@craigb244 A few things. With PID control, if the actual temperature is a long way away from the set point, then initially heating will start at 100% PWM and only reduce when the temperature approaches the set point. So in a way, your assumption is correct and you could possibly use a macro to reduce the maximum PWM value. One way would be like this (note that I haven't tried it so it might not work):

M307 H0 S0.5 ; sets the maxi PWM for the bed heater to 50%
M190 S30 ; Heat bed to 30 and wait for temperature to be reached.
M307 H0 S0.75; sets heater PWM to a maximum of 75%
M190 S40; Heat be to 40 and wait for temperature to be reached
M307 H0 S1.0; set heater to use up to 100% PWM
M190 S50; complete heating to 50.

But I don't think this will help your warping problem. The reason I say that is with the temperature sensor in the middle of the silicone heater, when you start to heat the bed, the temperature of the silicone heater rises rapidly to the set point but there hasn't been enough time for that heat to conduct through the rest of the plate. But the sensor and control system "don't care" so the heater gets turned off (or PWM kicks in at a very low value). PWM only increases once the temperature at the silicone heater drops below the set point. So in a way, having the sensor where it is in the centre of the heater, effectively reduces the PWM value for the aluminium plate as whole. By switching off early, it is already doing what you want to do by other means.

So I don't think any of that will help with your warping issue. i.e. - I don't think the warping is due to the rate of change of temperature and is simply a function of increased temperature.

As a side note, what size (wattage) is your heater? For info, mine is 800 Watt and that's a bit on the high side. I also have 12mm of semi rigid insulation underneath the heat pad. With my sensor in the edge of the bed as described, it take about 5 1/2 minutes to go from from ambient of around 22 to 50 degC. PWM actually cuts in at around 49 degrees or so. Before I moved the sensor, it would take more than twice that long for the top surface to reach that temperature due to the heater switching off prematurely.