Evaluate the accuracy of the RF model

Một phần của tài liệu Combining building information modeling (bim) and choosing by advantages (cba) method to select design construction solutions toward sustainable construction in viet nam (Trang 43 - 109)

The accuracy of the predictive model is evaluated by randomly selecting a set of result values and comparing them with the corresponding set of test values, the closer the predicted value is to the test value, the higher accuracy prediction has.

Besides, the study also uses the index Nash Sutcliffe Efficiency (NSE) to evaluate the predictive performance of the model:

NSE = 1- ∑ (𝑦𝑡−𝑥𝑡)

𝑛 2 1

∑ (𝑦𝑛1 𝑡−𝑦̅)2 n is the sample size

yt is the value selected for evaluation

xt is the predicted value

ȳ is the mean of yt in the sample

The closer the NSE value is to 1, the more accurate the model's prediction performance is.

Combining physical and statistical models using Machine Learning techniques as shown below:

In which: Training data is used for RF model, test data is used to check the accuracy of the predictive model after being trained.

3.7. Software used in the study

In the thesis, research support software has been used for specific purposes as shown in the table below:

Creating 3D models with the characteristics of a typical office by

DesignBuilder software

Identifying primary simulation variables and their ranges of value

Simulation processing by DesignBuilder software

Energy consumption datasets with different values of variables

Creating RF model to predict energy consumption

Defining design alternatives having the optimal energy consumption Splitting Training data and Testing data

Figure 3.5: Procedure of building a model to predict energy consumption

Table 3.3: List of software used in the study

No Name of software Objective

1 IBM SPSS Statistics 22 Inferential statistics

2 Microsoft Excel 2019 Descriptive statistics

3 DesignBuilder V6.1.0.006 Energy consumption

simulation

4 Python 3.10.9 and Spyder IDE 5.4.1

Creating RF model to predict the energy

consumption

ANALYSIS

1.1. Determining the factors

To analyze and present factors affecting the selection of materials - construction towards sustainable construction, the author has studied based on previous studies, specialized technical documents and consulted experienced experts in the field of green buildings and buildings with efficient energy use. Most of these experts have over 10 years of experience in designing, project management or contractor of projects applying green standards. They are also working for reputed companies in construction fields such as Investors, Design Consultants, Project Management Consultants, Contractors.

After referring to these sources, the study identified 12 main factors impacting the decision of selecting design options. Because the objective of this research is toward sustainable construction, the results include factors not only relating to material – construction characteristics but also energy consumption which are divided into 2 groups as follows:

Table 4.1: Factors affecting to decision of choosing design-construction options

No Factors Referred source

A. Factors relating to energy consumption

A1 Total annual electricity usage [18]

A2 Total annual CO2 emission Expert

A3 Total annual discomfort hours [18]

B. Factors relating to designing features

B1 Sound insulation capacity Personal experience

B2 Firefighting capacity Expert

B3 Aesthetic aspect Expert

B4 Impact to Structure Expert

83%

17%

Have joined Have not joined

No Factors Referred source

B5 Durability Expert

B6 Impact to method statement and construction schedule Personal experience

B7 Impact to operation and maintenance Expert

B8 Availability in market Personal experience

B9 Recycling capacity Expert

1.2. Questionnaire Design

Details of the Questionnaire are performed in Appendix 1 1.3. Survey results

Details of the Survey results are performed in Appendix 2 1.4. Analyzing the characteristics of the study sample

The suitability of the survey participants

Table 4.2: Percentage of participants who have ever joined in Green Building projects or energy efficient buildings

No Response Quantity Percentage Graph

1 Have joined 43 83%

2 Have not

joined 9 17%

Total 52 100%

Comment: The majority of survey subjects have ever participated in the process of implementing green buildings or energy efficient buildings, accounting for 82.7%, the rate of never participating accounts for 17.3%, however these people have researched and updated Update knowledge about green building, energy efficient building.

Table 4.3: Years of experience of the survey participants

No Response Quantity Percentage Graph

1 1-5 years 5 10%

2 5-10 years 23 44%

3 More than 10

years 24 46%

Total 52 100%

Comment: The percentage of respondents who have worked in the construction industry for more than 5 years accounts for 90%, showing that the majority of respondents have a lot of experience, which contributes to the respondents' responses.

very reliable.

