^{1}

^{*}

^{1}

^{1}

^{1}

^{1}

^{1}

^{2}

It is very important to consider proper intelligent integration and locations of renewable energy sources into the built environment for developing smart cities. Wind speed distribution study in the built environment is very essential for analyzing the wind turbine performance located in the built environment. In this work, the building layout like nozzle is proposed and the objective is to optimize the building layout for increasing electrical energy output of wind turbine, assumed to be installed in actual cities of Japan. The wind speed distribution across buildings is numerically simulated by using CFD-ACE+. Wind turbine power output is estimated using the power curve of a real commercial wind turbine and wind speed distribution is simulated using CFD software. The meteorological data of Fukushima city and Tsu city of Japan are utilized for evaluating the wind speed distribution profile across the building and for finding the electrical energy output from wind turbine. The proposed building models, which have the angle between two buildings like nozzle of 90 °, 135 ° and 180 °, can provide the wind acceleration at the back of buildings for the wind blowing from the main wind direction and the angle of 135 ° is optimum building layout. In the case of installing the proposed building model in Fukushima city and Tsu city, the wind energy output in winter season is higher while that in summer season is lower irrespective of the buildings’ angle. The interaction between the change in frequency distribution of wind speed and direction throughout the year and the location of open tip of building model decides the power generation characteristics of the proposed building model.

For city planning, it is very important to propose the building model for integrating and locating environment friendly energy systems. It is effective to construct a smart city, consisting proposed buildings orientations for not only meeting the energy demand but also decreasing the amount of CO_{2 }emitted by power generation. In developing environment friendly smart cities, it is needed to analyze the intelligent integration with proper location of intermittent renewable energy sources in the built environment [

The wind speed profiles on the tall building rooftops are presented in the reference [

It is observed through literature survey that no study has been reported yet for analyzing built environment wind turbine power output due to wind speed distribution considering building layouts [

In the present paper, the objective is to optimize building layout for increasing electrical energy output from wind turbine assumed to be installed in the actual city in Japan. In this work, the angle between two buildings, like nozzle (α), is changed by 90˚, 135˚ and 180˚. According to our previous study [

The wind speed data base of the Japan Meteorological Agency [

Two buildings are configured/arrayed as a nozzle (

In this work, the wind speed distribution across the buildings is developed/simulated through the commercial CFD software CFD-ACE+ (ESI Japan, Ltd.). The previous works [

The wind at the area at the back of buildings is thought to be available for power generation by wind turbine, since the wind speed would be accelerated by blowing through the buildings. Three points at the back of building which are apart from the building by 20, 30 and 40 m (x/L = 1.25, 1.875, 2.50) is assumed as the installation point of wind turbine. The wind speed for calculating the power generated by wind turbine is obtained on 1049 points located in the area where the rotor of wind turbine rotates, that is, the swept rotor area. The wind speed at each point on the swept rotor area is the averaged speed in the local area of 0.5 m × 0.5 m. By considering the wind speed distribution of this local wind speed, the wind energy can be calculated. Root mean cube wind speed to x axis direction is estimated by using the following equation:

where U_{rmc} is the root mean cube wind speed to x axis direction, Q_{x} is the summation of wind energy to x axis direction on each point for calculating wind speed distribution, N is the points for calculating wind speed distribution in the swept rotor area A (= 1049 points), ρ is the density of wind, A is the swept rotor area. Wind energy at each point on the swept rotor area is calculated by the following equation:

where Q_{x}_{,i} is the wind energy to x axis direction at each point, A_{i} is the area of each point which is equal to 0.5 m × 0.5 m, U_{i} is the wind speed to x axis direction at each point for calculating wind energy.V_{rmc} which is the root mean cube wind speed to y axis direction is estimated by the same calculation way of U_{rmc}. The average root mean square wind speed to horizontal surface of the swept rotor area U_{h}_{, ave} is calculated by the following equation:

Here, the wind speed and wind energy to z axis direction are ignored since the rotor of wind turbine can’t move toward z axis direction and wind energy to z axis direction can’t be utilized.

