KR20170080384A - System and method for predicting energy consumption based on cloud - Google Patents

Abstract

The present invention relates to a cloud based energy usage prediction system and method.
According to an aspect of the present invention, there is provided a cloud-based energy usage forecasting system including at least one facility management apparatus for providing facility status information on at least one individual facility installed in a building; A cloud server providing big data associated with at least one of said at least one facility management device and said building; And an energy management device for receiving the facility status information and the big data, and for predicting energy usage for the at least one individual facility and the building.

Description

[0001] SYSTEM AND METHOD FOR Predicting Energy Consumption Based on Cloud [

The present invention relates to a cloud-based energy usage forecasting system and method, and more particularly, to a cloud-based energy consumption forecasting system for estimating and managing energy consumption consumed in buildings, factories, and the like using cloud-based big data ≪ / RTI >

BEMS (Building Energy Management System) is a system that predicts energy consumption in buildings and factories and manages them efficiently.

The method of predicting the energy consumption of the BEMS according to the prior art is based on the operation information such as the operating state and the operating time of the individual facility (refrigerator, heat exchanger, cooling tower, etc.) Based on the predicted energy usage of the individual facilities and the combined energy usage estimates of these individual facilities, we estimate the total energy consumption in the building.

However, according to the conventional BEMS energy consumption prediction method, the prediction accuracy is somewhat reduced because the correlation between the predicted values of the energy consumption for individual facilities is not taken into consideration at all. Thus, the predicted value of the total energy consumption The efficiency of the energy management system is relatively low due to the lack of accuracy.

In addition, since the energy consumption predicting method according to the related art generates the energy consumption prediction value based on the past data, the prediction accuracy is deteriorated over time without considering various factors changing in real time in relation to the individual facility. There has been a problem in that it is not possible to provide an efficient energy consumption prediction function to users when the errors in the summer and the winter, which are the seasons in which power consumption of the heating and cooling increases, increase.

Korean Patent Publication No. 10-2014-0116619

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a system and method for realizing a real- (Hereinafter referred to as " energy consumption prediction system ").

It is another object of the present invention to provide a cloud-based energy usage forecasting system and method that can improve the prediction accuracy of energy consumption by analyzing the correlation between energy usage for individual facilities running in a building.

To this end, a cloud-based energy usage forecasting system according to an aspect of the present invention includes: at least one facility management apparatus for providing facility status information on at least one individual facility installed in a building; A cloud server providing big data associated with at least one of said at least one facility management device and said building; And an energy management device for receiving the facility status information and the big data, and for predicting energy usage for the at least one individual facility and the building.

An energy management apparatus for a cloud-based energy usage prediction system according to an embodiment of the present invention includes a data analysis unit for analyzing information transmitted from a cloud server; And setting a predictive model for at least one individual facility and a predictive model for the building in the building based on the results of the analysis of the at least one individual facility, And a prediction model management unit for estimating an energy usage amount by simulating the prediction model.

Meanwhile, a cloud-based energy consumption prediction method according to an exemplary embodiment of the present invention includes: setting a prediction model for at least one individual facility in a building and a prediction model for the building; Simulating a prediction model for the at least one individual facility based on information transmitted from the cloud server to firstly predict energy usage for the at least one individual facility; Simulating a prediction model for the building based on a result of the first energy estimation for the at least one individual facility to firstly predict an energy usage for the building; And finally estimating the energy usage for the at least one individual facility and the building by reflecting the correction variables derived based on the energy usage primary prediction results for the at least one individual facility and the building .

According to the present invention, by utilizing cloud-based big data to reflect various factors that change in real time in relation to an individual facility, it is possible to more accurately predict energy consumption consumed in a building and provide optimal facility operation guidance And has an effect of efficiently managing energy.

In addition, according to the present invention, the correlation between the energy consumption of individual facilities operating in the building is analyzed and reflected, thereby improving the prediction accuracy of the energy consumption further.

1 is a configuration diagram of a cloud-based energy usage forecasting system according to an embodiment of the present invention.
2 is a configuration diagram of an energy management apparatus according to an embodiment of the present invention.
3 is a configuration diagram of a prediction model management unit according to an embodiment of the present invention.
4 is a flowchart of a cloud-based energy usage forecasting method according to an embodiment of the present invention.
5 is a detailed flowchart of step S410 of FIG.
6 is a detailed flowchart of step S420 of FIG.
7 is a detailed flowchart of step S430 of FIG.
8 is a detailed flowchart of step S440 of FIG.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings and preferred embodiments. In the following description, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention unnecessarily obscure.

