The vibration quality evaluation of concrete dam is an important means for the quality control of the concrete dam construction. Traditional method combines artificial experience control with postmortem finite core sampling point detection to control the vibration quality of concrete dam;in addition, the existing vibration monitoring research on vibrator only uses a single rangefinder to range and lacks evaluation method of vibration quality. Taking into account the challenges and difficulties of real-time vibration quality dynamic evaluation and feedback control in concrete dam storehouse, this paper puts forward a real-time monitoring method of the vibration quality of concrete dam considering vibration quality acceptance management based on a ranging scheme with two rangefinders, and establishes a dynamic evaluation model for concrete dam vibration quality considering parameter uncertainty on the basis of this method. The model mainly includes the following contents: (1) conducting quantitative analysis of the uncertainty of concrete characteristic parameters by introducing the "information entropy" theory;(2) proposing the dynamic evaluation method of concrete compression strength based on random forest algorithm in the case of parameter uncertainty, and the evaluation results are applied to the whole dam storehouse surface vibration quality evaluation. The evaluation model are compared with the models built by multiple linear regression and artificial neural network algorithm, which verifies the consistency and superiority of the random forest model. The dynamic evaluation of the vibrating quality of the abnormal concrete area of a RCC dam project in southwest China is carried out by the evaluation method proposed in this paper. The results show that the method can accurately evaluate the vibration of concrete storehouse surface quality, and it also provides a theoretical basis for the vibration quality control of the concrete dam. © 2018, China Water Power Press. All right reserved.
