IDAS sensitivity yields absurd results

[gerrer@…]

I run forward sensitivity analysis using IDA (​https://openmodelica.org/doc/OpenModelicaUsersGuide/latest/parametersensitivity.html) on the following model:

model basic
parameter Real mu = 2;
parameter Real lambda = -1;
Real y(start=2, fixed=true);
equation
der(y) = mu*y + lambda;
end basic;

Analytic solution of this model is :
y(t) = (mu + lambda/mu)*exp(mu*t) - lambda/mu.

Analytic sensitivities are thus:
s_lambda(t) = (exp(mu*t) - 1/mu (--> increases with time)
s_mu(t) = (big nonzero thing)*exp(mu*t) + ctt (--> increases with time)

But sensitivity computation yields:
$Sensitivities.lambda.y = [0.0, ..., 0.0]
$Sensitivities.mu.y = [0.0, ..., 0.0].

[gerrer@…]

the model employed and corresponding result file

[Francesco Casella]

Rescheduled to 1.18.0

[Francesco Casella]

Ticket retargeted after milestone closed

[Francesco Casella]

1.18.0 blocker tickets moved to 1.19.0