy usage and power generation prediction apparatuses 100 that predict energy usage and power generation in different regions.

In addition, the database server 140 may include an energy consumption DB (Database) server 141, an energy production DB server 142, a cost DB server 143, and a policy DB server 144 as shown in FIG. have. A detailed configuration of the database server 140 will be described in detail with reference to FIG. 2 below.

The energy usage and generation amount prediction apparatus 100 may predict energy usage and generation amount by pre-processing and analyzing measurement data collected from energy devices and sensors and knowledge data collected from a big data platform. In addition, the energy usage and generation amount prediction apparatus 100 may increase the accuracy of the analysis result when predicting the energy usage and generation amount in the future by storing and managing the prediction results as knowledge.

2 is an example of database servers according to an embodiment of the present invention.

The energy consumption DB server 141 may store and manage data related to energy consumption prediction.

The energy production DB server 142 may store and manage data related to the prediction of the generation amount of each of various power generation devices such as renewable energy.

The cost DB server 143 may store and manage data related to a cost for generating energy from a power generation device, a cost related to energy consumption, and a profit related to energy production and consumption.

The policy DB server 144 may store and manage data related to government policies and current trends in energy technology.

Energy consumption DB server 141, energy production DB server 142, cost DB server 143, and policy DB server 144, respectively, as shown in FIG. It may consist of raw data including pattern data including trends, representative algorithm results, and analysis results received from the energy usage and power generation prediction apparatus 100, and raw data including raw data.

And, the energy consumption DB server 141, the energy production DB server 142, the cost DB server 143, and the data stored in the policy DB server 144 is the energy usage and generation amount prediction device 100 is the sensor 101, the Korea Meteorological Administration It is possible to increase the accuracy of the data analysis process by being used again in the preprocessing process together with the data collected in real time from the server 102 and the Internet 103 .

3 is a flowchart illustrating a method for predicting energy usage and generation amount according to an embodiment of the present invention.

In step 310 , the data collection unit 110 searches for and collects input data from at least one of the sensor 101 , the Korea Meteorological Administration server 102 , the Internet 103 , and the database server 140 .

In step 320 , the data preprocessor 120 may preprocess the data collected in step 310 so that the data analyzer 130 can analyze it. In this case, the data preprocessor 120 may check whether there is missing data in the input data. And, when there is missing data, the data preprocessor 120 may process the missing data.

In step 330 , the data analysis unit 130 may analyze the data preprocessed in step 320 to predict the target's energy usage and energy generation.

In operation 340 , the data analysis unit 130 may provide a visualization service by visualizing the analysis result of the data predicted in operation 330 .

In step 350 , the data analysis unit 130 stores the analysis result and the visualized data in the database server 140 to learn the analysis result and the visualized data.

4 is a flowchart illustrating a data pre-processing process of a method for predicting energy usage and generation according to an embodiment of the present invention. Steps 410 to 450 of FIG. 4 may be included in step 320 of FIG. 3 .

In step 410 , the data preprocessor 120 may check whether there is missing data in the input data, and if there is missing data, process the missing data.

Specifically, the data preprocessor 120 may perform an ingestion process for generating a data field by storing the data collected by the data collecting unit 110 . In this case, the data preprocessor 120 may check and process the missing or missing data from the input data. For example, when missing data occurs in the metering data, the data preprocessor 120 may process an average value of data adjacent to the missing data as a value of the missing data. Also, when missing data occurs in the state data, the data preprocessor 120 may process data of a previous time adjacent to the missing data or data of a later time adjacent to the missing data by replacing the missing data with a value of the missing data.

In step 420 , the data preprocessor 120 may search for data related to a target for estimating energy usage and energy generation from the input data from which the missing data has been processed in step 410 . In this case, the data searched by the data preprocessor 120 may include data related to a region where the target is installed.

In operation 430 , the data preprocessor 120 may synchronize the data searched for in operation 420 by aligning the data according to a time reference point or a spatial reference point.

Data collected through an independent energy device or sensor 101 needs time synchronization. However, since global time synchronization is difficult due to the time difference, the data preprocessor 120 may synchronize data based on an event. In addition, data synchronized based on the event may each have a correlation.

