Cement free calcium soft measurement method based on non-supervision and supervision learning

Abstract

The invention discloses a cement free calcium soft measurement method based on non-supervision and supervision learning. By analyzing a cement technology, variables are selected as input variables ofclinker fCaO soft measurement, and a time sequence of each variable is used as model input; by using the selected input variables, a prediction model combining non-supervision learning with supervision learning is established; sparse self-coding decoding layers are removed, coding layers are laminated to form a deep network structure, determined initial parameters are used for removing initialization deep network parameters, and a BP reverse error correction algorithm is adopted for supervision learning; the trained prediction model combining non-supervision learning with supervision learningis used for real-time prediction on cement clinker fCaO. According to the method, a greedy layer-wise non-supervision learning mode is adopted in the forward direction of the model, high-level characteristics of data are extracted; in combination with supervision reverse fine adjustment, the parameters are further optimized, and a trained deep network is used for real-time prediction on the clinker fCaO.

Description

technical field [0001] The invention relates to the field of industrial cement quality soft measurement monitoring, in particular to a cement free calcium soft measurement method based on unsupervised and supervised learning. Background technique [0002] The content of free calcium oxide (fCaO) in cement clinker is an important index to measure the quality of clinker in the production of new dry process cement. The clinker fCaO content not only affects the stability and clinker strength of cement, but also directly relates to the energy consumption of cement firing. At present, cement clinker fCaO is mainly measured by manual one-hour sampling and laboratory testing, but the off-line measurement results have obvious hysteresis during the cement firing process. optimization. The cement clinker firing process has the characteristics of large inertia, large time delay, multi-coupling, and few label samples, which makes it difficult to establish an accurate cement clinker fCa...

AI Technical Summary

Problems solved by technology

Considering that there are few labeled samples of clinker fCaO, the training of deep neural network with a small amount of sample data is prone to overfitting.

Images