OPERATE Options

OPERATE

The OPERATE menu shown in the figure (below). If the mouse is left hovering over an option a short description of the function will appear.

For these functions, the user selects a range of curves to be operated on. A range may be one or more curves, making it possible to operate on multiple curves, for example add 20 curves to 20 curves.

The options with the OPERATE menu are split into 3 groups. The first group require 2 sets of curves as input. The second group require a single set of curves as input. The third group also require a single set of curves as input but the output from these functions is a single curve. (See Curve Operations for more information on curve groups).

ABS	Produces the absolute y-values of a curve.
ADD (y)	Add the y axis values together for two curves or add a constant value to all the y-values. If two curves are being added together they must have identical x-axis values. If not, the resultant curve is generated by considering every x-coordinate on both curves and by interpolating the other curve as needed. Any duplicate points as well as points outside the range where the input curves overlap are culled as needed.
ADD (x)	Add the x axis values together for two curves or add a constant value to all the x-values. If two curves are being added together they must have identical y-axis values. If not, the nth x-value from the second curve is simply added to the corresponding x-value on the first curve and the y-value from the first curve gets used by the resultant as is.
AVE	Produces a single curve that is the average of the input curves.
CAT	Concatenate the second curve to the end of the first.
CLIP	Clip a curve to remove any points that exceed a set of specified minimum and maximum x & y axis value. The user is prompted for minimum and maximum values after the curves have been selected.. Instead of typing in values for the limits individual x and y axis minimum and maximum values can be selected by picking screen points. In addition to picking individual points an area can be dragged out interactively to set all 4 limits. When picking screen points the default is to allow any point to be selected. Snap to curve points can be used to select the point on the nearest curve instead of the screen coordinates.
COM	Two curves are combined to give a new curve. For example if a displacement/time curve is combined with a velocity/time curve a velocity/displacement curve will result. If the 2 curves do not contain points at the same x values then the curve with the larger x-axis intervals is automatically mapped on to the x-axis values of the other curve. If the curves do not start and finish at the same x-axis values then only the points for which the two curve x-axes overlap are mapped onto each other.
DIF	A curve is differentiated with respect to the x-axis variable.
DIV (y)	Divide the y axis values of the first curve by the y axis values of the second curve (or a constant). If two curves are being used they must have identical x-axis values. If not, the resultant curve is generated by considering every x-coordinate on both curves and by interpolating the other curve as needed. Any duplicate points as well as points outside the range where the input curves overlap are culled as needed.
DIV (x)	Divide the x axis values of the first curve by the x axis values of the second curve (or a constant). If two curves are being used they must have identical y-axis values. If not, the nth x-value on the first curve is simply divided by the corresponding x-value on the second curve and the y-value from the first curve gets used by the resultant as is.
ENV	Produces a single curve that bounds the maximum and minimum values of the group of input curves.
ERR	This option reports the degree of correlation between 2 input curves. The first curve selected is used as a reference curve and the following parameters are then reported : Maximum difference : Value & Time Value as a %age of reference curve Value as a %age of reference curve peak value. Average difference - Value %age of reference curve peak value Area Weighted Difference Correlation Parameter - 0 to 1 For more details on this function please see Appendix G.
INT	A curve is numerically integrated with respect to the x-axis variable using Simpson's rule.
LSQ	Fits a straight line through the points using the least squares method.
MAP	The second curve is mapped onto the first curve, the resulting curve has identical x-axis values to the reference (first) curve with y-axis values obtained from the mapped (second) curve.
MAX	Produces a single curve that bounds the maximum values of the group of input curves.
MIN	Produces a single curve that bounds the minimum values of the group of input curves.
MON	Sorts a curve into monotonically increasing x-axis values.
MUL (y)	Multiply the y axis values together for two curves or multiply all the y-values by a constant. If two curves are being multiplied together they must have identical x-axis values. If not, the resultant curve is generated by considering every x-coordinate on both curves and by interpolating the other curve as needed. Any duplicate points as well as points outside the range where the input curves overlap are culled as needed.
MUL (x)	Multiply the x axis values together for two curves or multiply all the x-values by a constant. If two curves are being multiplied together they must have identical y-axis values. If not, the nth x-values on the two curves are simply multiplied together and the y-value from the first curve gets used by the resultant as is.
