VoldKalmanParameters


Purpose

Vold-Kalman filter parameters class.

Syntax

VKP = VoldKalmanParameters(nshafts)

Description

VOLDKALMANPARAMETERS is a class that contains the description of the orders to extract using the Vold-Kalman filter. It defines the order numbers, associated shaft (tach) to use as a reference, order bandwidth, and revolutes.

NSHAFTS is the number of independent shafts/tachometers used for the analysis.

Several parameters need to be set in addition to the input argument. These are as follows:

Field Description
VKP.IsOrderBW Cell array (1xNSHAFTS) of logicals specifying the filter bandwidth type. If TRUE, the bandwidth is a fraction of the order. If FALSE, the bandwith is frequency. Each cell must be either a scalar or a vector the same size as the number of orders for that shaft.
VKP.BW Cell array (1xNSHAFTS) containing the bandwidth values for each order (either order fraction or Hz, depending on IsOrderBW). Each cell must be either a scalar or a vector the same size as the number of orders for that shaft.
VKP.Orders Cell array (1xNSHAFTS) containing the orders to extract for each shaft.
VKP.Revolute Cell array (1xNSHAFTS) containing the revolute (tach position time history) for each shaft.

 

Examples

This example has two shafts. We wish to extract one order from the first shaft (tach1), and two orders from the 2nd shaft (tach2). The order bandwidth is a fraction of the order for tach1, and fixed frequency for tach2.

VKP = VoldKalmanParameters(2);
VKP.IsOrderBW{1}  = true;                 % First shaft bandwidth type
VKP.IsOrderBW{2}  = false;                % Second shaft bandwidth type
VKP.BW{1}         = .05;                  % Bandwidth (/order)
VKP.BW{2}         = 5;                    % Bandwidth (Hz)
VKP.Orders{1}     = 1.23;                 % Shaft 1 extract order 1.23
VKP.Orders{2}     = [1.1 2.6];            % Extract these orders from shaft 2
VKP.Revolute{1}   = tach1.position(t)     % t is a vector of time values
VKP.Revolute{2}   = tach2.position(t)     % t is a vector of time values

See also

IMATVKFilter, VoldKalmanFilter