KR101683401B1 - Apparatus and method for evaluating energy performance of existing buildings using dynamic energy performance curve - Google Patents

Abstract

A method for evaluating the energy performance of a building using a computer according to embodiments of the present invention may include (a) a step of constructing a database with past energy usage values and building characteristics of a plurality of existing buildings; (B) a step of determining a benchmark based on the probability distribution of the energy usage values of existing buildings; (C) a step of selecting similar buildings having similar building characteristics to an evaluation target building among the existing buildings; (D) a step of acquiring each of dynamic energy performance curves which varies over a past period of time, based on the benchmark, for each of the selected similar buildings and the evaluation target building; and (e) a step of determining the trend of the dynamic energy performance curve of the evaluation target building against the dynamic energy performance curves of the selected similar buildings.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an energy performance evaluation system and a method of evaluating energy performance of a building using dynamic energy performance curves. 2. Description of the Related Art < RTI ID = 0.0 >

The present invention relates to a building energy performance evaluation technique.

In developed countries, more than 40% of total greenhouse gas emissions are emitted from the building sector. In Korea, the building sector accounts for about 25% of GHG emissions. In 2002, the European Union launched the Energy Performance of Buildings Directive (EPBD) to manage the energy performance of the building sector, which accounts for a significant portion of these greenhouse gas emissions. Within the European Union, the EPBD is issuing Energy Performance Certificates (EPC) for existing buildings or new buildings. Building EPCs are obligatory to be attached at the time of sale contract or lease contract, and accordingly, building EPC is an important factor in forming the sale price or rental price of the building according to this policy base.

Similarly, in Korea, the Law for Supporting Green Building Construction has been enacted, and the energy consumption certification system for buildings has been in operation. According to the building energy consumption certification system, a building energy efficiency grade based on the energy performance in the standard state according to the design specification of the building and a building energy use grade based on the actual energy usage history of the building are given. Likewise, by attaching a building energy evaluation, which shows energy consumption and actual usage of the building, to the contract, it is possible to induce the consumer to consider the energy consumption when the building is traded.

The building EPC can be divided into two categories: an asset rating system based on energy demand, and an operational rating system based on energy consumption. Studies have shown that the asset class system is suitable for new buildings where historical energy consumption data is not accumulated, and the operating class system is known to be suitable for existing buildings. For example, the UK issues Display Energy Certificates (DEC), which are assessed using an operational grade system for existing public buildings.

The UK's DEC reports energy ratings for the last three years for certain buildings. Buildings typically show a trend of increasing energy use as the number of years is increased due to aging or other reasons. The change in energy rating over three years is too short to recognize the trend of changes in the energy performance of buildings. In addition, it is difficult to know the relative position of the building through comparison with other buildings, and it is impossible to know how much effort should be made to raise the energy rating of the building.

Korea's energy consumption certification system is similar to DEC in Britain, but it is shorter than DEC in the UK because it is required to present only the energy consumption of the building in recent years (that is, the energy use rating of buildings). In addition, there is no way to know how much energy is used compared to other buildings. Korea's energy consumption certification system is similar to DEC in Britain.

Department of Energy & Climate Change, HM Government UK, "Display Enegry Certificate: How efficient is this building being used?"

SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus and method for evaluating building energy performance using a dynamic energy performance curve.

An object of the present invention is to provide a building energy performance evaluation apparatus and method for classifying buildings by categories according to areas, architectural structures, and the like, and evaluating energy performance of buildings in the categories.

An object of the present invention is to provide a building energy performance evaluation apparatus and method for evaluating energy performance of a given building using different category benchmarks for each category according to area, structure, and the like.

The present invention has been made to solve the above problems, and it is an object of the present invention to provide an apparatus and method for evaluating building energy performance, which can compare relative energy performance differences between buildings of similar types using dynamic energy performance curves.

The solution to the problem of the present invention is not limited to those mentioned above, and other solutions not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a method for evaluating building energy performance using a computer, the method comprising the steps of: (a) constructing a database using past energy usage values and building characteristics of a plurality of existing buildings; (b) determining a benchmark based on a probability distribution of energy usage values of the existing buildings; (c) selecting similar buildings having similar building characteristics to the evaluation target building from among the existing buildings; (d) obtaining dynamic energy performance curves that vary over a past period of time, respectively, based on the benchmark, with respect to each of the selected similar buildings and the to-be-rated building; And (e) determining a change trend of the dynamic energy performance curve of the evaluation target building against the dynamic energy performance curves of the selected similar buildings.

According to an exemplary embodiment, the step (b) comprises the steps of: classifying the existing buildings into a plurality of categories by clustering based on building characteristics and energy usage; And determining benchmarks for each category based on a probability distribution of energy usage values of existing buildings belonging to each category.

According to an exemplary embodiment, the step (c) may further include the steps of: (c-1) determining whether the candidate similar building (s) is related to the test building using the property similarity minimum criterion (MCAS) A step of firstly selecting a plurality of data; (c-2) adjusting the property similarity minimum criterion and the property weight range when the average prediction accuracy (APA) and the prediction rate (PR) do not satisfy all predetermined conditions with respect to the selected candidate similar structures; (c-3) secondarily selecting candidate similar buildings using the adjusted minimum likelihood and similarity weight ranges; (c-4) repeating the steps (c-1) to (c-3) until the average prediction accuracy and the prediction rate satisfy the predetermined conditions with respect to the candidate similar structures selected primarily or secondarily ; (c-5) The case-based reasoning model is set as the property similarity minimum criterion and the property weight range when the average prediction accuracy and the prediction rate both satisfy the predetermined condition with respect to the candidate similar structures selected primarily or secondarily step; And (c-6) selecting similar similar structures for the evaluation target building using the case-based model set.

According to an embodiment, the step (c-2) may include adjusting the attribute similarity minimum criterion and the attribute weight range based on a genetic algorithm (GA).

According to an embodiment, the step (d) includes the steps of: determining a maximum value, a representative value and a minimum value of the energy performance values of the similar buildings at each evaluation point in the past period; And acquiring dynamic energy performance curves respectively by connecting the maximum values, representative values and minimum values of the energy performance values of the similar structures over the past period, respectively.

