Continuous real-time prediction method of environmental temperature and humidity in tobacco production process based on LSTM

Abstract

The invention discloses an LSTM-based method for continuous real-time moisture prediction of ambient temperature and humidity in the shredded tobacco production process. In order to ensure the stability of moisture at the outlet of shredded tobacco after flavoring in a multi-batch and continuous production environment, the method based on LSTM is realized. Deep learning iterative prediction method, analyzing the impact of environmental temperature and humidity in the process of moisture analysis and prediction on the moisture of cut tobacco outlet after flavoring, and establishing a real-time prediction model for moisture content of cut tobacco after flavoring; through the model solution, the predicted environment can be obtained The influence trend of temperature and humidity on the moisture yield of shredded tobacco after flavoring was tested by comparing the predicted value of the model with the measured value. The invention can realize the prediction of the silk moisture, thereby improving the preparation and efficiency of the prediction.

Description

technical field [0001] The invention relates to the field of real-time prediction of intelligent manufacturing, and is specifically aimed at an LSTM-based real-time prediction method of ambient temperature and humidity in a continuous, multi-batch, feedback hysteresis environment in the cut tobacco production process. Background technique [0002] With the modernization of industry and the progress of science and technology, as an important part of my country's economic income, my country's tobacco industry has also entered a new stage of development. Cigarette production is a relatively complicated process, and each link has a significant impact on the quality of cigarettes and the consumption of materials. As an important link in the tobacco processing process, tobacco shred processing can effectively ensure the stability of cigarette quality during the processing process due to its strong continuity and correlation, as well as a variety of process and equipment factors. ...

AI Technical Summary

Problems solved by technology

However, in the tobacco shredded production process, multiple batches of tobacco leaves need to be predicted in real time in a continuous production environment. The control parameters of each batch are changing in real time. At the same time, research on the control of moisture content in shredded tobacco after flavoring Mainly focus on intelligent control, PID control and the combination of the two. Because the control effect has a certain hysteresis, the potential influence of ambient temperature and humidity cannot be considered, which will easily lead to a decrease in the stability of the moisture content of the tobacco after flavoring.

Therefore, the traditional LSTM algorithm is difficult to achieve real-time and accurate prediction under such hysteresis, continuous, and multi-batch conditions.

Images