| US 7,457,735 B2 | ||
| Method and system for automatic water distribution model calibration | ||
| Zheng Yi Wu, Waterbury, Conn. (US); Thomas M. Walski, Nanticoke, Pa. (US); Robert A. Gurrieri, Thomaston, Conn. (US); Gregg A. Herrin, Burlington, Conn. (US); and Robert F. Mankowski, Watertown, Conn. (US) | ||
| Assigned to Bentley Systems, Incorporated, Exton, Pa. (US) | ||
| Filed on Jan. 17, 2002, as Appl. No. 10/51,820. | ||
| Application 10/051820 is a continuation of application No. 09/990818, filed on Nov. 14, 2001. | ||
| Prior Publication US 2003/0093236 A1, May 15, 2003 | ||
| Int. Cl. G06G 7/48 (2006.01); G06G 7/50 (2006.01); G01D 18/00 (2006.01); G01D 21/00 (2006.01); G01M 19/00 (2006.01); G01P 21/00 (2006.01); G01R 35/00 (2006.01); G06F 19/00 (2006.01) | ||
| U.S. Cl. 703—9 [703/10; 702/85] | 21 Claims |
| 1. A method of automatically calibrating a water distribution model of a water distribution network, comprising the steps
of:
(A) selecting calibration parameters to be adjusted to calibrate the water distribution model, the calibration parameters
including link status and one or more of, pipe roughness and junction demand;
(B) collecting field observed data including a pipe flow measurement and a junction pressure measurement for at least one
point in the water distribution network, and including corresponding loading conditions and boundary conditions describing
settings of valves or pumps that existed in the network when said field observed data was collected and passing such information
to a genetic algorithm module;
(C) generating at said genetic algorithm module a population of calibration solutions that comprise a set of calibration results,
using a genetic algorithm;
(D) running multiple hydraulic simulations of each solution to obtain a set of predictions of pipe flows and junction pressures
at selected points in the network, corresponding to the loading conditions and associated boundary conditions when the field
observed data was collected;
(E) performing a calibration evaluation including computing a goodness-of-fit value for each calibration solution based upon
differences between field observed values and said predictions;
(F) repeating steps (C) through (E) until a user-selected desired goodness-of-fit value is obtained resulting in a corresponding
calibration solution for calibrating a water distribution model; and
(G) building a calibrated water distribution model using the desired set of calibration parameters.
|