OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING HOYOUNG MAENG 1 and KYUNG HOON HYUN 2 1,2 Department of Interior Architecture Design, Hanyang University 1,2 {roach555|hoonhello}@hanyang.ac.kr Abstract. This paper identifies the correlation between the stiffness of a tensile membrane and wind-induced vibration. Here, a means by which to optimize the tensile membrane architecture and a harvester for the collection of energy from wind-induced vibration is proposed. Because its material property, the tensile membrane architecture is suitable for harvesting energy. While high stiffness is desired to ensure the stability of the tensile membrane architecture, the stiffness is negatively correlated with the energy collection. Thus, optimization considering the stiffness and the amount of harvestable energy in tensile membrane structures is conducted, broken down into the following tasks: 1) investigate the energy generator on the tensile membrane to combine a membrane type of architecture and wind energy harvesting; 2) set-up the computational environment with which to analyze the energy efficiency of the tensile membrane architecture; 3) visualize the resulting datasets to determine the relationship between the stiffness and the energy efficiency; and 4) define optimized tensile membrane architectures. Keywords. Tensile Membrane Architecture; Design Optimization; Sustainable Architecture; Computational Design; Wind Energy Harvesting. 1. Introduction At present, energy harvesting from vibrations is a notable subject and is part of the response to the global concern over energy and environmental issues (Abdelkefi 2016; Zhao and Yang 2018; Orrego et al. 2017). However, implementing wind-induced vibration-based energy harvesting techniques with a tensile membrane architecture has yet to be explored. Conventionally, vibration is a factor to be minimized in building structures, and doing so requires a refined analysis (BSI 2002). According to Stranghöner et al. (2016), tensile structures also require reduced wind-induced vibration to ensure their proper functioning and to prevent failure of the membrane. However, energy harvesting through generators requires vibration to maximize the amount of power generated. Moreover, the shape of the tensile membrane has a significant influence on the load paths for the pressure in these systems. The membrane must be prestressed to enhance the stiffness and to support material loads such as wind and snow. In this respect, RE: Anthropocene, Proceedings of the 25 th International Conference of the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA) 2020, Volume 1, 569-578. © 2020 and published by the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA), Hong Kong.
OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING
Mar 31, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
HOYOUNG MAENG1 and KYUNG HOON HYUN2 1,2Department of Interior Architecture Design, Hanyang University 1,2{roach555|hoonhello}@hanyang.ac.kr
Abstract. This paper identifies the correlation between the stiffness of a tensile membrane and wind-induced vibration. Here, a means by which to optimize the tensile membrane architecture and a harvester for the collection of energy from wind-induced vibration is proposed. Because its material property, the tensile membrane architecture is suitable for harvesting energy. While high stiffness is desired to ensure the stability of the tensile membrane architecture, the stiffness is negatively correlated with the energy collection. Thus, optimization considering the stiffness and the amount of harvestable energy in tensile membrane structures is conducted, broken down into the following tasks: 1) investigate the energy generator on the tensile membrane to combine a membrane type of architecture and wind energy harvesting; 2) set-up the computational environment with which to analyze the energy efficiency of the tensile membrane architecture; 3) visualize the resulting datasets to determine the relationship between the stiffness and the energy efficiency; and 4) define optimized tensile membrane architectures.
Keywords. Tensile Membrane Architecture; Design Optimization; Sustainable Architecture; Computational Design; Wind Energy Harvesting.
1. Introduction At present, energy harvesting from vibrations is a notable subject and is part of the response to the global concern over energy and environmental issues (Abdelkefi 2016; Zhao and Yang 2018; Orrego et al. 2017). However, implementing wind-induced vibration-based energy harvesting techniques with a tensile membrane architecture has yet to be explored. Conventionally, vibration is a factor to be minimized in building structures, and doing so requires a refined analysis (BSI 2002). According to Stranghöner et al. (2016), tensile structures also require reduced wind-induced vibration to ensure their proper functioning and to prevent failure of the membrane. However, energy harvesting through generators requires vibration to maximize the amount of power generated. Moreover, the shape of the tensile membrane has a significant influence on the load paths for the pressure in these systems. The membrane must be prestressed to enhance the stiffness and to support material loads such as wind and snow. In this respect,
RE: Anthropocene, Proceedings of the 25th International Conference of the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA) 2020, Volume 1, 569-578. © 2020 and published by the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA), Hong Kong.
