The atmospheric boundary layer height（ABLH）is one of the most important parameters in the study of atmospheric environment，weather and climate. With the development of ground-based remote sensing technology，continuous monitoring of ABLH has become possible. However，the ABLH derived based on groundbased remote sensing depends on the inversion method used and is affected by complex atmospheric conditions. In this study，we use the data from ceilometer and rain gauge，weather records from 27th August 2020 to 1st Au‐ gust 2023 and radiosonding records obtained during the 2023 summer extensive observation period at the Pingliang Land Surface Process and Severe Weather Research Station，Chinese Academy of Sciences. The effective‐ ness of several commonly used algorithms for inverting ABLH based on backscatter profiles from ceilometer are evaluated by comparing with the ABLH identified by potential temperature profiles. A hybrid algorithm that employs different backscatter gradient inversion methods for daytime［08:00（Beijing time，same as after）-19:00］ and nighttime（from 20:00 to 07:00 the next day）is proposed with constrained retrieval heights tailored for the study area. The results reveal notable differences in the inversion results among various algorithms，including the maximum negative gradient method，the three-major negative gradient evaluation method，the percentage fluctuation method，the inflection point method，and the Flamant method. Specifically，the ABLHs derived by the Flamant method，the three-major negative gradient evaluation method，and the maximum negative gradient method correlated well with that determined by the potential temperature profile and give lower mean absolute deviations. In contrast，the ABLHs derived by the inflection point method and the percentage fluctuation method give large absolute deviations. Appropriate smoothing of the backscatter profiles，combined with the hybrid algorithm，significantly improved the accuracy of the derived ABLH. Among the investigated methods，the SG 25/ 25 smoothing scheme combined with the Flamant and maximum negative gradient hybrid algorithm yielded the best results，achieving a correlation of 0. 56 with the ABLH determined by the potential temperature profile and an average absolute deviation of approximately 406 m. The correlation between the ABLHs derived from the hybrid algorithm and that from the ceilometer’s internal algorithm is 0. 64. The hybrid algorithm can effectively capture the diurnal variation of ABLH. The proposed hybrid algorithm can be used to obtain continuous，highresolution ABLH information，serving as a valuable supplementary method for obtaining fundamental data on ABLH and related parameters.