Building design procedure has been explained in reference [

Characteristics | Figure |
---|---|

Rated power (kW) | 50 |

Start wind speed (m/s) | 3 |

Cut-in wind speed (m/s) | 3 |

Cut-out wind speed (m/s) | 25 |

Rotor diameter (m) | 18 |

Rotor speed (rpm) | 60 |

Hub height (m) | 30 |

Condition | Configuration |
---|---|

Density of wind at inlet (kg/m^{3}) | 1.166 |

Temperature of wind at inlet (K) | 293 |

Pressure of wind at inlet (MPa) | 0.10 |

Kinetic viscosity of wind (m^{2}/s) | 1.56 × 10^{−5} |

Wind speed at inlet (m/s) | U = U_{0} × (z/30)^{0.25} (U_{0} = 3.00 - 12.00) |

Slip on side wall of building (m/s) | V = (0.41 × |l|)^{0.25}U |

Turbulent flow model | RNG k-ε model |

Turbulent energy (m^{2}/s^{2}) | 0.025 |

Dissipation rate (m^{2}/s^{2}) | (1.58 × 10^{−3})/z |

Calculation number (-) | 10,000 |

Residue of each parameter (-) | <1.0 × 10^{−5} |

Calculation state | Steady state |

by RNG k-ε model. Calculation number is set 10,000 times. This calculation number should be appropriate since the residue of each parameter under each numerical simulation condition keeps a stable small value after 500 times calculation. Wind speed at inlet of the model is set by the following equation:

where U is the wind speed in x direction, U_{0} is the initial wind speed at z = 30 m, which is changed from 3.0 m/s to 12.0 m/s, z is height. U_{0} = 10.0 m/s the rated wind speed of AEOLOS wind turbine of 50 kW class [_{0} at z = 30 m which is the hub height of wind turbine when the wind reaches to the building.

In this work, wind speed data for Fukushima city and Tsu city are used from the Japan Meteorological Agency [_{0}:

where U_{o} is the wind speed at observation height for each city in the data base, z_{o} is the wind speed observation height for each city in the data base, a is the wind shear coefficient. In this study, α = 0.25, corresponding to the country area with many trees, is adopted; since Fukushima city and Tsu city have large country area as well as urban area.

In this study, the buildings are located as nozzle, therefore the wind inflow direction is important for obtaining the wind blowing through the buildings. The layouts of the buildings are decided based on the wind speed direction. The wind speed directions and buildings with α = 135˚ are given in

Assuming the symmetry to the main wind direction, the wind speed distributions around the buildings for the inflow angles (β) of 22.5˚, 45˚ and 67.5˚ are simulated to evaluate the effect of six angular inflows on the wind

speed distribution in the case of buildings with α = 135˚ and 180˚. When α is 90˚, the wind speed distributions around the buildings for β = 22.5˚ and 45˚ are simulated to evaluate the effect of four angular inflows on the wind speed distribution. Wind speed at inlet of the model is set by Equations (6) and (7), when the effect of inflow angle is considered.

where V is the wind speed in y direction, β is inflow angle.

The contour of wind speed distribution in x direction (U) around the buildings with α = 135˚ for U_{0} = 10.0 m/s at z = 30 m are given in _{0} of 10.0 m/s.

It is important to decide the location point of wind turbine in order to obtain the accelerated wind. In our previous studies [_{h}_{, ave} is an important wind speed which is used for the power curve to calculate the power output of wind turbine, Tables 3-5 list the

U_{0} (m/s) | 3.0 | 4.0 | 5.0 | |||
---|---|---|---|---|---|---|

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.10 | 1.35 | 1.10 | 3.70 | 1.10 | 7.80 |

1.875 | 1.09 | 1.29 | 1.09 | 3.53 | 1.09 | 7.43 |

2.5 | 1.06 | 1.16 | 1.06 | 3.21 | 1.06 | 6.77 |

U_{0} (m/s) | 6.0 | 7.0 | 8.0 | |||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.10 | 14.10 | 1.10 | 7.80 | 1.10 | 14.10 |

