Coal quality melting point prediction method based on multi-model fusion Stacking algorithm

Abstract

The invention discloses a coal quality melting point prediction method based on a multi-model fusion Stacking algorithm, and the method employs a machine algorithm to predict data, and comprises the steps: firstly determining a target variable and a feature variable, dividing a test database into a test A database and a test B database, and constructing a corresponding train A database and a train B database; carrying out maximum and minimum normalization processing on the data, carrying out Box-Cox transformation, adopting a '3 sigma criterion' to eliminate abnormal values and the like; and selecting a base model and a meta-model, separately training the base model, performing fitting training on the meta-model to obtain a final model A, and predicting the ash fusion point of the testA database by the model A. According to the coal quality melting point prediction method, the defects of time consumption, energy consumption and labor consumption are overcome, and large-scale samples can be predicted; the generalization of the model is high, and data prediction of characteristic variables exceeding a database can be processed; and the model fitting degree is high, the model is prevented from being over-fitted, the prediction accuracy is high, the prediction data variance is small, and the stability is good.

Description

technical field [0001] The invention relates to the technical field of coal quality ash melting point prediction, in particular to a coal quality melting point prediction method based on a multi-model fusion Stacking algorithm. Background technique [0002] The ash melting point of coal is one of the important measurement items of high temperature characteristics of coal. The melting temperature of coal is of great significance in industry whether it is a thermal power plant or a coal gasifier. The traditional coal ash melting point is determined by coal high-temperature experiments. The steps are relatively cumbersome, and the temperature needs to be gradually raised to a high temperature of 1500 degrees. [0003] With the development of computer technologies such as big data and machine learning algorithms, the method of using algorithms to fit historical databases has been applied to the prediction of coal ash melting point in recent years. However, there are still some ...

AI Technical Summary

Problems solved by technology

[0004] 1. There are few researches on feature engineering, and the selected feature variables are all but not precise, which affects the prediction accuracy

[0005] 2. The feature range matching between the forecast data and the historical database is not performed. If the feature variables of the forecast data are included in the historical database, the accuracy is better. If they are not included in the historical database, the accuracy is greatly reduced

[0006] 3. The existing coal ash melting point prediction method uses a single machine learning algorithm, and there is still room for improvement in fitting and accuracy

Images