Sensitivity analysis and control methodology for linear engine alternator 2019-01-0230

Linear engine generator design optimization traditionally has been difficult because each independent variable alters the motion with respect to time, and therefore alters the engine and alternator response to other governing variables. An analogy is drawn to an engine with a very light flywheel, where the rotational speed effectively is not constant. However, when springs are used in conjunction with a linear engine alternator (LEA), the motion becomes more consistent and more sinusoidal with increasing spring stiffness. This reduces some attractive features, such as variable compression ratio HCCI operation, but aids in reducing cycle to cycle variation for conventional combustion modes. To understand the cycle-to-cycle variations, we have developed a comprehensive model of an LEA with a 1kW target power in MATLAB®/Simulink, and an LEA corresponding to that model has been operated in the laboratory. The MATLAB®/ Simulink numerical model has been used to examine the sensitivity of the LEA dynamics and performance parameters (dependent variables) to changes in the design and operating inputs. Design variables such as the LEA translating mass and spring stiffness alter the operating frequency in an expected fashion, whereas input parameters such as heat release, friction and heat transfer model constants alter the efficiency and power density. The sensitivity analysis provides insight into the pathway for improving and optimizing the design, as well as an assessment of the effects of modeling assumptions on the reliability of predictions. A difficulty during the modeling is associated with the cycle to cycle energy balance for the LEA, and it is clear that this difficulty is reflected in real-world LEA control. If the alternator consumes more energy in a cycle than the engine provides, the system moves towards a stall. If the alternator consumes less energy, then the stroke, compression ratio and maximum translator velocity must rise steadily from cycle to cycle until ulterior energy (efficiency) losses curb the increase. We have recognized that the control of this energy balance in the model affects sensitivity analysis and must, therefore, mimic the real world intended control methodology. The energy balance control is analogous to the use of a governor to manage crankshaft rotational speed in a conventional engine. In this conventional case, frequency varies with constant compression ratio, whereas for the LEA the instinct is to control the compression ratio at substantially constant operating frequency.