570 H. MAENG AND K.H. HYUN
designers have explored various design concepts that are both structurally stable and aesthetically pleasing. Thus, it is crucial to find a balance between the amount of vibration and the degree of structural stability so as to maximize the energy harvesting performance while ensuring the serviceability of the tensile structure.
There are numerous related works on vibration-based energy harvesting through a membrane structure. A membrane-based energy harvesting device developed by Dong et al. (2015) had an adjustable resonant frequency of the membrane allowing the power output from vibration energy harvesting to be maximized. Allen and Smits (2001) investigated the feasibility of an “eel”-shaped piezoelectric membrane in underwater, aiming to maximize the strain energy and mechanical power. These studies concentrated on the development and evaluation of the efficiency of the generator itself. They focused on maximizing the amount of harvested energy, whereas shape optimization of the membrane was not considered. Therefore, the present paper aims to develop a method to simulate the energy harvesting performance and structural stability of a tensile membrane architecture depending on the shape and the membrane stiffness of the structure. To accomplish this, tensile membrane structure simulation software was developed with two major components. First, the proposed simulation software allows the generation of design variations based on three fundamental tensile structures: hyperbolic paraboloid (hypar), conic, and the barrel vault type. Second, the software estimates the amount of electrical power generated under various conditions, such as with different numbers of input sensors and different wind conditions. This paper focuses on developing a method to optimize the tensile membrane architecture for energy harvesting to assist designers when they make decisions about the design, including its form and the materials used.
2. Related Works 2.1. TENSILE ARCHITECTURE
Berger (1999) explained the advantages of a tensile structure, noting the “speed of construction, use of daylight, the reflection of heat from the sun, dispersal of interior sound, the beauty of the interior space and the excitement of the exterior sculpture.” Moreover, the high energy efficiency, possibility for daylight use, and heat-reflecting properties of a fabric structure can save energy and lower the general maintenance costs (Berger 1999). Tensile membrane structures offer efficient solutions that would not be possible with conventional building technologies. Tensile architectures have advantages in that they can be used in a variety of environments and may even be suitable for extreme environments such as that of the moon. Because the cost of transportation to the moon is extremely high, the structural materials must be as light as possible. The lunar structure must also be buildable with a minimum amount of manpower, as rapid construction lessens the time the workers are at risk. Tensile membranes meet these two requirements (Chow and Lin 1989). Therefore, the tensile structure is the optimal solution for the entire process of construction of a lunar structure, from logistics to installation. The soft architecture can be integrated with building systems or infrastructures and can act as an active building that communicates with
OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING
571
people with environmental or state information to which the architecture belongs (Davis 2011). Davis devised what is known as the sensing touch curtain, which uses electronics-embedded textile material programmed to glow or function as a type of insulation. It was shown that the tensile architecture can serve as not only a structure but also as an electronic medium of information.
The importance of the tensile membrane architecture is increasingly noted in modern urban environments. In contrast to conventional highly curved roof oriented membrane structures, recent developments have led to a wider range of shapes and uses. The tensile membrane structure is distinguished from other structures by its low self-weight, high flexibility, good deformability, and translucency. Given these properties, the membrane structure is an ideal option to express unique designs. When combined with other functions, such as electronics or software, it serves as a newmedium for sharing information with the environment. In this paper, the membrane architecture and the vibration-based energy harvesting technique are combined to function as an energy generation method as well as an architectural structure.
2.2. ENERGY HARVESTING
In response to increasing carbon emissions and the limited supply of fossil fuels, vibration energy and the harvesting techniques have been widely researched. Vibration-based energy generation techniques can be utilized to harness electrical energy from wind, ocean waves, human motion and rain.
Membrane energy harvesting is commonly realized with piezoelectric, triboelectric and electromagnetic generators. These devices generate electric energy from the mechanical energy of fluttering, bending, folding, rubbing, stretching movements, vibration and/or pressure. The windbelt is a typical example of a power generation application that harvests energy from aeroelastic fluttering. One type, for example, uses a magnet attached to a flexible membrane to generate an induced electrical current with a coil fixed onto a supporting structure (Quy et al. 2016). Simple integration of the membrane and the anchoring support allows the use of triboelectric generators based on fluttering as well (Phan et al. 2017; Zhang et al. 2015). Rubbing and contact mechanisms are widely used in wearable generators. Triboelectric generators can harvest energy from human motions such as foot pressing, arm swinging or walking (Pu et al. 2015; Jung et al. 2014). The vibration of the membrane can be converted into applicable energy using a piezoelectric cantilever on the fabric or piezoelectric fibers embedded within a structure (Wei et al. 2013; Swallow et al. 2008). Stretching, folding or crumpling of the membrane can generate energy with the help of both triboelectric and piezoelectric components. Tensile membranes come in various types, examples of which include piezoelectric strap twining around an elastic core, a textile triboelectric nanogenerator with a 3D orthogonal woven structure, and a PVDF-based piezoelectric harvester attached onto a garment (Kim and Yun 2017; Dong et al. 2017; Yang et al. 2016). A wearable fabric device for water energy harvesting has been studied as well, in which triboelectricity is created by water droplets (Xiong et al. 2017). The papers described above provide evidence of successful membrane energy harvesting in various environments.
