Multi-objective design optimization on building integrated photovoltaic with Trombe wall and phase change material based on life cycle cost and thermal comfort

https://doi.org/10.1016/j.seta.2021.101277Get rights and content

Abstract

This study performed design optimization on a building with integrated photovoltaic, Trombe wall and phase change material based on building life cycle cost (LCC) and indoor thermal comfort. A total of thirty-four design variables were considered as the optimization variables. The optimization process consists of four steps. Firstly, Monte Carlo Method was used to generate a set of design samples, and the LCC and indoor discomfort degree hour were obtained through computer simulation. Secondly, stepwise linear regression (SLR) and artificial neural network (ANN) models were developed as the prediction models. ANN models exhibited better prediction performance with errors of less than 6%. Thirdly, six optimization algorithms were selected to couple with the ANN models for design optimization and the Strength Pareto Evolutionary Algorithm II (SPEA II) presented the best performance. Finally, optimization with different groups of design variables were conducted by using the SPEA II algorithm. Compared with the reference building, the optimal solutions reduce life cycle cost up to 45.51% and improve thermal comfort level up to 43.81%. The recommended heating and cooling temperature setpoint is around 18 °C and 24 °C, respectively. The PCM type convergence value is BioPCM M182/Q21, and the east and west window-to-wall ratios are around 10%.

Introduction

The buildings consume 39% of the energy and lead to 38% of the carbon emission globally [1]. Due to the rapid expansion of building construction areas and demand for air-conditioning system to maintain indoor thermal comfort, the building energy consumption in China has increased sharply and become the second highest in the world after the USA [2], [3]. High energy consumption not only brings negative impact on the environment, such as air pollution, green house effects, urban heat island effect, but also has threatened the human health and economic growth [4]. Therefore, improvement on the building energy efficiency has become an important energy strategy for every country [5]. This has given rise to the need of developing building energy efficiency design standards that stipulate the minimum energy performance requirements for new and retrofit buildings [6].

Various renewable energy sources as well as advanced energy conversion and hybrid energy storage technologies are applied to buildings to relieve the energy crisis pressure [7], among which the photovoltaic power generation is a safe, clean and sustainable energy conversion technology. The application of photovoltaic systems can transform the role of buildings from pure energy consumers to energy producers [8], [9], [10]. At the same time, it can also help reduce the emissions of carbon oxides, nitrogen oxides and sulfur oxides [11], [12]. Due to the regional, seasonal and climatic impact, renewable energies are usually unstable and uncontrollable. The integration of phase change materials (PCMs) can help overcome the limitations of renewable energy utilization [13]. In addition, the use of PCMs to increase the thermal mass of the exterior wall has significant effects on improving building thermal insulation performance [14], [15], regulating indoor thermal environment [16], [17], and providing radiant cooling [16]. The combination of phase change materials and Trombe wall can help avoid the occurrence of sick building syndrome (SBS) and provide occupants with a healthy and comfortable indoor environment [18].