Expertise of the survey participants Table 4.4: Expertise of the survey participants

No Response Quantity Percentage Graph

1 Project

Manager 4 8%

2 Designer 21 40%

3 Supervisor 8 15%

4 Quantity

Surveyor 19 37%

Total 52 100%

10%

44%

46%

1-5 years 5-10 years

> 10 years

8%

40%

15%

37%

Project Manager Designer Supervisor Quantity Surveyor

Comment: The data shows that the expertise of the survey subjects is very diverse, which contributes to making the research results more multidimensional and objective.

In which, the proportion of architects, engineers and project managers accounts for a total of 48%, showing that the results come from subjects with deep knowledge of the research topic accounting for a high proportion.

Roles of the survey participants Table 4.5: Roles of the survey participants

No Response Quantity Percentage Graph

1 Client 17 33%

2 Project

Manager 7 13%

3 Design

Consultant 8 15%

4 Contractor 12 23%

5 QS Consultant 8 15%

Total 52 100%

Comment: Statistics show that the roles of survey subjects are very diverse, in which the Client and Contractor groups account for a higher proportion, which contributes to the multidimensional and objective response results.

33%

15% 13%

23%

15%

Client

Project Manager Design Consultant Contractor QS Consultant

Reliability of the scale is analyzed by IBM SPSS Statistics 22 Table 4.6: 1st results of reliability testing

No

Scale Mean if Item Deleted

Scale Variance if

Item Deleted

Corrected Item-Total Correlation

Cronbach’s Alpha if

Item Deleted

Cronbach’s Alpha A. Factors relating to energy consumption

A1 173.85 237.897 0.501 0.669

0.713

A2 181.00 202.353 0.602 0.544

A3 178.46 159.391 0.535 0.654

B. Factors relating to designing features

B1 455.87 5046.236 -0.052 0.877

0.859

B2 461.15 4452.094 0.515 0.852

B3 467.62 4435.339 0.436 0.856

B4 474.10 3889.736 0.819 0.823

B5 482.38 3536.633 0.837 0.815

B6 486.19 3584.629 0.764 0.824

B7 490.44 3656.095 0.665 0.836

B8 487.10 4119.696 0.399 0.867

B9 501.46 3701.038 0.763 0.824

As a result of the first test, the variable B1 – Sound insulation capacity is excluded because the total correlation coefficient is -0.052 < 0.3. After removing the variable B1, the second loop gives the following result:

Table 4.7: 2nd results of reliability testing

No

Scale Mean if Item Deleted

Scale Variance if

Item Deleted

Corrected Item-Total Correlation

Cronbach’s Alpha if

Item Deleted

Cronbach’s Alpha A. Factors relating to energy consumption

A1 173.85 237.897 0.501 0.669

0.713

A2 181.00 202.353 0.602 0.544

A3 178.46 159.391 0.535 0.654

B. Factors relating to designing features

B2 378.73 4486.632 0.493 0.877

0.877

B3 385.19 4459.256 0.426 0.881

No

Scale Mean if Item Deleted

Scale Variance if

Item Deleted

Corrected Item-Total Correlation

Cronbach’s Alpha if

Item Deleted

Cronbach’s Alpha

B4 391.67 3910.224 0.811 0.848

B5 399.96 3551.959 0.834 0.839

B6 403.77 3616.220 0.749 0.850

B7 408.02 3643.353 0.681 0.859

B8 404.67 4092.813 0.422 0.888

B9 419.04 3689.253 0.781 0.846

Comment:

The test results show that:

- Corrected Item-Total Correlations are suitable (All ≥ 0.3).

- Cronbach’s Alpha within 0,70 ≤ α ≤ 0.9 shows that the scale is good, meets the requirements of reliability.

1.6. Ranking factors

After evaluating the reliability of the scale, rank the factors based on the average value of the influence according to the response results of the survey subjects. This result will be converted to a 100-point scale to apply to the CBA assessment.

Table 4.8: Ranking factors through Mean values

Rank Variable Description Mean Std.

