RESUMO
Gas turbine vibration data may exhibit considerable differences under time-varying conditions, which poses challenges for neural network anomaly detection. We first propose a framework for a gas turbine vibration frequency spectra process under time-varying operation conditions, assisting neural networks' ability to capture weak information. The framework involves scaling spectra for aligning all frequency components related to rotational speed and normalizing frequency amplitude in a self-adaptive way. Degressive beta variational autoencoder is employed for learning spectra characteristics and anomaly detection, while a multi-category anomaly index is proposed to accommodate various operating conditions. Finally, a dataset of blade Foreign Object Damage (FOD) fault occurring under time-varying operating conditions was used to validate the framework and anomaly detection. The results demonstrate that the proposed method can effectively reduce the spectra differences under time-varying conditions, and also detect FOD fault during operation, which are challenging to identify using conventional methods.
RESUMO
Gas turbines are thermoelectric plants with various applications, such as large-scale electricity production, petrochemical industry, and steam generation. In order to optimize the operation of a gas turbine, it is necessary to develop system identification models that allow for the development of studies and analyses to increase the system's reliability. Current strategies for modeling complex and non-linear systems can be based on artificial intelligence techniques, using autoregressive neural networks of the NARX and LSTM type. In this context, this work aims to develop a model of a gas turbine capable of estimating the rotation speed of the turbine and simultaneously estimating the uncertainty associated with the estimation. These methodologies are based on artificial neural networks and the Monte Carlo dropout simulation method. The results were obtained from experimental data from a 215 MW gas turbine, getting the best model with a MAPE of 0.02% and an uncertainty associated with the turbine rotation speed of 2.2 RPM.
RESUMO
Adopting biomass energy as an alternative to fossil fuels for electricity production presents a viable strategy to address the prevailing energy deficits and environmental concerns, although it faces challenges related to suboptimal energy efficiency levels. This study introduces a novel combined cooling and power (CCP) system, incorporating an externally fired gas turbine (EFGT), steam Rankine cycle (SRC), absorption refrigeration cycle (ARC), and organic Rankine cycle (ORC), aimed at boosting the efficiency of biomass integrated gasification combined cycle systems. Through the development of mathematical models, this research evaluates the system's performance from both thermodynamic and exergoeconomic perspectives. Results show that the system could achieve the thermal efficiency, exergy efficiency, and levelized cost of exergy (LCOE) of 70.67%, 39.13%, and 11.67 USD/GJ, respectively. The analysis identifies the combustion chamber of the EFGT as the component with the highest rate of exergy destruction. Further analysis on parameters indicates that improvements in thermodynamic performance are achievable with increased air compressor pressure ratio and gas turbine inlet temperature, or reduced pinch point temperature difference, while the LCOE can be minimized through adjustments in these parameters. Optimized operation conditions demonstrate a potential 5.7% reduction in LCOE at the expense of a 2.5% decrease in exergy efficiency when compared to the baseline scenario.
RESUMO
Machine learning algorithms and the increasing availability of data have radically changed the way how decisions are made in today's Industry. A wide range of algorithms are being used to monitor industrial processes and predict process variables that are difficult to be measured. Maintenance operations are mandatory to tackle in all industrial equipment. It is well known that a huge amount of money is invested in operational and maintenance actions in industrial gas turbines (IGTs). In this paper, two variations of autoencoders were used to analyse the performance of an IGT after major maintenance. The data used to analyse IGT conditions were ambient factors, and measurements were performed using several sensors located along the compressor. The condition assessment of the industrial gas turbine compressor revealed significant changes in its operation point after major maintenance; thus, this indicates the need to update the internal operating models to suit the new operational mode as well as the effectiveness of autoencoder-based models in feature extraction. Even though the processing performance was not compromised, the results showed how this autoencoder approach can help to define an indicator of the compressor behaviour in long-term performance.