1.875 | 1.09 | 13.43 | 1.09 | 7.43 | 1.09 | 13.43 |

2.5 | 1.06 | 12.26 | 1.06 | 6.77 | 1.06 | 12.26 |

U_{0} (m/s) | 9.0 | 10.0 | 11.0 | |||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.09 | 49.86 | 1.09 | 52.36 | 1.09 | 51.46 |

1.875 | 1.07 | 46.72 | 1.07 | 52.55 | 1.07 | 51.64 |

2.5 | 1.03 | 41.77 | 1.03 | 52.87 | 1.04 | 51.95 |

U_{0} (m/s) | 12.0 | |||||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | ||||

1.25 | 1.10 | 50.57 | ||||

1.875 | 1.08 | 50.71 | ||||

2.5 | 1.05 | 50.98 |

U_{0} (m/s) | 3.0 | 4.0 | 5.0 | |||
---|---|---|---|---|---|---|

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.09 | 1.29 | 1.09 | 3.56 | 1.09 | 7.52 |

1.875 | 1.08 | 1.24 | 1.08 | 3.45 | 1.08 | 7.30 |

2.5 | 1.05 | 1.12 | 1.06 | 3.24 | 1.06 | 6.86 |

U_{0} (m/s) | 6.0 | 7.0 | 8.0 | |||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.08 | 13.23 | 1.09 | 22.34 | 1.09 | 34.12 |

1.875 | 1.07 | 12.84 | 1.08 | 21.69 | 1.07 | 32.18 |

2.5 | 1.05 | 12.08 | 1.05 | 19.83 | 1.04 | 29.41 |

U_{0} (m/s) | 9.0 | 10.0 | 11.0 | |||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.08 | 48.01 | 1.09 | 52.39 | 1.08 | 48.01 |

1.875 | 1.06 | 45.27 | 1.08 | 52.48 | 1.06 | 45.27 |

2.5 | 1.03 | 41.37 | 1.05 | 52.68 | 1.03 | 41.37 |

U_{0} (m/s) | 12.0 | |||||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | ||||

1.25 | 1.09 | 52.39 | ||||

1.875 | 1.08 | 52.49 | ||||

2.5 | 1.05 | 52.68 |

U_{0} (m/s) | 3.0 | 4.0 | 5.0 | |||
---|---|---|---|---|---|---|

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.15 | 1.57 | 1.15 | 4.30 | 1.15 | 9.02 |

1.875 | 1.15 | 1.56 | 1.15 | 4.25 | 1.15 | 8.97 |

2.5 | 1.13 | 1.48 | 1.13 | 4.07 | 1.13 | 8.54 |

U_{0} (m/s) | 6.0 | 7.0 | 8.0 | |||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.15 | 16.33 | 1.15 | 26.74 | 1.15 | 40.50 |

1.875 | 1.15 | 16.15 | 1.15 | 26.53 | 1.15 | 40.28 |

2.5 | 1.13 | 15.39 | 1.13 | 25.25 | 1.13 | 38.44 |

U_{0} (m/s) | 9.0 | 10.0 | 11.0 | |||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) | U_{h}_{,ave}/U_{0} (-) | P (kW) |

1.25 | 1.15 | 52.78 | 1.15 | 51.88 | 1.15 | 50.96 |

1.875 | 1.15 | 52.82 | 1.14 | 51.92 | 1.15 | 51.00 |

2.5 | 1.13 | 52.94 | 1.12 | 52.09 | 1.13 | 51.18 |

U_{0} (m/s) | 12.0 | |||||

x/L (-) | U_{h}_{,ave}/U_{0} (-) | P (kW) | ||||

1.25 | 1.15 | 50.04 | ||||

1.875 | 1.15 | 50.08 | ||||

2.5 | 1.13 | 50.27 |

variation of U_{h}_{, ave} in the swept rotor area at x/L = 1.25, 1.875 and 2.50 under the different U_{0} conditions for α = 90˚, 135˚ and 180˚, respectively. In these tables, U_{h}_{, ave} is estimated under the condition that β is 0˚. The power output of wind turbine P corresponding to each U_{h}_{, ave} is also listed in these tables.

