Key Technologies For Mobile Microwave Power Transmission In Coal Mine Tunnels

With the development of mine digitization and information,the usages of wireless sensors are surprisingly increasing.Due to the sensors are powered by battery,the energy of the node severely restricts the life cycle of sensor networks.Therefore,how to keep the adequacy of the node energy,especially before the disaster,is a key problem.The recent development of microwave power transmission(MPT)technology make it available to solve the energy limitation problem by MPT.In this paper,we take the energy transmission and energy allocation as the research contents,the main research works are shown as follows:1.The performance evaluation and influence factors of MPT in tunnels were analyzed.When the charging time is determined,the energy characteristics can be represented as power characteristics.The energy transmission distance in the tunnel environment is set to 10 meters in our paper,while the existing wave propagation models are invalid for this distance.Thus,we proposed a piecewise linear model fusing with mode theory,in which the power attenuation characteristics could be accurately described.Our model just need the values at the initial position and piecewise point position.The actual measurements in rectangular and arched tunnels verified the validity of our model.2.After the confirmation of the energy characteristics in tunnels,the main factor influencing energy replenishment of nodes is the rectifier.The rectifier efficiency of two rectifying type,single diode rectifier and voltage double rectifier,were analyzed.And the maximum problem equation of total rectifier output was given.We used multiple population genetic algorithm to solve the equation,thus to get the optimal charging position and the optimal parameters for the rectifier.Combining work 1 and 2,the system efficiency of MPT can be calculated.3.In tunnels,the mobile charger(MC)can only charge nodes moving along tunnel rails.According to this scenario,we theoretically gave the boundary conditions of the completely feasible strategy(CFS)that can ensure the whole network working perpetually.We proposed the maximum energy allocation model that satisfying CFS and solved the model using CVX tool.The whole sensor networks were divided into several sub-networks which satisfying CFS,and multiple MCs were arranged to charge every sub-networks periodically and orderly.By the way,when charging,we proposed a group nodes mechanism to expand the scale of sub-network.4.In order to ensure the CFS,the overcharging phenomenon would appear in group nodes.To solve this problem,we use beamforming mechanism to balance energy allocation in group nodes.In the solution process of beamforming,we used semi-definite relaxation technology to transform the original problem to a series of sub-problems according with semi-definite programming,thus iteratively solve the problem using the dichotomy.Moreover,we improved the traditional method which set the rectifier efficiency to be fixed,and introduced the probability constraint factor and the real rectifier efficiency value,thus to make the nodes attain more energy and closer to the real situation.For the noise channel,we use S-procedure lemma to convert the problem into equivalent and analogous determined channel problem,thereby obtaining the approximate optimal solution.A large number of comparative experiments showed the superiority of our algorithm.Combining work 3 and 4,the optimal charging strategy of MPT can be obtained.