RESUMO
The power demand from gas turbines in electrical grids is becoming more dynamic due to the rising demand for power generation from renewable energy sources. Therefore, including the transient data in the fault diagnostic process is important when the steady-state data are limited and if some component faults are more observable in the transient condition than in the steady-state condition. This study analyses the transient behaviour of a three-shaft industrial gas turbine engine in clean and degraded conditions with consideration of the secondary air system and variable inlet guide vane effects. Different gas path faults are simulated to demonstrate how magnified the transient measurement deviations are compared with the steady-state measurement deviations. The results show that some of the key measurement deviations are considerably higher in the transient mode than in the steady state. This confirms the importance of considering transient measurements for early fault detection and more accurate diagnostic solutions.
RESUMO
The application of identification techniques using artificial intelligence to the gas turbine (GT), whose nonlinear dynamic behavior is difficult to describe through differential equations and the laws of physics, has begun to gain importance for a little more than a decade. NARX (Nonlinear autoregressive network with exogenous inputs) is one of the models used to identify GT because it provides good results. However, existing studies need to show a systematic method to generate robust NARX models that can identify a GT with satisfactory accuracy. In this sense, a systematic method is proposed to design NARX models for identifying a GT, which consists of nine precise steps that go from identifying GT variables to obtaining the optimized NARX model. To validate the method, it was applied to a case study of a 215 MW SIEMENS TG, model SGT6-5000F, using a set of 2305 real-time series data records, obtaining a NARX model with an MSE of 1.945 × 10-5, RMSE of 0.4411% and a MAPE of 0.0643.
RESUMO
Dry-Low Emission (DLE) technology significantly reduces the emissions from the gas turbine process by implementing the principle of lean pre-mixed combustion. The pre-mix ensures low nitrogen oxides (NOx) and carbon monoxide (CO) production by operating at a particular range using a tight control strategy. However, sudden disturbances and improper load planning may lead to frequent tripping due to frequency deviation and combustion instability. Therefore, this paper proposed a semi-supervised technique to predict the suitable operating range as a tripping prevention strategy and a guide for efficient load planning. The prediction technique is developed by hybridizing Extreme Gradient Boosting and K-Means algorithm using actual plant data. Based on the result, the proposed model can predict the combustion temperature, nitrogen oxides, and carbon monoxide concentration with an accuracy represented by R squared value of 0.9999, 0.9309, and 0.7109, which outperforms other algorithms such as decision tree, linear regression, support vector machine, and multilayer perceptron. Further, the model can identify DLE gas turbine operation regions and determine the optimum range the turbine can safely operate while maintaining lower emission production. The typical DLE gas turbine's operating range can operate safely is found at 744.68 °C -829.64 °C. The proposed technique can be used as a preventive maintenance strategy in many applications involving tight operating range control in mitigating tripping issues. Furthermore, the findings significantly contribute to power generation fields for better control strategies to ensure the reliable operation of DLE gas turbines.
RESUMO
Thermoelectric (TE) waste heat recovery has attracted significant attention over the past decades, owing to its direct heat-to-electricity conversion capability and reliable operation. However, methods for application-specific, system-level TE design have not been thoroughly investigated. This work provides detailed design optimization strategies and exergy analysis for TE waste heat recovery systems. To this end, we propose the use of TE system equipped on the exhaust of a gas turbine power plant for exhaust waste heat recovery and use it as a case study. A numerical tool has been developed to solve the coupled charge and heat current equations with temperature-dependent material properties and convective heat transfer at the interfaces with the exhaust gases at the hot side and with the ambient air at the heat sink side. Our calculations show that at the optimum design with 50% fill factor and 6 mm leg thickness made of state-of-the-art Bi2Te3 alloys, the proposed system can reach power output of 10.5 kW for the TE system attached on a 2 m-long, 0.5 × 0.5 m2-area exhaust duct with system efficiency of 5% and material cost per power of 0.23 $/W. Our extensive exergy analysis reveals that only 1% of the exergy content of the exhaust gas is exploited in this heat recovery process and the exergy efficiency of the TE system can reach 8% with improvement potential of 85%.
