Payload estimation of weight bearing machinery using multiple model adaptive estimator system and method

1. A method of payload estimation of weight bearing machinery, the method comprising the steps of:

(a) creating a series of Kalman filter approximations to the weight bearing machinery, each approximation representing a system model of the weight bearing machinery having distinct levels of payload;

(b) determining from said series an overall approximation of the current operating parameters of the weight bearing machinery; and,

(c) utilizing the parameters derived from said step (b) to determine an estimated payload of said weight bearing machinery.

2. The method according to claim 1 wherein said weight bearing machinery comprises an electric mining shovel.

3. The method according to claim 1 wherein the inputs to one or more of the Kalman filters include weight bearing machinery drive control signals for controlling movement of the weight bearing machinery or cognate signals leading to direct or indirect measurement of hoist rope and crowd forces.

4. The method according to claim 1 wherein each Kalman filter approximation simulates increasing payloads on said weight bearing machinery.

5. The method according to claim 1 wherein said step (b) further comprises the steps of forming a current model probability for each system model and then forming a weighted sum of the models to determine the overall approximation of the current operating parameters of the weight bearing machinery.

6. The method according to claim 1 wherein said step (b) further comprises the steps of:

identifying one of the system models which most closely approximates the current operating parameters; and

setting the identified model as the overall approximation of the current operating parameters of the weight bearing machinery by assigning the identified model a probability larger than the probability assigned to the other system models.

7. A method of payload estimation of weight bearing machinery, the method including the steps of:

(a) utilizing a plurality of adaptive predictive models for predicting a future state of a series of time varying parameters associated with a rigid body dynamic equation of the machinery;

(b) determining for each model a difference measure between the projected future state and an actual measured state;

(c) utilizing said difference measure to adaptively improve model coefficients of each respective said adaptive predictive model; and,

(d) utilizing each said adaptive predictive model to produce said payload estimation.

8. The method according to claim 7 wherein each said adaptive predictive model is directly derived from a system model.

9. The method according to claim 7 wherein said model coefficients are improved such that variance of said difference measure is minimized in a mean squared sense.

10. The method according to claim 7 wherein each said adaptive predictive model is independent and has a distinct model.

11. The method according to claim 10 wherein said payload estimation is produced by a weighted sum of parameters predicted from each said adaptive predictive model.

12. The method according to claim 7 wherein each said adaptive predictive model is a Kalman filter and Kalman filter techniques are utilized to adaptively improve said model coefficients.

13. A method of payload estimation of weight bearing machinery, the method including the steps of:

utilizing a multi-model approach to build up a statistical inferencing system directed to the operation of the machinery, and

deriving therefrom a payload estimation, the method further including the steps of:

(a) creating a plurality of system models, each model representing the weight bearing machinery having distinct levels of payload;

(b) determining from said plurality of models an overall approximation of the current operating parameters of the weight bearing machinery; and

utilizing said overall approximation to produce said payload estimation.

14. The method according to claim 13 wherein the plurality of system models is approximated by a series of Kalman filters.

15. A system for payload estimation of weight bearing machinery, said system comprising:

one or more input sensors configured to measure input coefficients from said machinery; and

a processor configured to receive said input coefficients from said input sensors; said processor further configured to:

evaluate a plurality of adaptive predictive models for predicting a future state of a series of time varying parameters associated with a rigid body dynamic equation of the machinery,

determine for each model a difference measure between said future state and said measured input coefficients,

utilize said difference measure to adaptively improve model coefficients of each said adaptive predictive model, and

utilize each said adaptive predictive model to produce said payload estimation.

16. The system according to claim 15 wherein said processor is configured to simultaneously evaluate a plurality of adaptive predictive models for predicting said future state of said parameters.

17. The system according to claim 16 wherein each said adaptive predictive model is independent and has a distinct model.

18. The system according to claim 16 wherein said processor is configured to utilize each said adaptive predictive model by computing a weighted sum of parameters predicted from each said adaptive predictive model.

19. The system according to claim 15 wherein each said adaptive predictive model is a Kalman filter and said processor is configured to adaptively improve said filter coefficients using Kalman filter techniques.