Deviation

100-point scale converted 1 A1 Total annual electricity usage 92.81 6.937 100 2 A3 Total annual discomfort hours 88.19 9.959 95

3 A2 Total annual CO2 emission 85.65 7.690 90

4 B2 Firefighting capacity 77.13 7.603 85

5 B3 Aesthetic aspect 70.67 8.913 75

6 B4 Impact to Structure 64.19 10.175 70

7 B5 Durability 55.90 13.268 60

8 B6 Impact to method statement

and construction schedule 52.10 13.767 55

9 B8 Availability on market 51.19 14.010 55

Deviation

converted 10 B7 Impact to operation and

maintenance 47.85 14.499 50

11 B9 Recycling capacity 36.83 12.627 40

1.7. One sample T-Test

Table 4.9: Results of One-sample T-Test

One-Sample T-Test Test value = 50 Variable t df

A1 44.500 51 0.000 42.808 40.88 44.74

A2 33.434 51 0.000 35.654 33.51 37.79

A3 27.655 51 0.000 38.192 35.42 40.96

B2 25.736 51 0.000 27.135 25.02 29.25

B3 16.726 51 0.000 20.673 18.19 23.15

B4 10.058 51 0.000 14.192 11.36 17.03

B5 3.209 51 0.002 5.904 2.21 9.60

B6 1.098 51 0.277 2.096 -1.74 5.93

B7 -1.071 51 0.289 -2.154 -6.19 1.88

B8 .614 51 0.542 1.192 -2.71 5.09

B9 -7.523 51 0.000 -13.173 -16.69 -9.66

Comment:

The variables A1, A2, A3, B2, B3, B4, B5 and B9 have Sig. values (2-tailed) ≤ α = 0.05.

Therefore, rejecting H0 and accepting H1, that is, the average influence of all survey subjects on the influence of factors other than 50 (the mean value of the mean influence).

Variables B6, B7 and B8 have Sig values. (2-tailed) ≥ α = 0.05. Therefore, accepting H0 means accepting the assumption that the average influence of all survey subjects on the influence of the factors is equal to 50 (the mean value of the average influence).

1.8. Multi-sample testing

Mean difference analysis for the experience of the respondents Table 4.10: Mean difference analysis for the experience of the respondents

Test of Homogeneity of Variances

Var. Description Levene

Statistic df1 df2 Sig.

A1 Total annual electricity usage 0.369 2 49 0.694

A2 Total annual CO2 emission 0.910 2 49 0.409

A3 Total annual discomfort hours 3.645 2 49 0.033

B2 Firefighting capacity 0.039 2 49 0.962

B3 Aesthetic aspect 2.236 2 49 0.118

B4 Impact to Structure 2.818 2 49 0.069

B5 Durability 3.969 2 49 0.025

B6 Impact to method statement and

construction schedule 2.855 2 49 0.067

B7 Impact to operation and

maintenance 00.993 2 49 0.378

B8 Availability on market 3.217 2 49 0.049

B9 Recycling capacity 3.147 2 49 0.052

ANOVA

Var. Description Sum of

Squares df Mean

Square F Sig.

A1

Total annual electricity usage

Between Groups 101.826 2 50.913 1.061 0.354 Groups Within 2352.251 49 48.005

Total 2454.077 51 A2

Total annual CO2

emission

Between Groups 44.958 2 22.479 0.371 0.692 Groups Within 2970.812 49 60.629

Total 3015.769 51 A3

Total annual discomfort hours

Between Groups 279.573 2 139.786 1.433 0.248 Groups Within 4778.504 49 97.520

Total 5058.077 51 B2 Firefighting

capacity

Between Groups 184.816 2 92.408 1.639 0.205 Groups Within 2763.242 49 56.393

Total 2948.058 51

B3 Between Groups 465.980 2 232.990 3.184 0.050

Var. Description Sum of

Squares df Mean

Square F Sig.

Aesthetic aspect

Groups Within 3585.463 49 73.173 Total 4051.442 51

B4 Impact to Structure

Between Groups 717.840 2 358.920 3.855 0.028 Groups Within 4562.237 49 93.107

Total 5280.077 51 B5 Durability

Between Groups 1122.891 2 561.446 3.502 0.038 Groups Within 7855.628 49 160.319

Total 8978.519 51 B6

Impact to method &

construction

Between Groups 1061.152 2 530.576 3.021 0.058 Groups Within 8605.367 49 175.620

Total 9666.519 51 B7

Impact to operation &

maintenance

Between Groups 1526.292 2 763.146 4.067 0.023 Groups Within 9194.477 49 187.642

Total 10720.769 51 B8 Availability

on market

Between Groups 485.948 2 242.974 1.250 0.295 Groups Within 9524.129 49 194.370

Total 10010.077 51 B9 Recycling

capacity

Between Groups 796.664 2 398.332 2.661 0.080 Groups Within 7334.778 49 149.689

Total 8131.442 51

Robust Tests of Equality of Means

Var. Description Welch

Statistic df1 df2 Sig.