According to an embodiment, the step (d) includes the steps of: determining, at each evaluation point in the past period, an upper limit and a lower limit of the energy performance at the evaluation point, according to the distribution of the energy performance values of the similar buildings; And determining a maximum value, a representative value and a minimum value of the energy performance values of the remaining similar buildings, excluding the similar buildings deviating from the upper and lower limits of the energy performance, for each evaluation point in the past period; And acquiring dynamic energy performance curves respectively by connecting the maximum values, representative values and minimum values of the energy performance values of the similar structures over the past period, respectively.

According to one embodiment, the upper and lower bounds of the energy performance are greater by a mean absolute percentage error (MAPE) in the representative value of the energy performances and at an average absolute percentage error in the representative value of the energy performances It can be a smaller value.

According to an exemplary embodiment, the step (e) may include, at a predetermined evaluation time point, dynamic characteristics of the evaluation target building among sections divided by a maximum value, a representative value and a minimum value of dynamic energy performances of the selected similar buildings Identifying an interval in which the energy performance lies; And determining the change tendency of the dynamic energy performance of the evaluation target building at the evaluation time point as one of very rapid, rapid, gradual, and very gradual, according to the identified section.

According to another aspect of the present invention, there is provided an apparatus for evaluating building energy performance, comprising: a database constructed with past energy usage values and building characteristics of a plurality of existing buildings; A benchmark determining unit for determining a benchmark based on a probability distribution of energy usage values of the existing buildings; A similar structure selecting unit for selecting similar buildings having similar building characteristics to the evaluation target building from among the existing buildings; A dynamic energy performance determination unit for obtaining dynamic energy performance curves varying over the past period based on the benchmark, with respect to each of the similar buildings and the evaluation target building; And a change trend determining unit for determining a change trend of the dynamic energy performance curve of the evaluation target building against the dynamic energy performance curves of the selected similar buildings.

According to one embodiment, the benchmark determination unit classifies the existing buildings into a plurality of categories by clustering based on the building characteristics and the energy usage amount, and based on the probability distribution of the energy usage values of the existing buildings belonging to each category, May act to determine the benchmarks by themselves.

According to an embodiment, the similar building selection unit may include a case-based inference model calculation unit for selecting similar buildings with respect to a test building according to a case-based reasoning model having a property similarity minimum criterion (MCAS) and an attribute weight range (RAW); And a case-based reasoning model optimizing unit for optimizing the case-based reasoning model as the property similarity minimum criterion and the property weighting range when the average prediction accuracy (APA) and the prediction ratio (PR) satisfy predetermined conditions with respect to the selected similar buildings And the case-based reasoning model computing unit may operate to select similar similar buildings for the evaluation target building using the case-based reasoning model optimized by the case-based reasoning model optimizing unit.

According to one embodiment, the case-based inference model optimizer may operate to optimize the property similarity minimum criterion and the property weight range based on a genetic algorithm (GA).

According to one embodiment, the dynamic energy performance determination unit may determine a maximum value, a representative value, and a minimum value of the energy performance values of similar buildings at each evaluation point in the past period, And to obtain dynamic energy performance curves, respectively, by connecting the maximum values, representative values, and minimum values of the energy performance values, respectively.

According to one embodiment, the dynamic energy performance determination unit may determine the upper and lower limits of the energy performance of each evaluation point in accordance with the distribution of the energy performance values of the similar buildings at each evaluation point in the past period, And a filtering unit that excludes similar structures that are outside the upper and lower limits of the energy performance and may be operable to determine a maximum value, a representative value, and a minimum value, respectively, of the energy performance values of the remaining similarities.

According to one embodiment, the upper and lower limits of the energy performance are determined by the average absolute percent error (MAPE) in the representative values of the energy performances and the average absolute percentage error in the representative values of the energy performances It can be a smaller value.

According to one embodiment, the change trend determining unit may determine, at an evaluation time point, a dynamic energy performance of the evaluation target building among the sections classified by the upper limit, the representative value and the lower limit of the dynamic energy performances of the selected similar buildings And to determine a change tendency of the dynamic energy performance of the evaluation target building at the evaluation time point as one of very sudden, rapid, gradual, and extremely gradual, according to the identified section.

According to the apparatus and method for evaluating building energy performance of the present invention, it is possible to provide an apparatus and method for evaluating building energy performance using dynamic energy performance curves.

According to the apparatus and method for evaluating building energy performance of the present invention, it is possible to classify buildings according to categories according to areas, architectural structures, and evaluate the energy performance of buildings in categories.

According to the apparatus and method for evaluating building energy performance of the present invention, energy performance of a given building can be evaluated using different category benchmarks for each category according to area, structure, and the like.

According to the apparatus and method for evaluating building energy performance of the present invention, it is possible to compare relative energy performance differences of buildings of a similar type using dynamic energy performance curves.

The effects of the present invention are not limited to those mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the following description.

1 is a flowchart illustrating an energy performance evaluation method of a building according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating a method for evaluating energy performance of a building according to an embodiment of the present invention. In order to demonstrate that a dynamic energy performance curve based on a carbon dioxide discharge concentration is valid for an energy performance evaluation, This graph is a graph showing the carbon dioxide emission history over a long period of time with respect to buildings.
3 is a flowchart illustrating an S-CBR procedure among the energy performance evaluation methods of a building according to an embodiment of the present invention.
4 is a graph illustrating dynamic energy performance curves obtained from a method for evaluating building energy performance according to an embodiment of the present invention.
FIG. 5 is a flowchart specifically illustrating a procedure for acquiring dynamic energy performance curves among the energy performance evaluation methods of a building according to an embodiment of the present invention.
FIG. 6 is a graph illustrating dynamic energy performance curves obtained through the S-CBR procedure among dynamic energy performance evaluation methods according to an embodiment of the present invention and dynamic energy performance curves of the evaluation target building.
FIG. 7 is a graph comparing a kinetic energy performance curve of a building to be evaluated and carbon dioxide emission amount by energy source among the energy performance evaluation methods of buildings according to an embodiment of the present invention.
FIG. 8 is a graph illustrating dynamic energy performance curves obtained through the S-CBR procedure among dynamic energy performance evaluation methods according to an embodiment of the present invention and dynamic energy performance curves of the target building.
FIG. 9 is a graph comparing dynamic energy performance curves of the buildings to be evaluated among the energy performance evaluation methods of buildings according to an embodiment of the present invention and carbon dioxide emissions by energy source.
10 is a block diagram illustrating an apparatus for evaluating building energy performance according to an embodiment of the present invention.