Studies on the solar wall application, PCMs, solar air heating system and BIPV primarily focused on their thermal and energy performance. Stazi et al. [19] studied the behavior of Trombe wall (solar wall) with different insulation levels for a residential building under a Mediterranean climate. They concluded that the solar wall could provide heating in cold and intermediate weather, and the risk of overheating in summer can be overcome by shading and ventilation. Bojić et al. [20] found that a building with Trombe walls in Lyon, France, could save around 20% of the heating energy compared to the same building without Trombe walls. Ma et al. [21] proposed a ventilated composite Trombe wall and simulated its performance in an office building in Japan. They found that annual energy cost reduction could be up to 3.7% compared to the non-ventilated Trombe wall. Xu et al. [22] presented a study on a hybrid BIPV/T solar wall system that provided space air heating and electricity in winter, and hot water and electricity at other times of the year. They found that the system could provide passive cooling in summer and space heating efficiently in winter. Gao et al. [23] conducted an experiment on a building-integrated solar air heating system using unglazed transpired collectors (UTCs) in the severe cold weather region in China and found its average efficiency was much higher than most of the glazed flat-plate collectors. Paraschiv et al. [24] performed an analysis on the techno-economic-environmental performance of a solar air system integration in an insulated residential building wall, which could lead to a minimal thermal need reduction of 15.53%, simple payback of 14 years without government subsidies and CO2 reduction of between 37 t/year and 44 t/year. Luo et al. [25] carried out an experimental study on a PV-water/air solar wall (HPSW) system operated under three modes: air mode in winter, water mode in summer, and air/water mode during the transitional season. They found that the system could effectively improve indoor thermal comfort in winter and meet the energy needs during the transitional season. Kishore et al. [26] performed a numerical investigation on the performance of the PCM-integrated wall and found that it could lead to heat gain reduction of 3.5% to 47.2% and heat loss of −2.1% to 8.1% at different climate regions in the United States. Yoo [27] studied the performance of building integrated photovoltaic (BIPV) as a shading device and concluded that it could harvest 4% more energy than a traditional PV system achieve a reduction of 32–34% for cooling load in summer in Korea.

The use of high-performance energy-saving materials and high-efficiency energy equipment to improve building performance is often very costly. Since cost is an important decision factor for building design and construction, it is impractical to apply all the best technologies or measures. Therefore, energy-efficient building design must consider economic costs and benefits without compromising indoor thermal comfort.

When performing optimization, LCC and thermal comfort of the building are affected by various design parameters. The large number of design variables and nonlinear relationships among them make it extremely difficult to provide a low-cost and high-comfort building design solution [28]. Therefore, it is crucial to choose the right optimization tool/method. The most commonly used method for design optimization is to couple the simulation tools with optimization algorithms. The building simulation tools include TRNSYS [29], [30], [31], EnergyPlus [32], [33], [34], DOE-2 [35], IDA-ICE [36], [37], etc.

Thermal comfort is a key building performance indicator and has attracted attention from many researchers. Wright et al. [38] applied the multi-objective genetic algorithm (MOGA) for building mechanical design optimization, using operating energy cost and thermal comfort as the objective functions. Using thermal comfort and building energy consumption as two objective functions, Magnier et al. [29] coupled ANN with Non-dominated Sorting genetic algorithm II (NSGA-II) to find the optimal design values of 12 parameters regarding window-to-wall ratio (WWR), thermal mass, HVAC system and thermostat settings for a residential building. Asadi et al. [30] combined TRNSYS, GenOpt and the Tchebycheff optimization technology developed in MATLAB to find the trade-offs among renovation cost, energy saving and thermal comfort of a residential building. Hong et al. [32] developed an optimization model that could simultaneously consider occupants' thermal comfort, building energy consumption, and life cycle economic and environmental values. The optimized design solution was determined by using EnergyPlus and NSGA-II algorithm. Nguyen et al. [33] used EnergyPlus and the optimization program GenOpt to perform numerical optimization on building design and operation strategy, considering energy consumption, thermal comfort, and initial investment cost.

