Actuator updates #161
Actuator updates #161
Conversation
|
@galbramc these test failures seem like mosek issues on some of the VMs |
|
They sure do! Weird, I thought we just fixed those last month. |
| static.Qmax >= Q, | ||
| v <= static.V_max, | ||
| TCS([i >= Q*Kv+i0]), | ||
| TCS([v >= omega/Kv + i*R]) |
There was a problem hiding this comment.
Are you worried about these not being tight?
There was a problem hiding this comment.
I was at one point. It depends on how the model is being used. Is there a high cost to having that check?
There was a problem hiding this comment.
No, but in general I think we should only have TCS when there's a possible use case where it shouldn't be tight. Let's leave it for now.
|
|
||
| Variables | ||
| --------- | ||
| Qstar .5 [kg/(N*m)] motor specific torque |
|
|
||
| Variables | ||
| --------- | ||
| V_static 1 [cm/s] inflow velocity for static thrust |
There was a problem hiding this comment.
Is this the right way to do static? Should we ask Mark about this
There was a problem hiding this comment.
I believe it is but it's not actually used so I'm deleting it.
| def actuator_propulsor_test(): | ||
|
|
||
| test = Actuator_Propulsor_Test() | ||
| #sol = test.debug() |
There was a problem hiding this comment.
Can you delete all of these commented lines?
| #sol = test.debug() | ||
| #print sol.table() | ||
|
|
||
| def test(): |
There was a problem hiding this comment.
are you not going to test the individual motors here? I really like that idea
| return 2. - 1./c[self.etaadd] | ||
|
|
||
| def setup(self, static, state): | ||
| exec parse_variables(Propeller_Performance.__doc__) | ||
| def setup(self, static, state = None, rho = None, V = None): |
There was a problem hiding this comment.
Are rho and V necessary? Is = None necessary?
| lam >= V/(omega*R), | ||
| CT >= Tc*lam**2, | ||
| CP <= Q*omega/(.5*rho*(omega*R)**3*pi*R**2), | ||
| eta >= CT*lam/CP, |
There was a problem hiding this comment.
No; eta is determined by eta_v and eta_i and then used to constrain the ratio of output to input power (CT lam)/(CP)
There was a problem hiding this comment.
Oh right. It was over constrained in the other one.
| i = arange(0.0001,1,.001) | ||
| j = arccos(exp(-i)) | ||
| #P = Vdd**2 + 30*Vdd*exp(-(Vth-0.06*Vdd)/0.039) | ||
| #u = vstack((Vdd,Vth)) |
| y = log(j) | ||
| K = 1 | ||
|
|
||
| cstrt, rmsErr = fit(x,y,K,"SMA") |
There was a problem hiding this comment.
This gets used in the blade element model; could be deleted from this PR
|
@courtin looking good. Mostly minor changes. |
|
@mjburton11 should be fixed now |
|
sorry I'm working on upgrading the VM's and hashing out some of the hickups. |
|
test this please |
|
test models please |
1 similar comment
|
test models please |
|
looks like jenkins is back in business, and all the tests are passing for this PR. |
|
|
||
| test = Actuator_Propulsor_Test() | ||
| sol = test.solve() | ||
| #print sol.table() |
There was a problem hiding this comment.
let's delete all of print sol.table(). And just have test.solve()
| lam >= V/(omega*R), | ||
| CT >= Tc*lam**2, | ||
| CP <= Q*omega/(.5*rho*(omega*R)**3*pi*R**2), | ||
| eta >= CT*lam/CP, |
There was a problem hiding this comment.
Oh right. It was over constrained in the other one.
|
done |
|
test this please |
|
test models please |
1 similar comment
|
test models please |
|
@galbramc still something strange going on with the tests for this PR - it looks like one if being marked as failed despite all the tests passing. |
|
You have to go to console output to interpret those...one timed out (at 90 minutes! what kind of processor prioritization is going on on that machine anyway :p) while the other two lost internet access for a bit. |
|
test models please |
|
I was also missing a pip install cvxopt that I just added back in. Turns out that cvxopt accidentally was installed on the system for windows |
|
test models please |
|
finally. one of the VMs was busy all day. |
|
woo! |
Updates to the actuator disk propeller and electric motor models for integration into the solar model.