Bubbles are existent everywhere and of great importance for the daily life and industry process, such as heat exchange rate influenced by bubbles in the tube, battery life partially decided by bubbles of chemical reaction in it, etc. With the further requirement for miniaturization, physical mechanisms behind bubble behaviors in microchannels become crucial. In the present work, the lattice Boltzmann method is used to investigate the behavior of bubbles as they rise in complex microchannels under the action of buoyancy. The channel is placed with two asymmetric obstacles on its left and right side. Initially, the lattice Boltzmann model is tested for its reliability and accuracy by Laplace law. Then a few parameters of flow field, i.e. the Eötvös number, the viscosity ratio, the vertical distance between the obstacles, the horizontal distance between the obstacles, are employed to study the characteristics of the bubble during the movement, including the deformation, the rising speed, the residual mass, and the time of bubble passing through the channel. The results are shown below. First, the trend of the bubble's velocity changing with time in the process of passing through the channel corresponds to the change process of the dynamic behavior of the interface, i.e. the bubble velocity decreases when the bubble shape changes significantly under the same channel width. Second, with the increase of Eo number, the bubble deformation as well as the bubble velocity increases and the bubble residual mass decreases. Besides, the gas-to-liquid viscosity ratio has a significant effect on the bubble velocity. Under the condition of high viscosity ratio, the bubble shape is difficult to maintain a round shape, while the bubble rise velocity increases and the residual mass of the bubble decreases with the viscosity ratio. What is more, when the obstacle setting is changed, the longer the vertical distance between the two asymmetric obstacles, the shorter the bubble passing time is, and the faster it will return to the original shape after passing through the obstacle, while the residual mass of the bubble shows a change trend of approximately unchanged-increase-decrease-increase with the augment of the vertical distance between the obstacles. In the study of changing the horizontal spacing, two cases: the two obstacles are changed at the same time (Case A) and only the one-sided obstacle is changed (Case B), are considered. The results show that under the same small horizontal interval, the obstruction effect caused by changing only the length of one side obstacle is stronger. Finally, the study shows that when the width of the right obstacle is long enough, although the width of the obstacle continues to increase, the passing time of the bubble increases slowly, and the position of the bubble leaving from the obstacle is always approximately the same.