Abstract: Based on the cloud microphysical detection data of 10-sortie aircraft over southern Sichuan Basin in 2015 and 2017, the liquid water content measured by DMT (Droplet Measurement Technologies) hot-wire liquid water content sensor is examined, and abnormal values in maximum, minimum and negative values are found.There are 4 possible causes for the abnormal maximum, minimum and negative values of liquid water content. First, the errors are caused by multiple parameters such as temperature, air pressure and vacuum velocity, which may lead to the error superposition of calculated values. Second, the on-board operators didn't calibrate the zero before entering the cloud. Third, the on-board operators only calibrate the zero once before entering the cloud during the whole flight. Fourth, the interval between cloud entry and exit is too short, so that the manual zero calibration is inaccurate.Using cloud particle spectrum data from cloud droplet probe (CDP), cloud imaging probe (CIP) and precipitation imaging probe (PIP), three solutions are proposed for calibrating hot-wire liquid water content sensor. Solution 1 is to set the criteria for entering cloud as the concentration of particle above a certain size from CDP probe greater than 0. Solution 2 is to set the criteria for entering cloud as the number concentration of cloud particles greater than 10 cm^-3 from CDP probe. Solution 3 is to set the criteria for entering cloud as the number concentration from CDP, CIP and PIP probe greater than 0. The results show that when the number concentration is 0 from CDP, CIP and PIP probe, the original non-zero liquid water content problems are corrected by these solutions.To avoid the influence of ice phase particles on CDP number concentration, the verification is carried out in the positive temperature zone. All the test results show that the negative proportion of liquid water content is also significantly reduced compared with the original data. Solution 1 reduces the negative proportion of liquid water content, and make the minimum and maximum more reasonable than other scales. The liquid water content measured by Solution 1 are more reasonable than Solution 2 and 3.