A1 Total annual electricity usage 1.214 2 13.230 0.328

A2 Total annual CO2 emission 0.340 2 12.198 0.718

A3 Total annual discomfort hours 0.627 2 10.089 0.554

B2 Firefighting capacity 1.813 2 11.931 0.205

B3 Aesthetic aspect 2.135 2 10.273 0.168

B4 Impact to Structure 2.165 2 10.390 0.164

B5 Durability 1.944 2 10.240 0.192

B6 Impact to method statement and

construction schedule 1.700 2 10.486 0.229

B7 Impact to operation and

maintenance 2.680 2 10.816 0.113

B8 Availability on market 0.481 2 10.238 0.632

Robust Tests of Equality of Means

Var. Description Welch

Statistic df1 df2 Sig.

B9 Recycling capacity 1.716 2 9.785 0.230

Comment:

The number of working years of the survey respondents did not affect the assessment of the influence of the factors: A1, A2, A3, B2, B3, B5, B6, B8, B9.

The number of working years of the survey respondents affected the assessment of the influence of the factors: B4, B7.

Mean difference analysis for the expertise of the respondents Table 4.11: Mean difference analysis for the expertise of the respondents

Test of Homogeneity of Variances

Var. Description Levene

Statistic df1 df2 Sig.

A1 Total annual electricity usage 1.687 3 48 0.182

A2 Total annual CO2 emission 1.897 3 48 0.143

A3 Total annual discomfort hours 0.750 3 48 0.528

B2 Firefighting capacity 0.138 3 48 0.937

B3 Aesthetic aspect 1.140 3 48 0.343

B4 Impact to Structure 3.494 3 48 0.023

B5 Durability 2.797 3 48 0.050

B6 Impact to method statement and

construction schedule 3.523 3 48 0.022

B7 Impact to operation and

maintenance 6.751 3 48 0.001

B8 Availability on market 3.754 3 48 0.017

B9 Recycling capacity 7.714 3 48 0.000

Var. Description Sum of

Squares df Mean

Square F Sig.

A1

Total annual electricity usage

Between Groups 346.886 3 115.629 2.634 0.060 Groups Within 2107.191 48 43.900

Total 2454.077 51 A2

Total annual CO2

emission

Between Groups 89.150 3 29.717 0.487 0.693 Groups Within 2926.620 48 60.971

Total 3015.769 51 A3

Total annual discomfort hours

Between Groups 290.799 3 96.933 0.976 0.412 Groups Within 4767.278 48 99.318

Total 5058.077 51 B2 Firefighting

capacity

Between Groups 217.990 3 72.663 1.278 0.293 Groups Within 2730.068 48 56.876

Total 2948.058 51 B3 Aesthetic

aspect

Between Groups 147.339 3 49.113 0.604 0.616 Groups Within 3904.104 48 81.335

Total 4051.442 51 B4 Impact to

Structure

Between Groups 823.217 3 274.406 2.955 0.042 Groups Within 4456.860 48 92.851

Total 5280.077 51 B5 Durability

Between Groups 1946.039 3 648.680 4.428 0.008 Groups Within 7032.481 48 146.510

Total 8978.519 51 B6

Impact to method &

construction

Between Groups 1340.750 3 446.917 2.577 0.065 Groups Within 8325.769 48 173.454

Total 9666.519 51 B7

Impact to operation &

maintenance

Between Groups 3478.514 3 1159.505 7.685 0.000 Groups Within 7242.256 48 150.880

Total 10720.769 51 B8 Availability

on market

Between Groups 997.125 3 332.375 1.770 0.165 Groups Within 9012.952 48 187.770

Total 10010.077 51 B9 Recycling

capacity

Between Groups 2301.039 3 767.013 6.315 0.001 Groups Within 5830.404 48 121.467

Total 8131.442 51

Robust Tests of Equality of Means

Var. Description Welch

Statistic df1 df2 Sig.

