The Development of a Neural Network Based System for the Optimal Control of Chain-Grate Stoker-Fired Boilers

2000 ◽  
Author(s):  
A. Z. S. Chong ◽  
S. J. Wilcox ◽  
J. Ward
Keyword(s):  
Neural Network ◽  
Steady State ◽  
Pilot Scale ◽  
Combustion Efficiency ◽  
Fine Tuning ◽  
Pollutant Emissions ◽  
Test Facility ◽  
Test Results ◽  
Load Following ◽  
Online Implementation

Abstract A novel Neural Network Based Controller (NNBC) was developed following a comprehensive set of experiments carried out on a pilot-scale stoker test facility at CRE Group Ltd., U.K. The NNBC mimicked the actions of an expert boiler operator, by providing ‘near optimum’ settings of coal feed and air flow, as well as ‘staging’ these parameters during load following conditions, before fine tuning the combustion air under quasi-steady-state conditions. Test results from the online implementation of the NNBC have demonstrated that improved transient and steady-state combustion conditions were attained. The prototype NNBC thus provides both stoker manufacturers and users with a means of reducing pollutant emissions, as well as improving the combustion efficiency of this type of coal firing equipment.

Fuel Cell Gas Turbine Hybrid Simulation Facility Design

2002 ◽  
Author(s):  
David Tucker ◽  
Eric Liese ◽  
John VanOsdol ◽  
Larry Lawson ◽  
Randall S. Gemmen
Keyword(s):  
Power Generation ◽  
Steady State ◽  
Fuel Cell ◽  
Gas Turbine ◽  
Test Facility ◽  
System Response ◽  
Facility Design ◽  
Transient Event ◽  
Load Following ◽  
Transient Events

Fuel cell hybrid power systems have potential for the highest electrical power generation efficiency. Fuel cell gas turbine hybrid systems are currently under development as the first step in commercializing this technology. The dynamic interdependencies resulting from the integration of these two power generation technologies is not well understood. Unexpected complications can arise in the operation of an integrated system, especially during startup and transient events. Fuel cell gas turbine systems designed to operate under steady state conditions have limitations in studying the dynamics of a transient event without risk to the more fragile components of the system. A 250kW experimental fuel cell gas turbine system test facility has been designed at the National Energy Technology Laboratory (NETL), U.S. Department of Energy to examine the effects of transient events on the dynamics of these systems. The test facility will be used to evaluate control strategies for improving system response to transient events and load following. A fuel cell simulator, consisting of a natural gas burner controlled by a real time fuel cell model, will be integrated into the system in place of a real solid oxide fuel cell. The use of a fuel cell simulator in the initial phases allows for the exploration of transient events without risk of destroying an actual fuel cell. Fuel cell models and hybrid system models developed at NETL have played an important role in guiding the design of facility equipment and experimental research planning. Results of certain case studies using these models are discussed. Test scenarios were analyzed for potential thermal and mechanical impact on fuel cell, heat exchanger and gas turbine components. Temperature and pressure drop calculations were performed to determine the maximum impact on system components and design. Required turbine modifications were designed and tested for functionality. The resulting facility design will allow for examination of startup, shut down, loss of load to the fuel cell during steady state operations, loss of load to the turbine during steady state operations and load following.

Neural Network Modeling and Control of Stoker-Fired Boiler Plant

2007 ◽  
Author(s):  
S. M. Thai ◽  
S. J. Wilcox ◽  
A. Z. S. Chong ◽  
J. Ward
Keyword(s):  
Neural Network ◽  
Combustion Process ◽  
Pollutant Emissions ◽  
Neural Network Modeling ◽  
Test Results ◽  
Boiler Plant ◽  
Modeling And Control ◽  
Load Demand ◽  
Industrial Chain ◽  
And Control

The work described in this paper aims to address the development of a Neural Network Based Controller (NNBC) to control chain grate stoker fired boilers. Artificial Neural Networks (ANNs) were used to estimate future emissions from and control the combustion process. The resultant ANNs were able to characterise the dynamics of the process and delivered rational multi-step-ahead predictions wth test data collected at an industrial chain grate stoker at HM Prison Garth, Lancashire. This technique was built into a carefully designed control strategy, to control the industrial stoker. Test results showed that the developed NNBC was able to optimise the industrial stoker boiler plant whilst delivering the load demand required and in so doing, the NNBC also managed to maintain low pollutant emissions.