It is known that high performance buildings are usually associated with high investment cost. When considering economic factors, how to improve the building performance at the lowest cost is worth investigating. LCC is most commonly used by many researchers. Hirvonen et al. [36] utilized NSGA-II to identify the trade-off between the LCC and annual CO2 emissions for a Finnish apartment building renovation. It was found that the most cost-effective retrofit measures were ground source heat pumps, demand ventilation and solar power generation. Arabzadeh et al. [37] proposed several optimal combinations of solar collectors and hot water storage tanks based on the NSGA-II based optimization results using energy efficiency and LCC as the objective functions. It was found that the most cost-effective solar thermal system could lead to energy savings of 24%–34% and profits of 4€/m2–23€/m2. Harkouss et al. [31] used cooling energy consumption, heating energy consumption and LCC as the objective functions of optimization, and conducted NSGA-II based optimization on the thermal conductivity of external walls/roof/ground, WWR and window type. The results show that the cooling demand, heating demand and LCC of the optimized building were reduced by 54%, 87% and 52%, respectively, compared to the reference building. Grygierek et al. [34] coupled EnergyPlus with NSGA-II to find the impact of window size/type, building orientation, insulation thickness of the exterior walls/roof/ground, and infiltration on LCC and number of discomfort hours. It was found that after design optimization, the heating demand and LCC of the building were reduced by 40% and 22%, respectively. Dubrow et al. [35] coupled DOE-2 with genetic algorithm (GA) to find the impact of building form and building envelop design parameters on the LCC and identified rectangular and trapezoidal buildings with the lowest LCC under five different climatic regions.

A review of the existing optimization algorithms for BIPV, PCMs, and Trombe wall shows that various optimization methods, such as Genetic Algorithm (GA), Particle Swarm Optimization Algorithm (PSO), Taguchi method, customized mathematical models, etc., have been employed in previous studies. For example, Khaki et al. [39] used Genetic Algorithm (GA) to optimize the geometric parameters and air mass flow rate of an integrated photovoltaic/thermal (BIPV/T) system in Iran to maximize the annual energy and exergy efficiency. Pereira and Aelenei [40] also used Genetic Algorithm (GA) to optimize the ventilation (air change rate), air cavity thickness, PCM thickness, and latent heat of fusion to maximize the energy performance of a BIPV/T-PCM system. Tunçbilek et al. [41] performed a parametric study on the energy performance of an external office wall with PCM under intermittent cooling operation. Energy savings of up to 12.8% were attained for the PCM layer thickness of 23 mm as compared to a wall without any PCM. Liu et al. [42] proposed a hybrid system composed of a phase change materials-ventilated Trombe wall (PCMs-VTW) and a photovoltaic/thermal panel integrated with phase change material (PV/T-PCM). They used Taguchi method to maximize the equivalent overall output energy by optimizing the mass flow rate, diameter of water pipe, inlet cooling water temperature, and the thickness of PCM. Lin et al. [43] proposed a design optimization strategy for a thermal energy storage (TES) system using PCMs for a solar decathlon house. A hybrid Particle Swarm Optimization and Hooke–Jeeves (PSO-HJ) algorithm was used to optimize the thermal energy storage density, which was increased from 13.58 kWh/m3 to 26.47 kWh/m3 after optimization. Lin et al. [44] proposed a two-level model-based optimization strategy for a solar photovoltaic thermal collector coupled with phase change material thermal energy storage to improve the overall thermal energy efficiency and the latent TES capacity.

Based on the above literature review, it can be found that few of the studies have focused on the optimization of high-performance buildings that integrate different advanced building technologies such as the phase change materials, ventilated Trombe walls and photovoltaic panels. In addition, the previous studies focused only on one or two categories of the building envelope characteristic, HVAC system settings, or the type of renewable energy sources. Furthermore, the objectives functions of these studies were related to building energy consumption, thermal comfort, and building capital cost but relatively few were related to building LCC and thermal comfort. Likewise, although various optimization strategies have been applied for the optimization of BIPV, PCMs, and Trombe wall, most of the studies concentrated on thermal and energy performance of the system rather than LCC or thermal comfort.

In this paper, a total of 34 design parameters related to building envelope, phase change materials, ventilated Trombe walls and photovoltaic panels were selected to be optimized with the LCC and discomfort degree hour (DDH) as the objective functions. Firstly, a total of 1000 samples were generated using Mento Carlo approach and the associated LCC and DDH were obtained through building simulation and calculation. Secondly, SLR and ANN models were developed for the prediction of LCC and DDH. Thirdly, the prediction models with better performance were selected and coupled with six different optimization algorithms for design optimization and SPEA II was found to be the best algorithm. Finally, optimization with different groups of design variables were conducted and the associated outcomes were analyzed.