A1 Total annual electricity usage 2.691 3 18.354 0.076 A2 Total annual CO2 emission 0.459 3 16.675 0.714 A3 Total annual discomfort hours 8 3 21.849 0.001

B2 Firefighting capacity 1.064 3 11.383 0.402

Robust Tests of Equality of Means

Var. Description Welch

Statistic df1 df2 Sig.

B3 Aesthetic aspect 0.754 3 12.943 0.54

B4 Impact to Structure 1.535 3 15.925 0.244

B5 Durability 2.332 3 20.929 0.103

B6 Impact to method statement and

construction schedule 1.369 3 16.745 0.287

B7 Impact to operation and

maintenance 2.806 3 10.375 0.092

B8 Availability on market 0.882 3 18.894 0.468

B9 Recycling capacity 4.724 3 13.138 0.019

Comment:

The expertise of the survey respondents did not affect the assessment of the influence of the factors: A1, A2, A3, B2, B3, B4, B6, B7, B8.

The expertise of the survey respondents affected the assessment of the influence of the factors: B5, B9.

Mean difference analysis for the role of the respondents Table 4.12: Mean difference analysis for the role of the respondents

Test of Homogeneity of Variances

Var. Description Levene

Statistic df1 df2 Sig.

A1 Total annual electricity usage 1.009 4 47 0.412

A2 Total annual CO2 emission 3.782 4 47 0.010

A3 Total annual discomfort hours 2.366 4 47 0.066

B2 Firefighting capacity 1.174 4 47 0.334

B3 Aesthetic aspect 2.476 4 47 0.057

B4 Impact to Structure 2.913 4 47 0.031

B5 Durability 5.362 4 47 0.001

B6 Impact to method statement and

construction schedule 3.326 4 47 0.018

Var. Description Levene

Statistic df1 df2 Sig.

B7 Impact to operation and

maintenance 4.454 4 47 0.004

B8 Availability on market 2.738 4 47 0.040

B9 Recycling capacity 10.830 4 47 0.000

ANOVA

Var. Description Sum of

Squares df Mean

Square F Sig.

A1

Total annual electricity usage

Between Groups 175.578 4 43.895 0.905 0.469 Groups Within 2278.499 47 48.479

Total 2454.077 51 A2

Total annual CO2

emission

Between Groups 168.088 4 42.022 0.694 0.600 Groups Within 2847.681 47 60.589

Total 3015.769 51 A3

Total annual discomfort hours

Between Groups 490.318 4 122.579 1.261 0.299 Groups Within 4567.759 47 97.186

Total 5058.077 51 B2 Firefighting

capacity

Between Groups 118.671 4 29.668 0.493 0.741 Groups Within 2829.387 47 60.200

Total 2948.058 51 B3 Aesthetic

aspect

Between Groups 100.253 4 25.063 0.298 0.878 Groups Within 3951.190 47 84.068

Total 4051.442 51 B4 Impact to

Structure

Between Groups 638.649 4 159.662 1.617 0.186 Groups Within 4641.428 47 98.754

Total 5280.077 51 B5 Durability

Between Groups 1382.652 4 345.663 2.139 0.091 Groups Within 7595.868 47 161.614

Total 8978.519 51 B6

Impact to method &

construction

Between Groups 1070.669 4 267.667 1.464 0.228 Groups Within 8595.850 47 182.890

Total 9666.519 51 B7

Impact to operation &

maintenance

Between Groups 2389.348 4 597.337 3.370 0.017 Groups Within 8331.421 47 177.264

Total 10720.769 51 B8 Availability

on market

Between Groups 1568.958 4 392.239 2.184 0.085 Groups Within 8441.119 47 179.598

Total 10010.077 51 B9 Recycling

capacity

Between Groups 1753.764 4 438.441 3.231 0.020 Groups Within 6377.678 47 135.695

Total 8131.442 51

Robust Tests of Equality of Means

Var. Description Welch

Statistic df1 df2 Sig.

A1 Total annual electricity usage 0.958 4 19.163 0.453 A2 Total annual CO2 emission 0.436 4 20.016 0.781 A3 Total annual discomfort hours 1.54 4 18.294 0.232

B2 Firefighting capacity 0.713 4 20.259 0.593

B3 Aesthetic aspect 0.18 4 18.865 0.946

B4 Impact to Structure 1.246 4 19.908 0.324

B5 Durability 1.838 4 21.217 0.159

B6 Impact to method statement and

construction schedule 2.319 4 21.017 0.091

B7 Impact to operation and

maintenance 3.578 4 20.742 0.023

B8 Availability on market 1.715 4 18.43 0.189

B9 Recycling capacity 1.098 4 21.845 0.383

Comment:

The role of the survey respondents did not affect the assessment of the influence of the factors: A1, A2, A3, B2, B3, B4, B5, B6, B8, B9.

