Solver was unable to reduce the step size without violating the minimum step size value of 2.66454E-10 for 5 consecutive times

I have a Simscape simulation of a PV plant, battery, and electrolyzer. When the electrolyzer operates continuously at 100%, the simulation runs smoothly. However, when I introduce intermittent operation, the simulation encounters an issue when the electrolyzer turns off. The error message:

"Solver was unable to reduce the step size without violating the minimum step size value of 2.66454E-10 for 5 consecutive times at time 75000"

appears repeatedly.

I attempted to increase the number of consecutive minimum step violations, as suggested, but it did not resolve the issue. Notably, when I simulate only the electrolyzer with on/off cycles, the problem does not occur.

[Sam Chak]

Increasing the number of consecutive steps only addresses the symptoms rather than the root cause. If intermittent operation refers to the photovoltaic (PV) system operating by repeatedly starting and stopping in an alternating manner, it can create a situation that destabilizes the system.

For example, you may unknowingly stop a crucial stabilizing action long enough for the error signal to accumulate over time and grow unbounded. When the integration of the system's ODEs reaches a minimum step size value of for 5 consecutive instances, the solver can no longer perform the integration.

[Juan José]

This is the simulation, the system start good, working good distributing the energy for electrolyzer, and the charging and discharging baterry, for 20h time , the electroyzer turn off, and there is the problem...

[Sam Chak]

The storage current drops almost instantaneously from 5000 A to 0 A, which may imply that the issue is caused by this relatively high stiffness. Nevertheless, it is necessary to figure out exactly which part of the PV system is causing the problem.

Since it is suspected that the sudden transition from the 'ON' state to the 'OFF' state in the electrolyzer is the root cause, could you introduce a small delay using a first-order transfer function with a relatively small time constant T?

[Juan José]

thank for your comments,

I remove the PV plant, and working with only the electrolyzer and battery. however the erro remain, so I conclude the problem is on the battery

and

The baterry system has trouble to change for ON state to OFF.

where I should put the small delay using a first-order transfer function ?

[Sam Chak]

I am unfamiliar with Simscape. But I suspect that it may be related to the Stop Simulation block, which terminates the simulation altogether. If you 'comment out' (disable) this Stop Simulation block, does the error still occur?

[Juan José]

this part is only to the charge of battery is low than 30% , the simulation is finish. If I comment out this, nothing change with the current problem

[Sam Chak]

Okay, not the Stop block. How do you turn off the battery's state?

[Juan José]

With the pulse generator of the DC/DC converter, I can control the current required by the electrolyzer, while the battery should respond accordingly.

[Sam Chak]

I see, I overlooked the Pulse Generator block. If you modify the duty cycle to extend the 'ON' state, what happens?

[Juan José]

I solved the problem creating a smooth pulse, with time to rising and down , as the picture,

Answer 1

[Sam Chak]

Hi @Juan José

Good to hear that the custom smooth signal works. The masked subsystem is able to generate the trapezoidal pulse signal.

[Juan José]

how you use the internal time on the code? without using the clock block?

[Sam Chak]

Hi @Juan José

I forgot to show how the blocks appear under the Masked Subsystem. As you can see, I also used the Clock block.