Search

Heating, ventilation and air conditioning (HVAC) systems account for approximately 50% of the total energy consumption in buildings. Advanced control and optimal operation, seen as key technologies in reducing the energy consumption of HVAC systems, indispensably rely on an accurate prediction of the building’s heating/cooling load. Therefore, the goal of this research is to develop a model capable of making such accurate predictions. To streamline the process, this study employs sensitivity and correlation analysis for feature selection, thereby eliminating redundant parameters, and addressing distortion problems caused by multicollinearity among input parameters. Four model identification methods including multivariate polynomial regression (MPR), support vector regression (SVR), multilayer perceptron (MLP), and extreme gradient boosting (XGBoost) are implemented in parallel to extract value from diverse building datasets. These models are trained and selected autonomously based on statistical performance criteria. The prediction models were deployed in a nearly zero-energy office building, and the impacts of feature selection, training set size, and real-world uncertainty factors were analyzed and compared. The results showed that feature selection considerably improved prediction accuracy while reducing model dimensionality. The research also recognized that prediction accuracy during model deployment can be influenced significantly by factors like personnel mobility during holidays and weather forecast uncertainties. Additionally, for nearly zero-energy buildings, the thermal inertia of the building itself can considerably impact prediction accuracy in certain scenarios.

An accurate and fast building load prediction model is critically important for guiding building energy system design, optimizing operational parameters, and balancing a power grid between energy supply and demand. A physics-based simulation tool is traditionally used to provide the building load demand; however, it is constrained by its complex model development process and requirement for engineering judgments. Machine learning algorithms (i.e., data-driven models) based on big data can bridge this gap. In this study, we used the massive energy data generated by a physics-based tool (EnergyPlus) to develop three data-driven models (i.e., LightGBM, random forest (RF), and long-short term memory (LSTM)) and compared their prediction performances. The physics-based models were developed using office prototype building models as baselines, and ranges were provided for selected key input parameters. Three different input feature dimensions (i.e., six-, nine-, and fifteen-input feature selections) were investigated, aiming to meet different demands for practical applications. We found that LightGBM significantly outperforms the RF and LSTM algorithms, not only with respect to prediction accuracy but also in regard to computation cost. The best prediction results show that the coefficient of variation of the root mean squared error (CVRMSE), squared correction coefficient (R2), and computation time are 5.25%, 0.9959, and 7.0 s for LightGBM, respectively, evidently better than the values for the algorithms based on RF (18.54%, 0.9482, and 44.6 s) and LSTM (22.06%, 0.9267, and 758.8 s). The findings demonstrate that a data-driven model is able to avoid the process of establishing a complicated physics-based model for predicting a building’s thermal load, with similar accuracy to that of a physics-based simulation tool.

During the design and planning stage of a district energy system, the prediction of the cooling and heating loads is an important step. The accurate estimate of the load pattern can provide a basis for the configuration and optimization of the system. To meet the demand in practical application, this paper proposes a fast load prediction method for district energy systems based on a presimulated forward modelling database and KNN (K-nearest neighbor) algorithm and develops it into a practical tool. Owing to the absence of some design parameters at the planning stage, scenario analysis is also used to determine some input conditions for load prediction. In this paper, the scenarios cover three types of building: office, shopping mall and hotel. To test the performance of this new method, we randomly selected 15 virtual buildings (5 buildings for each type) with different design parameters and took their detailed BPS (building performance simulation) model as a benchmark to assess the prediction results of the new method. The index “ratio of the hours with effective prediction” is defined as the ratio of the hours whose relative error of hourly load prediction is less than 15% to the hours whose load is not 0 in the whole year, and the test result shows that this index is not less than 0.9 (90%) for the predicted cooling load of all 45 test cases and the predicted heating load of 25 of the 45 cases. As a research achievement with practical value, this paper accomplishes the programming work of the tool and makes it into a software. The application of this software in the actual project of district energy system is also presented. The performance of the new load prediction tool was compared with the traditional approach commonly used in engineering—the load estimation based on reference building models—and the result shows that the fast load estimate tool can provide the same level of prediction accuracy as traditional simulation methods.