OpenHPL provides several turbine models. The Francis component is especially useful for frequency-control studies because it explicitly couples hydraulic flow, pressure, guide-vane geometry, runner kinematics, shaft speed and losses.
The turbine receives hydraulic inlet/outlet conditions through physical connectors, a guide-vane control signal, and generator angular velocity. It produces shaft power and hydraulic outflow.
Mass flow and volume flow are related by
$$Q=\frac{\dot m}{\rho},$$or, when compressibility is enabled, by a pressure-dependent density. The turbine pressure drop is determined from the runner and optional guide-vane losses.
The key mechanical conversion is based on the Euler turbine equation. OpenHPL computes two power terms associated with inlet and outlet velocity triangles. In compact notation,
$$P_{Euler}=\dot m\,\omega\left(R_1c_{u1}-R_2c_{u2}\right).$$In the implementation these terms are constructed from runner radius, angular speed, volume flow, cross-sectional areas and blade/guide-vane angles.
The model separates shock, whirl and wall-friction losses:
$$P_{loss}=P_{shock}+P_{whirl}+P_{friction}.$$A simplified shaft-power expression is therefore
$$P_m=P_{Euler}-P_{loss}.$$At low load, OpenHPL switches to a stronger loss term to regularize and balance the turbine model.
The guide-vane command does not merely multiply flow. The servo position is converted through linkage geometry into guide-vane angle $\alpha_1$, which changes the inlet velocity triangle and therefore both flow and torque. Symbolically,
$$u_t\to Y\to\theta\to\alpha_1\to c_{u1}\to P_m.$$This is one reason the nonlinear OpenHPL turbine is richer than the textbook relation $P_m=\eta\rho gQH$.
A control-oriented approximation around an operating point $(Q^\star,H^\star,\omega^\star,y_g^\star)$ is
$$\Delta P_m\approx K_Q\Delta Q+K_H\Delta H+K_\omega\Delta\omega+K_y\Delta y_g,$$where the coefficients are Jacobian terms evaluated from the nonlinear turbine model.
The shaft torque is
$$T_m=\frac{P_m}{\omega},$$so the turbine is the constitutive bridge between the hydraulic DAE and the generator rotational dynamics.
The FCR loop acts primarily through $\Delta y_g$. Because the turbine relation depends on hydraulic states and speed, the effective gain
$$\frac{\Delta P_m}{\Delta y_g}$$is frequency dependent once coupled to the waterway. This is the transfer function we ultimately want to extract from the OpenHPL operating point.