Diane L. Peters

Control Proxy Functions for Sequential Design and Control Optimization

Optimal system design of “smart” products requires optimization of both the artifact and its controller. When the artifact and the controller designs are independent, the system solution is straightforward through sequential optimization. When the designs are coupled, combined simultaneous optimization can produce system-optimal results, but presents significant computational and organizational complexity. This paper presents a method that produces results comparable with those found with a simultaneous solution strategy, but with the simplicity of the sequential strategy. The artifact objective function is augmented by a control proxy function (CPF), representing the artifact’s ease of control. The key to successful use of this method is the selection of an appropriate CPF. Four theorems that govern the choice and evaluation of a CPF are given. Each theorem is illustrated using a simple mathematical example. Specific CPFs are then presented for particular problem formulations, and the method is applied to the optimal design and control of a micro-electrical mechanical system actuator. [DOI: 10.1115/1.4004792]

ON MEASURES OF COUPLING BETWEEN THE ARTIFACT AND CONTROLLER OPTIMAL DESIGN PROBLEMS

Optimization of smart products requires optimizing both the artifact design and its controller. The presence of coupling between the design and control problems is an important consideration in choosing the system optimization method. Several measures of coupling have been proposed based on different viewpoints of the system. In this paper, two measures of coupling, a vector based on optimality conditions and a matrix derived from an extension of the global sensitivity equations, are shown to be related under certain conditions and to be consistent in their coupling determination. The measures’ physical interpretation and relative ease of use are discussed using the example of a positioning gantry. A further relation is derived between one measure and a modified sequential formulation that would give results sufficiently close to the true solutions.Copyright

Optimal Component Sizing and Forward-Looking Dispatch of an Electrical Microgrid for Energy Storage Planning

Optimal design of an electrical microgrid and sizing of its components seeks to balance capital investment with expected operational cost while meeting performance requirements. Calculating operational cost requires scheduling each microgrid component over some time period (dispatching) for each design evaluated. Heuristic or rule-based dispatch strategies typically consider only single time instances and are computationally efficient but do not include scheduling energy storage for future time periods. In this paper, we propose to optimize microgrid designs using forward-looking optimal dispatch for future energy storage planning. We present a case study of an ‘islanded’ military base microgrid with renewable and non-renewable electricity generation, battery storage, and plug-in vehicles with electrical export power capability. The optimal design and forward-looking dispatch strategy are compared to results obtained using the publicly available rulebased dispatch strategy in HOMER Energy software. Results show that the forward-looking strategy uses storage batteries to plan for future energy shortfalls rather than simply as a buffer for variable renewable energy supply, resulting in a 7.8% reduction in predicted fuel use. For the given cost assumptions, sensitivity analysis of the optimal design with respect to fuel price shows that investment in renewable energy technology is justified at prices greater than

Organic vapor jet printing at micrometer resolution using microfluidic nozzle arrays

5 per gallon (

MODEL PREDICTIVE CONTROL OF A MICROGRID WITH PLUG-IN VEHICLES: ERROR MODELING AND THE ROLE OF PREDICTION HORIZON

1.32/liter) with an attendant reduction in fuel use of 3–30%.

Co-design of a MEMS actuator and its controller using frequency constraints

Organic vapor jet printing with a print head comprised of a microfluidic Si nozzle array is used to deposit parallel lines of an organic semiconductor thin film with a line width of 16 μm and edge resolution of 4 μm. Line width and feature size are functions of process conditions, depending strongly on nozzle-to-substrate separation distance. Experimental results are accurately characterized by a direct simulation Monte Carlo model. The model suggests that feature sizes of <1.5 μm are attainable by this printing process. The ability of the print head to codeposit doped films is demonstrated by growing the emissive layer of a green phosphorescent organic light emitting diode sandwiched between hole and electron transport layers deposited by vacuum thermal evaporation. This device had an external quantum efficiency of 8.8±1.3%, comparable to a similar device entirely grown by vacuum thermal evaporation.

Coupling Between Component Sizing and Regulation Capability in Microgrids

We demonstrate the use of model predictive control (MPC) for a microgrid with plug-in vehicles. A predictive model is developed based on a hub model of the microgrid, and the control is optimized for minimum generator fuel usage. A variety of horizons and levels of prediction error are used in the optimization. A new method to model expected load and error is presented based on radial basis functions. Results show that for a given prediction horizon, as the level of prediction error increases, the amount of fuel used increases. Results also show that in some cases there is little benefit in extending the prediction horizon. While an extended prediction horizon does result in increased use of battery storage, this does not necessarily produce significant decreases in fuel usage. This result is analyzed and explained in terms of battery charging and discharging limitations.Copyright

Generalized Coupling Management in Complex Engineering Systems Optimization

A MEMS actuator and its controller are jointly optimized, both sequentially and simultaneously. The sequential problem is formulated to account for controllability by means of a constraint on the actuator’s natural frequency. By varying the frequency constraint, sequential optimization generates a set of designs with significantly increased displacement, compared to the original non-optimized design, and with various settling times. In simultaneous optimization, a non-linearly weighted objective function combines the two objectives, and the relative weights are varied. The tradeoff between the two objectives shows that the use of the frequency constraint serves as an effective surrogate for controllability of the actuator.Copyright

Sequential Co-Design of an Artifact and its Controller Via Control Proxy Functions

Increasing energy security and reliability concerns are intensifying the interest in microgrids. In this setting, design optimization is vital to achieve a reliable infrastructure without overbuilding. This paper considers the impact of frequency and voltage regulation on the optimal design of a conceptual, autonomous military microgrid. This microgrid comprises a solar panel and vehicles as power sources, with each vehicle incorporating a battery and generator. The power output and terminal voltage of these inverter-based sources must be regulated. The paper investigates the effects of battery DC voltage variations on a decentralized regulation scheme, and the resulting influence on optimal component sizing. To this end, controllers are first designed based on the typical assumption that the voltage on the DC side of each inverter is constant. The battery internal resistance is then considered and its impact on regulation performance is investigated. The results show that the battery internal resistance can affect the performance of both frequency and voltage regulation, and consequently must be taken into account in component sizing decisions. Thus, the paper identifies an important coupling between regulation and component sizing problems through battery characteristics, and highlights the need for a combined sizing and regulation framework for microgrid design.

A procedure for evaluating the applicability of a control proxy function to optimal co-design

Decomposition-based design optimization strategies are used to solve complex engineering system problems that might be otherwise unsolvable. Yet, the associated computational cost can be prohibitively high due to the often large number of iterations needed for coordination of subproblem solutions. To reduce this cost one may exploit the fact that some systems may be weakly coupled and their interactions can be suspended with little loss in solution accuracy. Suspending such interactions is usually based on the analyst’s experience or experimental observation. This article introduces an explicit measure of coupling strength among interconnected subproblems in a decomposed system optimization problem, along with a systematic way for calculating it. The strength measure is then used to suspend weak couplings and, thus, improve system solution strategies such as the model coordination method. Examples show that the resulting strategy may decrease the number of required function evaluations significantly.