The role of the survey respondents affected the assessment of the influence of the factors: B7.

5.1. Simulation of building energy consumption using DesignBuilder 5.1.1. Procedure of building an energy simulation model

Simulation models can be built directly in DesignBuilder software, or during the design process, can take advantage of information models built by architecture, structural, and MEP disciplines to build energy simulation models. By assigning geolocation parameters and selecting weather stations to provide data for analysis. Then proceed to export the energy model with GBxml cloud to import into DesignBuilder software. The designer selects the parameters to be calculated, assigns the input parameter values and runs the simulation to get the results.

Preliminary architecture model

Energy model in GBxml format

Energy model in DesignBuilder

Figure 5.1: Procedure of building an energy simulation model in DesignBuilder

5.1.2. Model information

A typical office area 144 m2, height from floor to roof 3.8m, weather data at Tan Son Hoa station, Ho Chi Minh City.

5.1.3. Defining design variables and input data to DesignBuilder Defining design variables

Referring to the available studies, the ANSI/ASHRAE/IES Standard: Energy Standard for Buildings Except Low-Rise Residential Buildings, LEED standard, the author has determined the variable of the parameters used in the simulation model as shown in the table below:

Table 5.1: Define design variables

Item Variables Symbol Unit Reference

1 Ratio of Window on Wall WWR % [18], [33],

[34], [16]

2 Building Orientation BO (o) [33], [34],

[16]

3 Heat transfer rate (U-value) of external Wall

W-U W/(m2.K) [18], [33], [34]

4 Heat transfer rate (U-value) of Glazing

G-U W/(m2.K) [18], [33], [34]

5 Solar Heat Gain Coefficient of Glazing

SHGC [18], [33],

[34]

6 Heat transfer rate (U-value) of Roof R-U W/(m2.K) [18], [33], [34]

Figure 5.2: Input weather data

7 Cooling setting point CSP C [33], [34]

8 Cooling coefficient of performance CoP [33], [34], [16]

Explanation of technical terms:

- U-Value is the glazing parameter that characterizes the heat transfer through the central part of the glazing, i.e. without edge effects, and expresses the steady-state density of heat transfer rate per temperature difference between the environmental temperatures on each side. US Standard units are Btu/hrãft²ãF and SI / Metric units are W/m2 K.

- Solar Heat Gain Coefficient (SHGC) is the sum of the solar direct transmittance and the secondary heat transfer factor of the glazing towards the inside, the latter resulting from heat transfer by convection and longwave IR-radiation of that part of the incident solar radiation which has been absorbed by the glazing.

- The coefficient of performance or COP (sometimes CP or CoP) of a heat pump, refrigerator or air conditioning system is a ratio of useful heating or cooling provided to work (energy) required.

Input design variables to DesignBuilder:

Based on the author's experience, the values of the design variables mentioned above are determined in detail as follows:

Table 5.2: Design variables data

Item Variables Symbol Options Structure Description 1 Ratio of

Window on Wall

WWR (%)

30 – 60 (in steps of 5%)

2 Building Orientation

BO (o) 0 – 350 (in steps of 10o)

3 Heat transfer rate (U-

W-U

(W/(m2.K))

Hollow brick wall – U=1.743

- Outer layer: mortar 25mm thk.

Item Variables Symbol Options Structure Description value) of

external Wall

- Middle layer: 2 layers of burned clay brick – 180mm thk.

- Inner layer: mortar 15mm thk

AAC block wall – U = 0.957

- Outer layer: mortar 15mm thk.

- Middle layer: 1 layer of Autoclaved Aerated Concrete 200mm thk.

- Inner layer: mortar 15mm thk.

Concrete block wall – U = 0.983

- Outer layer: mortar 25mm thk.

- Middle layer: 1 layer of Aggregate concrete block 190mm thk.

- Inner layer: mortar 15mm thk.

4 Heat transfer rate (U- value) of Glazing

G-U

(W/(m2.K))

Laminated Solar control glass, 13.52mm - U = 5.1

- Solar control glass Tempered 6mm thk.

