The resolution of the Gaussian fused peaks using the simplex algorithm
The simplex algorithm is used to perform curve-resolution of fused peak systems. The gaussian function is chosen as the peak model. The algorithm basically involves moving from a region of poor response to a region of the best response. The movement is controlled by changing the guessed parameter va...
| Main Author: | |
|---|---|
| Format: | Others |
| Published: |
DigitalCommons@Robert W. Woodruff Library, Atlanta University Center
1985
|
| Online Access: | http://digitalcommons.auctr.edu/dissertations/3687 http://digitalcommons.auctr.edu/cgi/viewcontent.cgi?article=5210&context=dissertations |
| Summary: | The simplex algorithm is used to perform curve-resolution of fused peak systems. The gaussian function is chosen as the peak model. The algorithm basically involves moving from a region of poor response to a region of the best response. The movement is controlled by changing the guessed parameter values of the function and using the sum of the squares of the residuals as the calculated response. The resolution is achieved when the deviation of the observed fused peak from the calculated model peak is minimum. The program GAUSSEX which implements the simplex algorithm, was tested with synthesized data as well as experimental data. The fast fourier transform technique was used to do digital filtering if the peak array was noisey. GAU S S EX gives the plot of synthesized data or experimental data, the smoothed data, the fitted data and the components of each spectrum. The results in both synthesized spectra and experimental spectra were good. The deviation between the actual parameters and computed parameters were less than 5% |
|---|