Pneumatically　supported　membrane　structures　belong　to　wind-sensitive　structures,　and　the　on-site　measured　wind　pressure　data　is　a　very　valuable　and　intuitive　data　source　for　pneumatic　membrane　structures.　However,　the　acquisition　of　wind　pressure　data　is　difficult,　costly,　and　there　is　very　little　research　on　the　processing　of　wind　pressure　data　for　pneumatic　membrane　structures　with　long　duration.　In　this　paper,　all　wind　pressure　measurement　points　on-site　of　the　pneumatic　membrane　structure　were　taken,　and　based　on　the　characteristics　of　the　measured　wind　pressure　data,　the　data　was　preprocessed　by　dividing　the　data　into　abrupt　change　points　to　remove　baseline　shifts　caused　by　abrupt　changes.　The　results　show　that　the　algorithm　proposed　in　this　paper　is　efficient,　and　provides　flexibility　in　dividing　abrupt　change　points,　laying　a　foundation　for　further　extracting　useful　information　from　the　data.　This　algorithm　can　also　be　applied　to　data　processing　with　similar　characteristics　in　other　applications.