NOR (y)	Normalise a curve so that the y axis values lie in the range [-1, +1]. If the manual normalising range is checked then a custom value can be chosen to normalise with. The custom value chosen for normalising is calculated by taking the absolute maximum of the user-defined textbox values. In addition to this is the option to Lock to the specified Min and Max axis values which can be used for normalising the curve to the axis value.
NOR (x)	Normalise a curve so that the x axis values lie in the range [-1, +1]. If the manual normalising range is checked then a custom value can be chosen to normalise with. The custom value chosen for normalising is calculated by taking the absolute maximum of the user-defined textbox values. In addition to this is the option to Lock to the specified Min and Max axis values which can be used for normalising the curve to the axis value.
ORDER	Reverse the order of all the points in the curve.
REC	Produces the reciprocal of the y-values of a curve.
RES	Calculate the vector magnitude from a group of input curves.
REV	Reverses the x and y axes of a curve. For example if you start with a curve with displacement (y axis) against time (x axis) you end up with a curve of time (y axis) against displacement (x axis).
R-AVE	Produces a single curve of the running average on the input curve. If the time window is set to 0 then the y values for the output curve are the average value of all the point up to that point. If the time window is non-zero (T) then the y values at each point are calculated by averaging the values between -T/2 and +T/2.
SMO	A moving average technique is used to smooth (filter) a curve. The user will be prompted for a smoothing factor.  The integer refers to the number of points included in the averaging of each point. The value you want will depend on the number of points in the curve and the amount of smoothing required. A certain amount of trial and error is necessary to get the required result.
SQR	Take the square root of the y-values of a curve.
STRESS	Converts a stress / strain curve between True and Engineering Stress /Strain.
SUB (y)	Subtract the y axis value (or constant) of the second curve from the first curve. If two curves are being subtracted they must have identical x-axis values. If not, the resultant curve is generated by considering every x-coordinate on both curves and by interpolating the other curve the other curve as needed. Any duplicate points as well as points outside the range where the input curves overlap are culled as needed.
SUB (x)	Subtract the x axis value (or constant) of the second curve from the first curve. If two curves are being subtracted they must have identical y-axis values. If not, the nth x-value from the second curve is simply subtracted from the corresponding x-value on the first curve and the y-value from the first curve gets used by the resultant as is.
SUM	Calculates the sum of a group of curves. This "sums" up the y-axis values of a group of curves, and maps the result onto the x-axis of the first curve.
TRA	Translate a curve with respect to the x and y axes. The user is prompted for the x and y values.
VEC	Calculate the vector magnitude from three input curves.
VEC(2D)	Calculate the vector magnitude from two input curves.
WINDOW	This function is typically used to modify a curve before carrying out an FFT on it. The y-axis values for each of the input curves is multiplied by a factor between 0 and 1. Three different window shapes are available. The Store Window Curve option can be used to output the multiplying factors to a separate curve if required.
ZERO	Translate a curve so that the first data point is moved to (0,0). By default this option will translate the curve in both X and Y, alternatively the curve can be translated in X only or Y only.
dB	Converts a curve to dB. f(x) = 20log(y/ref)
dBA	Converts a curve from dB to dBA by applying "A" weighting factors to the curve values. Narrow band A weighting values are calculated using a formula. 1/3 Octave A weighting values are calculated from a lookup table.
Octave	Converts a curve from narrow band to either Octave bands or 1/3 rd Octave bands. The input curve can either be a curve that has already been converted to dB or it can be an unconverted "linear" curve. The output curve can also be generated using either Mean values or RMS values.
Regres	Fits data with either a linear, polynomial (degree 1-4), logarithmic or exponential regression curve, using least squares. The equation of the regression curve, as well as the value of Pearson's correlation coefficient, R2, can be found by right-clicking the output curve and selecting properties. The value of R2 gives a measure of the goodness of fit of the regression curve, with a value close to 1 corresponding to a good fit. In the case of linear regression, the standard deviation of the gradient, intercept and y values are also provided in the properties pop-up accessed through right-clicking the curve. Additionally, 95% confidence and prediction bands can be displayed around the linear regression curve by selecting Properties in the top-right Tools panel, then selecting All Curves and ticking Show CBands next to the output regression curve. The confidence band, which is the inner of the two bands, gives a 95% confidence interval at each x value for the best value of y. The prediction band gives a 95% confidence interval at each x value for predicting a new value of y.