572 H. MAENG AND K.H. HYUN
The tensile membrane architecture is open to the outdoor environment and the corresponding multi-ambient energy sources. This type of architecture utilizes a membrane material with a proven capacity to generate electrical energy. One important goal is to find an energy harvester suitable for the architecture that can also make use of wasted energy and thus contribute to green energy development. Existing structures can use attachment generators, while newly built structures can be composed of smart textiles or adequate generators depending on the environmental condition. This paper can help designers factor in new design considerations by suggesting membrane structures with energy harvesting capabilities.
2.3. DESIGN SUPPORT SYSTEM FOR OPTIMIZATION
Dream Lens was developed to explore large-scale generative design datasets (Matejka et al. 2018). Matejka et al. (2018) claimed that a large number of design solutions can be created with a generative design system by defining the goals, constraints or parameters. A parametric design with a limited condition can be used to create membrane architecture forms as well. Architects and engineers are able to make design decisions considering certain requirements, geometric forms and constraints, and desired materials for the most energy-efficient membrane architecture. Narangerel et al. (2016) explored building façade optimization for better daylighting performance. Parametric modelling and a multi-objective optimization algorithm were utilized to optimize the daylighting performance and BIPV efficiency. Datasets investigated with pre-defined design parameters were illustrated in a graph, and the three most outstanding alternatives were visualized.
In this paper, the parameters of the geometry of the tensile membrane architecture, the membrane stiffness and the wind are defined to find the optimized tensile membrane structure with the goal of stability of the structure and maximum amount of the harvested energy. The simulation results are presented in the form of graphs with an illustration of the optimized 3D model of the tensile membrane architecture. Providing such a design dataset will allow designers to make the optimal choices among various options.
3. Methodology One of the aims of this paper is to determine the correlation between the stability of the structure and the performance of wind-induced vibration-based energy harvesting. The second goal is to help designers make decisions with visualized optimized design solutions and graphs with the two following objectives: the membrane stiffness and the amount of energy harvested. Simulations were carried out to achieve these goals. This study was conducted using Kangaroo, Grasshopper’s plug-in, which is easy to use and widely used in relation to tensile structures for form-finding. Kangaroo has excellent accessibility, is capable of tests of wind actions on tensile structures, and allows real-time interaction. This makes Kangaroo a suitable tool for optimizing the tension and energy harvesting of membrane architectures in windy environments. While the average wind speed is constant at 1.7 m/s, the geometry of the membrane architecture changes by a
OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING
573
distance of 1 m inside the boundary of the 14m x 14m x 7m box as a variable in each simulation (Figure 1). Additionally, there is an interval of 500 kN/m starting from 1500 to 6000 kN/m in the other variable, i.e., the stiffness of membrane. The results considering these two objectives, stability and the amount of harvested energy, are organized and the optimized structures are visualized in this paper.
Figure 1. Hypar, conic and barrel vault types (arrows indicate the direction of the wind).
3.1. PARAMETERS
The three membrane structure types - the conic, hypar and barrel vault types - have different structural characteristics and main components affecting each structure (Bridgens and Birchall 2012). Analyses and optimization of these typical tensile forms were thus performed separately. The characteristics and parameters of the three forms considered in the experiment are as follows. A 14m x 14m x 7m boundary box is assumed, and all three aforementioned structures are formed inside the boundary zone. ‘Hypar’ features a geometry consisting of alternating high and low support points. The most basic form of the hypar type can take ‘Hypar 9’ in Figure 2 as an example which consists of two high points and two low points. More complex hypars can have a larger number of support points, but in this experiment, optimization was performed on a hypar with only four points, two of which are diagonally positioned and fixed onto the bottom. The other two non-fixed points are positioned in the 0-7m range along the z-axis. ‘Conic’ refers to a structure that consists of an upper ring shape and lower square supports. Regarding the conic shape, the membrane fails if the minimal surface and the desired shape are different (Bridgens and Gosling 2010). The ring height and radius are established by referring to the feasible bounds examined by Bridgens and Gosling. The radius is set within the range of 0-7m, the height between 0m and 7m and the radius is set to satisfy the feasible bounds. The Barrel Vault type consists of two parallel straight lines and elliptic curves at the end, anchoring the structure. The length of both straight lines is 14 m and the radius of each of the two ellipses is in the range of 0-7m. The distance between the two lines is fixed at 7m; therefore, if the radius approaches 0m, the barrel vault form would be similar to a flat panel. Design alternatives of the three tensile membrane architectures are visualized in Figure 2. Complete zero-height (flat) variables were not considered in this simulation, as they are unable to be referred to as hypar, conic or barrel
574 H. MAENG AND K.H. HYUN
vault types.
