We have a HP flare that unfortunately burns hydrocarbons badly, batch wise. I checked some scenarios like sudden change in gas flow rate (sweep gas, all source of gases sent to flare line) and no significant change in gas flow rate is seen once the flare batch-wisely burns hydrocarbons. I also checked the level of K.O. drum of flare for any disturbance during this problem and it was also constant. The air blower is also checked and they were good. I feel this problem originates from wind at the tip of flare but not sure. I appreciate your comments.