A Neural Network Simulator of a Gas Turbine With a Waste Heat Recovery Section

2000 ◽  
Author(s):  
C. Boccaletti ◽  
G. Cerri ◽  
B. Seyedan
Keyword(s):  
Neural Network ◽  
Gas Turbine ◽  
Waste Heat ◽  
Direct Method ◽  
Computing Time ◽  
Operating Conditions ◽  
Network Simulator ◽  
Back Propagation Algorithm ◽  
Waste Heat Boiler ◽  
Computing Technique

The objective of the paper is to assess the feasibility of the neural network (NN) approach in power plant process evaluations. A “feedforward” technique with a back propagation algorithm was applied to a gas turbine equipped with waste heat boiler and water heater. Data from physical or empirical simulators of plant components were used to train such a NN model. Results obtained using a conventional computing technique are compared with those of the direct method based on a NN approach. The NN simulator was able to perform calculations in a really short computing time with a high degree of accuracy, predicting various steady-state operating conditions on the basis of inputs that can be easily obtained with existing plant instrumentation. The optimization of NN parameters like number of hidden neurons, training sample size and learning rate is discussed in the paper.

A Neural Network Simulator of a Gas Turbine With a Waste Heat Recovery Section

2001 ◽  
Vol 123 (2) ◽  
pp. 371-376 ◽  
Author(s):  
C. Boccaletti ◽  
G. Cerri ◽  
B. Seyedan
Keyword(s):  
Neural Network ◽  
Gas Turbine ◽  
Waste Heat ◽  
Direct Method ◽  
Computing Time ◽  
Operating Conditions ◽  
Network Simulator ◽  
Back Propagation Algorithm ◽  
Waste Heat Boiler ◽  
Computing Technique

The objective of the paper is to assess the feasibility of the neural network (NN) approach in power plant process evaluations. A “feed-forward” technique with a back propagation algorithm was applied to a gas turbine equipped with waste heat boiler and water heater. Data from physical or empirical simulators of plant components were used to train such a NN model. Results obtained using a conventional computing technique are compared with those of the direct method based on a NN approach. The NN simulator was able to perform calculations in a really short computing time with a high degree of accuracy, predicting various steady-state operating conditions on the basis of inputs that can be easily obtained with existing plant instrumentation. The optimization of NN parameters like number of hidden neurons, training sample size, and learning rate is discussed in the paper.

Application of Neural Network in Monitoring an Industrial Plant for a Complex

2004 ◽  
Author(s):  
Babak Seyedan ◽  
Rory Hynes
Keyword(s):  
Neural Network ◽  
Energy Consumption ◽  
Direct Method ◽  
Computing Time ◽  
Hot Water ◽  
Computational Method ◽  
Energy Utilization ◽  
Industrial Plant ◽  
Back Propagation Algorithm ◽  
Computing Technique

The objective of the paper is to assess the feasibility of the neural network (NN) approach in industrial process facilities. The energy consumption of the plant can be improved by defining suitable operating levels of the various parallel components connected to the plant facility using computerized system. The concept of using a computerized procedure capable of recognizing the status of the equipment from monitoring systems and using that data to automatically optimize the plant operation could lead to significant economic and energy consumption improvements. To demonstrate this goal a “Feed Forward Neural Network” technique with a back propagation algorithm was applied to an existing facility equipped with a cogeneration system based on natural gas engines, hot water boilers, standby boilers and other heat sources. In this paper, the heat capacity of a typical installation is presented and a procedure to optimize energy utilization based on a computational model is developed, the plant existing condition is taken as a reference condition, a general block diagram of the system is presented and discussed and the installation heat load allocation is analyzed. Then the data from the physical model of the facility was used to train such a NN model. Results obtained using a conventional computing technique are compared with those of the direct method based on a NN approach. The NN simulator was capable of performing calculations in a very short computing time with a high degree of accuracy. The optimizations of neural network parameters such as the number of hidden neurons; training sample size and learning rate are discussed in the paper. Trained neural network outputs are compared with those of the computational method and discussed.

