Intelligent Building Control Systems

Most of the content of this volume highlights new research developments in building automation, with an emphasis on heating, ventilation, and air conditioning (HVAC) control. This research is motivated by important industry and societal imperatives for improving the energy efficiency, carbon footprint, occupant comfort, and economics of building operation. Much needs to be done and control scientists and engineers have an important role to play. In this chapter, we hope to “ground” the research. The chapter consists of two primary parts. We first discuss the systems aspect of building automation systems (BASs). State-of-the-art BASs are large, complex, distributed systems. They connect to sensors, actuators, and low-level controllers; provide interfaces for operational, engineering, and management staff; and, increasingly, interconnect with other computer systems for enterprise-level applications such as facility management, energy management, and computerized maintenance management systems. A trend we highlight is the move toward “connected” buildings—in today’s Internet-of-Things (IoT) age the BAS extends to the cloud. Interoperability is another driving force in building automation; solutions that enable building owners and operators to combine equipment and applications from different suppliers—to avoid vendor lock-in—are also discussed. The content in this part of the chapter is relevant across the buildings sector. For illustration purposes, we discuss Honeywell systems that we are familiar with, especially the Enterprise Building Integrator (EBI) platform and the Tridium Niagara framework. The penetration of modern BASs is still largely limited to medium- and large-scale commercial buildings. Few light commercial buildings have systems of this sophistication (and cost). We also outline typical building control systems in use for light commercial. This building sector represents a major opportunity for new innovation—advanced control promises huge impact on energy efficiency, for example, provided that the technology can be delivered at low cost and is easy to deploy and operate by non-control-experts. The second part of the chapter discusses recent research projects in advanced control and related technologies that we have been involved with, most of which have now matured to the point of being deployed in buildings. Results from operational installations are described where available. We believe the controls research community can benefit from being better informed about the state of the art in building automation and control, with regard to the system platform as it exists and how it is evolving as well as to the algorithmic innovations that are being explored and applied in industry. We hope that, within the limitations of our experience and understanding, the content of this chapter will serve this purpose.

In this chapter, we delve deeper into understanding modeling and control approaches for one of the important subsystems in an intelligent building, the HVAC system. Specifically, Vapor Compression Systems (VCS) are the primary energy systems in building air conditioning, heat pump, and refrigeration systems. We will discuss standard methods for constructing dynamic models of vapor compression systems, and their relative advantages for analysis, design, control design, and fault detection. The principal interests are moving boundary and finite-volume approaches to capture the salient dynamics of two-phase flow heat exchangers. We will present modeling approaches for auxiliary equipment, such as, valves, compressors, fans, dampers, and heating/cooling coils, allowing the reader to understand the construction of typical HVAC system models. We will then highlight limitations of such models and address advanced modeling approaches for challenging transient scenarios. Finally, we give a summary of single-input, single-output control strategies for HVAC system, with simulation and experimental examples to illustrate their effectiveness.

While the previous chapter presented modeling and control strategies for vapor compression systems in general, in this chapter, a model predictive controller is designed for a multi-zone vapor compression system. Controller requirements representing desired performance of production-scale equipment are provided and include baseline requirements common in control literature (constraint enforcement, reference tracking, disturbance rejection) and also extended requirements necessary for commercial application (selectively deactivating zones, implementable on embedded processors with limited memory/computation, compatibility with demand response events.). A controller architecture is presented based on model predictive control to meet the requirements. Experiments are presented validating constraint enforcement and automatic deactivation of zones.

In this chapter, we now address modeling and control for multi-zone and building-level temperature regulation. Many commercial and residential buildings have multiple independent zones requiring space conditioning. These zones may be uniform in size or may have a significant size distribution. Moreover, the space conditioning requirements may be quite disparate depending on the zone usage. This chapter provides an overview of multi-zone temperature control within buildings. We review modeling approaches for building zones and their interconnections. The primary modeling framework is a resistor–capacitor framework where thermal energy is stored in a zone in a capacitive sense and transferred between zones of different temperature through a thermally resistive path. The zones are fed by a cooling or heating system that could be liquid or air. Subsequently, we focus attention on control of multi-zone systems starting with an appropriate architecture for their representation. We present particular structures for energy flow that naturally decompose the dynamic properties of the building into hierarchical structures. These structures can then be used to create controllers or either a centralized or distributed variety. Simulations serve to illustrate the concepts and represent the results for both the modeling and dynamic control.