1 is a configuration diagram of a cloud-based energy usage prediction system according to an embodiment of the present invention.

1, a cloud-based energy usage forecasting system according to an embodiment of the present invention includes at least one facility management device 110, 120, or 130, a cloud server 200, an energy management device 300, or the like do.

The facility management devices 110, 120, and 130 are devices for managing individual facilities provided in a building. The facility management apparatuses 110, 120, and 130 may be installed in various locations in the building such as a refrigerator, a heat exchanger, a cooling tower, an air conditioner, a lighting, an entrance door, (Hereinafter referred to as " facility status information ") by directly receiving information such as operation time, etc., or sensing the operation state of the individual facility from various sensors installed in the building. The facility management devices 110, 120, and 130 transmit the collected equipment state information to the energy management device 300 through the cloud server 200. [ Of course, the facility status information may be transmitted from the facility management apparatuses 110, 120, and 130 directly to the energy management apparatus 300.

For reference, Table 1 below illustrates concrete data linkage between the facility management apparatuses 110, 120, and 130 and the energy management apparatus 300. [

[Table 1]

Meanwhile, the cloud server 200 includes various pieces of information (hereinafter, referred to as 'big data') that can be obtained in a cloud service environment including facility status information transmitted from the facility management devices 110, 120 and 130 It is a device that collects and provides.

Big data includes, for example, information transmitted from a terminal (e.g., a smart phone) of a user who is working or currently located in a building, information on the Internet that is relevant to the building, information on the Internet that is related to energy, In the case of the invention, information on the building human condition related to the number and specific location of the people in the current building, building environment information related to the weather, temperature, humidity and / or the temperature and humidity inside the building, Building energy information related to energy efficiency inherent to the building according to location, direction, exterior, insulation, window, heat capacity, building industry information related to purpose of use, use of building, type of industry, efficiency by energy resource, And the like.

The energy management apparatus 300 is provided in an energy integrated management center for managing all buildings (buildings, factories, etc.) in a predetermined region, for example, and receives various information related to buildings from the cloud server 200 State information, building personal status information, building environment information, building energy characteristic information, building industry information, and energy resource information), and predicts energy usage for the entire building and / or individual facilities based on the information. The energy management apparatus 300 generates the facility operation guidance based on the predicted energy usage value and transmits the generated guidance to the individual facility management apparatuses 110, 120, and 130 to efficiently and efficiently store the energy consumed in the entire building and / Management.

Specifically, the energy management apparatus 300 provides a high-quality service at low cost from a customer's viewpoint using the cloud technology, monitors energy consumption status of buildings in a predetermined area in real time, and employs a big data analysis technique, We analyze and predict the energy consumption forecasting model of each building and the energy consumption prediction model of each unit. In addition, it visualizes the buildings registered as the building group, monitors the energy consumption consumption consumed in each building, and thereby can calculate the total amount of energy consumed in the specific area.

In this regard, FIG. 2 is a configuration diagram of an energy management apparatus according to an embodiment of the present invention.

2, an energy management apparatus 300 according to an exemplary embodiment of the present invention includes a prediction model management unit 310, a data analysis unit 320, a data storage unit 330, a display unit 340, 350) and the like.

The Prediction Model Manager 310 manages a prediction model (hereinafter referred to as an 'individual facility prediction model') for individual facilities provided in a building based on various information collected through the cloud server 200, (Hereinafter, referred to as a 'building prediction model') to predict energy consumption, generate savings items and guidance, and perform efficient planning and modification of facilities in a building.

In this regard, FIG. 3 is a configuration diagram of a predictive model management unit according to an embodiment of the present invention.

3, the prediction model management unit 310 according to an embodiment of the present invention includes a life cycle management module 311, a correlation setting module 312, an input / output variable management module 313, A variable management module 314, a filtering module 315, a simulation module 316, a unit integration module 317, an MES integration module 318, and the like.

The lifecycle management module 311 manages the life cycle such as creation, modification and deletion of at least one individual facility (lower level) prediction model and / or building (upper level) prediction model. The lifecycle management module 311 may prioritize the individual facility prediction models at the time of generating the prediction model, and may define the one-level prediction model by combining the related prediction models.