Study on the Smart Diagnosis of Gas Turbine With Distributed Computing Technique

2019 ◽  
Author(s):  
Hang Wu ◽  
Jinwei Chen ◽  
Huisheng Zhang
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Gas Turbine ◽  
Short Term Memory ◽  
Computing Time ◽  
Physical Models ◽  
Data Driven ◽  
Computing Technique ◽  
Training Time ◽  
High Level

Abstract Monitoring and diagnosis of a gas turbine is a critical issue in equipment maintenance field. Traditional diagnosis methods are established on the basis of physical models. However, the complexity and degradation of gas turbine limit both comprehensiveness and accuracy of these physical models, making the diagnosis less effective. Therefore, data-driven models are introduced to supplement and revise previous models. Benefitting from the prosperous development of machine learning, neural network has been greatly improved and widely used in various fields of data mining. Three neural networks, Multilayer Perceptron, Convolutional Neural Network and Long Short-term Memory Network are applied in data-driven model establishment. Their training time and prediction accuracy are the two most important factors in judging the effectiveness. An active real time training which means training and predicting simultaneously is applied as the main modelling method for an on-line diagnosis system. Three periods are defined according to the time line: data preparation period, model establishing period and stable prediction period. From the three above neural networks, the most effective data-driven models that corresponding to the last two periods are tested and selected, the purpose is to ensure the high level of accuracy. When high level of accuracy is demanded, neural network always need large computing time and memory space in data learning process. To avoid prediction delay and keep rapid response for the coming fault, distributed training on a 1-master 2-workers computer cluster is designed and applied in this system. Two types of data parallelism are realized on the cluster through Apache Spark and Shell Script for Linux. Comparing with each other and the local training mode, the results shows that dispensing data at first and averaging parameters at last reaches a better outcome both in high accuracy and low training time.

scholarly journals A study of the dependence between fuel consumption of a heat gas turbine and variation of heat loading of regional consumers having various climatic conditions taking into account determination of structural characteristics of heat exchanging equipment for grid water heating

2021 ◽  
Vol 25 (4) ◽  
pp. 478-487
Author(s):  
E. L. Stepanova ◽  
P. V. Zharkov
Keyword(s):  
Gas Turbine ◽  
Fuel Consumption ◽  
Parametric Optimization ◽  
Waste Heat ◽  
Structural Characteristics ◽  
Climatic Conditions ◽  
Operating Conditions ◽  
Waste Heat Boiler ◽  
Heat Loading

The aim was to optimize the dependence between fuel consumption and heat loading of regional consumers varied due to climatic conditions, taking into account the determination of structural characteristics of heat exchanging equipment for grid water heating in a heat gas turbine. A heat gas turbine comprising two fuel combustion chambers, a waste-heat boiler and a contact heat exchanger to heat makeup grid water was investigated. Scheme and parametric optimization studies were carried out using a mathematic model of a gas turbine created using a software and hardware system developed at the Department of Heat Power Systems of the Melentiev Energy Systems Institute, Siberian Branch of the Russian Academy of Sciences. Th turbine operating conditions differing in heat loads in four suggested operating regions were studied. It was found that an increase in fuel consumption in the second combustion chamber was 29%– 84% compared to that in the first combustion chamber. This rise was recorded when the turbine heat loading was increasing in the considered regions. Data analysis of the scheme and parametric optimization studies showed that, for operating conditions with a higher heat loading, it seems reasonable to ensure the maximum possible heating of makeup grid water as the loading rises. It is also recommended to slightly increase the heat surface area of the makeup grid water heater whose structural materials are less expensive than in a waste-heat boiler. It was shown that the suggested technical solution slightly increases specific capital investments while fully providing electrical and heat power to consumers. The obtained results can be used to select optimal technical solutions ensuring competitiveness in the operation of a heat gas turbine in regions with various climatic characteristics.