Figure 2. Design alternatives (fixed anchors are shown in red and variables are shown in blue).
The stiffness of the material is represented by the strength parameter of the Kangaroo solver. The Kangaroo solver regards the unit line object of a membrane surface as a spring. The spring follows Hooke’s law; i.e., the strength of the spring is calculated as EA/L, where E is the Young’s modulus, A is the cross-sectional area of the spring and L is the rest length (Piker 2014). The stiffnesses of the structural components, which are in this case polyester fabric and glass fiber fabric, are 25,200N/cm and 54,720N/cm, respectively; thus, the range of the stiffness was established according to real numbers from 1,500 kN/m to 6,000 kN/m (Uhlemann 2016).
3.2. WIND TYPE / GOALS
All simulations were conducted under the assumption that the surrounding ground conditions are negligible to focus on the geometry, stiffness and vibration of the membrane under the wind condition. The wind load applied to the membrane corresponds to the wind speed defined, by a sine function where the midline is at 1.7 m/s, the annual average wind speed in Seoul, with an amplitude level of 2 m/s. The wind speed thus has a maximum of 3.7 m/s and minimum of -0.3 m/s towards a fixed direction on the y=x graph (the boundary box is assumed to be located in the first quadrant). The wind speed can then be converted into the load in Pa, as the unit of time in the simulation differs from the actual time. The Kangaroo solver is used for the calculation according to its own iteration, which varies with different computer loads. The simulation begins once the membrane object is stabilized and ends after the applied wind speed draws a complete sine period twice.
The membrane is composed of a UV surface, on which power generators are assumed to be placed at the vertices of the surface. Each form has a different number of vertices: 613 for the hypar and barrel vault types and 800 for the conic type. Displacement and acceleration values are the obtained experimental results. The displacement is the sum of the distances travelled by all of the vertices during the simulation, and the acceleration is the sum of the number of vertices when the acceleration values exceed a defined value. Because the Kangaroo solver runs according to its own iteration, the defined value for measuring the acceleration is set to show the simulation results as clearly as possible.
OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING
575
The result of the experiment represents the movement of the assumed energy harvesters. In the literature, acceleration is not exactly proportional to the amount of energy generated but can be said to have the same direction (Saha et al. 2008). The optimum generator type differs according to the acceleration or displacement pattern, for which the designer will be able to select the harvester that matches their design based on the results of this experiment. In addition, by analyzing the resulting values of each vertex alone or together, the designer can choose the number and the position of generators for the highest efficiency in their design.
4. Implementation Results Figure 3 shows graphs of the results and the optimized forms, where the x-axis indicates the stiffness and the y-axis is the acceleration and displacement for the three tensile membrane structure forms. The stressed lines in Figure 3 indicate the values of the optimized alternatives. In the hypar and conic cases, alternatives Figure 3-b and Figure 3-d are the optimized options when stiffness is in 1500 kN/m and in the rest Figure 3-a, Figure 3-c and Figure 3-e have the largest acceleration and displacement values among the forms. An inverse relationship between the two axes: the stiffness and acceleration can be found in all three types and between the stiffness and displacement as well. Given the outcomes shown in Figure 3, it can be inferred that the amount of generated energy decreases as the material becomes stiffer. In a comparison between the two factors, the decline of the displacement grows steeper, resulting in logarithmic graphs, while the shapes of stiffness-to-acceleration graphs vary from nearly linear to logarithmic.
Figure 3. Results graphs and optimized 3D models.
