RAWAvCon Methods Overview and new Aircraft Models

This blog outlines new aircraft models created in the last week and an overview of the RAWAvCon modelling approach. Aerodynamic, weight, propulsion and performance methods will be addressed in coming blogs.

New Aircraft Models

This week, I have modelled 2 modern wide-body aircraft families with which I have significant history. I was responsible for all aircraft level topics in the Rolls-Royce team that secured a position for the Trent 1000 on the Boeing 787 (and as the lead engine). I also provided aircraft performance support during the Trent XWB development programme that included ratings development, engine lifing profiles and marketing support. Both aircraft families provide good baselines for modelling actual and potential new aircraft developments from Airbus and Boeing.

Boeing 787: -8, -9 and -10:

1. Boeing’s latest wide-body family

2. Minimal model changes required across family members

3. Fuselage length, passenger count and design weights

4. Minor changes to systems assumptions - reflects common systems across the family

5. A good baseline for modelling the in-development 777-8 & -9 plus Boeing NMA options

Airbus A350: -900, -1000:

1. Airbus’s latest wide-body family

2. A good baseline to consider potential A350-1100 or A350-2000 models as a response to the 777-8 and -9.

RAWAvCon Methods overview

The RAWAvCon aircraft design system attempts to forecast the top-level attributes of aircraft that would be delivered by a 5-year development programme performed by huge, highly professional and dedicated design teams in the airframe and engine manufacturing companies and their respective supply chains. It is clearly impossible to model all the detail that these design teams address, particularly by one man and his laptop.

This objective is achieved using methods suited to the inputs available and guided by outcomes from previous designs, particularly where public domain data are available.

The methods used are only as complex as required to achieve the desired outcomes i.e. modelling performance within 1-2%. Increasingly complex methods are used only where ALL three of the following criteria are met:

1. appropriate increased complexity methods are available;

2. values for all the important input parameters of the more complex methods are available – ‘garbage in – garbage out’ will often result in less accurate results that the simper methods where the smaller number of inputs are more readily defined;

3. the more complex methods provide at least one of the following:

4. substantially more representative results than those achieved with relatively simpler methods.;

5. Increased functionality that is of some use

The methods used in for conventional transport aircraft are mostly well proven semi-empirical methods modified by my own understanding/experience and research of the major factors and drivers impacting aircraft design and how these have changed from the 1960’s through to the present. In some cases, new methods have been derived (or existing methods modified) where suitable public domain research or data are available.

Should a customer wish to consider a specific component or system in more detail than I have modelled, I would work with the customer to define and enhance RAWAvCon to forecast the relevant attributes to the required detail level. The retention of such capability for general use would depend on its general applicability (are the widely inputs available?) and any associated IP issues. If necessary, bespoke models would be created for use in specific customer studies.

Higher order or 'physics based' methods, e.g. CFD of FEM, would only be used indirectly in RAWAvCon. For example, the results of high order analysis on specific issues could be used to define the relationships between major parameters (i.e. response surfaces or Pareto fronts) that represent best practise for a design team. The response surfaces or Pareto fronts will be included in RAWAvCon with adjustment parameters. This is representative of what a ‘design team’ would deliver – they simply would not select a non-optimum solution without good reason. In such a case, the 'good reason' should be in be included in the response surfaces or represented by a Pareto front or as a limitation on certain input values.

This high order analysis can be performed by a third party, e.g. a customer, published research papers or partners) or my own limited capability. I have previously modelled the Avro Vulcan broad delta configuration in the NASA OpenVSP tool VSPAERO. This could be used to derive incompressible and low speed aerodynamic characteristics for such aircraft.