دانلود مقاله ISI انگلیسی شماره 6345
ترجمه فارسی عنوان مقاله

به حداقل رساندن مصرف انرژی یک واحد بررسی هوایی با رویکرد هوش محاسباتی

عنوان انگلیسی
Minimizing energy consumption of an air handling unit with a computational intelligence approach
کد مقاله سال انتشار تعداد صفحات مقاله انگلیسی
6345 2013 9 صفحه PDF
منبع

Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)

Journal : Energy and Buildings, Volume 60, May 2013, Pages 355–363

ترجمه کلمات کلیدی
واحد بررسی هوایی - داده کاوی - گروه - الگوریتم مانند الکترومغناطیس - عملکرد پنالتی پویا
کلمات کلیدی انگلیسی
پیش نمایش مقاله
پیش نمایش مقاله  به حداقل رساندن مصرف انرژی یک واحد بررسی هوایی با رویکرد هوش محاسباتی

چکیده انگلیسی

A data-mining approach is applied to optimize the energy consumption of an air handling unit. A multi-perceptron ensemble algorithm is used to model a chiller, a pump, and the supply and return fans. A non-linear model is developed to minimize the total energy consumption of the air-handling unit while maintaining the temperature of the supply air and the static pressure in a predetermined range. A dynamic, penalty-based, electromagnetism-like algorithm is designed to solve the proposed model. In all, 200 test data points are used to validate the proposed algorithm. The computational results show that the energy consumed by the air-handling unit is reduced by almost 23%.

مقدمه انگلیسی

The world's consumption of energy has increased over the years. Heating, ventilation, and air-conditioning (HVAC) systems use as much as 60% of the energy consumed in buildings [1], and they account for approximately 30% of the total energy consumption in the United States [2]. The energy efficiency of HVAC systems is being considered as a vehicle for accomplishing energy savings. Many research efforts related to the modeling and optimization of HVAC systems have been reported in the literature. Lu et al. [3] and [4] formulated a mixed-integer, non-linearly constrained model for minimizing the energy consumption of HVAC systems. Engdahl and Johansson [5] minimized the energy use of a system with variable air volumes by setting the supply air temperature optimally in response to load, fan power, coefficient of performance of the chiller, outdoor temperature, and outdoor relative humidity. They showed that the recommended control strategy was more energy-efficient than requiring a constant temperature for the supply air. Numerous studies involving multi-objective optimization of HVAC systems have been performed to determine the most effective trade-offs between total energy cost and indoor thermal comfort [6], [7], [8] and [9]. Nassif et al. [6] and [9] proposed a supervisory control strategy for online optimization of energy use and thermal comfort by adjusting set points of local-loop controllers in a multi-zone HVAC system. Wright et al. [8] applied a genetic algorithm to seek optimal settings for the temperature and flow rate of the supply air for a single-zone HVAC system.

نتیجه گیری انگلیسی

In research reported in this paper, the energy consumed by an air handling unit was minimized by combining data-mining with optimization. The multi-layer perceptron ensemble algorithm was applied to develop predictive models of the energy consumption the chiller, the pump, and the supply and return fans. To minimize the energy consumption while maintaining the temperature and static pressure of the supply air at predetermined levels, a non-linear, constrained optimization model was developed. A dynamic penalty function-based electromagnetism-like algorithm was proposed to solve the model. Computational results showed that the energy consumed by the chiller could be reduced significantly by slightly increasing the energy consumption of the pump and fans. The total energy consumption of the AHU system was reduced by almost 23%. Future research involving optimization criteria is needed to accommodate different supply air requirements (e.g., humidity and CO2 concentration). The research presented in the paper has focused on single zone modeling. Modeling different zones calls for new experiments and data which were not available in this research. The single-zone approach presented in the paper is generalizable to a multi-zone optimization provided that the suitable data is available.