A System and Method for Modelling System Behaviour

• US 20170147722A1
• Filed: 06/29/2015
• Published: 05/25/2017
• Est. Priority Date: 06/30/2014
• Status: Active Grant

• Current Assignee: Evolving Machine Intelligence Pty Ltd.
• Original Assignee: Evolving Machine Intelligence Pty Ltd.
• Inventors: Greenwood, Nigel John Conrad
• Granted Patent: US 10,565,329 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 17/00 Digital computing or data processing equipment or methods, specially adapted for specific functions information retrieval, database structures or file system structures therefor G06F16/00
• CPC Class Codes: G06F 17/13 Differential equations using digital differential analysers G06F7/64
• CPC Class Codes: G06F 2111/06 Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
• CPC Class Codes: G06F 30/00 Computer-aided design [CAD]
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G16B 5/20 Probabilistic models
• CPC Class Codes: G16B 99/00 Subject matter not provided for in other groups of this subclass
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

A method of modelling system behaviour of a physical system, the method including, in one or more electronic processing devices obtaining quantified system data measured for the physical system, the quantified system data being at least partially indicative of the system behaviour for at least a time period, forming at least one population of model units, each model unit including model parameters and at least part of a model, the model parameters being at least partially based on the quantified system data, each model including one or more mathematical equations for modelling system behaviour, for each model unit calculating at least one solution trajectory for at least part of the at least one time period; determining a fitness value based at least in part on the at least one solution trajectory; and, selecting a combination of model units using the fitness values of each model unit, the combination of model units representing a collective model that models the system behaviour.

Claims

1. ) A method of modelling system behaviour of a physical system, the method including, in one or more electronic processing devices:

a) obtaining quantified system data measured for the physical system, the quantified system data being at least partially indicative of the system behaviour for at least a time period;

b) forming at least one population of model units, each model unit including model parameters and at least part of a model, the model parameters being at least partially based on the quantified system data, each model including one or more mathematical equations for modelling system behaviour;

c) for each model unit;

i) calculating at least one solution trajectory for at least part of the at least one time period;

ii) determining a fitness value based at least in part on the at least one solution trajectory; and

,

d) selecting a combination of model units using the fitness values of each model unit, the combination of model units representing a collective model that models the system behaviour.

2. ) A method according to claim 1, wherein the method includes:

a) forming a plurality of populations; and

,

b) at least one of;

i) selecting a combination of model units including model units from at least two populations;

ii) exchanging, between the populations, at least one of;

(1) model parameters;

(2) at least part of a model; and

,

(3) model units;

iii) selectively isolating a number of model units in an isolated population; and

,

iv) selectively reintroducing model units from the isolated population.

5. ) A method according to claim 1, wherein the method includes, modifying at least one of:

a) the model units;

i) iteratively over a number of generations;

ii) by at least one of inheritance, mutation, selection, and crossover;

iii) in accordance with the fitness value of each model unit;

iv) in accordance with speciation constraints;

v) by exchanging or modifying at least part of a chromosome;

b) at least part of the model; and

,

c) the model parameters.

7. ) A method according to claim 1, wherein each model unit includes:

a) at least part of the model;

b) a chromosome indicative of the model parameters and fitness value; and

,

c) details indicative of behaviour of at least some solution trajectories calculated for the model unit.

9. ) A method according to claim 1, wherein the method includes, at least one of:

a) determining the fitness value at least in part using a behaviour of the at least one solution trajectory;

b) determining the fitness of a model unit by comparing the at least one solution trajectory to quantified system data indicative of system behaviour;

c) determining the fitness value at least in part using a texture set representing a geometrical set in state space;

d) determining the fitness value at least in part using a texture set representing a geometrical set in state space, wherein at least one of;

i) the properties of the texture sets are initially defined independently of the chromosomes, the chromosomes including model parameters at least partially based on the quantified system data; and

,

ii) the texture sets are built from mathematical formulations of the interactions between trajectory elements and observed data;

e) in accordance with the behaviour of at least one solution trajectory, at least one of;

i) introducing the texture set into state space;

ii) deleting the texture set from state space;

iii) moving the texture set through state space;

iv) modifying the texture set by at least one of;

(1) enlarging;

(2) compressing;

(3) deforming; and

,

(4) subdividing;

f) determining the fitness value at least in part using fitness criteria based on;

i) qualitative behaviour of solution trajectories in state space;

ii) quantitative behaviour of solution trajectories in state space;

iii) collision of solution trajectories with sets in state space;

iv) capture of solution trajectories by sets in state space;

v) avoidance of sets in state space by solution trajectories;

vi) divergence of solution trajectories in state space; and

,

vii) intersection with a geometrical surface in state space;

g) modifying fitness criteria for successive generations of model units;

h) at least one of selectively activating and deactivating fitness criteria for at least one generation of model units; and

,

i) determining the fitness value at least in part using texturing achieved by at least one of;

i) using one or more solution trajectories achieving one or more discretely-distributed qualitative objectives in state space, defined in discrete terms of the trajectories'"'"' collision, capture or avoidance with, by, or of one or more specified sets in state space;

ii) using one or more solution trajectories exhibiting specified behaviour in respect of qualitative objectives in state space described by a continuously distributed function;

iii) using contours permeating state space and parameter space that are associated with one or more Lyapunov functions and/or the derivatives of Lyapunov functions;

iv) mapping the fitness to a mathematical structure that captures geometrical information with respect of at least one of;

(1) trajectories corresponding with distinct chromosomes, which remain geometrically close or diverge in state space; and

,

(2) the value of a pattern over a geometrical surface or hypersurface, generated by a patterning algorithm and mapped to the complex plane and inserted or mapped into the state space, at the point where it is intersected by the trajectory.