576 H. MAENG AND K.H. HYUN
The hypar was simulated with the combination of stiffness of 1500-6000 kN/m and a height of 0-7m. In the acceleration graph, outstanding values were achieved from hypar structures with anchors of the height of 7m and the form containing relatively tall anchors tend to be ranked highly. The remaining examples at the lower end are a combination of anchors with heights of 1-3m. Symmetrical forms such at the indication of a 0m / 7m anchored hypar and another 7m / 0m anchored hypar show minor differences in both the acceleration and displacement, demonstrating that the diametrical opposite directionality of the wind does not have a significant impact on the result. Additionally, the wind when blowing towards the inside of the hypar structure always scored slightly higher compare to when it was blowing towards the outside. Some of the forms show greater decreasing rates than others, indicating that it is desirable to use more flexible materials. The 7m / 0m anchored hypar has themost distinctive values of generated energy, while the 7m / 7m anchored hypar peaked at the 1500 kN/m material. Thus, the optimized structures are the two: 7m / 0m anchored hypar (Figure 3-a) at majority of stiffness value and 7m / 7m anchored hypar (Figure 3-b) at stiffness value around the 1500 kN/m. The displacement values indicate a clear hierarchy according to the geometric properties. The exception of an intersection between resulting graph lines did not occur. The displacement also shows larger…
Abstract. This paper identifies the correlation between the stiffness of a tensile membrane and wind-induced vibration. Here, a means by which to optimize the tensile membrane architecture and a harvester for the collection of energy from wind-induced vibration is proposed. Because its material property, the tensile membrane architecture is suitable for harvesting energy. While high stiffness is desired to ensure the stability of the tensile membrane architecture, the stiffness is negatively correlated with the energy collection. Thus, optimization considering the stiffness and the amount of harvestable energy in tensile membrane structures is conducted, broken down into the following tasks: 1) investigate the energy generator on the tensile membrane to combine a membrane type of architecture and wind energy harvesting; 2) set-up the computational environment with which to analyze the energy efficiency of the tensile membrane architecture; 3) visualize the resulting datasets to determine the relationship between the stiffness and the energy efficiency; and 4) define optimized tensile membrane architectures.
Keywords. Tensile Membrane Architecture; Design Optimization; Sustainable Architecture; Computational Design; Wind Energy Harvesting.
1. Introduction At present, energy harvesting from vibrations is a notable subject and is part of the response to the global concern over energy and environmental issues (Abdelkefi 2016; Zhao and Yang 2018; Orrego et al. 2017). However, implementing wind-induced vibration-based energy harvesting techniques with a tensile membrane architecture has yet to be explored. Conventionally, vibration is a factor to be minimized in building structures, and doing so requires a refined analysis (BSI 2002). According to Stranghöner et al. (2016), tensile structures also require reduced wind-induced vibration to ensure their proper functioning and to prevent failure of the membrane. However, energy harvesting through generators requires vibration to maximize the amount of power generated. Moreover, the shape of the tensile membrane has a significant influence on the load paths for the pressure in these systems. The membrane must be prestressed to enhance the stiffness and to support material loads such as wind and snow. In this respect,
RE: Anthropocene, Proceedings of the 25th International Conference of the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA) 2020, Volume 1, 569-578. © 2020 and published by the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA), Hong Kong.
570 H. MAENG AND K.H. HYUN
designers have explored various design concepts that are both structurally stable and aesthetically pleasing. Thus, it is crucial to find a balance between the amount of vibration and the degree of structural stability so as to maximize the energy harvesting performance while ensuring the serviceability of the tensile structure.
There are numerous related works on vibration-based energy harvesting through a membrane structure. A membrane-based energy harvesting device developed by Dong et al. (2015) had an adjustable resonant frequency of the membrane allowing the power output from vibration energy harvesting to be maximized. Allen and Smits (2001) investigated the feasibility of an “eel”-shaped piezoelectric membrane in underwater, aiming to maximize the strain energy and mechanical power. These studies concentrated on the development and evaluation of the efficiency of the generator itself. They focused on maximizing the amount of harvested energy, whereas shape optimization of the membrane was not considered. Therefore, the present paper aims to develop a method to simulate the energy harvesting performance and structural stability of a tensile membrane architecture depending on the shape and the membrane stiffness of the structure. To accomplish this, tensile membrane structure simulation software was developed with two major components. First, the proposed simulation software allows the generation of design variations based on three fundamental tensile structures: hyperbolic paraboloid (hypar), conic, and the barrel vault type. Second, the software estimates the amount of electrical power generated under various conditions, such as with different numbers of input sensors and different wind conditions. This paper focuses on developing a method to optimize the tensile membrane architecture for energy harvesting to assist designers when they make decisions about the design, including its form and the materials used.
