Research On Prediction Method For Multi-objective Optimization Resource Scheduling Results For Cloud Manufacturing

Manufacturing industry is the material basis and main industry of the national economy.It is an important symbol of the country's scientific and technological level and comprehensive strength.In recent years,with the transformation of manufacturing industry,humanized and intelligent manufacturing methods have attracted more and more attention.The concept of cloud manufacturing is proposed in response to the development of the times.Cloud manufacturing is based on the concept of manufacturing-as-a-service,which combines technologies such as cloud computing,big data and Internet of Things.Task scheduling and resource allocation for cloud manufacturing has always been a hot topic in cloud manufacturing research.Different from cloud computing resource scheduling,which focuses on computing power,cloud manufacturing resource scheduling is based on state information of virtual manufacturing resources in the cloud platform and real-time information of manufacturing tasks to generate an optimal task execution plan.With the development of big data and Internet of Things,how to find the optimal result in the shortest time is the core problem of cloud manufacturing resource scheduling and distribution.It is a feasible path to use the deep learning model to train the optimized data to achieve direct prediction of the scheduling results.This paper mainly focuses on the optimization of scheduling algorithm and the improvement of deep learning model,as follows:1.Aiming at the problem of cloud manufacturing resource scheduling,the scheduling model with the shortest total task execution time based on QoS constraints is established.2.Based on the cloud manufacturing task scheduling model,an improved new bat algorithm is proposed to solve the multi-objective optimization resource scheduling problem of cloud manufacturing.The improved new bat algorithm adds a second-order oscillating link and integrates the difference based on the original algorithm,the speed and accuracy of the improved algorithm have been significantly improved.3.On the basis of the above,the deep learning model is introduced for the optimization of scheduling efficiency,and the learning rate of the deep learning model is improved.The training speed of the model is accelerated by improving the learning rate to speed up the training of the model.Then use the scheduling data set to train the improved deep learning model to predict the scheduling results.4.Finally,the paper firstly simulates the improved new bat algorithm,uses the improved algorithm to solve the proposed cloud manufacturing scheduling model and compares it with other algorithms,then uses the obtained scheduling data set to train the improved deep learning model,and then use the training.A good model predicts the scheduling results and compares them with the scheduling time of the traditional methods,so that the proposed method is practical.