From these tables, it is seen that U_{h}_{, ave} is greater than U_{0} for every condition. Hence, it is revealed that the proposed building model can provide the wind acceleration under the investigated conditions. Considering the location point of wind turbine, the highest U_{h}_{, ave} is obtained at x/L = 1.25 for every α and U_{0} condition though P is relatively smaller compared to the other x/L under the condition that U_{0} is over 10.0 m/s. Since the rated wind speed of supposed wind turbine is 10.0 m/s, the P decreases with the increase in wind speed gradually under the condition that U_{0} is over 10.0 m/s according to the power curve [

Since the wind blows from different directions actually, it is necessary to evaluate the effect of wind direction on wind speed distribution around buildings. In this study, the wind speed distributions around the buildings for β = 22.5˚, 45˚ and 67.5˚ are simulated. As an example of the simulation results, the contour of wind speed U distribution around buildings for U_{0} = 10.0 m/s at z = 30 m on x?y cross section for angular inflow direction of β = 45˚ are given in

To optimize the building layout for increasing electrical energy output of wind turbine assumed to be installed in actual cities of Japan, the wind speed data of five years (from 2010 to 2014) for Fukushima city and Tsu city are used from the Japan Meteorological Agency [

The hourly data on wind speed and direction are obtained for Fukushima city and Tsu city from the Japan Meteorological Agency [

day in January is shown as an example. From these figures, it is observed that the higher power energy of wind turbine is obtained in the daytime. This result is also confirmed from the other months. In these two cities, the wind energy is mainly obtained in the daytime. However, the hourly power energy output of wind turbine in the case of Tsu city has peaks in nighttime as well as daytime while there is only one peak in the case of Fukushima city. Since Tsu city is located facing Isewan bay, the wind blowing between land and bay is occurred resulting from the temperature difference between land and bay in the middle of daytime as well as that of nighttime.

From these tables, it is observed that the wind power energy of wind turbine in the building layout with α = 135˚ is the highest among the investigated α conditions. Though U_{h}_{, ave}/U_{0} obtained under the different U_{0} conditions for the building layout with α = 180˚ is relatively higher than that with α = 90˚ and 135˚ according to Tables 3-5, the difference among them is small. Therefore, it is believed that the optimum building layout is decided by the wind blowing from angular inflow directions.

_{0} = 10.0 m/s condition are shown. It is found that P for β = 45˚obtained under α = 135˚ condition is much higher compared to the other α conditions.

In addition, it is clarified from

α(˚) | January (kW∙h) | February (kW∙h) | March (kW∙h) | April (kW∙h) |
---|---|---|---|---|

90 | 8044 | 5545 | 6978 | 6430 |

135 | 8193 | 5734 | 7216 | 6685 |

180 | 5512 | 5940 | 9516 | 7608 |

α(˚) | May (kW∙h) | June (kW∙h) | July (kW∙h) | August (kW∙h) |

90 | 3417 | 1033 | 354 | 333 |

135 | 3826 | 1034 | 361 | 341 |

180 | 6100 | 855 | 299 | 1600 |

α(˚) | September (kW∙h) | October (kW∙h) | November (kW∙h) | December (kW∙h) |

90 | 1412 | 4993 | 4969 | 8086 |

135 | 1518 | 5367 | 5037 | 8342 |

180 | 1177 | 3294 | 3634 | 3604 |

α(˚) | Year (kW∙h) | |||

90 | 51,595 | |||

135 | 53,655 | |||

180 | 49,140 |

α(˚) | January (kW∙h) | February (kW∙h) | March (kW∙h) | April (kW∙h) |
---|---|---|---|---|

