1 | /** |
---|
2 | This software was developed by the University of Tennessee as part of the |
---|
3 | Distributed Data Analysis of Neutron Scattering Experiments (DANSE) |
---|
4 | project funded by the US National Science Foundation. |
---|
5 | |
---|
6 | If you use DANSE applications to do scientific research that leads to |
---|
7 | publication, we ask that you acknowledge the use of the software with the |
---|
8 | following sentence: |
---|
9 | |
---|
10 | "This work benefited from DANSE software developed under NSF award DMR-0520547." |
---|
11 | |
---|
12 | copyright 2008, University of Tennessee |
---|
13 | */ |
---|
14 | |
---|
15 | |
---|
16 | #include <math.h> |
---|
17 | #include "parameters.hh" |
---|
18 | #include <stdio.h> |
---|
19 | #include <stdlib.h> |
---|
20 | using namespace std; |
---|
21 | #include "fuzzysphere.h" |
---|
22 | |
---|
23 | extern "C" { |
---|
24 | #include "libSphere.h" |
---|
25 | } |
---|
26 | |
---|
27 | // scattering from a uniform sphere w/ fuzzy surface |
---|
28 | // Modified from FuzzySpheres in libigor/libSphere.c without polydispersion: JHC |
---|
29 | static double fuzzysphere_kernel(double dp[], double q){ |
---|
30 | double pi,x,xr; |
---|
31 | double radius,sldSph,sldSolv,scale,bkg,delrho,fuzziness,f2,bes,vol,f; //my local names |
---|
32 | |
---|
33 | pi = 4.0*atan(1.0); |
---|
34 | x= q; |
---|
35 | scale = dp[0]; |
---|
36 | radius = dp[1]; |
---|
37 | fuzziness = dp[2]; |
---|
38 | sldSph = dp[3]; |
---|
39 | sldSolv = dp[4]; |
---|
40 | bkg = dp[5]; |
---|
41 | delrho=sldSph-sldSolv; |
---|
42 | |
---|
43 | xr = x*radius; |
---|
44 | //handle xr==0 separately |
---|
45 | if(xr == 0.0){ |
---|
46 | bes = 1.0; |
---|
47 | }else{ |
---|
48 | bes = 3.0*(sin(xr)-xr*cos(xr))/(xr*xr*xr); |
---|
49 | } |
---|
50 | vol = 4.0*pi/3.0*radius*radius*radius; |
---|
51 | f = vol*bes*delrho; // [=] A |
---|
52 | f *= exp(-0.5*fuzziness*fuzziness*x*x); |
---|
53 | // normalize to single particle volume, convert to 1/cm |
---|
54 | f2 = f * f / vol * 1.0e8; // [=] 1/cm |
---|
55 | |
---|
56 | f2 *= scale; |
---|
57 | f2 += bkg; |
---|
58 | |
---|
59 | return(f2); //scale, and add in the background |
---|
60 | } |
---|
61 | |
---|
62 | FuzzySphereModel :: FuzzySphereModel() { |
---|
63 | scale = Parameter(0.01); |
---|
64 | radius = Parameter(60.0, true); |
---|
65 | radius.set_min(0.0); |
---|
66 | fuzziness = Parameter(10.0); |
---|
67 | fuzziness.set_min(0.0); |
---|
68 | sldSph = Parameter(1.0e-6); |
---|
69 | sldSolv = Parameter(3.0e-6); |
---|
70 | background = Parameter(0.001); |
---|
71 | } |
---|
72 | |
---|
73 | /** |
---|
74 | * Function to evaluate 1D scattering function |
---|
75 | * The NIST IGOR library is used for the actual calculation. |
---|
76 | * @param q: q-value |
---|
77 | * @return: function value |
---|
78 | */ |
---|
79 | double FuzzySphereModel :: operator()(double q) { |
---|
80 | double dp[6]; |
---|
81 | |
---|
82 | // Fill parameter array for IGOR library |
---|
83 | // Add the background after averaging |
---|
84 | dp[0] = scale(); |
---|
85 | dp[1] = radius(); |
---|
86 | dp[2] = fuzziness(); |
---|
87 | dp[3] = sldSph(); |
---|
88 | dp[4] = sldSolv(); |
---|
89 | dp[5] = 0.0; |
---|
90 | |
---|
91 | // Get the dispersion points for the radius |
---|
92 | vector<WeightPoint> weights_radius; |
---|
93 | radius.get_weights(weights_radius); |
---|
94 | |
---|
95 | // Get the dispersion points for the fuzziness |
---|