Initial Operation and Analysis of a Cogeneration Laboratory

2002 ◽  
Author(s):  
Andrew Banta
Keyword(s):  
Performance Analysis ◽  
Gas Turbine ◽  
Waste Heat ◽  
Gas Turbine Engine ◽  
California State University ◽  
State University ◽  
Absorption Chiller ◽  
Waste Heat Boiler ◽  
Turbine Engine ◽  
Thermodynamic Performance

California State University, Sacramento, has constructed and put into service a stand alone cogeneration laboratory. The major components are a 75 kW gas turbine and generator, a waste heat boiler, and a 10 ton absorption chiller. Initial testing has been completed with efforts concentrating on the gas turbine engine and the absorption chiller. A two part thermodynamic performance analysis procedure has been developed to analyze the cogeneration plant. A first law energy balance around the gas turbine determines the heat into the engine. A Brayton cycle analysis of the gas turbine engine is then compared with the measured performance. While this engine is quite small, this method of analysis gives very consistent results and can be applied to engines of all sizes. Careful attention to details is required to obtain agreement between the calculated and measured outputs; typically they are within 10 to 15 percent. In the second part of the performance analysis experimental operation of the absorption chiller has been compared to that specified by the manufacturer and a theoretical cycle analysis. While the operation is within a few percent of that specified by the manufacturer, there are some interesting differences when it is compared to a theoretical analysis.

scholarly journals The Effects of Steam Injection in Combustion Air on the Thermal Efficiency and the Specific Work Output of a Stationary Gas Turbine Plant

1989 ◽  
Author(s):  
K. S. Varma ◽  
Asgharali I. Khandwawala ◽  
S. A. Asif
Keyword(s):  
Gas Turbine ◽  
Waste Heat ◽  
Steam Injection ◽  
Specific Work ◽  
Work Output ◽  
Waste Heat Boiler ◽  
Turbine Plant ◽  
Gas Turbine Plant ◽  
Cycle Efficiency ◽  
The Impact

In the present study a stationary open cycle gas turbine plant, including a thermal regenerator has been theoretically analyzed to assess the impact of steam addition in combustion air, on its performance. the effect of varying steam upto 15% air at different pressure ratios and turbine inlet temperatures have been reported. Mixing of steam in air results in higher values of cycle efficiency and increased specific work output, feasibility to generate steam needed for the purpose in a waste heat boiler have also been studied.

Neural Network Based Predictive Emission Monitoring Module for a GE LM2500 Gas Turbine

2010 ◽  
Author(s):  
K. K. Botros ◽  
M. Cheung
Keyword(s):  
Neural Network ◽  
Gas Turbine ◽  
Operating Conditions ◽  
Training Data ◽  
Pipeline System ◽  
Emission Data ◽  
Ambient Temperatures ◽  
Emission Monitoring ◽  
Data Points ◽  
Monitoring Module

A Predictive Emission Monitoring (PEM) model has been developed for a non-DLE GE LM2500 gas turbine used on a natural gas compressor station on the TransCanada Pipeline System in Alberta. The PEM model is based on an optimized Neural Network (NN) architecture which takes four fundamental engine parameters as input variables. The model predicts NOx emission in ppmv-dry-O2 corrected and in kg/hr as NO2. The NN was trained using Continuous Emission Monitoring (CEM) measurements comprising two sets of actual emission data collected over two different dates in 2009, when the ambient ambient temperatures were vastly different (∼1° C and 24 °C), respectively. These training data were supplemented by other emission data generated by GE ‘Cycle-Deck’ tool to generate emission data at different ambient temperatures ranging from −30 to +30 °C. The outcome is a total of 1872 emission data of engine emissions at different operating conditions covering the range of the engine operating parameters (402 data points from CEM and 1470 data points from GE Cycle-Deck). The PEM model comprises a simple single hidden layer perceptron type NN with only two neurons in it. The performance of the NN-based model showed a correlation coefficient greater than 0.99, and error standard deviation of 4.5 ppmv of NOx and 1.4 kg/hr as NO2. Uncertainty analysis was conducted to assess the effects of uncertainties in the engine parameters on the NOx predictions by PEM. It was shown that for uncertainty in the ambient temperature of ±1 °C, the uncertainty in the NOx prediction is ± 0.9 to ±3.5%. Uncertainties of the order of ±1% in the other three input parameters results in uncertainties in NOx predictions by ±2.5 to ±6%. Finally, the PEM model was implemented in the station CEHM (Compressor Equipment Health Monitoring) system and NOx prediction were reported online on a minutely basis. These data are presented here over the first three months since implementation.

