Research On Dynamic Characteristics Of Trees And New Methods Of Measurement

Trees can not only regulate the climate,avoid soil erosion,but also improve the natural ecological environment,and play a very important role in urban air purification and city appearance improvement.But trees are also fragile,prone to snap and fall under strong winds.Therefore,it is necessary to study the dynamic characteristics of trees as the basis of wind induced dynamic response analysis and wind disaster protection research.Firstly,the modal identification and analysis of a willow tree were carried out under hammer excitation and natural wind excitation.It was found that some modes were missing in the modal recognition results under the force hammer excitation,while the main modes of willow could be recognized better under the natural wind excitation.The first ten measured modes of willow were determined by combining the modal recognition results under the two excitation conditions.According to the shape of the real willow tree and adopting reasonable material parameters,a finite element analysis model was constructed for the willow tree.The order of a large branch in the model is used to divide the frequency range of each order of the willow finite element model,and the first ten natural vibration frequencies and mode shapes are obtained.It is found that there is a good one-to-one correspondence between the willow finite element simulation results and the willow measured modal results,which shows that the construction of the willow finite element model is correct.It was found that the first mode shape of willow was mainly the branch vibration,and the trunk vibration was very small.Then,a maple tree with more complex structure was measured by force hammer excitation mode.It was found that five of the six low-frequency modes can be obtained by measurement because the structure of maple tree is relatively slender and the stiffness of trunk and branch is similar.The finite element model was constructed according to the actual maple tree structure.It is found that due to the complex branch system,there is a large difference in morphology and natural frequency between each branch,and it is necessary to retain two main branches to basically maintain the main vibration frequency of the maple tree.According to the vibration of these two large branches,the frequency range of the maple finite element model is divided,and the first six-order frequency and mode shape are obtained,and the results correspond well to the measured results.Since the stiffness of the branches of the maple tree is close to that of the trunk,the vibration of the maple trunk is not small compared to the branch structure at the first-order mode shape.Finally,in order to study the vibration characteristics of tree crown structure,a center of gravity calculation algorithm based on video matting is proposed.A cantilever beam structure was used to discuss the accuracy and influencing factors of the method,and it was found that if you want to obtain the high-order vibration mode of the object,you must excite the object with obvious high-order vibration as much as possible.Then,a natural wind excitation test was performed on a Douban boxwood tree in the wind tunnel,and its dynamic characteristics were analyzed using the center of gravity calculation algorithm based on video matting.It is found that the algorithm can be applied to the structure of trees and can basically obtain the first two vibration modes of Douban boxwood.The measured results of the modal are consistent with the frequency results obtained by the force balance,and are similar to the finite element simulation results.It is found that the Douban boxwood exhibits the first-order and second-order modes of the cantilever beam in the first and second modes,which shows that it is feasible to study the canopy structure as a whole.However,it can also be seen that when part of the tree branches produce relatively large local vibrations,the resulting modal results will be disturbed.When the deformation is large,the layered center of gravity of Douban boxwood will also move down,which will affect the modal recognition results.However,by setting different layering methods and dividing each layer into a curved shape,the situation can be significantly improved.