| With the wide application of power cables in urban power transmission networks,the development and application of power cable tunnel monitoring systems has also become an important part to ensure the safe and stable operation of power systems.However,the sensor monitoring module and network communication module depend on the condition of the power cable tunnel,and most mobile inspection robots do not have amphibious movement capabilities or sufficient unstructured terrain coping capabilities.Therefore,the existing fixed power cable tunnel monitoring system cannot effectively respond to emergencies such as floods in power cable tunnels.Consequently,it is necessary to develop a mobile power cable tunnel monitoring system that can effectively deal with emergencies such as floods that occur in the power cable tunnel.The main research work of this thesis is as follows:Firstly,this thesis designs the overall scheme of the power cable tunnel amphibious environment monitoring robot.Aiming at the current situation that the existing power cable tunnel monitoring system lacks the ability to deal with emergencies,this thesis designs a set of mobile power cable tunnel amphibious environment monitoring robot system that is suitable for both routine and emergency conditions.During routine inspection,the monitoring system realizes the communication through the original or preset communication network of the power cable tunnel;during emergency monitoring,the monitoring system realizes the communication through arranging the base stations at the cable wellhead to construct an emergency communication network.Based on the hexapod robot system platform,the thesis designs a AmHexaBot amphibious robot to be treated as the mobile terminal of the power cable tunnel monitoring system.In terms of mechanical structure,the research focuses on the waterproof and sealing design of AmHexaBot.In addition,An integrated amphibious drive mechanism that combines arched feet and fins is proposed to simplify the mechanical structure and enable AmHexaBot to be competent for amphibious monitoring tasks.Then,the control system of AmHexaBot is designed based on the idea of layered modularization.Accordingly,this thesis builds the software system and carry out the hardware layout and integration.Because the traditional model-based control method has the shortcomings of high complexity and poor robustness,this thesis adopts a bionic-based CPG neural network to control the motion and behavior of AmHexaBot.In order to imitate the pattern of the neural network that generates and controls biological rhythmic motions,this thesis sets up six pairs of Hopf oscillators to simulate the mutual excitation and inhibition between biological neuron pairs and generate self-excited oscillation signals to control the behavior of AmHexaBot.Based on the CPG neural network control method,this thesis designs the amphibious gait of AmHexaBot on land and water.Additionally,the concepts of average stability margin and period average stability margin are proposed to analyze the stability of AmHexaBot quantitatively.Furthermore,a smooth gait switching strategy is formulated to achieve smooth and stable gait switching process.The robot’s motion mechanism is likely to be damaged due to the accidents in the power cable tunnel.Therefore,based on the CPG neural network control method,this thesis designs a highly robust gait used to deal with the fault of the AmHexaBot motion mechanism.The nerve terminal feedback is introduced to monitor the fault of the AmHexaBot amphibious robot.If and only if one foot of the AmHexaBot is disabled,the AmHexaBot can still maintain high robustness to ensure the normal movement.The robust gait improves AmHexaBot’s adaptability to the external environment and the robustness of control.Finally,this thesis designs a physical prototype of AmHexaBot and carries out experiments are to verify and analyze the correctness of the proposed control method,including the rutine gait experiments on land and water,smooth gait switching strategy experiments and robust gait experiments.Comparing the expectational and real motion trajectories of each foot,the experiment results validate the correctness and effectiveness of the theoretical methods by calculating the tracking errors and analyzing the reasons. |