- PVB 1.52mm thk.

- Clear glass

Tempered 6mm thk.

Laminated Low E glass, 13.52mm - U = 5.1

- Full tempered glass with Low E coating 6mm thk.

- PVB 1.52mm thk.

- Clear glass

Tempered 6mm thk.

Laminated Clear glass, 13.52mm - U = 5.3

- Clear glass

Tempered 6mm thk.

- PVB 1.52mm thk.

- Clear glass

Tempered 6mm thk.

5 Solar Heat Gain

Coefficient of Glazing

SHGC Laminated Solar control glass,

13.52mm - SHGC = 0.45

Laminated Low E glass, 13.52mm - SHGC = 0.43 Laminated Clear glass, 13.52mm – SHGC = 0.82 6 Heat transfer

rate (U- value) of Roof

R-U

(W/(m2.K))

Concrete roof insulated by Rockwool – U = 0.339

- Ceiling tiles: 80mm thk.

- Cement screeding:

20mm thk.

- Rockwool: 50mm thk.

- Cement screeding:

20mm thk.

- Waterproofing layer:

20mm thk

Item Variables Symbol Options Structure Description - Roof concrete:

150mm thk.

Concrete roof insulated by XPS form – U = 0.282

- Ceiling tiles: 80mm thk.

- Cement screeding:

20mm thk.

- XPS foam: 50mm thk.

- Cement screeding:

20mm thk.

- Waterproofing layer:

20mm thk - Roof concrete:

150mm thk.

Concrete roof without insulation – U = 0.535

- Ceiling tiles: 80mm thk.

- Cement screeding:

20mm thk.

- Waterproofing layer:

20mm thk - Roof concrete:

150mm thk.

7 Cooling setting point

CSP (oC) 24 – 26 (in steps of 0.5 degree)

8 Cooling coefficient of performance

CoP 2.5 – 4.5 (in steps of 0.1)

Figure 5.3: Design variables setting in DesignBuilder Define objectives and additional outputs

DesignBuilder software runs simulations based on input variables information and gives optimal results for objective functions. The software allows the input of up to 2 objective functions. The remaining output information if necessary, can be set in Additional Output.

Based on the survey results on the importance of design elements towards sustainable construction, the author chooses two objective functions as follows:

1. Minimize electricity usage 2. Minimize Discomfort hours

Remaining factor – CO2 emission will be included in Additional output.

Figure 5.4: Objectives and Outputs setting

5.1.4. Simulation results:

After entering the design parameter values, the model is run to generate a complete data set of results for each parameter set and find the optimal parameter sets of the objective functions as follows:

Table 5.3: Simulation results in DesignBuilder

No. WW

%

CST

(°C) CoP BO

(°) Glazing E-Wall Roof Electricity (kWh)

Discomfort

(hr) CO2 (kg)

