Question: Parameters in dsolve as variables for another Equation

Hello everyone.

I have 2 systems of differential equations which depend on variables from each other, of course the differential variable for each is different. 

Not to get into the details of the system, since they are rather large, the problem is as follows:

I've managed to get with the parameters option a solution for x(z), in essence x(parameters=[a,b])(z). I've managed in turn to generate a procedure that would give me the avg value of x (xm),which I need for the other system, over a closed range of z. Keep in mind a and b are variables to the other system and not to the x(z) system, that's why I thought parameters was a good fit.

All that I need for my other system thus is xm(a,b), to be used it in the following sense: diff(a,t)=xm(a). However since the parameters option requires these parameters to be values I am unable to use it in the way I hope I could (as variables instead of values).

Is there anyway of getting around this?

I've considered doing a regresion for the different values of xm(a,b) but since there are several values to be tweaked in the system it would be rather tedious solution

In advance Thank your for your help!