Section snippets

Objective functions

Two objective functions, i.e., annual accumulative DDH and LCC, are used in this study. The optimization can be described as:Minf1x-,f2x-,x-=x1,x2,,xn

where f1 is the annual accumulative DDH [45], in °C⋅h, and f2 is the LCC, in ¥. The DDH can be calculated by:f1x-=i=18760tix--tH|tix->tH+tL-tix-|tix-<tL

where ti is the indoor air temperature, obtained from the simulation results using DesignBuilder [46], with a time step of 0.5 h, in °C; tH is the higher indoor temperature limit, taken as 26 °C,

Reference building and design parameters

The reference building is in the city of Shanghai which has a hot summer and cold winter climate. Table 3 lists the climatic information of Shanghai city, according to the EnergyPlus weather file [55].

The reference building is a 10 m × 10 m × 4 m single-story concrete frame residential building, and the building orientation of which is 15°east to south, according to the recommended optimal value from JGJ 134–2010 [47]. DesignBuilder is used to create the reference building model (Fig. 1a). The

Sample database creation

Adequate sample size is very important to ensure the stability and accuracy of the prediction model. This study adopts the Monte Carlo Sampling Method (MCSM) [60] to create the sample data sets. MCSM is a random sampling method, which is independent of the dimensions of errors and variables and has been widely applied in various fields. Fig. 3 presents the visualization of selected design variables. It can be observed that they are evenly distributed among the variable space. A total of 34

Multi-objective optimization algorithm

Six different multi-objective optimization algorithms were coupled with the ANN prediction models for design optimization. The results are compared to find the best optimization algorithm. The details about the six optimization algorithms are discussed as follows.

Comparison on the performance of different optimization algorithms

All the six optimization algorithms, NSGA-II, MOPSO, SPEA-II, PESA-II, MOWOA, and MSSA, were coupled with the ANN models trained in section 3.3 and ran repeatedly for many times to collect as many optimal solutions as possible. Table 11 presents the comparison on the running time for each algorithm to find the optimal solutions. It can be found that both the running time and number of Pareto solution for all the algorithms are in the same order of magnitude. The NSGA-II algorithm was the

Conclusion

In this paper, a novel building design that integrates phase change materials, ventilated Trombe wall and photovoltaic panel was studied. The building performance was optimized based on life cycle cost and discomfort degree hour. A total of thirty-four design variables were considered in the optimization. Mento-Carlo approach was used for sampling, and stepwise linear regression and artificial neural network data mining approaches were used to develop the prediction models for life cycle cost

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (86)

  • X. Sun et al.

    Energy and economic analysis of a building enclosure outfitted with a phase change material board (PCMB)

    Energy Convers Manage

    (2014)
  • Z. Wang et al.

    A remote integrated energy system based on cogeneration of a concentrating solar power plant and buildings with phase change materials

    Energy Convers Manage

    (2019)
  • M.T. Plytaria et al.

    Numerical simulation of a solar cooling system with and without phase change materials in radiant walls of a building

    Energy Convers Manage

    (2019)
  • W. Sun et al.

    Two types of composite phase change panels containing a ternary hydrated salt mixture for use in building envelope and ventilation system

    Energy Convers Manage

    (2018)
  • F. Stazi et al.

    The behaviour of solar walls in residential buildings with different insulation levels: An experimental and numerical study

    Energy Build

    (2012)
  • M. Bojić et al.

    Optimizing energy and environmental performance of passive Trombe wall

    Energy Build

    (2014)
  • Q. Ma et al.

    Optimizing energy performance of a ventilated composite Trombe wall in an office building

    Renewable Energy

    (2019)
  • L. Xu et al.

    Study of a hybrid BIPV/T solar wall system

    Energy

    (2020)
  • L. Gao et al.

