DQN-based real-time optimization method for material delivery in uncertain workshop environment

Abstract

The invention discloses a DQN-based real-time optimization method for material delivery in an uncertain workshop environment, and the method comprises the following steps: modeling the uncertain workshop environment, and building a material demand dynamic time window equation and a real-time path resistance coefficient equation; converting a material distribution real-time decision optimization problem in an uncertain workshop environment into a semi-Markov decision problem, and designing key model elements such as a state space, a global action space, a local action space and a reward function; designing two Q networks of the DQN by using a full-connection neural network; wherein the DQN continuously interacts with the environment to carry out trial and error learning until Q value network training is stable; transmitting workshop key state data sensed in real time and disturbance data obtained by calculation of the environment model to the DQN which is stably trained; wherein the DQN calculates the optimal safety action in the current state, and transmits the optimal safety action to the AGV to guide the AGV to respond to disturbance in real time in an uncertain workshop environment and make proper action selection, so that a material distribution task is quickly completed at relatively low cost.

Description

technical field [0001] The invention relates to the technical field of material distribution in a discrete manufacturing workshop, in particular to a DQN-based real-time optimization method for material distribution in an uncertain workshop environment. Background technique [0002] Improving production efficiency by improving the internal production logistics of the enterprise workshop has become an important competitive factor for enterprise development. With the rapid development of Industrial Internet of Things (IoT) and Artificial Intelligence (AI), the production mode of manufacturing enterprises has begun to change to the direction of informatization and intelligence. Workshop Material Delivery (MD) optimization problem is an important research branch of Production Logistics (PL) optimization problem, and timely delivery of materials is the key to ensure the smooth progress of workshop production activities. However, there are often various uncertain factors in the p...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a DQN-based material distribution in an uncertain workshop environment that can effectively solve the problems of weak dynamic response capability, low distribution accuracy and insufficient real-time decision-making in the uncertain workshop environment. real-time optimization method

Images