The correlation setting module 312 analyzes the correlation between the individual facility prediction models and / or the building prediction model to set the correlation coefficient. Correlation analysis is performed, for example, by using Pearson correlation analysis, Sprearman's rank correlation analysis, partial correlation analysis, and the like for the input and output variables of each prediction model .

The input / output variable management module 313 sets an input variable and an output variable applied to the individual facility and / or the building prediction model, and calculates the variable value. For example, in the case of a lower level refrigerator prediction model, the input variable may be the operating state of the refrigerator, the operating time, and the like, and the output variable may be the power consumption of the refrigerator. Also, in the case of the present invention, by applying a correlation between prediction models of individual facilities (e.g., refrigerator, heat exchanger, cooling tower), the output variable of the first facility prediction model can be set as an input variable of the second facility prediction model have. For example, the amount of power consumption, which is an output variable of the heat exchanger prediction model, can be used as an input variable of the refrigerator prediction model. On the other hand, the input variable of the upper level building prediction model can be the output variable of the lower level prediction models belonging to the building.

The correction parameter management module 314 sets correction parameters to be applied to the individual facility prediction model and / or the building prediction model based on the information obtained through the analysis of the big data, and calculates the variable values. Examples of correction parameters are the weather, temperature, humidity, temperature inside the building, humidity in the place where the building is located, the weather, the number of people in the building, the location, orientation, exterior, insulation, capacity, purpose of use of the building, use, type of business, and the like.

The filtering module 315 generates a filtered result using a Kalman filter, a Gaussian filter, or the like when filtering is required for input / output variables and / or correction variables. For reference, the Kalman filter can generate continuous measurement values assuming that the state at a specific point in time has a linear relationship with the state at the previous point. Gaussian filter A stochastic representation of the result can be obtained.

The simulation module 316 includes an algorithm (e.g., an algorithm for estimating energy usage, an algorithm for calculating the unit price for each real-time energy type, an algorithm for calculating the cost of the operating facility for each hour) applied to the individual facility and / To simulate the individual plant and / or building predictive model.

The unit integration module 315 calculates the energy usage by applying unified unit standards (for example, Tonnage of Oil Equivalent (TOE) and carbon emission (kgCo2 / kWh)) between prediction models.

The MES interworking module 318 sets independent variables that can be interworked with the MES (Manufacturing Execution System).

Referring again to FIG. 2, the data analysis unit 320 analyzes information received from the cloud server 200. As described above, the information received from the cloud server 200 includes the equipment status information transmitted by the facility management apparatuses 110, 120, and 130 and the big data obtained in the cloud service environment, Building information, building energy information, building industry information, and energy resource information. Accordingly, the data analysis unit 320 extracts specific information for calculating the input variable and / or the correction variable for the individual facility and / or building using the big data analysis technique or the like.

The data storage unit 330 stores information managed by the prediction model management unit 310 (e.g., prediction model, input variable, output variable, correction variable) and information analyzed by the data analysis unit 320.

The display unit 340 displays various results (for example, energy consumption and energy consumption comparison) implemented in the energy management apparatus 300 so that the user can view and manage the real-time energy usage amount and the prediction result of the entire building and / , Energy usage statistical analysis, management cost analysis, facility operation cost).

The communication unit 350 includes a wire / wireless communication module and communicates with the cloud server 200, the facility management devices 110, 120, and 130, and communicates with an individual terminal through a web, a mobile, So that the user or manager can view the relevant information anytime, anywhere.

The control unit 360 controls the entirety of the building and / or each facility according to the overall control of the prediction model management unit 310, the data analysis unit 320, the data storage unit 330, the display unit 340, Perform energy management and usage forecasting.

Meanwhile, FIG. 4 is a flowchart of a cloud-based energy management method according to an embodiment of the present invention. 4 is a detailed flow chart of step S420 of FIG. 4, FIG. 7 is a detailed flowchart of step S430 of FIG. 4, and FIG. 8 is a detailed flowchart of step S430 of FIG. A detailed flowchart of step S440.

In step S410, the energy management apparatus sets up a prediction model for individual facilities (lower level) and building (upper level).

Specifically, referring to FIG. 5, in step S412, the energy management apparatus sets input / output variables of the individual facility prediction model. For example, in the case of the low-level refrigerator prediction model, the operating state and the operating time of the freezer are set as input variables and the power consumption of the freezer is set as an output variable. In the case of the low-level heat exchanger prediction model, the operating state and the operating time of the heat exchanger are set as input variables, and the power consumption of the heat exchanger is set as an output variable. In the case of the lower-level cooling tower prediction model, the operating state and the operating time of the cooling tower are set as input variables, and the power consumption of the cooling tower is set as an output variable.