2. Related Works 2.1. TENSILE ARCHITECTURE
Berger (1999) explained the advantages of a tensile structure, noting the “speed of construction, use of daylight, the reflection of heat from the sun, dispersal of interior sound, the beauty of the interior space and the excitement of the exterior sculpture.” Moreover, the high energy efficiency, possibility for daylight use, and heat-reflecting properties of a fabric structure can save energy and lower the general maintenance costs (Berger 1999). Tensile membrane structures offer efficient solutions that would not be possible with conventional building technologies. Tensile architectures have advantages in that they can be used in a variety of environments and may even be suitable for extreme environments such as that of the moon. Because the cost of transportation to the moon is extremely high, the structural materials must be as light as possible. The lunar structure must also be buildable with a minimum amount of manpower, as rapid construction lessens the time the workers are at risk. Tensile membranes meet these two requirements (Chow and Lin 1989). Therefore, the tensile structure is the optimal solution for the entire process of construction of a lunar structure, from logistics to installation. The soft architecture can be integrated with building systems or infrastructures and can act as an active building that communicates with
OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING
571
people with environmental or state information to which the architecture belongs (Davis 2011). Davis devised what is known as the sensing touch curtain, which uses electronics-embedded textile material programmed to glow or function as a type of insulation. It was shown that the tensile architecture can serve as not only a structure but also as an electronic medium of information.
The importance of the tensile membrane architecture is increasingly noted in modern urban environments. In contrast to conventional highly curved roof oriented membrane structures, recent developments have led to a wider range of shapes and uses. The tensile membrane structure is distinguished from other structures by its low self-weight, high flexibility, good deformability, and translucency. Given these properties, the membrane structure is an ideal option to express unique designs. When combined with other functions, such as electronics or software, it serves as a newmedium for sharing information with the environment. In this paper, the membrane architecture and the vibration-based energy harvesting technique are combined to function as an energy generation method as well as an architectural structure.
2.2. ENERGY HARVESTING
In response to increasing carbon emissions and the limited supply of fossil fuels, vibration energy and the harvesting techniques have been widely researched. Vibration-based energy generation techniques can be utilized to harness electrical energy from wind, ocean waves, human motion and rain.
Membrane energy harvesting is commonly realized with piezoelectric, triboelectric and electromagnetic generators. These devices generate electric energy from the mechanical energy of fluttering, bending, folding, rubbing, stretching movements, vibration and/or pressure. The windbelt is a typical example of a power generation application that harvests energy from aeroelastic fluttering. One type, for example, uses a magnet attached to a flexible membrane to generate an induced electrical current with a coil fixed onto a supporting structure (Quy et al. 2016). Simple integration of the membrane and the anchoring support allows the use of triboelectric generators based on fluttering as well (Phan et al. 2017; Zhang et al. 2015). Rubbing and contact mechanisms are widely used in wearable generators. Triboelectric generators can harvest energy from human motions such as foot pressing, arm swinging or walking (Pu et al. 2015; Jung et al. 2014). The vibration of the membrane can be converted into applicable energy using a piezoelectric cantilever on the fabric or piezoelectric fibers embedded within a structure (Wei et al. 2013; Swallow et al. 2008). Stretching, folding or crumpling of the membrane can generate energy with the help of both triboelectric and piezoelectric components. Tensile membranes come in various types, examples of which include piezoelectric strap twining around an elastic core, a textile triboelectric nanogenerator with a 3D orthogonal woven structure, and a PVDF-based piezoelectric harvester attached onto a garment (Kim and Yun 2017; Dong et al. 2017; Yang et al. 2016). A wearable fabric device for water energy harvesting has been studied as well, in which triboelectricity is created by water droplets (Xiong et al. 2017). The papers described above provide evidence of successful membrane energy harvesting in various environments.
572 H. MAENG AND K.H. HYUN
The tensile membrane architecture is open to the outdoor environment and the corresponding multi-ambient energy sources. This type of architecture utilizes a membrane material with a proven capacity to generate electrical energy. One important goal is to find an energy harvester suitable for the architecture that can also make use of wasted energy and thus contribute to green energy development. Existing structures can use attachment generators, while newly built structures can be composed of smart textiles or adequate generators depending on the environmental condition. This paper can help designers factor in new design considerations by suggesting membrane structures with energy harvesting capabilities.