For the embodiments of the invention disclosed herein, specific structural and functional descriptions are set forth for the purpose of describing an embodiment of the invention only, and it is to be understood that the embodiments of the invention may be practiced in various forms, The present invention should not be construed as limited to the embodiments described in Figs.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals are used for the same constituent elements in the drawings and redundant explanations for the same constituent elements are omitted.

The present invention proposes a trend of dynamic energy performance of an existing building in operating the energy consumption certification system of a building, thereby it is possible to derive an optimal plan for improving the energy performance of an existing building, It can be used as a decision support tool to encourage voluntary participation in the energy conservation movement of existing buildings.

As will be described later, the present invention presents past trends of dynamic energy performance in the buildings and similar structures to be evaluated as existing buildings.

First, from the point of view of the building to be evaluated, the existing building energy consumption certification system provides residents with information on the energy consumption of the building for the last one year (that is, the energy use rating of the building) It is difficult to grasp the trend of change in performance over a long period of time. Accordingly, it is difficult to provide a clear and intuitive basis for the proposed method by the expert even if the owner of the building to be evaluated has a willingness to improve the energy performance. In order to solve these problems, the building energy consumption rating (the actual energy consumption of the building) is given in the building energy consumption certification system, and the past trend of the dynamic energy performance of the evaluation target building is presented together, The present invention can be utilized as a decision support tool to intuitively evaluate the absolute deterioration trend of a curve and to make decisions on energy performance improvement.

Second, from the viewpoint of similar buildings, since the existing building energy consumption certification system only provides the building owner with information on the energy consumption of the building for the last one year (that is, the energy use rating of the building) It is difficult to objectively grasp the relative phase of energy performance. Furthermore, it is difficult for building owners and residents to clearly recognize the improvement goals that can be achieved through participation in energy conservation campaigns. In order to solve these problems, we propose a generalized distribution of dynamic energy performance through comparative analysis with similar buildings, while giving a building energy use rating (actual energy consumption of buildings) in the building energy consumption certification system The general dynamic energy performance of similar buildings is categorized into a good case, an average case and a bad case, so that the relative degradation rate of the dynamic energy performance curve of the target building can be intuitively evaluated, The present invention can be utilized as a decision support tool.

Accordingly, the present invention can provide an optimal solution for improving the energy performance of the target building in consideration of the absolute or relative deterioration rate of the dynamic energy performance curves of existing buildings with clear grounds.

For example, when the relative degradation rate of the dynamic energy performance curve of the building to be evaluated is more serious than other similar buildings, it is possible to intuitively identify the residents as to how severe the relative degradation rate is, It is possible to visually understand that the degradation rate of the dynamic energy performance curve can be alleviated when the energy performance of the target building is improved through the application.

Further, in the present invention, if the degradation rate of the dynamic energy performance curve of the target building is gentler than that of other similar buildings, the building owner or residents use renewable energy policy to replace the energy consumed in the target building And can assist in adopting measures.

An energy evaluator or an appraiser who operates and manages a building energy consumption certification system can understand the dynamic characteristics of the energy performance of existing buildings based on the actual built database using the present invention.

Furthermore, the present invention can be applied to real-time energy use information based on real-time energy usage information, that is, when a base on which energy usage information of a building can be measured, collected and analyzed in real- Thereby enabling a dynamic energy performance curve management system and method.

1 is a flowchart illustrating an energy performance evaluation method of a building according to an embodiment of the present invention.

Referring to FIG. 1, a method for evaluating building energy performance using a computer may start from a step S11, in which the computer constructs a database with past energy usage values and building characteristics for a plurality of existing buildings.

Energy usage values can be stored as either power consumption or fuel consumption, but can be converted to carbon dioxide emissions and stored in a database.

The building characteristics to be included in the database can be illustrated with various properties that can describe the building so as to estimate similar buildings. These properties are independent of each other and are independent of other properties, so they are used as independent variables in case - based reasoning.

For example, the characteristics of buildings are classified as administrative factors such as administrative districts, building types such as ownership / residence / work type, structure type (steel frame, reinforced concrete, etc.), occupancy factor safety grade, building age, building area, (Or unit space per unit of use) per unit space and carbon dioxide emission concentration (CO ₂ ) as a global warming potential (GWP) factor, emission density is an attribute determined by the above independent variables and is used as a dependent variable in case-based reasoning.

Referring to FIG. 2 to demonstrate the validity of the energy performance evaluation based on the carbon dioxide emission concentration, FIG. 2 is a graph showing the carbon dioxide emission history over a long period of time for specific buildings.

2, the abscissa is the number of years, and the ordinate is the concentration of carbon dioxide in the buildings of a given type. The annual statistics of education issued by the Seoul Metropolitan Office of Education provides data on the energy consumption of schools in each year. Based on this data, the concentration of CO2 emissions from energy use of elementary schools operated in Seoul from 1999 to 2010 Can be calculated.

2, it can be seen that the concentration of carbon dioxide emission increases steadily over time. In FIG. 2, since the significance level is fixed at about 0.05, the mean difference between different groups can be said to be valid. According to the observation result of FIG. 2, it can be said that the dynamic change of the carbon dioxide emission concentration of a certain building over time directly represents the dynamic energy performance change of the building.

Referring back to FIG. 1, in step S12, the computer may determine a benchmark based on a probability distribution of energy usage values of existing buildings.

Here, the benchmark refers to the quantified energy usage of a typical building that can represent a plurality of buildings.

At this time, the energy consumption can be quantified with the carbon dioxide emission concentration so that it can be practically applied to almost all buildings. Unless buildings are 100% based on renewable energy, whether energy consumption in the form of electric energy based on fossil fuel or nuclear fuel, or direct consumption of fossil fuels such as gas, It can be converted to emission concentration.