90 | 14,900 | 13,220 | 12,696 | 11,212 |

135 | 15,898 | 14,608 | 13,367 | 12,164 |

180 | 14,408 | 12,249 | 12,022 | 10,376 |

α(˚) | May (kW∙h) | June (kW∙h) | July (kW∙h) | August (kW∙h) |

90 | 9004 | 936 | 4761 | 5223 |

135 | 9954 | 958 | 4833 | 5438 |

180 | 8563 | 843 | 4418 | 4824 |

α(˚) | September (kW∙h) | October (kW∙h) | November (kW∙h) | December (kW∙h) |

90 | 3953 | 7506 | 8756 | 13,866 |

135 | 4258 | 7905 | 9130 | 14,719 |

180 | 3792 | 6980 | 8158 | 13,020 |

α(˚) | Year (kW∙h) | |||

90 | 106,034 | |||

135 | 113,233 | |||

180 | 99,652 |

Comparing the annual wind energy of output of turbine in the building layouts between Fukushima city and Tsu city, the annual wind energy for Tsu city is approximately twice as much as that for Fukushima city. According to the wind speed data of the Japan Meteorological Agency [

From the investigation in this study, it is revealed that the proposed building model performs the highest power generation production under α = 135˚ condition. The power generation characteristics of wind turbine in the building layouts assumed to be located in actual cities have been clarified. The actual wind condition influences the power generation performance of proposed building model. This study will continue to verify the feasibility of the proposed model with more wind speed data in the near future.

β(˚) | 22.5 | 45 | 67.5 |
---|---|---|---|

α(˚) | P (kW) | P (kW) | P (kW) |

90 | 5.29 | 0.62 | - |

135 | 5.29 | 1.26 | 0.21 |

180 | 5.34 | 0.21 | 0.37 |

In this study, building topologies/orientations in a smart city are investigated for finding out the wind speed profile in the built environment. The analysis of wind speed distribution in the built environment is very important for not only to find the mechanical wind stress but also to find the energy output from the built environment located wind turbines. This analysis is also useful for designing the building layouts in such a way to make a nozzle of the wind by using wind directions and then finding out the proper location of the wind turbine in the built environment. In this work, the objective is to optimize the building layout for increasing electrical energy output of wind turbine assumed to be installed in the actual cities of the Japan. The wind speed distribution across the buildings is simulated using CFD software. The output power of wind turbine is estimated by using the power curve of a real commercial wind turbine and the wind speed distribution around buildings using the wind speed data for Fukushima city and Tsu city. As a result, the following conclusions have been obtained:

1) The proposed building models with α = 90˚, 135˚, 180˚ can provide the U_{h}_{, ave} over U_{0} at x/L = 1.25, 1.875 and 2.50 under the different U_{0} conditions for the wind blowing from the main wind direction, i.e., β = 0˚.

2) Though the wind blowing for angular inflow direction is not accelerated well by the proposed building models, the optimum α which is 135˚ is decided by the wind blowing from angular inflow direction.

3) In the case of installing the proposed building model in Fukushima city and Tsu city, the wind power energy output in winter season is higher while that in summer season is lower irrespective of α. The interaction between the change in frequency distribution of wind speed direction throughout the year and the location of open tip of building model decides the power generation characteristics of the proposed building model.

4) It is revealed that the annual wind power energy of wind turbine in the building layouts is influenced by the annual mean wind speed significantly.

This study was financially supported by MAYEKAWA HOUONKAI FOUNDATION. The authors thank for its cooperation.

AkiraNishimura,MasanobuKakita,JunsukeMurata,ToshitakeAndo,YasunariKamada,MasafumiHirota,Mohan LalKolhe, (2015) Optimization of Building Layouts to Increase Wind Turbine Power Output in the Built Environment Assumed to Be Installed at Fukushima City and Tsu City in Japan. Smart Grid and Renewable Energy,06,279-292. doi: 10.4236/sgre.2015.69023