2.3. DESIGN SUPPORT SYSTEM FOR OPTIMIZATION
Dream Lens was developed to explore large-scale generative design datasets (Matejka et al. 2018). Matejka et al. (2018) claimed that a large number of design solutions can be created with a generative design system by defining the goals, constraints or parameters. A parametric design with a limited condition can be used to create membrane architecture forms as well. Architects and engineers are able to make design decisions considering certain requirements, geometric forms and constraints, and desired materials for the most energy-efficient membrane architecture. Narangerel et al. (2016) explored building façade optimization for better daylighting performance. Parametric modelling and a multi-objective optimization algorithm were utilized to optimize the daylighting performance and BIPV efficiency. Datasets investigated with pre-defined design parameters were illustrated in a graph, and the three most outstanding alternatives were visualized.
In this paper, the parameters of the geometry of the tensile membrane architecture, the membrane stiffness and the wind are defined to find the optimized tensile membrane structure with the goal of stability of the structure and maximum amount of the harvested energy. The simulation results are presented in the form of graphs with an illustration of the optimized 3D model of the tensile membrane architecture. Providing such a design dataset will allow designers to make the optimal choices among various options.
3. Methodology One of the aims of this paper is to determine the correlation between the stability of the structure and the performance of wind-induced vibration-based energy harvesting. The second goal is to help designers make decisions with visualized optimized design solutions and graphs with the two following objectives: the membrane stiffness and the amount of energy harvested. Simulations were carried out to achieve these goals. This study was conducted using Kangaroo, Grasshopper’s plug-in, which is easy to use and widely used in relation to tensile structures for form-finding. Kangaroo has excellent accessibility, is capable of tests of wind actions on tensile structures, and allows real-time interaction. This makes Kangaroo a suitable tool for optimizing the tension and energy harvesting of membrane architectures in windy environments. While the average wind speed is constant at 1.7 m/s, the geometry of the membrane architecture changes by a
OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING
573
distance of 1 m inside the boundary of the 14m x 14m x 7m box as a variable in each simulation (Figure 1). Additionally, there is an interval of 500 kN/m starting from 1500 to 6000 kN/m in the other variable, i.e., the stiffness of membrane. The results considering these two objectives, stability and the amount of harvested energy, are organized and the optimized structures are visualized in this paper.
Figure 1. Hypar, conic and barrel vault types (arrows indicate the direction of the wind).
3.1. PARAMETERS
The three membrane structure types - the conic, hypar and barrel vault types - have different structural characteristics and main components affecting each structure (Bridgens and Birchall 2012). Analyses and optimization of these typical tensile forms were thus performed separately. The characteristics and parameters of the three forms considered in the experiment are as follows. A 14m x 14m x 7m boundary box is assumed, and all three aforementioned structures are formed inside the boundary zone. ‘Hypar’ features a geometry consisting of alternating high and low support points. The most basic form of the hypar type can take ‘Hypar 9’ in Figure 2 as an example which consists of two high points and two low points. More complex hypars can have a larger number of support points, but in this experiment, optimization was performed on a hypar with only four points, two of which are diagonally positioned and fixed onto the bottom. The other two non-fixed points are positioned in the 0-7m range along the z-axis. ‘Conic’ refers to a structure that consists of an upper ring shape and lower square supports. Regarding the conic shape, the membrane fails if the minimal surface and the desired shape are different (Bridgens and Gosling 2010). The ring height and radius are established by referring to the feasible bounds examined by Bridgens and Gosling. The radius is set within the range of 0-7m, the height between 0m and 7m and the radius is set to satisfy the feasible bounds. The Barrel Vault type consists of two parallel straight lines and elliptic curves at the end, anchoring the structure. The length of both straight lines is 14 m and the radius of each of the two ellipses is in the range of 0-7m. The distance between the two lines is fixed at 7m; therefore, if the radius approaches 0m, the barrel vault form would be similar to a flat panel. Design alternatives of the three tensile membrane architectures are visualized in Figure 2. Complete zero-height (flat) variables were not considered in this simulation, as they are unable to be referred to as hypar, conic or barrel
574 H. MAENG AND K.H. HYUN
vault types.
Figure 2. Design alternatives (fixed anchors are shown in red and variables are shown in blue).
The stiffness of the material is represented by the strength parameter of the Kangaroo solver. The Kangaroo solver regards the unit line object of a membrane surface as a spring. The spring follows Hooke’s law; i.e., the strength of the spring is calculated as EA/L, where E is the Young’s modulus, A is the cross-sectional area of the spring and L is the rest length (Piker 2014). The stiffnesses of the structural components, which are in this case polyester fabric and glass fiber fabric, are 25,200N/cm and 54,720N/cm, respectively; thus, the range of the stiffness was established according to real numbers from 1,500 kN/m to 6,000 kN/m (Uhlemann 2016).
