Simulating Energy Efficient Control of Multiple-Compressor Compressed Air Systems

In many industrial facilities it is common for more than one air compressor to be operating simultaneously to meet the compressed air demand. The individual compressor set-points and how these compressors interact and respond to the facility demand have a significant impact on the compressed air system total power consumption and efficiency. In the past, compressors were staged by cascading the pressure band of each compressor in the system. Modern automatic sequencers now allow more intelligent and efficient staging of air compressors. AirSim, a compressed air simulation tool, is now able to simulate multiple-compressor systems with pressure band and automatic sequencer controls. AirSim can simulate a current compressed air system and a proposed system with changes to the equipment and/or controls. Thus, quickly and accurately, users can calculate the energy and cost savings expected from many proposed compressed air system upgrades. INTRODUCTION Nearly every industrial plant contain compressed air systems. In many industrial facilities air compressors use more electricity than any other single type of equipment. Commonly referred to as the “fourth utility”, compressed air systems can typically be optimized to decrease the energy use of the system by 20% to 50%. In addition to energy and cost savings, an energy efficient compressed air system can reduce maintenance, extend the useful life of the system components, and improve system reliability [6]. Compressed air controls match the compressed air supply with the facility demand and can be one of the most important determinants in overall system energy efficiency. Compressed air systems are sized for the maximum expected plant air demand, thus these systems typically operate only partially loaded. Compressed air system controls coordinate how individual compressors operate and how multiple compressors interact to deliver the required pressure and volume of air to the facility in the most reliable and efficient manner. Systems with multiple compressors contain greater opportunity for controls optimization. The three main types of multiplecompressor control strategies which will be discussed in this paper are: pressure band control, network sequencer control, and automatic sequencer control (also referred to as system master control) [13]. Compressor air component manufactures are acutely aware of the potential for energy savings from multiple-compressor controls. Atlas Copco, Kaeser, and Quincy all market compressed air system central controllers to optimize system efficiency [1] [10] [12]. Furthermore, the 2013 California Building Energy Efficiency Standard, which became law July 1, 2014, requires a central controller for multiplecompressor compressed air systems with total rated power over 100-hp. This standard also requires a variable speed drive (VSD) trim compressor [8]. As will be discussed later, these two requirements cannot be met with pressure band control or network sequencer control. Only automatic sequencer control allows a trim compressor to always meet the partload marginal system demand. This paper begins by reviewing the basics of simulating individual air compressors, fundamental to the compressed air simulation tool AirSim. Next, the basic principles and control algorithms are detailed for pressure band control, network sequencer control, and automatic sequencer control strategies for multiple-compressor compressed air systems. Finally, a case study is presented demonstrating the use of the improved compressed air simulation tool, AirSim [9], to quickly and accurately model multiple-compressor compressed air systems. SIMULATING SINGLE AIR COMPRESSOR PERFORMANCE Individual air compressors can be controlled in several ways. Schmidt and Kissock describe these control methods as generalized linear relationships between fraction full-load power (FP) and fraction rated capacity (FC) [3]. Using linear generalizations and assigning FP0 as the fraction of full-load power consumed when the compressor is producing no compressed air, the relationship between FP and FC can be modeled as: FP = FP0 + (1 – FP0) × FC (1) The normalized power and capacity coefficients in Equation 1 are the actual power and capacity divided by the maximum power and capacity: FP = P / FLP (2) FC = C / FLC (3) FP0 = P0 / FLP (4) P is the actual compressor power, FLP is the full-load compressor power, C is the actual compressed air output, FLC is the full-load compressor output capacity, and P0 is the compressor power when producing no compressed air. Schmidt and Kissock originally graphed the linear relationships between FP and FC for different control methods [3]. Figure 1 shows these FP and FC relationships for several common compressor control methods with added insight. While Equation 1 can be used to model the part-load efficiencies of these different control types, it is important to notice the variations which occur for load/unload, variable speed, and on/off control. Load/unload and on/off only operate at full-load, 100% capacity and 100% power, or no-load, 0% capacity and FP0. Variable speed control can operate on the continuum between full-load and about 25% FC. Blow off and modulation control operate continuously between full-load and no-load. Figure 1. FP vs FC for Common Compressor Control