scholarly journals Experimental Evaluation of a Two-Stage Slagging Combustor Design for a Coal-Fueled Industrial Gas Turbine

1992 ◽  
Author(s):  
L. H. Cowell ◽  
R. T. LeCren
Keyword(s):  
Gas Turbine ◽  
Combustion Zone ◽  
Combustion Process ◽  
Coal Ash ◽  
Pollutant Emissions ◽  
Test Facility ◽  
Inlet Temperature ◽  
Water Mixture ◽  
Test Results ◽  
Industrial Gas Turbine

A full-size combustor for a coal-fueled industrial gas turbine engine has been tested to evaluate combustion performance prior to integration with an industrial gas turbine. The design is based on extensive work completed through one-tenth scale combustion tests. Testing of the combustion hardware is completed with a high pressure air supply in a combustion test facility at the Caterpillar Technical Center. The combustor is a two-staged, rich-lean design. Fuel and air are introduced in the primary combustion zone where the combustion process is initiated. The primary zone operates in a slagging mode inertially removing coal ash from the gas stream. Four injectors designed for coal-water mixture (CWM) atomization are used to introduce the fuel and primary air. In the secondary combustion zone additional air is injected to complete the combustion process at fuel-lean conditions. The secondary zone also serves to reduce the gas temperatures exiting the combustor. The combustor has operated at test pressures of 7 bars with 600K inlet temperature. Tests have been completed to set the air flow split and to map the performance of the combustor as characterized by pollutant emissions, coal ash separation, and temperature profile. Test results with a comparison to subscale test results are discussed. The test results have indicated that the combustor operates at combustion efficiencies above 98% and with pollutant emissions below design goals.

Operational Determination of the Activated Sludge Process Using Neural Networks

1992 ◽  
Vol 26 (9-11) ◽  
pp. 2461-2464 ◽  
Author(s):  
R. D. Tyagi ◽  
Y. G. Du
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Steady State ◽  
Activated Sludge ◽  
Feedforward Neural Network ◽  
Training Data ◽  
Activated Sludge Process

A steady-statemathematical model of an activated sludgeprocess with a secondary settler was developed. With a limited number of training data samples obtained from the simulation at steady state, a feedforward neural network was established which exhibits an excellent capability for the operational prediction and determination.

scholarly journals Combustion Optimization for Coal Fired Power Plant Boilers Based on Improved Distributed ELM and Distributed PSO

Energies ◽  
2019 ◽  
Vol 12 (6) ◽  
pp. 1036 ◽  
Author(s):  
Xinying Xu ◽  
Qi Chen ◽  
Mifeng Ren ◽  
Lan Cheng ◽  
Jun Xie
Keyword(s):  
Power Plant ◽  
Combustion Process ◽  
Optimization Method ◽  
Combustion Efficiency ◽  
Pollutant Emissions ◽  
Combustion Model ◽  
Nox Emissions ◽  
Coupling Characteristics ◽  
Learning Machine ◽  
Combustion Optimization

Increasing the combustion efficiency of power plant boilers and reducing pollutant emissions are important for energy conservation and environmental protection. The power plant boiler combustion process is a complex multi-input/multi-output system, with a high degree of nonlinearity and strong coupling characteristics. It is necessary to optimize the boiler combustion model by means of artificial intelligence methods. However, the traditional intelligent algorithms cannot deal effectively with the massive and high dimensional power station data. In this paper, a distributed combustion optimization method for boilers is proposed. The MapReduce programming framework is used to parallelize the proposed algorithm model and improve its ability to deal with big data. An improved distributed extreme learning machine is used to establish the combustion system model aiming at boiler combustion efficiency and NOx emission. The distributed particle swarm optimization algorithm based on MapReduce is used to optimize the input parameters of boiler combustion model, and weighted coefficient method is used to solve the multi-objective optimization problem (boiler combustion efficiency and NOx emissions). According to the experimental analysis, the results show that the method can optimize the boiler combustion efficiency and NOx emissions by combining different weight coefficients as needed.

