v.3.1.0

[ChRen95]

What's Changed

• Fixed Bug in flatten() method of PassSequence by @ChRen95 in Fixed Bug in flatten() method of PassSequence #338
• Added Engine Class by @ChRen95

Full Changelog: v3.0.5...v.3.1.0

---

This discussion was created from the release v.3.1.0.

[ChRen95]

With this new release, pyroll introduces a new Engine class.
This can be incorporated into the RollPass as follows:

RollPass(
        label="Oval I",
        roll=Roll(
            groove=CircularOvalGroove(depth=8e-3, r1=6e-3, r2=40e-3),
            nominal_radius=160e-3,
            neutral_point=-20e-3,
        ),
        engine=Engine(
            rotational_frequency=100,
            gear_ratio=2,
            maximum_power=500e3,
            spindle_efficiency = 0.98
            gear_box_efficiency = 0.95,
            idle_torque=0
        ),
        gap=2e-3,
    )

The spindle_efficiency as well as gear_box_efficiency rate how much power is passed from the engine to the roll.
The ìdle_torque represents the torque, which is needed to drive the roll pass without rolling.
When the rotational_frequency as well as gear_ratio is given in the RollPass, the rotational_frequency of the Roll is calculated.
The implementation was made with respect to also support backward compability.

[sag-kalibreur]

More problematic is the calculation of the important variable "Power Mean", since I believe that PyRolL does not calculate the time during which the roll/motor is under load (length of the rod)?

Here, a simplified formula could also be used, assuming that the torque and speed are constant over time or known as average values.

P avg = M avg ⋅ ω avg

However, this is rather inaccurate.

[ChRen95]

Well, when given an initial length, this calculation can be easily added.
So you mean the gap time (in our case transport time) as well as the roll time?

[ChRen95]

More problematic is the calculation of the important variable "Power Mean", since I believe that PyRolL does not calculate the time during which the roll/motor is under load (length of the rod)?

Here, a simplified formula could also be used, assuming that the torque and speed are constant over time or known as average values.

P avg = M avg ⋅ ω avg

However, this is rather inaccurate.

Okay, I just checked. All required hooks are already there.
The biggest problem will be, to combine this values for e.g. a finishing block were one classically only has one engine.
Maybe we will have to add some sort of intermediate grouping layer in between. Like a special kind of ``PassSequence`.

[sag-kalibreur]

Well, when given an initial length, this calculation can be easily added.
So you mean the gap time (in our case transport time) as well as the roll time?

Perfect, since we don't have any torque or power curves for a rolling pass, we would assume that the power is constant over time. Or do we have different torques/powers for a rolling pass in the calculations? Otherwise we can calculate both the time and the transport/gap...

[sag-kalibreur]

Okay, I just checked. All required hooks are already there.
The biggest problem will be, to combine this values for e.g. a finishing block were one classically only has one engine.
Maybe we will have to add some sort of intermediate grouping layer in between. Like a special kind of ``PassSequence`.

Of course, for a finishing block with only a single engine, the challenge is then to combine the power of several rolling stands and relate this to the engine. Do you think this is possible using a special type of pass sequence?