In step S414, the energy management apparatus establishes a correlation between individual facility prediction models. For example, the correlation coefficient is set by analyzing the correlation between the power consumption amount of the refrigerator prediction model, the power consumption amount of the heat exchanger prediction model, and the power consumption amount of the cooling tower prediction model. Then, for example, the output variable of the heat exchanger (first facility) prediction model is set as an input variable of the refrigerator (second facility) prediction model in accordance with the correlation between the individual facility prediction models. For reference, step S414 may be performed prior to or simultaneously with step S412.

In step S416, the energy management apparatus sets input / output variables of the building prediction model. For example, the output variable of the refrigerator prediction model, the output variable of the heat exchanger prediction model, and the output variable of the cooling tower prediction model are set as input variables of the building prediction model, and the power consumption of the whole building is set as the output variable of the building prediction model.

Referring again to FIG. 4, in step S420, the energy management device performs energy usage prediction (primary prediction) for the individual facility prediction model.

Specifically, referring to FIG. 6, in step S422, the energy management apparatus analyzes information transmitted from the cloud server 200. FIG. As described above, the information transmitted from the cloud server 200 includes the equipment status information transmitted by the facility management apparatuses 110, 120, and 130 and the big data obtained in the cloud service environment. Building information, building energy information, building industry information, and energy resource information.

In step S424, the energy management apparatus calculates the input variable value for the individual facility prediction model based on the information analyzed in step S422.

Then, in step S426, the energy management apparatus firstly predicts the energy usage for the individual facility by simulating the individual facility prediction model by applying the input variable value calculated in step S424.

Referring again to FIG. 4, in step S430, the energy management device performs energy usage prediction (primary prediction) on the building prediction model.

Specifically, referring to Fig. 7, in step S432, the energy management apparatus calculates an input variable value for the building prediction model based on the information analyzed in step S422 and the energy use primary prediction result for the individual facility calculated in step S426 .

Then, in step S434, the energy management apparatus firstly estimates the energy usage for the entire building by simulating the building prediction model by applying the input variable value calculated in step S432.

Referring again to FIG. 4, in step S440, the energy management device makes a final prediction of the energy usage for the individual facility and the building prediction model by reflecting the correction variables.

Specifically, referring to FIG. 8, in step S442, the energy management device derives the energy usage pattern of the individual facility and / or building prediction model based on the predicted results in steps S426 and S434. For reference, the energy use pattern can be derived in a predetermined time unit (for example, every ten minutes, one hour, one day, one week, one month, one year).

Then, in step S444, the energy management apparatus sets (selects) a correction variable based on the energy usage pattern of the individual facility and / or the building prediction model derived in step S442, and calculates the correction variable value. For example, the energy management apparatus compares the energy usage pattern for the current time (i.e., past) with the actual energy usage in the energy usage pattern of the individual facility and / or the building prediction model derived in step S442, (I.e., the correction parameter) is analyzed and extracted from the information included in the big data. Specifically, the energy management apparatus is configured to calculate a correlation between changes in weather, temperature, humidity, and the like, actual energy consumption, changes in the number of people in the building, actual energy consumption, Correlation variables and their weights are determined by analyzing the correlation between direction, exterior, insulation, windows, heat capacity and actual energy usage, purpose of building, use, industry and actual energy usage. And calculates the corresponding correction variable value.

Then, in step S446, the energy management apparatus performs energy use final prediction on the individual facility and the building prediction model by reflecting the correction variables

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. ≪ / RTI > are to be understood in all respects only as illustrative and not restrictive.

It is to be understood that both the foregoing general description and the following detailed description of the present invention are exemplary and explanatory and are intended to provide further explanation of the invention as claimed. .

