Irfan Qaisar

Buildings are responsible for approximately 40% of the world’s energy consumption and 36% of the total carbon dioxide emissions. Building occupancy is essential, enabling occupant-centric control for zero emissions and decarbonization. Although existing machine learning and deep learning methods for building occupancy prediction have made notable progress, their analyses remain limited when applied to complex real-world scenarios. Moreover, there is a high expectation for Transformer algorithms to predict building occupancy accurately. Therefore, this paper presents an occupancy prediction Transformer network (OPTnet). We fused and fed multi-sensor data (building occupancy, indoor environmental conditions, HVAC operations) into a Transformer model to forecast the future occupancy presence in multiple zones. We performed experimental analyses and compared it to different occupancy prediction methods (e.g., decision tree, long short-term memory networks, multi-layer perceptron) and diverse time horizons (1, 2, 3, 5, 10, 20, 30 min). Performance metrics (e.g., accuracy and mean squared error) were employed to evaluate the effectiveness of the prediction algorithms. Our OPTnet method achieved superior performance on our experimental two-week data compared to existing methods. The improved performance indicates its potential to enhance HVAC control systems and energy optimization strategies.

Buildings are responsible for approximately 40% of the world’s energy consumption and 36% of the total carbon dioxide emissions. Building occupancy is essential, enabling Occupant-Centric Control for zero emissions and decarbonization. Although existing machine learning and deep learning methods for building occupancy prediction have achieved remarked progress, their analyses remain limited when applied to complex real-world scenarios. Besides, there is a high expectation for Transformer algorithms to predict building occupancy accurately. Therefore, this paper presents an Occupancy Prediction Transformer network (OPTnet). We fuse and feed multi-sensor data (building occupancy, indoor environmental conditions, HVAC operations) into a Transformer model to forecast the future occupancy presence in multiple zones. We perform experimental analysis and compare it to different occupancy prediction methods (e.g., Decision Tree, Long Short-Term Memory networks, Multi-layer perceptron) and diverse time horizons (1,2,3,5,10,20,30mins). The performance metrics (e.g., Accuracy and Mean Squared Error) are employed to evaluate the effectiveness of prediction algorithms. Our OPTnet method achieves superior performance on our experimental two-week data compared to existing methods. The improved performance signifies its potential to enhance HVAC control systems and energy optimization strategies. We will make the code publicly available at https://github.com/kailaisun/occupancy-prediction-binary to promote transparency and reproducibility.

Multiple model methods for nonlinear dynamical system control are appealing because local models can be simple and obvious, and global dynamics can be studied in terms of transitions between small operating zones. In this study, we propose that using qualitative models strengthens the multiple model method even more by enabling each local model to explain a huge class of effective nonlinear dynamical systems. Furthermore, reasoning using qualitative models reveals weak necessary conditions sufficient to verify qualitative features like stability analysis. The authors show the method by creating a global controller for the free pendulum. In addition, local controllers are specified and validated by comparing their patterns to basic general qualitative models. Our proposed procedure establishes qualitative limitations on controller designs that are sufficient to ensure the necessary local attributes and to establish feasible transitions between local areas for the existing problems. As a result, the continuous phase picture may be reduced to a simple transitional graph. The degrees of freedom in the system that are not bound by the qualitative description are still accessible to the designer for optimization for any other purpose. An example of a pendulum plant illustrates the effectiveness of the proposed method.

This paper considers an adaptive event‐triggered robust H_∞ control for the Takagi–Sugeno (T‐S) fuzzy under the networked Markov jump systems (NMJSs) with time‐varying delay. First, a new adaptive event‐triggered scheme is developed to guarantee the T‐S fuzzy NMJSs, and as a result, communication energy consumption reduced while device efficiency is maintained. Besides, an asynchronous operation method is adopted to deal with the mismatched premise variables between the fuzzy system and the fuzzy controller. One of the main objectives of this article is to construct the fuzzy state‐feedback controller (mode‐dependent) in a closed‐loop form for stochastic stability for all admissible parameter uncertainties with an H_∞ performance index. Different from the conventional triggering mechanism, in this paper, the parameters of the triggering function are based on a new adaptive law that is obtained online rather than a predefined constant. To achieve the less conservative control design, a new type of stochastic Lyapunov–Krasovskii functional is designed by decomposing method, in which the delay interval transforms into various equidistant subintervals in terms of linear matrix inequalities. An example of a truck‐trailer application is used to demonstrate the effectively of the proposed algorithms.