Therefore, the benchmark determined in step S12 may be a value obtained by quantifying the energy usage of a typical building selected for the benchmark as a carbon dioxide emission concentration.

Typical buildings for benchmarking should be selected to represent a group of existing buildings. For this purpose, the present invention can set the median of the carbon dioxide emission concentration values of existing buildings as a benchmark. In accordance with an embodiment, the mean or mode of CO2 emission concentration values of existing buildings may be determined as a benchmark.

According to the inventors' original observation, carbon dioxide emission concentration values of buildings have a positive negative correlation to area values per unit space. For example, according to the educational statistical yearbook of the Seoul Metropolitan Office of Education, the larger the average classroom size of a certain school, the lower the concentration of carbon dioxide.

If benchmarking as a typical building energy use value can not be properly selected if the median of the statistics is distorted because of the average classroom area is considerably broader than most other schools and thus some schools with significantly lower CO2 emissions .

Accordingly, in step S12, prior to determining the benchmark, the existing structures may be properly clustered to form a plurality of categories. A benchmark may be separately determined for each of a plurality of categories formed.

Specifically, the step S12 includes the steps of the computer classifying the existing buildings into a plurality of categories by clustering based on the building characteristics and the energy usage amount, and a step of calculating, based on the probability distribution of the energy usage values of the existing buildings belonging to each category And determining benchmarks for each category.

Clustering can be performed, for example, according to a Decision Tree technique. More specifically, two categories can be formed by performing a decision tree technique with an area per unit space and a carbon dioxide emission concentration as a dependent variable as independent variables of each building. At this time, the decision tree technique can be performed based on the CHAID (Chi-Squared Automatic Interaction Detection) algorithm since the dependent variable is a continuous scale.

For example, according to the Education Statistics Yearbook of the Seoul Metropolitan Office of Education, 418 elementary school buildings are clustered into categories of 335 buildings with an average classroom size of less than 328.875 and 83 categories with an average classroom size of more than 328.875 , And the benchmark of each category, ie, the typical concentration of CO2 equivalent to the median value, can be determined to be 23.272 and 19.878, respectively.

The energy performance of buildings belonging to each category can be quantified, for example, with an operational rating or graded with a letter rating.

Since the operating score is absolute and hard to place on a physical basis, it can be given as a percentage of the energy usage value of a typical building selected for benchmarking and the numerator of the energy usage value of any building being evaluated. The amount of energy used can be quantified as the carbon dioxide emission concentration, as illustrated above.

The benchmark for a particular category is the median of the CO2 emission concentration values for that category. Accordingly, the operating score of a typical building selected for the benchmark can be quantified to 100. The operation score of the building subject to evaluation is the percentage obtained by dividing the carbon dioxide emission concentration value of the building by the benchmark of the category.

Therefore, the lower the operation score, the better the energy performance of any building to be evaluated. In particular, if the operating score is in the range of 0 to 100, it means that the building to be evaluated has better energy performance than the building corresponding to the benchmark. If the operating score is greater than 100, the building to be evaluated corresponds to the benchmark This means that the energy performance is worse than that of buildings.

The character class of the 7th grade adopted by the EPC is assigned the highest character grade A when the operating score is 0 to 25, the character grade B when the character is 26 to 50, the character class C when the range is 51 to 75, 101 to 125, character class F is assigned if character class E is 126 to 150, and character class G is assigned if character class F is 151 or more.

Subsequently, in step S13, the computer can select similar buildings having similar building characteristics to existing buildings to be evaluated.

According to an embodiment, in step S13, the computer may select similar structures by case-based reasoning (CBR), in particular by S-CBR (Simplified CBR) technique.

3 is a flowchart illustrating a method of evaluating building energy performance according to an exemplary embodiment of the present invention. Referring to FIG. 3, to be.

In Fig. 3, step S13 may be concretely composed of steps S131 to S134.

In step S131, the computer calculates a candidate similarity structure (CBR) using a minimum criterion for scoring the attribute similarity (MCAS) and a range of attribute weight (RAW) according to a case- Can be selected first.

The case-based reasoning methodology will be briefly described below. In general, basic case-based reasoning (CBR) is a methodology that selects project cases with high similarity through comparative analysis of projects based on project characteristics. The core of the CBR methodology is the attribute similarity (AS), the attribute weight (AW), and the case similarity (CS). The case similarity CS between a real case and a test case can be quantified as the sum of the products of the property similarities AS and the property weights AW.

Where AS is the attribute similarity, AW is the attribute weight, CS is the case similarity, n is the number of attributes, and m is the number of cases.

The attribute similarity (AS) is based on the difference between the independent variables of the same kind between the actual and test cases, for each project characteristic, that is independent variables (which may in some cases also include dependent variables) . If the data of the independent variable is a nominal scale, the property similarity is 1 if they are equal to each other and 0 otherwise. If the data of the independent variable is a numerical scale, the attribute similarity AS can be expressed by the following equation (2).

f _AS indicates the attribute similarity level calculation, and AV _{Test _Case} is the value of the specified property of the test cases, AV _{Retrieved _Case} is a specific property value of the retrieved cases. f _AS is closer to 100% as the attributes are similar. The MCAS is the minimum criterion for the quantification of the property similarity. The property similarity value must remain higher than the MCAS value, otherwise the value is discarded. The MCAS may be set to, for example, 10% and may be optimized as described below.

The similarity degree (CS) of a test case can be expressed by the following equation (3) by using the property similarity and attribute weight of all independent variables between the test case and the retrieval case, and how the test case is similar to each retrieval case.

f _CS is a function for calculating the case similarity (CS), f _AW is a function or the result for calculating the attribute weight (AW), f _AS is a function or the result of calculating the attribute similarity (AS) n is the number of attributes. The attribute similarity AS is calculated as shown in Equation 2, and the attribute weight AW can be empirically determined or mathematically optimized through other optimization methods. The case similarity (CS) can be illustrated as a value obtained by dividing the weighted property similarity obtained by adding the attribute similarity to the property weight by the sum of the property weights.

One or more cases with the highest case similarity (CS) can be selected as similar cases and the value of the dependent variable of the test case can be predicted by the average or weighted average of the dependent variable values of the estimated similar cases.

