Potential of Model-Free Control for Demand-Side Management Considering Real-Time Pricing

Abstract

This paper presents a detailed description of data predictive control (DPC) applied to a demand-side energy management system. Different from traditional model-based predictive control (MPC) algorithms, this approach introduces two model-free algorithms of artificial neural network (ANN) and random forest (RF) to make control strategy predictions on system operation, while avoiding the huge cost and effort associated with learning a grey/white box model of the physical system. The operating characteristics of electrical appliances, system energy consumption, and users’ comfort zones are also considered in the selected energy management system based on a real-time electricity pricing system. Case studies consisting of two scenarios (0% and 15% electricity price fluctuation) are delivered to demonstrate the effectiveness of the proposed approach. Simulation results demonstrate that the DPC controller based on ANN pays only 0.18% additional bill cost to maintain users’ comfort zones and system economy standardization while using only 0.096% optimization time cost compared with the MPC controller.

Data-driven Predictive Control for Unlocking Building Energy Flexibility: A Review

“…However, whether there are potential performance improvements when weather forecasts are used as part of the state in the RL problem is not known and should be investigated. Zhou et al [155] trained data-driven models based on past data from an MPC controller with purely past inputs (ambient temperature and electricity price). In most cases, offline predictions are used (which rely on data which has already been measured) with exact forecasts of disturbances used in testing the controller.…”

A Data-Driven Scheduling Approach for Hydrogen Penetrated Energy System Using LSTM Network

“…Other devices should primarily meet the relationship between power and efficiency or energy efficiency ratio (EER) as shown in Equation (20). The electro-thermal efficiency of devices should also satisfy the upper and lower limits as shown in Equations (21) and (22). The power exchanged between the grid and the HPES also needs to meet the upper and lower limit of Equations (21) and (22).…”

“…Therefore, when online scheduling and control of the HPES is needed, the mathematical-driven approaches sometimes are not able to meet the speed requirements. Thus, some data-driven based approaches are proposed that learn the scheduling experience offline and control the system online [19][20][21][22]. In article [19], the authors proposed a reinforcement learning control approach for building heating systems.…”