Abstract:It is particularly important to accurately predict NOx emission concentration in cement production. As the cement industry is a process industry, there are many chemical reactions, coupling of front and back processes, and large fluctuations in production, resulting in time-delayed denitration data, strong coupling, and uncertainty, which makes it difficult to adequately extract the characteristics of the denitration data to achieve accurate prediction of NOx concentration. At the same time, excessive ammonia is used in production to reduce NOx concentration, resulting in increased ammonia escape and secondary hazards to the environment, therefore it is necessary to perform accurate multi-step prediction of NOx concentration and ammonia escape. To address these issues, this paper presented a new multicriteria deep learning hybrid model MI-ITBA. The model is constructed by connecting an improved causal dilated temporal convolutional network (ITCN) with bidirectional gated recurrent units (BiGRU) and a special attention mechanism, which enhances the ability to extract features from denitration sequence, using the attention mechanism to weigh the importance of different time steps and highlight the influence of key features on the prediction results. In addition, the time series (TS) of denitration containing time-delay information is designed in the input of the model to minimize the influence of time delay. Dynamic information support for denitration control was provided by multi-step prediction of NOx concentration at seven moments and ammonia escape value at three moments to retain the dynamics of the data. Compared with TCN, BiGRU, and TCN-BiGRU, MI-ITBA has a high prediction accuracy, which can provide an effective guide to realizing NOx emission reduction and accurate ammonia injection.

Multivariate time series based variable prediction in air separation process

Ultra-early Prediction of the Process Parameters of Coal Chemical Production

Application of Multi-variable Model Predictive Control to Automatic Variable-loading of Air Separation Unit

On-line diagnosis of abnormal conditions of air separation process by dynamic PCA

A Novel Streaming Variational Bayesian Supervised Factor Analysis For Industrial Adaptive Soft Sensor Modeling

A Predictive Adaptive Learning Method for Multivariable Time Series With Mooney Viscosity Prediction as an Application Case

A method for the early prediction of abnormal conditions in chemical processes combined with physical knowledge and the data-driven model

Data Fusion Combined with Echo State Network for Multivariate Time Series Prediction in Complex Electromechanical System

A Divide-and-Rule Combined Learning Method for Truly Multivariate Time Series Prediction

A highly accurate method for forecasting the compressor geometric variable system based on the data-driven method

Multivariable Time Series Prediction for the Icing Process on Overhead Power Transmission Line

Predicting Natural Gas Prices Based on a Novel Hybrid Model with Variational Mode Decomposition

Decomposition-Based Multi-Step Forecasting Model for the Environmental Variables of Rabbit Houses

Multivariate linear-regression variable parameter spatio-temporal zoning model for temperature prediction in steel rolling reheating furnace

A novel multi-step prediction model for process monitoring

A Multi-Indicator Prediction Method for NOx Emission Concentration and Ammonia Escape value for Cement Calciner System

Research on Multi-step Mixed Predictiom Model of Coal Gasifier Furnace Temperature Based on Machine Learning

Multivariate Chaotic Time Series Prediction Based on Improved Grey Relational Analysis

Multi-level recursive adaptive prediction for non-stationary states of stack gas turbines

Compound Autoregressive Network for Prediction of Multivariate Time Series

Multi-Parameter Prediction for Steam Turbine Based on Real-Time Data Using Deep Learning Approaches