Research And Implementation Of Production Resource Refined Prediction Model Pool Oriented To Industrial Internet

Economic development and the rise of industrial reform begin to boost the change and promotion of the industry.At the same time,the development of information technology represented by the Internet greatly change our life and bring great opportunities for the transformation of traditional industry.Industry is developing toward digitalization,networking and intelligence.The traditional factory resource management and calculation method can obviously not meet the dynamic and fine production demand in the production scene,and it is hard to manage a lot of large quantities of industrial production data.Therefore,in the dynamic production environment,it is very necessary to realize digital management of industrial big data and accurate fine-grained prediction.On the basis of fully investigating the industrial Internet platform and core data modeling technology,we first present the overall design of the industrial resource data prediction system and the design of each functional module according to the application scenarios and user needs.Based on the analysis of industrial data,the prediction issue is raised as the modeling issue of time series data,and the corresponding algorithm model is proposed for the key problem of fine data prediction in the system:multi-frequency and multi-variable long short time memory network model and multi-level deep decomposition network model for fine-grained prediction of single-dimensional industrial time series data.We design and integrate a production resource prediction model pool containing diverse time series data models combining the diversity of production scenarios.It can not only achieve fine-grained prediction of industrial data,but also provide service such as rapid feedback prediction results and predictive maintenance through calling different models.The whole system adopts modular design.Through the interaction from one module to another and the calls of database and model pool to implement digital factory resource data management and forecast service.We design and implement a forecasting system,which is an important part of the Industrial Internet Intelligent Cloud Collaboration Platform.As a national key scientific research project,the platform aims at providing intelligent and flexible industrial production mode.It is verified that we achieve the desired effect.Now the system is successfully put into service,and it will be put into actual production after the completion of the overall platform development.