Optimization control method of multi-site conveyor belt feeding production processing site system

Abstract

The invention discloses an optimization control method of a multi-site conveyor belt feeding production processing site system. A reaction item and a diffusion item of a performance function are respectively constructed according to the site self information and the interactive information of adjacent sites, wherein the reaction item is regarded as the feedback of the site self information, and the diffusion item is regarded as the feedback of downstream sites to upstream sites. Compared with a method only relying on the site self information, by introducing a reaction diffusion ideal and designing a new performance function, the invention favorably improves the load balance of the sites, obviously improves the working piece processing rate of the whole system and can effectively solve the problems of dimension disaster and modeling difficulty of a large-scale multi-agent system. Meanwhile, aiming to the problems of two performance criterions of discount and average, through the function constructed in the invention, a uniform continuous-time multi-agent study formula and an optimization algorithm can be established.

Description

technical field [0001] The invention relates to the field of automatic control, in particular to an optimal control method for a multi-station conveyor belt material production and processing station system. Background technique [0002] In some production and processing enterprises in the real world, there is a kind of production line with production and processing stations as the main body of processing, such as some robot assembly lines in advanced manufacturing, in which the processing stations are processed by conveying workpieces by conveyor belts. Such a type of system is called Conveyor-Serviced Production Station (CSPS) for conveyor belt material production. The station contains a processing subject, which can be a robot or a robot arm, etc., or even an assembly worker. The station is equipped with two warehouses, one is used to store workpieces unloaded from the conveyor belt to be processed, which is called a buffer warehouse, and its capacity is limited. The ot...

AI Technical Summary

Problems solved by technology

[0003] 1. If all the stations are considered as a whole, that is, an agent (Agent) performs centralized control, and combines the buffer space of each station to form the state of the system, and the look-ahead distance of each station to form the system action. It will inevitably cause the system state space and action set to be huge, leading to the problems of "curse of dimensionality" and "difficulty in modeling", and the complexity of the solution will increase exponentially with respect to the number of agents

[0004] 2. The communication ability or perception ability between sites may be limited by objective conditions, often only the information of their surrounding environment can be observed or perceived

Therefore, when learning control, the learning convergence speed of the upstream site is always faster than that of the downstream site, and the learning changes of the downstream site cannot be fed back to the upstream site.

This working mode tends to cause the load of the upstream station of the conveyor belt to be too high and the load of the downstream station to be light, which is not conducive to the load balance between the stations, and thus affects the improvement of the system processing rate

[0006] 4. In some literatures, the most considered cases are discrete system models, that is, the reward or cost is instantaneous and has nothing to do with time

At the same time, most of the existing research on performance optimization problems under the two performance criteria of discount and average are separated, and there is no unified expression or optimization algorithm

Images