In step S132, the computer can determine whether Average Prediction Accuracy (APA) and Prediction Rate (PR) satisfy predetermined conditions with respect to the selected candidate similar structures.

In this step S132, the case-based reasoning model is verified by quantifying how well the selected similar cases are selected.

The prediction rate (PR) of the case-based reasoning model can be verified as follows. For example, a mean absolute percentage (MAPE) as a dependent variable predictability (PR) from each difference between the actual values (AV) of the dependent variable and the previously calculated predicted values (PV) error can be calculated by the following equation (4).

Where f _MAPE is a function to obtain MAPE, AV is the actual value of the dependent variable, PV is the predicted value of the dependent variable, and m is the number of similar cases.

Another validation index, APA, can be calculated as 100-f _MAPE .

For example, the case-based reasoning model can be said to be verified if the condition that the PR based on the MAPE is 95% or more and the APA is the maximum is satisfied, and otherwise, the property similarity minimum criterion and the property weight range can be adjusted.

If both the APA condition and the PR condition are satisfied, the procedure advances from step S132 to step S134, and if not, the procedure proceeds to step S133 to adjust the attribute similarity minimum criterion and the attribute weight range.

According to the embodiment, in step S133, MCAS and RAW can be adjusted based on the genetic algorithm (GA).

The procedure returns to step S131 and the computer can secondarily select the candidate similar structure using the adjusted minimum similarity property and the adjusted property weight range.

Steps S131 to S133 can be repeated until the average prediction accuracy and the prediction rate satisfy the predetermined conditions with respect to the candidate similar structures that are primarily or secondarily selected.

In step S134, the computer can set the case-based reasoning model as the property similarity minimum criterion and the property weight range when the average prediction accuracy and the prediction ratio satisfy predetermined conditions with respect to the selected candidate similar structures.

In step S135, the computer can select similar similar structures for the evaluation target building using the set case-based model.

Referring back to Fig. 1, in step S14, the computer can obtain dynamic energy performance curves, which vary over the past period, based on the benchmark, for each of the selected similar buildings and the building being evaluated .

In the present invention, the dynamic energy performance curve is a curve for visually displaying a change in energy performance over a long period of time, rather than a limited energy performance at a specific point in time.

Since the benchmark of the category has been determined in advance, the energy performance values of each of the similar buildings selected to be similar within the category to which the particular evaluation target belongs can be given on a benchmark basis at each evaluation point over a long period of time in the past. In the present invention, these energy performance values are referred to as dynamic energy performance curves.

Referring to FIG. 4 for illustrating a dynamic energy performance curve, FIG. 4 is a graph illustrating dynamic energy performance curves obtained from a method for evaluating building energy performance according to an embodiment of the present invention.

4, the horizontal axis of the graph is the evaluation year, and the vertical axis is the Dynamic Operational Rating (DOR).

The dynamic energy performance curves are each composed of a curve of a maximum value, a curve of a representative value (for example, an average value) and a curve of a minimum value.

The energy performance of a building is a function of the energy performance of the building, as evidenced by its gradual deterioration due to normal use, in other words, unless there are individual and special circumstances such as large-scale repairs, equipment replacement, remodeling, .

Beyond this basic reasoning, the present invention proposes, at each point in time, that the energy performances of similar buildings are generally within a certain range, and that maximum, representative, and minimum values defining such a typical range. The representative value may be an average value, a weighted average value, a median value, or a mode value.

The curves of the maximum values connecting the maximum values, the curves of the representative values connecting the representative values, and the curve of the minimum value connecting the minimum values may be referred to as generalized historical trends of energy performance, respectively.

In addition, depending on the location where the energy performance of the building to be evaluated is relatively placed relative to the energy performances of similar buildings, it can be understood how the evaluation target building is placed in comparison with similar buildings.

In FIG. 4, if the energy performance of the building to be evaluated is in the area 1 (Area 1), which is larger than the maximum value of the energy performance curves of the similar buildings, the evaluation target building is degraded "very suddenly" Can be interpreted.

If the energy performance of the building to be evaluated is in Area 2 (Area 2), which is smaller than the maximum value of the energy performance curves of similar buildings but larger than the representative value, the building to be evaluated is interpreted as "abruptly degrading energy performance" .

If the energy performance of the building to be evaluated is in Area 3 (Area 3), which is larger than the minimum value of the energy performance curves of similar buildings but smaller than the representative value, the building to be evaluated is interpreted as "moderately" deteriorating energy performance .

If the energy performance of the building to be evaluated is in Area 4 (Area 4), which is smaller than the minimum value of the energy performance curves of similar buildings, the building to be evaluated can be interpreted as "very gently" deteriorating energy performance.

Although the above-described analysis is made only with the energy performance of the immediately preceding point, according to the embodiment, the deterioration trend of the energy performance can be interpreted based on the change of the energy performance for the immediately preceding plurality of points of time.

These dynamic energy performance curves and interpretations allow residents to intuitively understand not only the relative ratings of the present time but also the dynamic trends of the grade with respect to the energy performance of the building.

Furthermore, grades A through D that show better energy performance than the benchmark can be described as an incentive zone, and grades E through G that exhibit poorer energy performance than benchmarks can be referred to as penalty zones. have.

Specifically, referring to FIG. 5 to illustrate a procedure for obtaining such dynamic energy performance curves in step S14, FIG. 5 is a graph showing dynamic energy performance curves among methods of evaluating building energy performance according to an embodiment of the present invention. And a procedure for acquiring the data.

In Fig. 5, step S14 may include steps S141 to S143.

In step S141, the computer can determine an upper limit and a lower limit of the energy performance at the evaluation point, respectively, at each evaluation point in the past period, for example, every year, according to the distribution of the energy performance values of the similar buildings.

Depending on the embodiment, the upper limit of the energy performance may be set to a value that is larger by an average absolute percentage error (MAPE) as shown in equation (4) based on the median value.

According to an embodiment, the lower limit of the energy performance may be set to a value that is smaller by an average absolute percentage error (MAPE) as shown in equation (4) based on the median value.