16. ) A method according to claim 9, wherein at least one of:

a) the sets in state space are texture sets;

b) the texture sets undergo changes within generations of a population of model units; and

,

c) the properties of texture sets are initially defined independently of chromosomes including model parameters being at least partially based on the quantified system data.

17. ) A method according to claim 1, wherein the method includes:

a) determining control variable values representing control inputs; and

,

b) determining the solution trajectory in accordance with the control variable values.

18. ) A method according to claim 1, wherein the method includes at least one of:

segmenting quantified system data into respective time period segments and

a) at least one of;

i) determining respective model units for each time period segment; and

,

ii) selecting a combination of model units so as to include model units for each of the time period segments using an optimisation technique; and

,

b) subdividing quantified system data into smaller subsets of types of quantified system data across the same time period segment and determining respective model units for each subset, and then combining these models into a collective model to more accurately model system behaviour across this same time period segment, and wherein the method optionally includes establishing an optimal model for each subset.

21. ) A method according to claim 1, wherein the method includes:

a) selecting a number of model units in accordance with the fitness values;

b) determining a plurality of candidate combinations of model units;

c) determining a combination fitness value for each candidate combination; and

,

d) selecting the combination in accordance with the combination fitness value.

22. ) A method according to claim 1, wherein the method includes:

a) defining a plurality of circuits, each circuit including a sequence of Blocks, each Block representing a respective population of model units;

b) transferring at least one of model units and model parameters between the Blocks in accordance with circuit connections; and

,

c) using the circuit to select at least one of the combination of model units and model parameters.

23. ) A method according to claim 22, wherein the method includes modifying the circuits by modifying model units of the Blocks.

24. ) A method according to claim 1, wherein the method includes:

a) generating a network including a number of nodes;

b) associating parts of a model with a node;

c) for each node on a path through the network;

i) determining at least one model unit using parts of a model associated with the node; and

,

ii) optimising model parameters of the at least one model unit; and

,

d) generating a candidate model based on the model units and model parameters.

25. ) A method according to claim 24, wherein the method includes at least one of:

a) optimising the path to determine the combination;

b) optimising the path by;

i) traversing at least one path;

ii) assessing the resulting candidate combination; and

,

iii) traversing modified paths based on the assessment of the candidate combination;

c) optimising the path using a swarm optimisation technique.

28. ) A method according to claim 22, wherein the method includes:

a) defining a plurality of circuits, each circuit including a sequence of Blocks;

b) for each node in the network populating each Block with parts of models associated with the node;

c) transferring at least one of model units and model parameters between the Blocks in accordance with circuit connections to thereby optimise model parameters of the model units.

29. ) A method according to claim 22, wherein the method includes:

a) transferring a plurality of circuits along respective paths in the network;

b) at each node, calculating a respective circuit fitness; and

,

c) selecting an optimum path using the circuit fitness, the selected path representing the combination.

30. ) A method according to claim 1, wherein the method includes using the model to at least one of:

a) develop a control program including one or more control variable values indicative of control input to the system to modify system behaviour in a desirable manner;

b) derive system data that cannot be physical measured; and

,

c) determine a scenario or likelihood of system behaviour resulting in a failure by comparing at least one of system data and system behaviour to defined limits; and

,

d) manipulating or altering the behaviour of the physical system at least partially by applying the control program.

32. ) A method according to claim 30, wherein the method includes:

a) determining a selected model parameter corresponding to system data that cannot be at least one of directly manipulated and directly measured;

b) comparing the selected model parameter to a respective operating range; and

c) in response to the selected model parameter falling outside the respective range, determining a control input to at least one of;

i) prevent the selected model parameter falling outside the respective range; and

,

ii) restore the selected model parameter to within the range once the excursion has occurred.

33. ) A method according to claim 1, wherein the method includes, using a modular system including at least one of:

a) a Model Selector module for selecting the at least part of a model;

b) an Identifier module for identifying the combination of model units by requesting parts of a model from the Model Selector; and

,

c) a Controller module that uses the combination of model units to determine a control program.

36. ) A method according to claim 1, wherein the method includes:

a) defining meta-model units representing at least one of;

i) different combinations of model units; and

,

ii) different Blocks, each representing a respective population of model units; and

,

b) optimising the meta-model units to determine at least one of;

i) a combination of model units; and

,

ii) an optimal Block.

3. ) (canceled)

4. ) (canceled)

6. ) (canceled)

8. ) (canceled)

10. ) (canceled)

11. ) (canceled)

12. ) (canceled)

13. ) (canceled)

14. ) (canceled)

15. ) (canceled)

19. ) (canceled)

20. ) (canceled)

26. ) (canceled)

27. ) (canceled)

31. ) (canceled)

34. ) (canceled)

35. ) (canceled)

37. ) (canceled)

38. ) A method of modelling system behaviour of a physical system, the method including, in one or more electronic processing devices:

a) generating a network including a number of nodes;

b) associating parts of a model with a node;

c) for each node on a path through the network determining at least one model unit, each model unit including model parameters and at least part of a model, the model parameters being at least partially based on quantified system data, each model including one or more mathematical equations for modelling system behaviour, and the model unit being determined using the parts of the model associated with the node;

d) optimising model parameters of each model unit; and

,

e) generating a candidate combination of model units based on the model units and model parameters, the candidate combination of model units representing a candidate collective model that models the system behaviour.

39-70. -70) (canceled)