( Dong et al., 2016) developed a hybrid model which coupled a data-driven model and a thermal network model for predicting the total energy consumption of residential areas and compared its prediction performance to artificial neural networks (ANN), support vector machines (SVM) and least square support vector machine (LSSVM)-based models. Grey-box models may also show better performance compared to black-box and white-box models. In grey-box models, some internal parameters and equations are physically interpretable ( Eom et al., 2012 Amasyali and El-Gohary, 2018). Grey-box modelling approaches offer a combination of physical and data-driven prediction models, leveraging the advantages and minimizing the disadvantages of both approaches. However, some detailed architectural data may not be readily available to researchers, resulting in an inability to provide accurate inputs and thus leading to poor predictive performance. ( Zhu et al., 2012 Said, 2016) compared the Dest, Energy Plus and DOE-2 simulation software calculation methods, and their research results showed that the difference of load between the simulation results of Dest and Energy Plus was less than 10%. Its accuracy depends on the input parameters and the selected simulation software. The construction of the physical model requires a large number of physical parameters related to the building and a detailed setting of the system operation. White-box physics-based models rely on thermodynamic rules for detailed energy modelling and analysis. However, building electricity forecasting continues to be a challenging effort due to the variety of factors that affect energy consumption, such as building structure, equipment, weather conditions, and energy-use behaviours of the building occupants.īuilding electricity consumption predictions can be divided into three methods according to the type of data input and processing method used: White-box physics-based models, grey-box reduced-order models and black-box data-driven models. Scientists have explored various methods for predicting building electricity consumption, aiming to achieve intelligent energy management and energy-saving building reconstruction based on predicted energy consumption. From a global perspective, building energy consumption accounts for about 40% of the global energy consumption, and this proportion is likely to increase in the future. Among energy sources, building electricity consumption accounts for a large proportion of total social energy consumption. Increasing demand for energy is gradually drawing attention to energy conservation issues around the world. In the three control groups mentioned above, the R 2 value of the hybrid model improved by 10, 3 and 3%, respectively, the values of the mean absolute error (MAE) decreased by 48.9, 41.4 and 35.6%, respectively, and the root mean square error (RMSE) decreased by 54.7, 35.5 and 34.1%, respectively.Įnergy is critical in modern society, and energy consumption is a major issue that has long plagued humanity. The experimental results on the measured data of an office building in Qingdao show that the proposed hybrid model can improve the prediction accuracy and has better robustness compared to VMD-MIM-LSTM. In order to verify the performance of the proposed model, three categories of contrast methods were applied: 1) Comparing the hybrid model to a single predictive model, 2) Comparing the hybrid model with the backpropagation neural network (BPNN) to the hybrid model with the LSTM and 3) Comparing the hybrid model using mRMR and the hybrid model using mutual information maximization (MIM). In the forecasting module, the long short-term memory (LSTM) neural network model was used to predict power consumption. In the feature selection, the maximum relevance minimum redundancy (mRMR) algorithm was chosen to analyse the correlation between each component and the individual features while eliminating the redundancy between individual features. For data preprocessing, the variational mode decomposition (VMD) technique was used to used to decompose the original sequence into more robust subsequences. To address this difficulty, a hybrid prediction model based on modal decomposition was proposed in this paper. However, it is difficult for general machine learning models to handle complex time series data such as building energy consumption data, and the results are often unsatisfactory. School of Mechanical Engineering, Tongji University, Shanghai, ChinaĮnergy consumption prediction is a popular research field in computational intelligence.Yingjun Ruan, Gang Wang, Hua Meng and Fanyue Qian*
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |