Research On Energy-Efficient Multi-Level Collaborative Optimization Of Industrial Robot Equipment Operation

With the widely used of industrial robot equipment in the manufacturing industry,the energy consumed by them gradually increases.How to realize the energy-efficient optimization of industrial robot equipment operation while ensuring productivity has become a focus of many scholars and enterprises.According to the form and granularity of manufacturing capability in factory floor,industrial robotic manufacturing system can be divided into three levels,including manufacturing cell level,production line level and workshop level.Most of existing works about energy-efficient optimization of industrial robot equipment operation were mainly carried out on the first two levels individually,without considering the propagation characteristics of performance fluctuation between the two levels.However,manufacturing cells are the component parts of a production line,optimization at any level will inevitably affect the operation of another level and even cause unknown performance disturbances.Therefore,it is of great significance to study the energy-efficient multi-level collaborative optimization of industrial robot equipment operation.Towards service-oriented manufacturing model,energy-efficient optimization of industrial robot equipment operation in manufacturing cell level and production line level are studied respectively,then a multi-level collaborative optimization model and decision-making methods are further studied.The main work done in this paper is as follows:(1)A method based on Deep Q-Network(DQN)for energy-efficient control of industrial robot equipment operation speed and a method based on adaptive bees algorithm for industrial robot equipment services composition optimization are proposed respectively.Since industrial robot energy consumption modeling is complex and cumbersome,DQN is applied to the field of industrial robot equipment speed optimization,then a model-free method for energy-efficient speed control of industrial robot equipment operation is proposed.In the production line level,neighborhood search strategy of the bees algorithm is improved based on Q-learning,and an adaptive bees algorithm is proposed to solve the problem of industrial robot equipment services composition optimization.(2)An energy-efficient multi-level collaborative optimization model of industrial robot equipment operation is constructed.Considering the uncertain events in the production environment and elements,behaviors,and rules for the operation of industrial robotic manufacturing systems in both levels,the model is constructed based on the ideas of event-driven,double-decision,and cross-level coordination.The functional modules and operation mechanism of the model are designed carefully,and characteristics and key implementation technologies of the model are explained in detail.(3)Energy-efficient multi-level collaborative decision-making methods of industrial robot equipment operation are proposed.Towards the proposed model,taking into account the mutual influence of performance between the two levels,multi-level collaborative decision-making methods are designed based on the three principles of greed,performance protection and minimum cost.The proposed decision-making methods make decisions and coordination for optimization behaviors of industrial robotic manufacturing systems in manufacturing cell level and production line level,and finally achieves overall improvement of energy-efficient performance.