eta_s for turbocompressor as UDE(v)

[JanKochanski]

Hi!
I have a compressor-map, that is based in the volume-flow. If I understand correctly, the char maps only support mass flow. My idea was to bypass this through a user defined equation, so that $\eta_s = f(\dot{v})$. Do you have any ideas on that, or are there better ways to deal with this situation?
Thanks in advance!
Best regards, Jan

[fwitte]

Hi @JanKochanski,

if it is only $\eta_s=f\left(\dot V\right)$ then that is quite simple:

from tespy.tools.fluid_properties import isentropic


def calculate_eta_s(vol_flow):
    # eta_s map lookup method
    return 0.8


def eta_s_ude(ude):
    c1, c2 = ude.conns
    vol_flow = c1.calc_vol() * c1.m.val_SI
    eta_s = calculate_eta_s(vol_flow)
    return (
        (c2.h.val_SI - c1.h.val_SI) * eta_s - (
            isentropic(
                c1.p.val_SI,
                c1.h.val_SI,
                c2.p.val_SI,
                c1.fluid_data,
                c1.mixing_rule,
            ) - c1.h.val_SI
        )
    )


def eta_s_ude_dependents(ude):
    c1, c2 = ude.conns
    return [c1.m, c1.p, c1.h, c2.p, c2.h]    

If you need frequency in there as well, you could have a look at the implementation of the efficiency map in the Pump class:

tespy/src/tespy/components/turbomachinery/pump.py

Lines 614 to 638 in 90f3ba7

	def eta_flow_frequency_group_func(self):
	r"""
	Equation for efficiency :math:`\eta` over flow and frequency
	:math:`\omega` map
	Returns
	-------
	float
	Residual value of equation.
	.. math::
	0 = \eta \left(\omega,\dot{V}_\text{in}\cdot v_\text{in} \right)
	\cdot \left(h_\text{out} - h_\text{in}\right)
	- v_\text{in}\cdot\left(p_\text{out} - p_\text{in}\right)
	"""
	i = self.inl[0]
	o = self.outl[0]
	vol = i.calc_vol()
	flow = i.m.val_SI * vol
	frequency = self.frequency.val_SI
	eta = self.eta_flow_map.char_func.evaluate(frequency, flow)
	return (
	eta * (o.h.val_SI - i.h.val_SI) - vol * (o.p.val_SI - i.p.val_SI)
	)

. It would be quite easy to just copy that implementation, I believe :).

Best

[fwitte]

Note, the pump approximates the isentropic process with $\dot V \cdot \int dp$, which is possible for incompressible media, while it is $\int \dot V\left(p\right) dp$ for gases

[JanKochanski]

Hi @fwitte,
thanks for the quick reply! Actually, $\eta_s = f(\dot{v}, PR)$, since I need both axis of the compressor map to locate the correct point. But I guess, since your code returns the residual of $\eta_s$, I could just provide PR to the lookup function as well.
Thank you for your help.
Best regards, Jan

from tespy.tools.fluid_properties import isentropic


def calculate_eta_s(vol_flow, pr):
    # eta_s map lookup method
    return 0.8


def eta_s_ude(ude):
    c1, c2 = ude.conns
    vol_flow = c1.calc_vol() * c1.m.val_SI
    pr = c2.p.val_SI / c1.p.val_SI
    eta_s = calculate_eta_s(vol_flow, pr)
    return (
        (c2.h.val_SI - c1.h.val_SI) * eta_s - (
            isentropic(
                c1.p.val_SI,
                c1.h.val_SI,
                c2.p.val_SI,
                c1.fluid_data,
                c1.mixing_rule,
            ) - c1.h.val_SI
        )
    )


def eta_s_ude_dependents(ude):
    c1, c2 = ude.conns
    return [c1.m, c1.p, c1.h, c2.p, c2.h] 

[fwitte]

Yes that should work!