1 50 24 4.2 0 Glazing 1 Wall 3 Roof 2 33,047.0 51.9 20,026.5

2 45 24 3.7 110 Glazing 3 Wall 3 Roof 3 35,039.0 148.6 21,233.6 3 55 25 4.5 230 Glazing 2 Wall 1 Roof 1 32,121.5 1,176.6 19,465.6 4 60 25 4.1 340 Glazing 3 Wall 2 Roof 3 32,932.3 1,097.8 19,957.0 5 55 26 3.9 220 Glazing 2 Wall 2 Roof 2 32,418.1 2,008.9 19,645.3 6 45 25 2.8 300 Glazing 1 Wall 1 Roof 2 36,931.1 1,140.6 22,380.2 7 55 25.5 4.3 270 Glazing 1 Wall 2 Roof 1 31,856.2 1,597.0 19,304.8 8 45 25 3.7 300 Glazing 2 Wall 2 Roof 2 33,529.0 852.5 20,318.5 9 45 25 4.1 160 Glazing 3 Wall 3 Roof 1 33,209.9 1,187.8 20,125.2 10 40 24 3.9 130 Glazing 1 Wall 2 Roof 1 33,976.1 48.8 20,589.5 11 60 24.5 4.1 170 Glazing 1 Wall 3 Roof 3 33,146.2 236.3 20,086.6 12 45 25 3.3 220 Glazing 2 Wall 2 Roof 1 34,853.8 925.3 21,121.4 13 35 25.5 3.9 330 Glazing 1 Wall 1 Roof 2 32,860.1 1,653.4 19,913.2 14 60 24.5 3.7 0 Glazing 2 Wall 3 Roof 3 33,964.3 219.5 20,582.4 15 50 25 2.6 340 Glazing 3 Wall 2 Roof 2 38,102.5 1,013.9 23,090.1 16 45 25 4.5 190 Glazing 2 Wall 3 Roof 1 31,933.8 912.5 19,351.8 17 45 24 4.4 270 Glazing 1 Wall 2 Roof 3 32,722.3 58.4 19,829.7 18 30 26 2.5 210 Glazing 3 Wall 1 Roof 2 38,051.9 2,107.1 23,059.4 19 40 24 3 140 Glazing 3 Wall 1 Roof 2 37,905.9 234.9 22,970.9 20 60 26 2.7 120 Glazing 2 Wall 2 Roof 2 36,313.5 2,044.6 22,006.0 21 50 24 4.2 130 Glazing 1 Wall 3 Roof 1 33,252.9 49.4 20,151.2 22 50 24.5 3.9 310 Glazing 1 Wall 2 Roof 1 33,403.0 205.8 20,242.2 23 30 26 4.4 130 Glazing 1 Wall 2 Roof 1 31,513.2 2,048.0 19,097.0 24 45 26 3.8 330 Glazing 3 Wall 2 Roof 2 32,868.3 2,008.5 19,918.2 25 55 26 3.9 180 Glazing 1 Wall 3 Roof 2 32,438.1 2,046.6 19,657.5 26 35 24.5 4.4 250 Glazing 1 Wall 1 Roof 1 32,619.9 437.3 19,767.7

… … … … … … … … … … …

3332 60 25 2.5 50 Glazing 1 Wall 2 Roof 2 38,300.1 892.6 23,209.8 3333 30 24.5 3.5 210 Glazing 2 Wall 2 Roof 2 34,640.1 208.0 20,991.9 3334 60 24 2.5 170 Glazing 3 Wall 2 Roof 2 40,327.9 123.8 24,438.7 3335 60 24 4.5 290 Glazing 2 Wall 2 Roof 2 32,452.5 54.6 19,666.2 3336 60 24 2.7 50 Glazing 2 Wall 2 Roof 3 38,397.8 56.6 23,269.0 3337 50 24 4.2 30 Glazing 2 Wall 2 Roof 2 33,112.3 51.9 20,066.0 3338 45 24 2.5 70 Glazing 2 Wall 2 Roof 2 39,468.5 48.4 23,917.9 3339 55 24 3.5 110 Glazing 1 Wall 3 Roof 2 35,088.0 48.8 21,263.3 3340 55 24 4.4 120 Glazing 2 Wall 3 Roof 2 32,809.7 49.3 19,882.7 3341 60 26 4.1 90 Glazing 1 Wall 2 Roof 2 31,999.0 2,038.0 19,391.4 3342 40 24 2.5 20 Glazing 2 Wall 2 Roof 2 39,363.5 52.1 23,854.3 3343 45 25 3.7 20 Glazing 1 Wall 3 Roof 2 33,553.8 854.4 20,333.6 3344 30 24 2.7 20 Glazing 2 Wall 2 Roof 2 38,215.3 52.1 23,158.5 3345 45 25.5 4.3 350 Glazing 1 Wall 2 Roof 2 31,821.1 1,568.9 19,283.6

Algorithm by Python Programming Language.

5.2.1. Procedure of creating an energy prediction model

The first step of creating a prediction model is transforming qualitative variables such as Glazing, External Wall and Flat roof into quantitative variables. As described in Table 5.1 and 5.2, Glazing types have two quantitative variables (SHGC and U-value), External wall and Flat roof have one quantitative variable (U-value).

The model is developed on Spyder editor with Python programming language version 3.10 using tools available in Scitkit-learn library for Random Forest algorithm as below procedure:

Table 5.4: Procedure of creating a prediction model for energy consumption

Step 1: Import libraries

Step 2: Import and split data

Step 3: Train RF model

Step 4: Evaluate accuracy of model after training

Một phần của tài liệu Combining building information modeling (bim) and choosing by advantages (cba) method to select design construction solutions toward sustainable construction in viet nam (Trang 43 - 109)

Tải bản đầy đủ (PDF)

(109 trang)