Setting the upper and lower limits of this energy performance is contrary to the "generalized trend" of the energy performance of other similar buildings where the energy performance value is too high or too low compared to the median value, even though the building properties are similar This is because it can be seen.

Thus, in step S142, the computer stores the maximum value, the representative value, and the minimum value of the energy performance values of the remaining similar buildings, excluding the similar buildings that deviate from the upper and lower limits of the energy performance, You can decide.

At this time, excluding all of the similar buildings that deviate from the upper limit and the lower limit even once at each evaluation point in the past period, the number of remaining similar buildings may become insufficient to represent the "generalized trend ". Therefore, similar buildings can be independently judged whether they are outside the upper limit or lower limit at each evaluation time. For example, if analogous building A is out of the 2010 ceiling but is within the upper limit in 2011, analogous building A may be excluded from all evaluation points, but excluded only at the maximum value of 2010 It may be included in determining the maximum value of the year.

According to the embodiment, it is possible not to go through the step S141 and to not perform the exclusion of the similar building in the step S142.

In step S143, the computer may acquire dynamic energy performance curves, respectively, by connecting the maximum values, representative values, and minimum values of the energy performance values of the similar structures over all or a part of the past period, respectively.

Depending on the embodiment, the energy performance curve may be obtained for the entire past period in which the energy usage values are being investigated, but may be obtained for a portion of the past period, such as for the last 5 years.

Referring back to FIG. 1, in step S15, the computer can determine a change trend in the dynamic energy performance curve of the target building against the dynamic energy performance curves of the selected similar buildings.

According to the embodiment, step S15 is a step S15 in which, at a predetermined evaluation time point, the computer calculates the dynamic property of the evaluation target building among the sections divided by the maximum value, the representative value and the minimum value of the dynamic energy performances of the selected similar buildings The step of identifying the section where the energy performance is placed and the step of changing the dynamic energy performance of the building to be evaluated at the evaluation time point in a very rapid, And a step of determining as one.

According to the trend of dynamic energy performance change, the building manager can adopt the corresponding strategy as shown in Table 1 below.

domain Change trend Energy saving strategy Renewable energy strategy Energy performance> Maximum value Very abrupt Very fit incongruity Maximum value ≥ Energy performance> Representative value precipitation Fairly fit fitness Representative value ≥ Energy performance> Minimum value Gentle fitness Fairly fit Minimum value ≥ Energy performance Very gentle incongruity Very fit

6 and 7 to illustrate dynamic energy performance curves and their interpretations in an exemplary case, FIG. 6 is a graph illustrating a dynamic energy performance curve obtained through an S-CBR procedure of a building energy performance evaluation method according to an embodiment of the present invention FIG. 7 is a graph showing dynamic energy performance curves of dynamic structures and dynamic energy performance curves of buildings to be evaluated. FIG. 7 is a graph comparing dynamic energy performance curves of buildings to be evaluated and carbon dioxide emissions by energy sources.

In Fig. 6, the building to be evaluated has an energy performance of B grade in 1999, and the energy performance deteriorates as time elapses. However, recent trends are alleviated.

The energy performance of the rated building is B grade in 1999, which seems to be good at first, but according to the general tendency of similar buildings, it was the worst among similar buildings from the beginning. The energy performance of the target building is less deteriorating than similar buildings since 2003 and has better energy performance than the minimum value in 2009 and 2010. Though energy performance has deteriorated sharply in 2010, it can be inferred from the energy performance curve of other similar buildings, for example, that due to climatic conditions, energy use has increased in society as a whole.

The change in the dynamic energy performance curve of the target buildings in 2010 is interpreted as "gradual" or "very gradual".

Because the energy performance of the target building is close to the energy performance of the benchmark in 2010, it is predicted that if the measures are not taken, it will not maintain the D grade in the next year and will deteriorate to E grade. Nevertheless, It can be predicted that good performance will be maintained.

In Fig. 7, it can be seen that the amount of electricity used was rapidly increased in 2009 and 2010, but the energy performance deteriorated sharply.

According to the change trend analysis of the energy performance curves of Figs. 6 and 7 and the table 1, the managers of the building to be evaluated are more likely to use the electric energy in the system instead of the electric energy to be consumed It is desirable to choose a strategy that produces and self-supports itself from energy.

8 and 9 to illustrate an example of a building having a dynamic energy performance lower than that of similar buildings. FIG. 8 is a flowchart illustrating an S-CBR procedure of a building energy performance evaluation method according to an embodiment of the present invention. FIG. 9 is a graph comparing dynamic energy performance curves of buildings to be evaluated with carbon dioxide emissions according to energy sources. FIG. 9 is a graph illustrating dynamic energy performance curves obtained from the buildings to be evaluated and dynamic energy performance curves of buildings to be evaluated.

In FIG. 8, the target building to be evaluated is B grade in energy performance in 1999, but the energy performance greatly deteriorated immediately afterwards, and the trend has been relaxed in the past. However, the trend is remarkably deteriorated in the last two years.

The energy performance of the building to be evaluated seems to be good at first because it is B grade in 1999, but it was neglected very much compared with the general tendency of similar buildings from the following year. The energy performance of the target building is much worse than similar buildings since 2000, but it shows better energy performance than the minimum value in 2007 and 2008. In 2010 and 2011, the energy performance deteriorated sharply and again to a greater extent than the maximum values of the energy performance curves of other similar buildings.

The changing trend of the dynamic energy performance curve of the target buildings in 2010 is interpreted as "sudden" or "very sudden".

Since the energy performance of the target building is already worse than the energy performance of the benchmark in 2010, it can be predicted that if the measures are not taken, the F grade can not be maintained in the next year and the G grade can be deteriorated.

In FIG. 9, it can be seen that, in 2009 and 2010, the amount of electricity used was rapidly increased instead of rapidly decreasing the amount of gas used, and as a result, the energy performance deteriorated sharply.

According to the change trend analysis of the energy performance curves of FIGS. 8 and 9 and the Table 1, the managers of the building to be evaluated are more likely to be supplied with electricity from the building, It is desirable to select a strategy that reduces the electric energy to be consumed.

10 is a block diagram illustrating an apparatus for evaluating building energy performance according to an embodiment of the present invention.