This chapter focuses on advanced control design, specifically for forced air HVAC systems. Such advanced control schemes incorporate predictions of weather, occupancy, renewable energy availability, and energy price signals in order to deliver performance-driven automated decision making at a hierarchy of levels. The chapter covers thermal modeling for controls, predictive control design, and implementation of such controllers in real-world buildings. An overview of standard computational platforms and communication systems in buildings is reported. Our main objective is to discuss how advanced control relates to the existing building practices; in particular, a distributed control logic “Trim and Respond” is described in detail. The “Trim and Respond” logic is shown to match a one-step explicit distributed model predictive controller. The chapter concludes with an algorithm for advanced distributed model predictive control that is implementable on existing distributed building control architectures.

While the primary purpose of most buildings is to provide a safe and comfortable environment for its occupants, most of the current studies and solutions developed for building thermal control have been designed independent of the occupant feedback. An acceptable temperature range for the occupancy level is estimated, and control input is designed to maintain temperature within that range during occupancy hours. Consider office floors, conference rooms, student dorms, homes, and other multi-occupant spaces where temperature is chosen irrespective of the number of occupants and their individual preferences. This existing approach is not only non- user-centric but also suboptimal from both energy consumption and occupant satisfaction/productivity perspectives. It is thus highly desirable for such multi-occupant spaces to have a mechanism that would take into account each occupant’s individual comfort preference and the energy cost, to come up with optimal thermal setting. Individual occupant’s feedback and preference can be obtained through wearable sensors or smart phone applications. In this chapter, we propose algorithms that take into account each occupant’s preferences along with the thermal correlations between different zones in a building, to arrive at optimal thermal settings for all zones of the building in a coordinated manner. First, we present a control algorithm that uses binary occupant feedback based on singular perturbation theory to minimize aggregate user discomfort and total energy cost. A consensus algorithm for attaining a common temperature setpoint in a typical multi-occupant space is presented next that uses Alternating Direction Method of Multipliers (ADMM) to solve the consensus problem. We use our Watervliet, NY-based test facility to demonstrate the performance of our algorithms.

In this chapter , we turn our attention to another subsystem in an intelligent building, the illumination (or lighting) system. While traditional lighting systems were designed to provide illumination for the occupants of the building, with the advent of light-emitting diode (LED) technology, lighting systems are now far more sophisticated and responsive to the needs of the occupants. In this chapter, we first present a broad overview of modeling and control strategies for smart and autonomous lighting systems. One of the key challenges in development of lighting control strategies is the lack of standardized test scenarios, benchmarking, and the wide diversity of available hardware. To highlight this, we will describe and compare four lighting control algorithms that are applicable for different lighting scenarios and compare and contrast them. We also demonstrate and benchmark these algorithms on a full-scale testbed (a conference room). Finally, we present a decision tree to guide the reader on the choice of a suitable lighting control scheme.

The next-generation electric grid needs to be smart and sustainable to simultaneously deal with the ever-growing global energy demand and achieve environmental goals. In this context, the role of residential and commercial buildings is crucial, due to their large share of primary energy usage worldwide. In this chapter, we describe energy management frameworks for buildings in a smart grid scenario. An Energy Management System (EMS) is responsible for optimally scheduling end-user smart appliances, heating systems, ventilation units, and local generation devices. We discuss the performance and the practical implementation of novel stochastic MPC schemes for HVAC systems, and illustrate how these schemes take into account several sources of uncertainties, e.g., occupancy and weather conditions. Furthermore, we show how to integrate local generation capabilities and storage systems into a holistic building energy management framework.

There is an increasing interest in connecting things into network (known as the Internet of Things) to improve the performance and to provide novel services. Control plays a big role here not just to help connecting things together, but also to make things “smarter”. In this chapter, we will focus on a particular type of Internet of Things, namely the smart buildings. There is an increasing demand for energy efficiency, comfort, and safety in buildings. It is possible to achieve these different and sometimes conflicting objectives in the same time. Occupant-oriented wireless sensor network plays a key role, which collects information on the demand (what the occupant wants), the supply (what the building can offer), and how the two parts may coordinate with each other (the elasticity of the demand and the supply). We will briefly review the state of the art and the state of practice in this field. In particular, we will see how to control such an Internet of Things to achieve energy saving and fast evacuation in smart buildings.