scholarly journals Performance Evaluation of a Combined Heat and Power Plant in the Niger Delta of Nigeria

2019 ◽  
Vol 4 (4) ◽  
pp. 17-23
Author(s):  
Barikuura Gbonee ◽  
Barinyima Nkoi ◽  
John Sodiki
Keyword(s):  
Power Plant ◽  
Gas Turbine ◽  
Thermal Efficiency ◽  
Heat Balance ◽  
Niger Delta ◽  
Heat Recovery ◽  
Waste Heat ◽  
Combined Heat And Power ◽  
Steam Flow ◽  
Waste Heat Boiler

This research presents the performance assessment of a combined heat and power plant operating in the Niger Delta region of Nigeria. The main focus is to evaluate the performance parameters of the gas turbine unit and the waste heat recovery generator section of the combined-heat-and-power plant. Data were gathered from the manufacturer’s manual, field and panel operator’s log sheets and the human machine interface (HMI) monitoring screen. The standard thermodynamic equations were used to determine the appropriate parameters of the various components of the gas turbine power plant as well as that of the heat exchangers of the heat recovery steam generator (HRSG). The outcome of all analysis indicated that for every 10C rise in ambient temperature of the compressor air intake there is an average of 0.146MW drop in the gas turbine power output, a fall of about 0.176% in the thermal efficiency of the plant, a decrease of about 2.46% in the combined-cycle thermal efficiency and an increase of about 0.0323 Kg/Kwh in specific fuel consumption of the plant. In evaluating the performance of the Waste Heat Boiler (WHB), the principle of heat balance above pinch was applied to a single steam pressure HRSG exhaust gas/steam temperature profile versus exhaust heat flow. Hence, the evaporative capacity (steam flow) of the HRSG was computed from the total heat transfer in the super-heaters and evaporator tubes using heat balance above pinch. The analysis revealed that the equivalent evaporation, evaporative capacity (steam flow) and the HRSG thermal efficiency depends on the heat exchanger’s heat load and its effective maintenance.

scholarly journals Phishing Website Detection using Neural Network and PCA based on Feature Selection

2020 ◽  
Vol 8 (6) ◽  
pp. 1150-1152
Keyword(s):  
Neural Network ◽  
Principal Component Analysis ◽  
Feature Selection ◽  
Computing Time ◽  
Back Propagation ◽  
Principal Component ◽  
Criminal Activity ◽  
Back Propagation Algorithm ◽  
Propagation Algorithm ◽  
Network Method

Phishing is a criminal activity that tries to steal user account password or other confidential information by tricking user into believing they are on the actual website. In order to phishing, they must get user to go from an email to a website. User can also land on phishing site by mistyping a URL (web address). However, the numbers of phishing attacks have been growing and need the protection technique. Neural network and Principal Component Analysis (PCA) can be combined to detect phishing website. This study uses back-propagation algorithm based on neural network method and PCA based on feature selection to reduce large attributes into small attributes. Neural network without using PCA will be compared with neural network using PCA. The result shows that neural network using PCA has better accuracy in 55.67% and neural network without using PCA only reaches 54.43% accuracy. However neural network without using PCA has faster computing time than neural network using PCA. This study can be used as a phishing protection technique.

Sign in / Sign up

Export Citation Format

Share Document