RESUMO
A gas path analysis approach of dynamic modelling was used to examine the gas turbine performance. This study presents an investigation of the effect of physical faults on the performance of a three-shaft gas turbine at full-load and part-load operation. A nonlinear steady state performance model was developed and validated. The datasheet from the engine manufacturer was used to gather the input and validation data. Some engineering judgement and optimization were used. Following validation of the engine performance model with the engine manufacturer data using physical fault and component health parameter relationships, physical faults were implanted into the performance model to evaluate the performance characteristics of the gas turbine at degradation state at full- and part-load operation. The impact of erosion and fouling on the gas turbine output parameters, component measurement parameters, and the impact of degraded components on another primary component of the engine have been investigated. The simulation results show that the deviation in the output parameters and component isentropic efficiency due to compressor fouling and erosion is linear with the load variation, but it is almost nonlinear for the downstream components. The results are discussed following the plots.
Assuntos
Dinâmica não Linear , Simulação por ComputadorRESUMO
In this article, we study the possibility of gas turbine unit (GTU) monitoring using interferometric fiber optic sensors. We used the Mach-Zehnder interferometer (MZI) scheme, which can be easily implemented and simply installed on the turbine, and also allows us to solve the problem of phase unwrapping conveniently. In this research, the following main steps were carried out: an experimental scheme based on the MZI was assembled, and its sensitive arm was fixed on the GTU under study; data on various operation modes of the GTU was collected; the data were subjected to frequency FFT analysis, based on which the main rotational speeds of the turbine were identified. With FFT analysis, we also demonstrated multiples harmonics, which appear in the case of GTU after operating time, caused by the number of blades. The possibility of GTU monitoring and analysis using a non-invasive compact fiber-optic sensor is demonstrated: spectral analysis is used to detect the rotor speed, as well as the presence or absence of high-order multiple frequencies indicating blade and bearing defects, which are determined by the number of GTU's blades and rolling bearing used as turbines rotor supports.
RESUMO
The gas turbine was one of the most important technological developments of the early 20th century, and it has had a significant impact on our lives. Although some researchers have worked on predicting the performance of three-shaft gas turbines, the effects of the deteriorated components on other primary components and of the physical faults on the component measurement parameters when considering the variable inlet guide valve scheduling and secondary air system for three-shaft gas turbine engines have remained unexplored. In this paper, design point and off-design performance models for a three-shaft gas turbine were developed and validated using the GasTurb 13 commercial software. Since the input data were limited, some engineering judgment and optimization processes were applied. Later, the developed models were validated using the engine manufacturer's data. Right after the validation, using the component health parameters, the physical faults were implanted into the non-linear steady-state model to investigate the performance of the gas turbine during deterioration conditions. The effects of common faults, namely fouling and erosion in primary components of the case study engine, were simulated during full-load operation. The fault simulation results demonstrated that as the severity of the fault increases, the component performance parameters and measurement parameters deviated linearly from the clean state. Furthermore, the sensitivity of the measurement parameters to the fault location and type were discussed, and as a result they can be used to determine the location and kind of fault during the development of a diagnosis model.
RESUMO
On the basis of the established irreversible simple closed gas turbine cycle model, this paper optimizes cycle performance further by applying the theory of finite-time thermodynamics. Dimensionless efficient power expression of the cycle is derived. Effects of internal irreversibility (turbine and compressor efficiencies) and heat reservoir temperature ratio on dimensionless efficient power are analyzed. When total heat conductance of two heat exchangers is constant, the double maximum dimensionless efficient power of a cycle can be obtained by optimizing heat-conductance distribution and cycle pressure-ratio. Through the NSGA-II algorithm, multi-objective optimizations are performed on the irreversible closed gas turbine cycle by taking five performance indicators, dimensionless power density, dimensionless ecological function, thermal efficiency, dimensionless efficient power and dimensionless power output, as objective functions, and taking pressure ratio and heat conductance distribution as optimization variables. The Pareto frontiers with the optimal solution set are obtained. The results reflect that heat reservoir temperature ratio and compressor efficiency have greatest influences on dimensionless efficient power, and the deviation indexes obtained by TOPSIS, LINMAP and Shannon Entropy decision-making methods are 0.2921, 0.2921, 0.2284, respectively, for five-objective optimization. The deviation index obtained by Shannon Entropy decision-making method is smaller than other decision-making methods and its result is more ideal.
