Complex manufacturing system regression scheduling method

Abstract

The invention discloses a complex manufacturing system regression scheduling method, and the method achieves the obtaining of a scheduling rule weight combination which is generated in real time and adapts to a production state through the construction of a complex manufacturing system regression scheduling model based on an extreme learning machine, achieves the multi-target optimization of a manufacturing system, solves an output weight matrix of a hidden layer node by using a proportional integral differential (PID)-based gradient descent algorithm to complete training of an extreme learning machine model, so that the generalization performance of the algorithm can be improved under the conditions of reducing the calculation burden and improving the calculation efficiency.

Description

technical field [0001] The invention relates to the technical field of dynamic scheduling of complex manufacturing systems, in particular to a regression scheduling method of complex manufacturing systems. Background technique [0002] The production process of complex manufacturing system is complex and consists of multiple related production processes. When the manufacturing process is relatively stable, the original scheduling strategy can continuously ensure the optimization of system production performance; however, when the manufacturing system has disturbances such as machine failures, the manufacturing environment changes, the previously adopted scheduling strategy becomes invalid, and ultimately the expected production performance cannot be obtained . Therefore, how to dynamically determine an effective scheduling strategy according to the state of the production process is the key to improving the operation performance of complex manufacturing systems. This metho...

AI Technical Summary

Problems solved by technology

[0005] The dynamic scheduling of complex manufacturing systems is a multi-constraint, multi-objective optimization problem. Classification-based scheduling models usually use simple heuristic scheduling rules as the scheduling strategy, which is difficult to meet the requirements of multi-objective and multi-constraint manufacturing systems; traditional extreme learning machine based The regression model uses the ridge regression method to obtain the optimal regularization coefficient through the trial and error method to obtain the output weight matrix. This method has heavy computational burden and low computational efficiency, which reduces the generalization performance. Therefore, the design of an efficient The regression model is of great significance to the dynamic scheduling of complex manufacturing systems

Images