About script:
In most cases, the basic simulation modules available within OlympIOs offer sufficient capabilities for optimization of the relevant component parameters. However, in some cases lengthy and complex optimization procedures require more flexibility. The script module offers this increased optimization flexibility.

Script module benefits:
The script module allows for batch processing of many sequenential simulation job. The scripting language provides direct access to most of the internal capabilities of the software and includes conditional statements according to its c-like syntax. Design engineers can now automatically change devices based on the results of previous simulations to create optimum design solutions that are practically impossible to achieve using the standard one, two variables at a time optimization strategy.

Features:

• Batch processing of simulation sequences
• Compatibility with all other OlympIOs modules
• Direct access to simulation results via a variety of built-in functions
• Flexible optimization control (vary parameters with conditional statements)<

Script detailed examples
The script language gives access to large portions of OlympIOs. This can be used for a variety of applications, but the section below will focus on a few main tasks which are often applied. A few topics for more complex applications are also addressed.

What is a script?
A script within OlympIOs works comparable to other scripting languages. A series of commands is stored in a text file and can be used later to perform the same actions again. If those actions are defined correctly, you can also apply them on other designs and thus have a general tool. The OlympIOs scripts look a lot like C, so if you are familiar with this language this will help a lot in using scripts. The focus of a script is thus to automate things which you use often or are too repeatative for human interaction. The OlympIOs scripts give access to many parts of the program and using this you can customize it. The following paragraphs show potential scripts covering a wide range of applications.

Menu actions
The script language gives access to most of the menues by using the special script functions. A few examples script will be given which show the different menues. The examples are not general, since this would increase the size of the example without adding usefull information.

Open a file, start a simulation and save results
A typical application is to run a batch of simulations, which run during the night. To reduce on the required amount of memory, it is easier to start those things after each other then at the same time on a machine. The following script shows this, for 2 files.

// The first run
file::open("/batch/file1.dev");
sim::bpm();
result::saveas("/batch/file1.ica");
// And the second
file::open("/batch/file2.dev");
sim::bpm();
result::saveas("/batch/file2.ica");

var i; // An internal script variable
series (i in [0,1,3,7]) {
A=i+2;B=10*i-7;C=cos(i); // Update our structure based on i
sim::bpm();
}

// Use EIM method
sim::mode(); // Start a mode calculation
sim::fieldDim(); // and run a field dimension fit
sim::modeopt(opt_method='bend'); /* And change the mode solver */
sim::mode();
sim::fieldDim();

f=open(file,"result.txt"); // Open the required file
fprintf(f,"i\tA\tout1\tout2\tout1/out2\n");
for(i=0;i<10;i++) {
A=5+i*3; // update structure
result::clear();
sim::bpm();
var o1=res_Plane_Overlap(0,"out1",0),o2=res_Plane_Overlap(0,"out2",0);
fprintf(f,"%g\t%g\t%g\t%g\t%g\n",i,A,o1,o2,o1/o2);
}
close(f);

var i,cnt=0,aa[double,100],o1[double,100],o2[double,100];
for(i=0;i<10;i++) {
aa[cnt]=A=5+i*3; // update structure
result::clear();
sim::bpm();
o1[cnt]=res_Plane_Overlap(0,"out1",0);
o2[cnt]=res_Plane_Overlap(0,"out2",0);
cnt++;
}
var f;
f=open(file,"result.txt"); // Open the required file
fprintf(f,"A");
for(i=0;i<cnt;i++) fprintf(f,"\t%g",aa[i]);
fprintf(f,"\no1");
for(i=0;i<cnt;i++) fprintf(f,"\t%g",o1[i]);
fprintf(f,"\no2");
for(i=0;i<cnt;i++) fprintf(f,"\t%g",o2[i]);
fprintf(f,"\n");
close(f);

var i,j,cnt,res[double,10][4];
for(i=0;i<10;i++) {
cnt=0;
series (j in [0,2,6,7]) {
A=5+i*3; // update structure
result::clear();
sim::bpm();
var o1=res_Plane_Overlap(0,"out1",0),o2=res_Plane_Overlap(0,"out2",0);
res[i][cnt++]=abs(o1-o2)/(o1+o2);
}
}
var f;f=open(file,"result.txt");
for(i=0;i<10;i++) {
for(j=0;j<cnt;j++)
fprintf(f,"%g\t",res[i][j]);
fprintf(f,"\n");
}
close(f);

outmax=find (a in [0,100] max(res)) {
if(a<0) { // invalid range
res=0;
continue;
}
result::clear();
A=a;sim::bpm();
res=res_Plane_Overlap(0,"out",0);
}
printf("Maximal value %g for A=%g\n",outmax,a);

outmax=find (a in [2,100] max(res)) {
result::clear();
// first simulation, normal waveguide width
B=a; sim::bpm();r1=res_Plane_Overlap(0,"out",0);
// first simulation, expanded waveguide width
B=a*1.001;sim::bpm();r2=res_Plane_Overlap(0,"out",0);
res=abs(r1-r2);
}
printf("Maximal value %g for A=%g\n",outmax,a);

var nchan=Nout,a,b,c,res,oi[nchan];
find (a in [10,1],
b in [1,0.1], c in [5,2]
max(res)) {
if(a<2 || a>50 || b<-5 || b>100) { // invalid range
res=0;
continue;
}
result::clear();
A=a;B=b;C=c;sim::bpm();
var i,sum=0,maxdif=0,avg;
for(i=0;i<nchan;i++) {
oi[i]=res_Plane_Overlap(0,"out"+i,0);
sum=sum+oi[i];
}
avg=sum/nchan;
for(i=0;i<nchan;i++) {
var d=abs(avg-oi[i]);
if(d>maxdif) maxdif=d;
}
// The response function. We appreciate variation in power less then total power
res=1/(maxdif+1E-3)*sum;
}

var vars[10],i
for(i=0;i<10;i++) vars[i]=i;

function write_dev_file(filename) {
var f;f=open(file,filename);
fprintf(f,"// DEV file from script\n");
fprintf(f,"include \"material.dev\";\n");
var i;
for(i=0;i<10;i++)
fprintf(f," variable(Name=\"A%d\",Value=%g);\n",i,vars[i]);
fprintf(f,"include \"structure.dev\";\n");
}
function genetic() {
// update vars[0..10] using a genetic algoritm
}
// Just run a 1000 iterations
for(i=0;i<1000) {
write_dev_file("/tmp/generated.dev");
file::open("/tmp/generated.dev");
sim::bpm();
// get simulation results, see other examples
getresults();
genetic();
}