METHOD FOR OPTIMISATION OF AN AVIONICS PLATFORM

• US 20100292969A1
• Filed: 05/17/2010
• Published: 11/18/2010
• Est. Priority Date: 05/18/2009
• Status: Active Grant

• Current Assignee: Airbus Operations (European Aeronautic Defence and Space Company EADS NV)
• Original Assignee: Airbus Operations (European Aeronautic Defence and Space Company EADS NV)
• Inventors: Minot, Frédéric
• Granted Patent: US 8,423,331 B2
• Time in Patent Office: Days
• US Class Current: 703/2
• CPC Class Codes: G06F 30/00 Computer-aided design [CAD]

Abstract

The invention relates to a method for optimisation of the avionics platform of an aircraft, intended to implement a set of avionics functions. The association between the platform'"'"'s functional architecture elements (functions, functional links) and the elements of its equipment architecture (equipment, network paths, direct physical links) are formalised by sets of constraints in the form of linear inequalities of decision variables. Certain segregation or co-location functional constraints are also envisaged. Optimisation of the avionics platform is achieved through the minimisation of cost functions which are linearly dependent on said decision variables. This minimisation under constraints is achieved in an effective manner by the Simplex algorithm.

Claims

1. Computer-assisted method for optimisation of aircraft'"'"'s avionics platform, said method being intended to implement a set of avionics functions, where each avionics function is associated with one or more software module instances, where any two software module instances can exchange messages by means of a link, called a functional link instance, where said avionics platform additionally comprises equipment linked together by unidirectional or bidirectional physical links which belong to different types of networks, paths used to link said equipment, where each path is made up either of a single direct physical link between an item of source equipment and destination equipment, or of a first physical link between said source equipment and a first network node, a second physical link between said destination equipment and a second network node and a network path between said first and second nodes, characterised in that on the one hand functional link instances are associated with paths intended to implement these instances, by means of a first set of constraints and, on the other hand, software modules instances with equipment intended to host these instances, by means of a second set of constraints, where said constraints of the first and the second sets are formulated as linear inequalities of decision variables, and in that multiple cost functions formulated as linear combinations of at least certain of said decision variables are minimised in order to obtain at least one optimum solution for implementation of the platform.

2. Computer assisted method for optimisation, according to claim 1, characterised in that one or more power sources is also associated with each item of equipment by means of said second set of constraints.

3. Computer-assisted method for optimisation, according to claim 1 or 2, characterised in that prior to minimisation of the cost functions, a third set of constraints is envisaged, formulated as linear inequalities of decision variables, allowing instances of functional links to be co-located or segregated by means of sets of network paths that are, respectively, identical and disjoint.

4. Computer-assisted method for optimisation, according to claim 3, characterised in that the third set of constraints allows software module instances to be co-located or segregated using sets of equipment that are, respectively, identical and disjoint.

5. Computer-assisted method for optimisation, according to claim 3 or 4, characterised in that the third set of constraints allows software module instances to be required to be hosted by equipment supplied by power sources that are identical or distinct.

6. Computer-assisted method for optimisation, according to any of the previous claims, characterised in that prior to minimisation of the cost functions, a fourth set of constraints may also be envisaged, formulated as linear inequalities of decision variables, which can be used beforehand, to require that an instance of a functional link passes through a given network path, or to require that an instance of a software module be hosted by a given item of equipment.

7. Computer-assisted method for optimisation, according to any of the previous claims, characterised in that the cost functions are minimised conjointly over a set of possible values of said decision variables.

8. Computer-assisted method for optimisation, according to any of previous claims, characterised in that the cost functions are minimised successively, with a first cost function being minimised over the set of possible values of said decision variables and where each of the other cost functions are minimised over the sub-set of said values obtained through the previous cost function minimisation.

9. Computer-assisted method for optimisation, according to claims 1, 3 and 6, characterised in that at least one cost function includes an additional decision variable which does not occur in the set of constraints in the first, second, third or fourth sets of constraints, and that a fifth set of constraints in the form of an inequality, which is linearly dependent on at least said additional decision variable, is envisaged.

10. Computer-assisted method for optimisation, according to any of the previous claims, characterised in that said cost functions belong to the following list.

number of items of equipment connected to several communication networks;

weighted combination of the number of communication nodes of various types;

maximum number of communication node connections;

sum of the real distances between the items of equipment and the communications nodes to which they are connected;

number of power sources supplying the various items of equipment;

sum of the maximum levels of utilisation of equipment resources by the software module instances;

number of paths associated with a given functional link;

topological length of network paths associated with a given functional link;

number of communication nodes common to the network paths associated with a given functional link;

total number of communication nodes used by the network paths associated with a given functional link.

11. Method for the creation of an avionics platform, characterised in that the platform architecture is optimised by means of the optimisation method according to one of the previous claims, and that

terminals equipped with the resources determined by the optimum solution are envisaged;

the software module instances are stored in these terminals.

12. Method for creating an avionics platform, according to claim 11, characterised in that:

network and physical links, either directly between terminals or between terminals and network nodes, are envisaged;

routing/commutation tables are stored in said nodes in order to implement the network paths determined by said optimum solution.