10, the building energy performance evaluation apparatus 100 includes a database 110, a benchmark determination unit 120, a similar building selection unit 130, a dynamic energy performance determination unit 140, (150).

The database 110 may be constructed with historical energy usage values and building characteristics for a plurality of existing buildings.

The energy usage value may be a usage amount expressed in electricity consumption or fossil fuel consumption, or a consumption amount converted to carbon dioxide emission.

For example, the characteristics of buildings are classified as administrative factors such as administrative districts, building types such as ownership / residence / work type, structure type (steel frame, reinforced concrete, etc.), occupancy factor safety grade, building age, building area, (Or unit space per unit area) per unit space, unit group size per unit space (or unit space per use group), and may include carbon dioxide emission concentration as a global warming index (GWP) factor.

The benchmark determination unit 120 can determine a benchmark based on a probability distribution of energy usage values of existing buildings.

Benchmark refers to the quantified energy usage of a typical building that can represent multiple buildings. At this time, the energy consumption can be quantified with the carbon dioxide emission concentration.

According to the embodiment, the benchmark determining unit 120 can quantify the benchmark as the carbon dioxide emission concentration, using the energy consumption of the typical building selected for the benchmark.

According to the embodiment, prior to determining the benchmark, the benchmark determination unit 120 may first cluster existing structures to form a plurality of categories, and determine a benchmark for each of the plurality of formed categories .

Specifically, the benchmark determination unit 120 classifies the existing buildings into a plurality of categories by clustering existing buildings based on the building characteristics and the energy usage amount, and calculates the probability distribution of the energy usage values of the existing buildings belonging to each category Benchmarks can be determined on a per-category basis.

Depending on the embodiment, clustering may be performed, for example, according to a decision tree technique. More specifically, two categories can be formed by performing a decision tree technique with an area per unit space and a carbon dioxide emission concentration as a dependent variable as independent variables of each building. At this time, the decision tree technique can be performed based on the CHAID algorithm since the dependent variable is a continuous scale.

According to the embodiment, the benchmark for each category is the median of the carbon dioxide emission concentration values for that category. Accordingly, the operating score of a typical building selected for the benchmark can be quantified to 100. The operation score of the building subject to evaluation is the percentage obtained by dividing the carbon dioxide emission concentration value of the building by the benchmark of the category.

The similar building selection unit 130 can select similar buildings having similar building characteristics to existing buildings to be evaluated.

According to an embodiment, the similar building selection unit 130 may include a case based reasoning based model calculation unit 131 and a case based reasoning model optimization unit 132 in detail.

The case-based reasoning model computing unit 131 can select similar buildings with respect to the test building according to the case-based reasoning model having the property similarity minimum criterion (MCAS) and the property weighting range (RAW).

On the other hand, the case-based reasoning model optimizing unit 132 optimizes the case-based reasoning model driven by the case-based reasoning model computing unit 131. [ In particular, the case-based reasoning model optimizing unit 132 optimizes the case-based reasoning model with respect to the selected similar buildings as the property similarity minimum criterion and the property weighting range when the average prediction accuracy (APA) and the prediction rate (PR) You can optimize the model.

According to the embodiment, the case-based reasoning model optimizer 132 can optimize the property similarity minimum criterion and the property weight range based on the genetic algorithm (GA).

Accordingly, the case-based reasoning model computing unit 131 can select similar similar buildings for the evaluation target building using the case-based reasoning model optimized by the case-based reasoning model optimizing unit.

The dynamic energy performance determination unit 140 can acquire dynamic energy performance curves that vary over the past period based on the benchmark, with respect to each of the similar buildings and the evaluation target building.

Specifically, according to the embodiment, the dynamic energy performance determination unit 140 determines an upper limit and a lower limit of the energy performance at each evaluation point, respectively, according to the distribution of the energy performance values of the similar buildings at each evaluation point in the past period And a filtering unit 141 that excludes similar structures that are outside the upper and lower limits of the energy performance.

According to an embodiment, the upper and lower limits of the energy performance are a value that is larger by an average absolute percentage error (MAPE) in the representative value of the energy performances at each past evaluation point and smaller by an average absolute percentage error in the representative value of the energy performances .

The dynamic energy performance determination unit 140 determines a maximum value, a representative value, and a minimum value of the energy performance values of similar buildings at each evaluation point in the past period, and determines a maximum value, a representative value, Values, representative values, and minimum values, respectively, to obtain dynamic energy performance curves, respectively.

The change trend determining unit 150 can determine the change trend of the dynamic energy performance curve of the target building in comparison with the dynamic energy performance curves of the selected similar buildings.

According to the embodiment, the change tendency determination unit 150 determines, at the evaluation time point, the dynamic energy of the evaluation target building among the intervals classified by the upper limit, the representative value, and the lower limit of the dynamic energy performances of the selected similar buildings It is possible to identify the section where the performance is placed and determine the change tendency of the dynamic energy performance of the evaluation target building at the evaluation time point as one of very rapid, rapid, gradual, and very gradual, depending on the identified section.

According to the estimated change trend and the dynamic energy performance curve and Table 1, the building energy performance evaluating apparatus 100 of the present invention is designed to improve the energy performance of the building to be evaluated, .

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. It will be understood that variations and specific embodiments which may occur to those skilled in the art are included within the scope of the present invention.

Further, the apparatus according to the present invention can be implemented as a computer-readable code on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the recording medium include ROM, RAM, optical disk, magnetic tape, floppy disk, hard disk, nonvolatile memory and the like. The computer-readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner.