96 | vector<WeightPoint> weights_fuzziness; |
---|
97 | fuzziness.get_weights(weights_fuzziness); |
---|
98 | |
---|
99 | // Perform the computation, with all weight points |
---|
100 | double sum = 0.0; |
---|
101 | double norm = 0.0; |
---|
102 | double norm_vol = 0.0; |
---|
103 | double vol = 0.0; |
---|
104 | |
---|
105 | // Loop over radius weight points |
---|
106 | for(size_t i=0; i<weights_radius.size(); i++) { |
---|
107 | dp[1] = weights_radius[i].value; |
---|
108 | // Loop over fuzziness weight points |
---|
109 | for(size_t j=0; j<weights_fuzziness.size(); j++) { |
---|
110 | dp[2] = weights_fuzziness[j].value; |
---|
111 | |
---|
112 | //Un-normalize SphereForm by volume |
---|
113 | sum += weights_radius[i].weight * weights_fuzziness[j].weight |
---|
114 | * fuzzysphere_kernel(dp, q) * pow(weights_radius[i].value,3); |
---|
115 | //Find average volume : Note the fuzziness has no contribution to the volume |
---|
116 | vol += weights_radius[i].weight |
---|
117 | * pow(weights_radius[i].value,3); |
---|
118 | |
---|
119 | norm += weights_radius[i].weight * weights_fuzziness[j].weight; |
---|
120 | norm_vol += weights_radius[i].weight; |
---|
121 | } |
---|
122 | } |
---|
123 | |
---|
124 | if (vol != 0.0 && norm_vol != 0.0) { |
---|
125 | //Re-normalize by avg volume |
---|
126 | sum = sum/(vol/norm_vol);} |
---|
127 | return sum/norm + background(); |
---|
128 | } |
---|
129 | |
---|
130 | /** |
---|
131 | * Function to evaluate 2D scattering function |
---|
132 | * @param q_x: value of Q along x |
---|
133 | * @param q_y: value of Q along y |
---|
134 | * @return: function value |
---|
135 | */ |
---|
136 | double FuzzySphereModel :: operator()(double qx, double qy) { |
---|
137 | double q = sqrt(qx*qx + qy*qy); |
---|
138 | return (*this).operator()(q); |
---|
139 | } |
---|
140 | |
---|
141 | /** |
---|
142 | * Function to evaluate 2D scattering function |
---|
143 | * @param pars: parameters of the sphere |
---|
144 | * @param q: q-value |
---|
145 | * @param phi: angle phi |
---|
146 | * @return: function value |
---|
147 | */ |
---|
148 | double FuzzySphereModel :: evaluate_rphi(double q, double phi) { |
---|
149 | return (*this).operator()(q); |
---|
150 | } |
---|
151 | |
---|
152 | /** |
---|
153 | * Function to calculate effective radius |
---|
154 | * @return: effective radius value |
---|
155 | */ |
---|
156 | double FuzzySphereModel :: calculate_ER() { |
---|
157 | double rad_out = 0.0; |
---|
158 | |
---|
159 | // Perform the computation, with all weight points |
---|
160 | double sum = 0.0; |
---|
161 | double norm = 0.0; |
---|
162 | |
---|
163 | // Get the dispersion points for the radius |
---|
164 | // No need to consider the fuzziness. |
---|
165 | vector<WeightPoint> weights_radius; |
---|
166 | radius.get_weights(weights_radius); |
---|
167 | // Loop over radius weight points to average the radius value |
---|
168 | for(size_t i=0; i<weights_radius.size(); i++) { |
---|
169 | sum += weights_radius[i].weight |
---|
170 | * weights_radius[i].value; |
---|
171 | norm += weights_radius[i].weight; |
---|
172 | } |
---|
173 | if (norm != 0){ |
---|
174 | //return the averaged value |
---|
175 | rad_out = sum/norm;} |
---|
176 | else{ |
---|
177 | //return normal value |
---|
178 | rad_out = radius();} |
---|
179 | |
---|
180 | return rad_out; |
---|
181 | } |
---|
182 | double FuzzySphereModel :: calculate_VR() { |
---|
183 | return 1.0; |
---|
184 | } |
---|