Accordingly, the data preprocessor 120 may set reference points of time series data and spatial data according to an energy event. For example, the data preprocessor 120 may set an event such as an explosion in energy usage or an energy device failure or climate change as a reference point for time series data and spatial data.

In step 440 , the data preprocessor 120 may label the data synchronized in step 430 to give the data an event meaning.

In operation 450 , the data preprocessor 120 may extract data corresponding to the target by filtering the data labeled in operation 440 . In this case, the data preprocessor 120 may extract data corresponding to the purpose of analysis from among the labeled data.

The present invention can predict energy usage and power generation by pre-processing and analyzing measurement data collected from energy devices and sensors and knowledge data collected from a big data platform. In addition, the present invention can increase the accuracy of the analysis result when predicting the amount of energy consumption and power generation in the future by storing and managing the prediction results as knowledge.

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

Software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or apparatus, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

As described above, although the embodiments have been described with reference to the limited drawings, those skilled in the art may apply various technical modifications and variations based on the above. For example, the described techniques are performed in an order different from the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

110: data collection unit
120: data preprocessor
130: data analysis unit

Claims (9)

retrieving and collecting input data from energy devices, sensors, and big data platforms;
pre-processing the input data; and
Predicting the energy use and energy generation of the target to predict the amount of energy generation by analyzing the pre-processed input data
including,
The input data is
It includes measurement data collected from energy devices and sensors and knowledge data collected from big data platforms,
The pre-processing step is
determining whether there is missing data in the input data, and if there is missing data, processing the missing data;
retrieving data related to a target for estimating energy usage and energy generation amount from input data in which missing data is processed;
synchronizing the retrieved data according to a time reference point or a spatial reference point;
labeling the synchronized data; and
Filtering the labeled data to extract data corresponding to the target
including,
The predicting step is
An energy usage and generation amount prediction method for predicting the energy usage and energy generation amount of the target by analyzing the extracted data. delete According to claim 1,
The step of processing the missing data includes:
When missing data occurs in the metering data, an energy usage and generation amount prediction method that treats the average value of the data adjacent to the missing data as the value of the missing data. According to claim 1,
The step of processing the missing data is
When missing data occurs in the state data, an energy usage and generation amount prediction method that replaces the data of the previous time adjacent to the missing data or the data of the later time adjacent to the missing data with the value of the missing data. According to claim 1,
The big data platform is
A database server that receives, stores, and manages the predicted energy usage and energy generation of the target.
A method of predicting energy usage and power generation to predict, including. 6. The method of claim 5,
The database server,
an energy consumption DB server for storing and managing data related to the target's energy consumption prediction;
an energy production DB server that stores and manages data related to the prediction of the generation amount of each of the power generation devices;
Cost DB server for storing and managing data related to cost for generating energy from the power generation device, cost related to energy consumption, and profit related to energy production and consumption; and
Policy DB server that stores and manages data related to government policies and current energy technology trends
Energy usage and generation amount prediction method comprising at least one of. According to claim 1,
The knowledge data is
A method of predicting energy use and power generation, comprising at least one of weather-related climate data collected from the Meteorological Agency server, situation data collected through social media and the Internet, policy data collected from a database server, and knowledge data. 8. The method of claim 7,
The context data includes local data related to a local location,
The policy data is an energy usage and generation amount prediction method including power unit cost and facility unit cost and depreciation cost. a data collection unit that retrieves and collects input data from an energy device, a sensor, a meteorological agency server, the Internet, and a database server;
a data preprocessor for preprocessing the input data; and
A data analysis unit that analyzes the preprocessed input data to predict the energy usage and energy generation of the target to predict the energy generation amount
including,
The input data is
Includes measurement data collected from energy devices and sensors and knowledge data collected from Meteorological Agency servers, Internet and database servers,
The data preprocessor,
Check whether the input data contains missing data, if there is missing data, process the missing data;
retrieving data related to a target for estimating energy usage and energy generation from the missing data-processed input data, and synchronizing the retrieved data according to a temporal reference point or a spatial reference point;
Label the synchronized data, filter the labeled data to extract data corresponding to the target,
The data analysis unit,
An energy usage and power generation prediction device for predicting the energy usage and energy generation amount of the target by analyzing the extracted data.

Images