The machine connects the relationship between humans and the natural world. Due to the development of the engineering industry and the emergence of Machine Learning, humans began to use machines to understand the natural world and explore the unknown. Machines are no longer the useless iron sheets and parts, they have become part of the natural ecosystem. Humans act as intervenors to manipulate and influence machines to think. As new members of nature, machines are likely to spread all over the world in the future, forming a new group through continuous self-learning and evolution, just as different species have their stable ecological chains in nature and break the limits of human intervention. Complexity and creativity of machine are ideas that can be generated from an unconscious, meaningless and automated process which is natural selection and co-evolution. We are all in the process of evolution and selection, which is an undetected experience, and this intangible experience has inspired endless possibilities. Machines can get rid of human control, like animals without a leader in swarming behavior. In the meantime, they can rely on Collective Intelligence create new 'ethnicity' in nature. Human cognition of the natural world should not be limited to the stage of data analysis, building a new type of machine network should focus on data sharing and machine self-learning. Data can be a tool in the hands of creators to change the ecological environment which also provides a new definition of the disappearing history. I installed the two sets of fans on the opposite side, and let the audience to select the city to achieve real-time data transfer. I will retrieve the data of wind strength from the weather API and get the final wind speed by comparing the wind strength with the fan's PWM value and speed value. When one side of the fan receives the audience's command, the other side of the fan will learn the same wind speed. Under normal circumstances, the wind anemometer will rotate according to the wind on one side, but the wind anemometer in this device will be balanced, because machines take data and energy from nature, and use convective winds form a balance in the process of self-learning. I tried to blur the boundary between the machine and nature, proposed the idea of building a ""Mechanical Eco-chain"" in order to give the machine more autonomy. Data sharing and self-learning are the core of Mechanical Eco-chain. The impact is not only reflected in the diversification of data, but also in the creation of a new ecological relationship between people, machines and nature. Perhaps we can come up with a new hypothesis: human beings can create new ecological models, machines, creatures and various natural energies co-evolving in nature. In other words, humans promote technological advancement and try to use machines to build a broader field of control. Nature can observe itself and naturally evolve itself.