Knowledge Transferred Fine-Tuning for Anti-Aliased Convolutional Neural Network in Data-Limited Situation

2021 ◽  
Author(s):  
Satoshi Suzuki ◽  
Shoichiro Takeda ◽  
Ryuichi Tanida ◽  
Hideaki Kimata ◽  
Hayaru Shouno
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Fine Tuning

scholarly journals Fine-tuning of a generative neural network for designing multi-target compounds

2021 ◽  
Author(s):  
Thomas Blaschke ◽  
Jürgen Bajorath
Keyword(s):  
Neural Network ◽  
Small Molecules ◽  
De Novo ◽  
Pharmaceutical Research ◽  
Generative Models ◽  
Fine Tuning ◽  
Data Sets ◽  
Single Target ◽  
Fine Tune ◽  
Target Activity

AbstractExploring the origin of multi-target activity of small molecules and designing new multi-target compounds are highly topical issues in pharmaceutical research. We have investigated the ability of a generative neural network to create multi-target compounds. Data sets of experimentally confirmed multi-target, single-target, and consistently inactive compounds were extracted from public screening data considering positive and negative assay results. These data sets were used to fine-tune the REINVENT generative model via transfer learning to systematically recognize multi-target compounds, distinguish them from single-target or inactive compounds, and construct new multi-target compounds. During fine-tuning, the model showed a clear tendency to increasingly generate multi-target compounds and structural analogs. Our findings indicate that generative models can be adopted for de novo multi-target compound design.

scholarly journals Research on Real-Time Multiple Single Garbage Classification Based on Convolutional Neural Network

2020 ◽  
Vol 2020 ◽  
pp. 1-6
Author(s):  
Jian-ye Yuan ◽  
Xin-yuan Nan ◽  
Cheng-rong Li ◽  
Le-le Sun
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Real Time ◽  
Environmental Science ◽  
Fine Tuning ◽  
Classification Model ◽  
Classification Models ◽  
Data Set ◽  
Classification Tasks ◽  
Accuracy Rates

Considering that the garbage classification is urgent, a 23-layer convolutional neural network (CNN) model is designed in this paper, with the emphasis on the real-time garbage classification, to solve the low accuracy of garbage classification and recycling and difficulty in manual recycling. Firstly, the depthwise separable convolution was used to reduce the Params of the model. Then, the attention mechanism was used to improve the accuracy of the garbage classification model. Finally, the model fine-tuning method was used to further improve the performance of the garbage classification model. Besides, we compared the model with classic image classification models including AlexNet, VGG16, and ResNet18 and lightweight classification models including MobileNetV2 and SuffleNetV2 and found that the model GAF_dense has a higher accuracy rate, fewer Params, and FLOPs. To further check the performance of the model, we tested the CIFAR-10 data set and found the accuracy rates of the model (GAF_dense) are 0.018 and 0.03 higher than ResNet18 and SufflenetV2, respectively. In the ImageNet data set, the accuracy rates of the model (GAF_dense) are 0.225 and 0.146 higher than Resnet18 and SufflenetV2, respectively. Therefore, the garbage classification model proposed in this paper is suitable for garbage classification and other classification tasks to protect the ecological environment, which can be applied to classification tasks such as environmental science, children’s education, and environmental protection.

The Oxide Defect of C70250 Alloy

2014 ◽  
Vol 936 ◽  
pp. 1168-1172 ◽  
Author(s):  
Zhong Ping Chen ◽  
Chao Jian Xiang ◽  
Hua Qing Li
Keyword(s):  
High Temperature Oxidation ◽  
Rolling Process ◽  
Pilot Scale ◽  
Processing Temperature ◽  
Skin Defect ◽  
Test Results ◽  
Temperature Oxidation ◽  
Reducing Atmosphere ◽  
Hot Rolling Process ◽  
Oxide Skin

The oxide skin defect during hot rolling process for Cu-Ni-Si alloy strip was investigated. Oxide skin defects were analyzed by means of alloy elements detection and microstructures characterization. The characterization and test results showed that high temperature oxidation and silicon segregation are the main causes of the oxide skin defect. Pilot scale tests indicated that hot processing temperature for C70250 alloy should be lower than 950°C. Reducing atmosphere is recommended during the thermal treatment of Cu-Ni-Si alloys.

Sign in / Sign up

Export Citation Format

Share Document