3.2. WIND TYPE / GOALS
All simulations were conducted under the assumption that the surrounding ground conditions are negligible to focus on the geometry, stiffness and vibration of the membrane under the wind condition. The wind load applied to the membrane corresponds to the wind speed defined, by a sine function where the midline is at 1.7 m/s, the annual average wind speed in Seoul, with an amplitude level of 2 m/s. The wind speed thus has a maximum of 3.7 m/s and minimum of -0.3 m/s towards a fixed direction on the y=x graph (the boundary box is assumed to be located in the first quadrant). The wind speed can then be converted into the load in Pa, as the unit of time in the simulation differs from the actual time. The Kangaroo solver is used for the calculation according to its own iteration, which varies with different computer loads. The simulation begins once the membrane object is stabilized and ends after the applied wind speed draws a complete sine period twice.
The membrane is composed of a UV surface, on which power generators are assumed to be placed at the vertices of the surface. Each form has a different number of vertices: 613 for the hypar and barrel vault types and 800 for the conic type. Displacement and acceleration values are the obtained experimental results. The displacement is the sum of the distances travelled by all of the vertices during the simulation, and the acceleration is the sum of the number of vertices when the acceleration values exceed a defined value. Because the Kangaroo solver runs according to its own iteration, the defined value for measuring the acceleration is set to show the simulation results as clearly as possible.
OPTIMIZING TENSILE MEMBRANE ARCHITECTURE FOR ENERGY HARVESTING
575
The result of the experiment represents the movement of the assumed energy harvesters. In the literature, acceleration is not exactly proportional to the amount of energy generated but can be said to have the same direction (Saha et al. 2008). The optimum generator type differs according to the acceleration or displacement pattern, for which the designer will be able to select the harvester that matches their design based on the results of this experiment. In addition, by analyzing the resulting values of each vertex alone or together, the designer can choose the number and the position of generators for the highest efficiency in their design.
4. Implementation Results Figure 3 shows graphs of the results and the optimized forms, where the x-axis indicates the stiffness and the y-axis is the acceleration and displacement for the three tensile membrane structure forms. The stressed lines in Figure 3 indicate the values of the optimized alternatives. In the hypar and conic cases, alternatives Figure 3-b and Figure 3-d are the optimized options when stiffness is in 1500 kN/m and in the rest Figure 3-a, Figure 3-c and Figure 3-e have the largest acceleration and displacement values among the forms. An inverse relationship between the two axes: the stiffness and acceleration can be found in all three types and between the stiffness and displacement as well. Given the outcomes shown in Figure 3, it can be inferred that the amount of generated energy decreases as the material becomes stiffer. In a comparison between the two factors, the decline of the displacement grows steeper, resulting in logarithmic graphs, while the shapes of stiffness-to-acceleration graphs vary from nearly linear to logarithmic.
Figure 3. Results graphs and optimized 3D models.
576 H. MAENG AND K.H. HYUN
The hypar was simulated with the combination of stiffness of 1500-6000 kN/m and a height of 0-7m. In the acceleration graph, outstanding values were achieved from hypar structures with anchors of the height of 7m and the form containing relatively tall anchors tend to be ranked highly. The remaining examples at the lower end are a combination of anchors with heights of 1-3m. Symmetrical forms such at the indication of a 0m / 7m anchored hypar and another 7m / 0m anchored hypar show minor differences in both the acceleration and displacement, demonstrating that the diametrical opposite directionality of the wind does not have a significant impact on the result. Additionally, the wind when blowing towards the inside of the hypar structure always scored slightly higher compare to when it was blowing towards the outside. Some of the forms show greater decreasing rates than others, indicating that it is desirable to use more flexible materials. The 7m / 0m anchored hypar has themost distinctive values of generated energy, while the 7m / 7m anchored hypar peaked at the 1500 kN/m material. Thus, the optimized structures are the two: 7m / 0m anchored hypar (Figure 3-a) at majority of stiffness value and 7m / 7m anchored hypar (Figure 3-b) at stiffness value around the 1500 kN/m. The displacement values indicate a clear hierarchy according to the geometric properties. The exception of an intersection between resulting graph lines did not occur. The displacement also shows larger…
Related Documents
