I got two figures and answers which are seemingly contrary. In wikipedia, it says the the location of the centre of pressure of a cambered airfoil is just behind the quarter chord point at maximum lift or high angle of attack. But in other websites, I see the centre of pressure well ahead of the quarter chord point. Which of the below two pictures is correct?
The lower one is better. Details depend on the camber.
A flat plate or a symmetric airfoil have their center of pressure at the quarter chord point. Angle of attack changes change the lift produced, but the center of pressure stays at the quarter point (as long as the flow does not separate). This is exactly true for two-dimensional flow; as soon as we look at real wings with limited span, the exact location of the center of pressure moves forward at the same angle of attack with lower aspect ratio. The center of pressure of a slender body is approaching the leading edge. But your question is about airfoils, so the rest of this answer looks at 2D flow again.
Positive camber contributes lift independent of angle of attack, and its center of pressure location depends on the shape of the camber. A camber line shaped like a segment of a circle has its center of pressure at 50% of chord, and the camber line of rear-loaded airfoils is even farther back.
The resulting location of the center of pressure is the sum of both, camber and angle of attack effects. At low angle of attack the camber line dominates this location, so it is somewhere at mid chord. At high angle of attack the center of pressure shifts forward towards the 25% point. Note that this behavior will make the isolated airfoil unstable in pitch.
A negative camber will reduce lift and result in a center of pressure location ahead of the 25% point. Increasing the angle of attack will now shift the center of pressure back, again towards the 25% point. Now you have a stable pitch behavior (when the airfoil pitches up, the center of pressure moves back and creates a correcting moment change), such that a conventional airplane which loses its horizontal tail can still fly on when inverted. The center of gravity will determine the trim speed, and control may still be possible either by moving the wing flaps or by shifting the center of gravity. Theoretically, at least. I have witnessed an experienced model aircraft pilot doing this after the horizontal tail of his plane was lost in flutter, but the resulting glide speed did not lend itself to a smooth landing.
I think it will move forward. If you look at the picture below, you'll see that the pressure increases at all top surface locations with increasing angle of attack. But the suction peak at the front does this in a much faster fashion.
As a consequence the centre of pressure moves forward.
I can image that this might not be true for all airfoils, but this is the development that I saw most of the times.
