The CPS-Based Simulation of Indoor Thermal Comfort Control with Energy Saving

Introduction

Since the concept of cyber-physical system (CPS) was first proposed by U.S. National Committee in 2006, more and more CPSs are created and provide services for human beings in many important domains, such as smart building, traffic transportation, industry automation and remote patient care. However, due to the characteristics of distribution, heterogeneity and autonomy of CPS, the simulation and modeling for complex CPS are still challenging and significant work, especially in the domain of indoor comfort control.

CPS is defined as a controllable, credible and scalable network physical device system deeply integrated with the ability of computation, communication and control on the basis of environmental perception. A typical CPS is the integrations of both physical and cyber components. The physical components usually consist of sensors, controllers, actuators, controlled objects and so on, and the cyber components usually consist of computers, networks. The physical world follows dynamic ruled by the physics laws, while the cyber part computes, communicates and control the physical processes with feedback loops. Physical processes affect computations and vice versa.

There are mainly two differences between CPSs and traditional systems. One is the human’s role in the system. In CPS, human is not only the designer, supervisor and user of the system as in tradition, but also one part of the control loop. For example, Functional locked-in individuals with cognitive capability are able to control the robotic wheelchairs by playing the role of watching or speaking sensors in human-in-the-loop cyber-physical system (HilCPS) (Gunar et al., 2012; Zeiler et al, 2011). The other difference is the autonomy in CPS. The CPS senses the environment, gathers and transfers the data, computes and controls autonomously. The data gathered from external environment which include human, room and other devices is transferred to computing unit and processed into the information to control the system. By real-time computing, communication, and control, CPS fulfills effective operation between the information world and the physical world. In this way, the networked systems, such as industrial control system, wireless sensor network system (WSN) and embedded system, all belong to the research domain of CPS (Derler et al., 2012). The structure of CPS is shown in Figure 1.

With the development of computer science, communication technology and control technology, the applications of CPSs are getting widely in practice. Smart grid becomes more intelligent, efficient and robust (Li et al., 2014). Patients get timely treatment due to real-time monitoring of CPS (He et al., 2012). Humans will live and work in smart buildings with comfort and less energy cost (Cheng et al., 2013).

Figure 1.

The structure of CPS

For a newly built complex system, especially for a CPS, the simulation and modeling are significant and necessary to reduce the development effort and cost. Developers can estimate the behavior of the target system in the simulation system without building the whole actual target hardware. As systems are becoming more and more complex, the importance of simulation and modeling grows bigger and bigger. However, due to the characteristics of uncertain interactions, heterogeneity, real-time communication and distribution in CPS, it’s more complicated to simulate the CPS compared to traditional computing system. To simulate the CPS precisely, the following challenges and difficulties should be considered (Kang et al., 2012; Wan et al., 2010):