Abstract: 【Objective】Tomato bacterial wilt, caused by Ralstonia solanacearum, is a major disease in tomato production. The ecological risks of traditional pesticide control have prompted the adoption of biological control, which has received extensive attention. Research has revealed that the interaction between arbuscular mycorrhizal (AM) fungi and beneficial microorganisms such as Burkholderia is an important mechanism for biological control of plant diseases. However, the effect of these interactions on soil microbial structure remains elusive. 【Method】In this experiment, the interaction of tomato-AM fungi-Birkholderia-Ralstonia solanacearum was established in the greenhouse to explore the prevention and control effect of AM fungi and Burkholderia alone and in combination on tomato bacterial wilt and its rhizosphere microbial regulation mechanism.【Result】It was found that inoculation with AM fungi and Burkholderia could reduce the incidence of tomato by 27.22 % and 22.11 %, respectively, and the incidence of combined inoculation was 34.78 % lower than that of the control (without inoculation of beneficial microorganisms). After inoculation with Ralstonia solanacearum, the biomass of tomato treated with AM fungi and Burkholderia was significantly increased, and the combined treatment was significantly better than single inoculation in promoting tomato biomass. Combined inoculation can better optimize the rhizosphere microbial community structure. The relative abundance of Actinobacteriota and Gemmatimonadota in the rhizosphere soil increased by 22.64 % and 12.09 %, respectively, compared with the control. The abundance of Gemmatimonadota increased by 34.64 %, and the abundance of Acidobacteriota decreased by 21.17 % on the 7th day after inoculation. The abundance of Chloroflexi increased by 49.40 % on the 14th day after inoculation. In addition, the co-inoculation increased the positive correlation connection ratio of the rhizosphere bacterial co-occurrence network from 51.56 % of the control to 73.76 %, the average connectivity increased from 1.966 to 2.484, and the modularity index decreased from 0.988 to 0.985. The neutral community model showed that the network complexity of the combined inoculation treatment was 23.2 % lower than that of the control on the 7th day after inoculation, and returned to the same level as that of the control on the 14th day. The results showed that combined inoculation achieved synergistic disease resistance by regulating the abundance of specific beneficial bacteria, optimizing the microbial interaction network, and dynamically adjusting the community stability. This demonstrates a new microbial combination strategy for the biological prevention and control of tomato bacterial wilt, and provides theoretical support for green and efficient production of facility agriculture.