Claims (16)

As a cloud-based energy usage forecasting system,
At least one facility management device for providing facility status information for at least one individual facility in a building;
A cloud server providing big data associated with at least one of said at least one facility management device and said building; And
And an energy management device for receiving the facility status information and the big data and predicting the energy usage for the at least one individual facility and the building. The method according to claim 1,
Wherein the energy management device is configured to set a predictive model for the at least one individual facility and a predictive model for the building and to generate a predictive model for the at least one individual facility and the building based on the building state information and the Big Data Estimating the amount of energy usage for the at least one individual facility and the building based on a correction variable derived based on the energy use primary prediction result after firstly predicting the energy use amount by simulation, Energy usage prediction system. 3. The method of claim 2,
Wherein the energy management device establishes a correlation between prediction models for the at least one individual facility to primarily predict energy usage for the at least one individual facility and the building. 3. The method of claim 2,
Wherein the energy management apparatus derives an energy use pattern of a prediction model for the at least one individual facility and the building based on the energy use primary prediction result and calculates an energy use pattern for a past time among the derived energy use patterns To the actual energy usage to derive the correction variable. 5. The method according to any one of claims 1 to 4,
The big data includes building environment information related to the number of people inside the building, building environment information related to at least one of weather, temperature and humidity at a place where the building is located, building energy related to the energy efficiency inherent to the building Characteristic information, building industry information related to the business type of the building, and energy resource information related to the energy resource. As an energy management device for a cloud-based energy usage prediction system,
A data analysis unit for analyzing information transmitted from the cloud server; And
Setting a predictive model for at least one individual facility in a building and a predictive model for the building based on the results analyzed by the data analysis unit and a predictive model for the at least one individual facility, And a prediction model management unit for estimating an energy usage amount by simulating a prediction model. The method according to claim 6,
The prediction model management unit,
A lifecycle management module for managing generation, modification, and deletion of prediction models for the at least one individual facility and the building;
A correlation setting module for analyzing and setting a correlation between prediction models for the at least one individual facility;
An input / output variable management module for managing input / output variables of the predictive model for the at least one individual facility and the building;
A correction variable management module for managing correction variables applied to the at least one individual facility and a prediction model for the building; And
And a simulation module for executing an energy usage prediction algorithm for the at least one individual facility and the prediction model for the building. 8. The method of claim 7,
The prediction model management unit,
Further comprising a filtering module that performs filtering on input parameters of a prediction model for the at least one individual facility and the building. 8. The method of claim 7,
The prediction model management unit,
Further comprising a unit integration module for uniting units of input / output variables between the at least one individual facility prediction model. 10. The method according to any one of claims 6 to 9,
The information transmitted from the cloud server includes information on building human condition related to the number of people in the building, building environment information related to at least one of weather, temperature, and humidity at a location where the building is located, And information on building energy related to the business type of the building, and energy resource information related to the energy resource. The energy management apparatus for a cloud-based energy usage prediction system according to claim 1, As a cloud-based energy usage prediction method,
Setting a prediction model for at least one individual facility and a prediction model for the building in the building;
Simulating a prediction model for the at least one individual facility based on information transmitted from the cloud server to firstly predict energy usage for the at least one individual facility;
Simulating a prediction model for the building based on a result of the first energy estimation for the at least one individual facility to firstly predict an energy usage for the building; And
And finally estimating the energy usage for the at least one individual facility and the building by reflecting the correction variables derived based on the energy use primary prediction results for the at least one individual facility and the building A cloud-based energy usage forecasting method characterized. 12. The method of claim 11,
Wherein the step of setting the prediction model comprises:
Setting a correlation between prediction models for the at least one individual facility and setting input and output variables of the prediction model for the at least one individual facility; And
And setting an input / output variable of a predictive model for the building. 12. The method of claim 11,
The step of predicting energy usage for the at least one individual facility comprises:
Calculating an input variable value of a prediction model for the at least one individual facility based on the information transmitted from the cloud server; And
And simulating a prediction model for the at least one individual facility by reflecting an input parameter value of the prediction model for the at least one individual facility. 12. The method of claim 11,
The first step of predicting the energy usage of the building includes:
Calculating an input variable value of a prediction model for the building based on information transmitted from the cloud server and a result of first-order prediction of energy usage for the at least one individual facility; And
And estimating a prediction model for the building by reflecting an input parameter value of the prediction model for the building. 15. The method according to any one of claims 11 to 14,
The final prediction of energy usage for the at least one individual facility and the building may include:
Deriving an energy usage pattern based on a result of first-order prediction of energy usage for the at least one individual facility and the building;
Setting a correction variable based on the derived energy usage pattern and calculating a corresponding correction variable; And
And finally estimating the energy usage of the at least one individual facility and the building based on the correction parameters. 16. The method of claim 15,
Wherein the correction parameter is set by comparing and analyzing the derived energy usage pattern and the information transmitted from the cloud server.

Images