Yude Qin

Model predictive control is theoretically suitable for optimal control of the building, which provides a framework for optimizing a given cost function (e.g., energy consumption) subject to constraints (e.g., thermal comfort violations and HVAC system limitations) over the prediction horizon. However, due to the buildings’ heterogeneous nature, control-oriented physical models’ development may be cost and time prohibitive. Data-driven predictive control, integration of the “Internet of Things”, provides an attempt to bypass the need for physical modeling. This work presents an innovative study on a data-driven predictive control (DPC) for building energy management under the four-tier building energy Internet of Things architecture. Here, we develop a cloud-based SCADA building energy management system framework for the standardization of communication protocols and data formats, which is favorable for advanced control strategies implementation. Two DPC strategies based on building predictive models using the regression tree (RT) and the least-squares boosting (LSBoost) algorithms are presented, which are highly interpretable and easy for different stakeholders (end-user, building energy manager, and/or operator) to operate. The predictive model’s complexity is reduced by efficient feature selection to decrease the variables’ dimensionality and further alleviate the DPC optimization problem’s complexity. The selection is dependent on the principal component analysis (PCA) and the importance of disturbance variables (IoD). The proposed strategies are demonstrated both in residential and office buildings. The results show that the DPC-LSBoost has outperformed the DPC-RT and other existing control strategies (MPC, TDNN) in performance, scalability, and robustness.

This paper describes innovative research on the data-driven control policy for building energy-saving based on a cloud platform under four-story building energy Internet of Things (IoT) architecture. The platform devotes to provide data representation and interpretability analysis for different stakeholders to achieve flexibility and scalability; simultaneously, to improve thermal comfort while reducing building power consumption. The platform manages data from the IoT devices and builds energy management systems using a cloud-based, user-friendly human-computer interaction interface. Decisions on how to optimize the operation of building energy systems are becoming increasingly complex. Data-driven predictive control (DPC) is a control technology that replaces conventional strategies, such as rule-based control and model-based predictive control (MPC). When applied to complex building operations, control-oriented data-driven models are used to implement finite time receding horizon control. This paper adopts the CatBoost algorithm for model prediction, which is highly interpretable and easy to operate for different stakeholders. A numerical simulation case study is utilized to compare the control performance, such as the residential building power consumption, R Squares, RMSE, and Test time, with other strategies: XGBoost, TDNN, LightGBM. The results show that the investigated algorithm’s performance outperforms others’, while significantly reducing the complexity and implementation cost.

Abstract This paper makes innovative research on developing a data-driven control strategy under the Energy Internet of Things architecture. On the one hand, the platform aims to provide data representation and interpretable analysis for different stakeholders (end users, construction operators or managers) to realize the flexibility and scalability of the platform; on the other hand, it can improve the thermal comfort and also reduce the power consumption of buildings. However, to process vast amounts of data, it is critical to select appropriate control methods and design optimization issues. Data-driven predictive control (DPC) is a control technology that replaces model-based predictive control (MPC). When applied to complex building operations, MPC is implemented by using the control-oriented data-driven model. The key to DPC technology is the use of CatBoost algorithm, which is highly interpretable and easy to be operated by stakeholders. This paper chooses TDNN, LightGBM, and CatBoost to compare and analyze building energy consumption. Numerical simulation results show that the CatBoost algorithm’s performance is better than other algorithms, and the complexity and implementation cost is significantly reduced.

The problem of detecting a spontaneous abrupt change in the mathematical expectation (disorder) of a Gaussian time series using the Moving Average or MA-algorithm is considered. It is noted that the probabilistic characteristics of this algorithm necessary for its practical use have been obtained with the necessary completeness relatively recently and concerned only the case when the time series elements (its observations) are uncorrelated. The purpose of this work is a full-scale study of the MA-algorithm characteristics under the conditions of correlated observations, with a view to ultimately obtain the reference material necessary for synthesizing the optimal procedure for detecting a disorder. Simulation was carried out, which made it possible to reveal the features of choosing a decision threshold for a given value of the average time between false alarms depending on the smoothing window width of the controlling MA-algorithm and the maximum correlation interval of its observations. In a similar way, the values of the average delay time in producing an alarm signal when a disorder of a given fixed level occurs have been determined, as well as the dependences of the control procedure efficiency indicator on the procedure parameters. A comparison of the effectiveness of MA-algorithms for uncorrelated and correlated observations is carried out.