公开(公告)号：US20240143872A1

公开(公告)日：2024-05-02

申请号：US18407170

申请日：2024-01-08

Applicant: BEIJING UNIVERSITY OF TECHNOLOGY

Inventor： Jian TANG ,  JiaKun Chen ,  Heng XIA ,  Junfei QIAO

IPC: G06F30/20 ,  G06F111/10 ,  G06F119/08

CPC classification number: G06F30/20 ,  G06F2111/10 ,  G06F2119/08

Abstract: A simulation analysis system for dioxin concentration in furnace of municipal solid waste incineration process includes an area division module, the area division module is connected with a numerical simulation module, the numerical simulation module is connected with a single-factor analysis module, the single-factor analysis module includes an orthogonal test analysis module, and the orthogonal test analysis module is connected with a control module; the area division module is used for dividing areas in the incinerator, the numerical simulation module is used for conducting modeling simulation on the divided areas, the single-factor analysis module is used for conducting single-factor analysis according to the output of the numerical simulation module, and the orthogonal test analysis module is used for conducting orthogonal test analysis according to the output of the numerical simulation module.

公开(公告)号：US20240419872A1

公开(公告)日：2024-12-19

申请号：US18727294

申请日：2023-04-26

Applicant: Beijing University of Technology

Inventor： Jian TANG ,  Heng XIA ,  CanLin CUI ,  Junfei QIAO

IPC: G06F30/27 ,  G06F111/10

Abstract: The invention provides a soft measurement method for dioxin emission of grate furnace MSWI process based on simplified deep forest regression of residual fitting mechanism. The highly toxic pollutant dioxin (DXN) generated in the solid waste incineration process is a key environmental index which must be subjected to control. The rapid and accurate soft measurement of the DXN emission concentration is an urgent affair for reducing the emission control of the pollutants. The method comprises the following steps: firstly, carrying out feature selection on a high-dimensional process variable by adopting mutual information and significance test; then, constructing a simplified deep forest regression (SDFR) algorithm to learn a nonlinear relationship between the selected process variable and the DXN emission concentration; and finally, designing a gradient enhancement strategy based on a residual error fitting (REF) mechanism to improve the generalization performance of a layer-by-layer learning process. The method is superior to other methods in the aspects of prediction precision and time consumption.

公开(公告)号：US20240302341A1

公开(公告)日：2024-09-12

申请号：US18276179

申请日：2022-10-27

Applicant: BEIJING UNIVERSITY OF TECHNOLOGY

Inventor： Jian TANG ,  Heng XIA ,  Canlin CUI ,  Junfei QIAO

IPC: G01N33/00 ,  G06N20/20

CPC classification number: G01N33/0075 ,  G06N20/20

Abstract: A broad hybrid forest regression (BHFR)-based soft sensor method for DXN emission in a municipal solid waste incineration (MSWI) process, including: based on a broad learning system (BLS) framework, constructing a BHFR soft sensor model for small sample high-dimensional data by replacing a neuron with a non-differential base learner, where the BHFR soft sensor model includes a feature mapping layer, a latent feature extraction layer, a feature incremental layer and an incremental learning layer, and the method includes: mapping a high-dimensional feature; extracting a latent feature from a feature space of a fully connected hybrid matrix, and reducing model complexity and computation consumption based on an information measurement criterion; enhancing a feature representation capacity by training the feature incremental layer based on an extracted latent feature; and constructing the incremental learning layer based on an incremental learning strategy, obtaining a weight matrix with a Moore-Penrose pseudo-inverse, and implementing high-precision modeling.

公开(公告)号：US20230297736A1

公开(公告)日：2023-09-21

申请号：US18201578

申请日：2023-05-24

Applicant: BEIJING UNIVERSITY OF TECHNOLOGY

Inventor： Jian TANG ,  TianZheng WANG ,  Heng XIA ,  Junfei QIAO

IPC: G06F30/27 ,  G06Q10/30

CPC classification number: G06F30/27 ,  G06Q10/30 ,  G06F2111/10

Abstract: A hardware-in-loop simulation experiment platform of multiple input and multiple output loop control for MSWI process includes a real equipment layer and a virtual object layer, where in the real equipment layer and the virtual object layer realize communication through hard wirings and data acquisition cards, the real equipment layer and virtual object layer realize communication in OPC mode through Ethernet; the real equipment layer comprises monitoring equipment and control equipment, and the virtual object layer comprises an MSWI actuator model, an MSWI instrument device model and an MSWI process object model which are respectively operated in different industrial personal computers. The hardware-in-loop simulation experiment platform of multiple input and multiple output loop control for MSWI process provided by the invention is used for providing a reliable engineering verification environment for MSWI process control.

公开(公告)号：US20220092482A1

公开(公告)日：2022-03-24

申请号：US17544213

申请日：2021-12-07

Applicant: Beijing University of Technology

Inventor： Jian TANG ,  Heng XIA ,  Junfei QIAO ,  Zihao GUO

IPC: G06N20/20 ,  G06N5/02

Abstract: A method for predicting dioxin (DXN) emission concentration based on hybrid integration of random forest (RF) and gradient boosting decision tree (GBDT). A random sampling of a training sample and an input feature is performed on a modeling data with a small sample size and a high-dimensional characteristic to generate a training subset. J RF-based DXN sub-models based on the training subset are established. J×I GBDT-based DXN sub-models are established by performing I iterations on each of the RF-based DXN sub-models. Predicted outputs of the RF-based DXN sub-model and the GBDT-based DXN sub-model are combined by a simple average weighting method to obtain a final output.