100 Building energy performance evaluation device
110 database
120 Benchmark decision part
130 Construction of similar buildings
Case-based reasoning-based model operator
132 Case-Based Reasoning Model Optimization
140 Dynamic Energy Performance Determination Department
141 filtering section
150 change tendency determining section

Claims (17)

A method for evaluating building energy performance using a computer,
(a) building a database with historical energy usage values and building characteristics for a plurality of existing buildings; (b) determining a benchmark based on a probability distribution of energy usage values of the existing buildings; (c) selecting similar buildings having similar building characteristics to the evaluation target building from among the existing buildings; (d) obtaining dynamic energy performance curves that vary over a past period of time, respectively, based on the benchmark, with respect to each of the selected similar buildings and the to-be-rated building; And (e) determining a trend of the dynamic energy performance curve of the building subject to evaluation against the dynamic energy performance curves of the selected similar buildings,
The step (c) may be performed by first selecting candidate similar structures with respect to the test building using the property similarity minimum criterion (MCAS) and the property weight range (RAW) according to the case-based reasoning (CBR) step; (c-2) adjusting the property similarity minimum criterion and the property weight range when the average prediction accuracy (APA) and the prediction rate (PR) do not satisfy all predetermined conditions with respect to the selected candidate similar structures; (c-3) secondarily selecting candidate similar buildings using the adjusted minimum likelihood and similarity weight ranges; (c-4) repeating the steps (c-1) to (c-3) until the average prediction accuracy and the prediction rate satisfy the predetermined conditions with respect to the candidate similar structures selected primarily or secondarily ; (c-5) The case-based reasoning model is set as the property similarity minimum criterion and the property weight range when the average prediction accuracy and the prediction rate both satisfy the predetermined condition with respect to the candidate similar structures selected primarily or secondarily step; And (c-6) selecting similar similar buildings to the evaluation target building using the case-based reasoning model set. The method of claim 1, wherein step (b)
Classifying the existing buildings into a plurality of categories by clustering based on building characteristics and energy usage; And
Determining benchmarks for each category based on a probability distribution of energy usage values of existing buildings belonging to each category. delete The method of claim 1, wherein the step (c-2)
And adjusting the property similarity minimum criterion and the property weight range based on a genetic algorithm (GA). The method of claim 1, wherein step (d)
Determining a maximum value, a representative value, and a minimum value of the energy performance values of the similar buildings at each evaluation point in the past period; And
And acquiring dynamic energy performance curves respectively by connecting maximum values, representative values, and minimum values of the energy performance values of the similar buildings over the past period, respectively. The method of claim 1, wherein step (d)
Determining an upper limit and a lower limit of the energy performance at the evaluation point in accordance with the distribution of the energy performance values of the similar buildings at each evaluation point in the past period; And
Determining a maximum value, a representative value, and a minimum value of the energy performance values of the remaining similar buildings, excluding the similar buildings deviating from the upper and lower limits of the energy performance, at each evaluation point in the past period; And
And acquiring dynamic energy performance curves respectively by connecting maximum values, representative values, and minimum values of the energy performance values of the similar buildings over the past period, respectively. 7. The method of claim 6 wherein the upper and lower limits of energy performance are
(MAPE) at a representative value of the energy performances at each evaluation point in the past period, and a value that is smaller by an average absolute percentage error at the representative value of the energy performances. 6. The method of claim 5, wherein step (e)
Identifying a section in which the dynamic energy performance of the building to be evaluated lies in the sections divided by the maximum value, the representative value and the minimum value of the dynamic energy performances of the selected similar buildings at a predetermined evaluation time point; And
And determining the change tendency of the dynamic energy performance of the evaluation target building at the evaluation time point as one of very sudden, rapid, gradual, and very gradual, according to the identified section . A program for a computer, recorded in a computer-readable recording medium, which is created by a computer to perform the steps of the method for evaluating building energy performance according to any one of claims 1, 2, and 4 to 8. A database constructed with historical energy usage values and building characteristics for a plurality of existing buildings; A benchmark determining unit for determining a benchmark based on a probability distribution of energy usage values of the existing buildings; A similar structure selecting unit for selecting similar buildings having similar building characteristics to the evaluation target building from among the existing buildings; A dynamic energy performance determination unit for obtaining dynamic energy performance curves varying over the past period based on the benchmark, with respect to each of the similar buildings and the evaluation target building; And a change trend determining unit for determining a change trend of a dynamic energy performance curve of the evaluation target building against the dynamic energy performance curves of the selected similar buildings,
The similar building selection unit
A case-based reasoning model computing unit for selecting similar buildings with respect to the test building according to the case-based reasoning model having the property similarity minimum criterion (MCAS) and the property weighting range (RAW); And a case-based reasoning model optimizing unit for optimizing the case-based reasoning model as the property similarity minimum criterion and the property weighting range when the average prediction accuracy (APA) and the prediction ratio (PR) satisfy predetermined conditions with respect to the selected similar buildings And the case-based reasoning model operating unit operates to select similar similar buildings for the evaluation target building using the case-based reasoning model optimized by the case-based reasoning model optimizing unit. . 11. The apparatus of claim 10, wherein the benchmark determination unit
The existing buildings are clustered on the basis of the building characteristics and the energy usage amount to divide them into a plurality of categories,
And to determine benchmarks for each category based on a probability distribution of energy usage values of existing buildings belonging to each category. delete The apparatus of claim 10, wherein the case-based inference model optimizer is operative to optimize the property similarity minimum criterion and the property weight range based on a genetic algorithm (GA). The system of claim 10, wherein the dynamic energy performance determiner comprises:
A representative value and a minimum value of energy performance values of similar buildings are determined for each evaluation point in the past period,
Wherein the controller is operative to obtain dynamic energy performance curves respectively by connecting maximum values, representative values and minimum values of energy performance values of similar buildings over the past period. 15. The system of claim 14,
Determining an upper limit and a lower limit of the energy performance at each evaluation point in accordance with the distribution of the energy performance values of the similar buildings for each of the evaluation points during the past period and filtering out the similar buildings falling outside the upper and lower limits of the energy performance, Further,
Representative values and minimum values of the energy performance values of the remaining similar buildings, respectively. 16. The method of claim 15 wherein the upper and lower limits of energy performance are
(MAPE) at a representative value of energy performances at each past evaluation point and a value smaller by an average absolute percentage error at a representative value of energy performances. 15. The apparatus of claim 14, wherein the change trend determining unit
A section in which the dynamic energy performance of the building subject to evaluation is placed is identified among sections divided by the upper limit, the representative value and the lower limit of the dynamic energy performances of the selected similar buildings,
Wherein the operation of determining the change tendency of the dynamic energy performance of the evaluation target building at the evaluation time is determined to be one of very sudden, rapid, gradual, and very gradual, according to the identified section.

Images