I am trying to fully understand what is happening as it pertains to bypass diodes. When a single cell is shaded, the diode acts by decreasing the module voltage in exchange for preserving current.
As an example, if the module current was 9.18A module Voc was 45.1 and one cell was shaded, then in theory the amperage would remain 9.18 and the approximate module voltage would decrease proportionally to 30.1 (one third/diode)? Or is the truth somewhere in the middle? Would the effect be the same if two cells were shaded in the same series wiring/diode? Three?
Your example is correct if you were using Vmp and Imp in the example. With a typical 60 cell module, we have 3 groups of 20 cells. If 1 to 20 cells in that group gets shaded enough to kick in the bypass diode, then you will lose 1/3 of the voltage of that module.
The problem with using Voc, is that Voc is when the module is turned off (open circuited) and there is no current in that condition, so there will be no reason for the bypass to bypass current.
If you take a PV module into the sun and measure the open circuit voltage, then shade a cell, usually that shaded cell will have some degree of light in the shade and you will get close to full voltage. No bypass diodes kick in, because there is no current in the open circuit condition. A PV module or solar cell in the shade will have close to the same Voc as the module will in full hot sunlight.
Then if you short circuit the module and shade a cell, you will get full current, because the bypass diode will kick in. In the short circuited condition, you will have no voltage, because positive is connected to negative.
An inverter will maximum power point track the IV curve and will work at the voltage that is most efficient for making power. When there is a shadow on a cell that is slight, the inverter will decide to work at that voltage that makes the most power and the bypass diode will not kick in. If the shadow on the affected cell is progressively larger, then at some point the inverter will work at a lower voltage, which will cause the bypass diode to kick in. This all works well with a single string on a single MPP, however with multiple strings on a single MPPT, the inverter should be more efficient at making power working at the voltage of the multiple modules that are unshaded, leaving the shaded module voltage at the mercy of the unshaded strings. If we have a 1000V inverter and a string of 24 modules in series, then the voltage difference will be 24 modules x 3 diode segments per module = 72 segments, so we would only be talking about 1/72 of the voltage being missing, which is not a big deal for being on the peak of the IV curve. However if we have a charge controller with multiple strings that can only go up to a maximum system voltage of 150V, then we often have 3 modules in series. 3 modules x 3 bypass diode segments = 9, so that shaded cell would be taking out 1/9th of the voltage, which will have a much greater effect on the maximum power point for that string with the shaded cell, than in the 1000V inverter example.
One of the gray areas is the quality of the shadow and when the diode will kick in and bypass current. Shadows can be from near objects, far away objects, on hazy days with a lot of diffuse light, days with reflected light, etc. This is one of those aspects of PV that is difficult to model and the best PV software has trouble modeling bypass diodes.
The best idea.... No shade!