    Experimental study of a building-integrated solar air heating system in cold climate of China

    Energy Build

    (2013)
  • S. Paraschiv et al.

    Technical and economic analysis of a solar air heating system integration in a residential building wall to increase energy efficiency by solar heat gain and thermal insulation

    Energy Rep

    (2020)
  • K. Luo et al.

    Seasonal experimental study of a hybrid photovoltaic-water/air solar wall system

    Appl Therm Eng

    (2020)
  • R.A. Kishore et al.

    Optimizing PCM-integrated walls for potential energy savings in U.S

    Buildings. Energy Build

    (2020)
  • S.-H. Yoo

    Optimization of a BIPV system to mitigate greenhouse gas and indoor environment

    Sol Energy

    (2019)
  • E. Znouda et al.

    Optimization of Mediterranean building design using genetic algorithms

    Energy Build

    (2007)
  • L. Magnier et al.

    Multiobjective optimization of building design using TRNSYS simulations, genetic algorithm, and Artificial Neural Network

    Build Environ

    (2010)
  • E. Asadi et al.

    A multi-objective optimization model for building retrofit strategies using TRNSYS simulations GenOpt and MATLAB

    Build Environ

    (2012)
  • F. Harkouss et al.

    Passive design optimization of low energy buildings in different climates

    Energy

    (2018)
  • T. Hong et al.

    A multi-objective optimization model for determining the building design and occupant behaviors based on energy, economic, and environmental performance

    Energy

    (2019)
  • J.A. Wright et al.

    Optimization of building thermal design and control by multi-criterion genetic algorithm

    Energy Build

    (2002)
  • R. Pereira et al.

    Optimization assessment of the energy performance of a BIPV/T-PCM system using Genetic Algorithms

    Renewable Energy

    (2019)
  • E. Tunçbilek et al.

    Thermal performance based optimization of an office wall containing PCM under intermittent cooling operation

    Appl Therm Eng

    (2020)
  • X. Liu et al.

    Optimization of a New Phase Change Material Integrated Photovoltaic/Thermal Panel with The Active Cooling Technique Using Taguchi Method

    Energies

    (2019)
  • W. Lin et al.

    Optimal design of a thermal energy storage system using phase change materials for a net-zero energy Solar Decathlon house

    Energy Build

    (2020)
  • Y. Zhang et al.

    Ideal thermophysical properties for free-cooling (or heating) buildings with constant thermal physical property material

    Energy Bulid

    (2006)
  • J.C. Lam et al.

    Regression analysis of high-rise fully air-conditioned office buildings

    Energy Build

    (1997)
  • S. Asadi et al.

    On the development of multi-linear regression analysis to assess energy consumption in the early stages of building design

    Energy Build

    (2014)
  • S.S. Amiri et al.

    Using multiple regression analysis to develop energy consumption indicators for commercial buildings in the U.S

    Energy Build

    (2015)
  • D. Hsu

    Comparison of integrated clustering methods for accurate and stable prediction of building energy consumption data

    Appl Energy

    (2015)
  • Y. Li et al.

    Towards a comprehensive optimization of engine efficiency and emissions by coupling artificial neural network (ANN) with genetic algorithm (GA)

    Energy

    (2021)
  • W. Yu et al.

    Application of multi-objective genetic algorithm to optimize energy efficiency and thermal comfort in building design

    Energy Build

    (2015)
  • S. Attia et al.

    Assessing gaps and needs for integrating building performance optimization tools in net zero energy buildings design

    Energy Build

    (2013)
  • Y. Yusoff et al.

    Overview of NSGA-II for optimizing machining process parameters

    Procedia Eng

    (2011)
  • M. Kumar et al.

    The elitist non-dominated sorting genetic algorithm with inheritance (i-NSGA-II) and its jumping gene adaptations for multi-objective optimization

    Inf Sci

    (2017)
  • Cited by (0)

    View full text