دانلود رایگان مقاله برآورد شدت مصرف انرژی ساختمان با یکپارچه سازی GIS و فناوری داده های بزرگ

abstract

Buildings are the major source of energy consumption in urban areas. Accurate modeling and forecasting of the building energy use intensity (EUI) in the urban scale have many important applications, such as energy benchmarking and urban energy infrastructure planning. The use of Big Data technology is expected to have the capability of integrating a large number of predictors and giving an accurate prediction of the energy use intensity of buildings in the urban scale. However, past research has often used Big Data technology in estimating energy consumption of a single building rather than the urban scale, due to several challenges such as data collection and feature engineering. This paper therefore proposes a geographic information system integrated data mining methodology framework for estimating the building EUI in the urban scale, including preprocessing, feature selection, and algorithm optimization. Based on 216 prepared features, a case study on estimating the site EUI of 3640 multi-family residential buildings in New York City, was tested and validated using the proposed methodology framework. A comparative study on the feature selection strategies and the commonly used regression algorithms was also included in the case study. The results show that the framework was able to help produce lower estimation errors than previous research, and the model built by the Support Vector Regression algorithm on the features selected by Elastic Net has the least cross-validation mean squared error.

4. Conclusions

To conclude, this study proposes a methodology framework to estimate the building energy use intensity on the urban scale by integrating GIS and Big Data technology. The framework addressed the major challenges of data mining in the urban scale, including preprocessing, feature selection, and algorithm optimization. A case study on estimating the energy use intensity of 3640 multifamily residential buildings in NYC was conducted to test the effectiveness of the proposed methodology framework. The results showed that the framework was able to help produce lower estimation error than in previous research. The contributions of the work can be summarized into three aspects. First is the methodology framework. Feature engineering is known as the key in data mining [20]. The integration of GIS into this most important data mining step has shown lots of benefits, such as connecting high dimension geo based datasets conveniently, filling the missing values nicely, and helping generate many useful features which may be easily overlooked using traditional feature management tools in urban scale energy related problems. In addition, data mining is such a practical tool that by changing the features, the target, and the algorithms, the framework can be easily extended to solve other urban scale research problems, such as estimating the urban scale air quality distribu tion and the noise pollution. Second is the case study on the residential buildings in NYC. By using the extensive open data in NYC, the study was able to estimate the energy use intensity of residential buildings with less error than previous research, and the estimated maps can be useful references for energy planning and policy making. Last is the comparative study on feature selection strategies and regression models. The case study shows that the filter methods can significantly reduce the feature size and the computation time, and some even produce better performance (filter features using Elastic Net). The wrapper methods generally have better performance than using the full set of features but dramatically increase the computation time. The comparison between different algorithms shows that SVR is an efficient and effective regression algorithm for non-linear models. ANN, particularly the feed-forward neural network, underperformed SVR and consumed longer computation time according to the case study. However, ANN is still believed to be a worth-trying regression tool for nonlinear problems considering its extensive possibilities for further adjustment.