Alexander Pollok

Using multi-objective optimization to balance system-level model complexity

Deciding what to include in a model is the essence of physical modeling. If this kind of decisions is done incorrectly, it results in a model that is both less accurate and offers less performance than others. This paper presents a method to deal with these decision problems and quantify the results. A workflow is proposed, where the modeler first builds a model and includes several sets of replaceable subsystems. A multi-objective optimization algorithm then identifies selections of subsystems that are pareto optimal regarding accuracy and computation time. Finally, the modeler can pick one of these selections based on the modeling intent. An implementation of the proposed method using Modelica and Python is presented and potential pitfalls are explained. Special consideration is given to the quantification of the error or distance between models of different structures. The method is then illustrated by the use of two examples, one of them a complex model of an avionics cooling system cold plate. Suggestions are given regarding the embedding of the method into typical modeling workflows by the use of custom annotations.

Control strategies for an advanced aircraft-cabin temperature-system

Thermal regulation of aircraft cabins requires controlling the temperature of supplied fresh air. State of the art plant architectures support only a small number of temperature zones. In this paper we consider a novel architecture that allows an arbitrary number of temperature zones. This is bought at the expense of a more complex control problem. Control challenges connected to this novel architecture are identified and possible control approaches are presented. They are benchmarked against high-fidelity models in the equation-based object-oriented modelling language Modelica. Results show that control input normalization offers significant advantages for any kind of control system, while the choice between PID-based and LQG-based control is somewhat ambiguous: The former shows better performance in the nominal case, the latter exhibits better robustness.

The use of Ockham's Razor in object-oriented modeling

From the perspective of a practitioner, the development of perfect equation-based models is limited by language, hardware, and ones own mind. While the first two aspects are covered extensively in literature, only little attention has been given to the third one. We make a case for simple models, with a focus on two aspects: use of inheritance and creation of flexible models. Both can have adverse side-effects if used without restriction. To exemplify this discussion, two versions of a library as used in the aerospace industry are compared. The old version made heavy use of inheritance and tried to conduct everything with a minimal number of components. It was completely redesigned after maintenance efforts became too high. A psychological experiment was performed, where the effect of inheritance on the ability of participants to understand a model was analyzed. Results showed that each level of hierarchy significantly increases the time to understand a model by 26.65 s, when correcting for total model length. This supports our hypothesis that flat models are easier to understand than deeply nested models.

Representations of equation-based models are not created equal

For equation-based modelling languages, modelling experts have many degrees of freedom when building a model from scratch. One of the most basic choices the expert faces is the mode of representation. The same system can be represented for instance as a block-diagram, by writing down the physical equations, by writing an algorithm, or by graphically connecting ready-made subcomponents. To give some guidance in this aspect, an experiment was conducted to measure the effects of different representations on various tasks. Participants had to identify models and predict their transient response. Both the time to execute the task and the correctness of the answer were measured. Participants also had to rate their confidence regarding the models. Results showed that tasks were executed much faster for graphical representations than for block-digrams. Equation-based and algorithm-based models can be grouped in the middle. The same results hold for rated confidence. Interestingly, the amount of errors was similar for all representations. Apparently, modelling experts largely compensate for difficulty by taking their time.

Modelling and simulation of self-regulating pneumatic valves

ABSTRACT In conventional aircraft energy systems, self-regulating pneumatic valves (SRPVs) are used to control the pressure and mass flow of the bleed air. The dynamic behaviour of these valves is complex and dependent on several physical phenomena. In some cases, limit cycles can occur, deteriorating performance. This article presents a complex multi-physical model of SRPVs implemented in Modelica. First, the working principle is explained, and common challenges in control-system design-problems related to these valves are illustrated. Then, a Modelica-model is presented in detail, taking into account several physical domains. It is shown, how limit cycle oscillations occurring in aircraft energy systems can be reproduced with this model. The sensitivity of the model regarding both solver options and physical parameters is investigated.

Exploitation Strategies of Cabin and Galley Thermal Dynamics

The thermal inertia of aircraft cabins and galleys is significant for commercial aircraft. The aircraft cabin is controlled by the Environment Control System (ECS) to reach, among other targets, a prescribed temperature. By allowing a temperature band of ± 2 K instead of a fixed temperature, it is possible to use this thermal dynamic of the cabin as energy storage. This storage can then be used to reduce electrical peak power, increase efficiency of the ECS, reduce thermal cooling peak power, or reduce engine offtake if it is costly or not sufficiently available. In the same way, also the aircraft galleys can be exploited. Since ECS and galleys are among the largest consumers of electrical power or bleed air, there is a large potential on improving energy efficiency or reducing system mass to reduce fuel consumption of aircraft. This paper investigates different exploitation strategies of cabin and galley dynamics using modelling and simulation. Modelica models of the thermal and the electrical system are used to assess and compare these different strategies. Potential impacts on passenger comfort are discussed. Additionally, the gained performance is compared to more conventional storage elements like electrical batteries. Finally, the potential of fuel reduction will be quantified using a reference aircraft model and the optimal strategy is selected.

Advanced Temperature Control in Aircraft Cabins - A Digital Prototype

For thermal cabin control of commercial aircraft, the cabin is usually divided into a small number of temperature zones. Each zone features its own air supply pipe. The necessary installation space for ducting increases significantly with the number of zones. This requires the number of temperature zones to be low. Factors such as seating layout, galley placement and passenger density result in deviations in heat flux throughout the cabin. These deviations cannot be compensated by the control system, if they occur within the same temperature zone. This work presents a novel temperature regulation concept based on local mixing. In this concept, two main ducts span the complete cabin length, and provide moderately warm and cold air. At each temperature zone, cabin supply air is locally mixed using butterfly valves. In this way, the number of temperature zones can be individually scaled up without any additional ducting, only requiring additional valves for each temperature zone. The access to hot trim air at the site of the mixing chamber can therefore be omitted. The advantages of this concept are bought at the expense of a more complex control system and a more sophisticated management of failure modes. Pneumatic coupling between temperature zones is much more pronounced compared with the traditional architecture. A centralized control architecture is required. A prototype architecture is presented, as well as a corresponding control system candidate. Possible strategies for the handling of failure modes are discussed. High fidelity simulations using the object-oriented equation-based modelling language Modelica are used to demonstrate robust performance under a wide range of boundary conditions.