RESUMO
This paper addresses a design optimization of a gas turbine (GT) for marine applications. A gain-scheduling method incorporating a meta-heuristic optimization is proposed to optimize a thermodynamics-based model of a small GT engine. A comprehensive control system consisting of a proportional integral (PI) controller with additional proportional gains, gain scheduling, and a min-max controller is developed. The modeling of gains as a function of plant variables is presented. Meta-heuristic optimizations, namely a genetic algorithm (GA) and a whale optimization algorithm (WOA), are applied to optimize the designed control system. The results show that the WOA has better performance than that of the GA, where the WOA exhibits the minimum fitness value. Compared to the unoptimized gain, the time to reach the target of the power lever angle is significantly reduced. Optimal gain scheduling shows a stable response compared with a fixed gain, which can have oscillation effects as a controller responds. An effect of using bioethanol as a fuel has been observed. It shows that for the same input parameters of the GT dynamics model, the fuel flow increases significantly, as compared with diesel fuel, because of its low bioethanol heating value. Thus, a significant increase occurs only at the gain that depends on the fuel flow.
RESUMO
Operational modes of a process are described by a number of relevant features that are indicative of the state of the process. Hundreds of sensors continuously collect data in industrial systems, which shows how the relationship between different variables changes over time and identifies different modes of operation. Gas turbines' operational modes are usually defined regarding their expected energy production, and most research works either are focused a priori on obtaining these modes solely based on one variable, the active load, or assume a fixed number of states and build up predictive models to classify new situations as belonging to the predefined operational modes. However, in this work, we take into account all available parameters based on sensors' data because other factors can influence the system status, leading to the identification of a priori unknown operational modes. Furthermore, for gas turbine management, a key issue is to detect these modes using a real-time monitoring system. Our approach is based on using unsupervised machine learning techniques, specifically an ensemble of clusters to discover consistent clusters, which group data into similar groups, and to generate in an automatic way their description. This description, upon interpretation by experts, becomes identified and characterized as operational modes of an industrial process without any kind of a priori bias of what should be the operational modes obtained. Our proposed methodology can discover and identify unknown operational modes through data-driven models. The methodology was tested in our case study with Siemens gas turbine data. From available sensors' data, clusters descriptions were obtained in an automatic way from aggregated clusters. They improved the quality of partitions tuning one consistency parameter and excluding outlier clusters by defining filtering thresholds. Finally, operational modes and/or sub-operational modes were identified with the interpretation of the clusters description by process experts, who evaluated the results very positively.
RESUMO
Maintenance is the process of preserving the good condition of a system to ensure its reliability and availability to perform specific operations. The way maintenance is nowadays performed in industry is changing thanks to the increasing availability of data and condition assessment methods. Soft sensors have been widely used over last years to monitor industrial processes and to predict process variables that are difficult to measured. The main objective of this study is to monitor and evaluate the condition of the compressor in a particular industrial gas turbine by developing a soft sensor following an autoencoder architecture. The data used to monitor and analyze its condition were captured by several sensors located along the compressor for around five years. The condition assessment of an industrial gas turbine compressor reveals significant changes over time, as well as a drift in its performance. These results lead to a qualitative indicator of the compressor behavior in long-term performance.
RESUMO
The interaction between the film-cooling jet and vortex structures in the turbine passage plays an important role in the endwall cooling design. In this study, a simplified topology of a blunt body with a half-cylinder is introduced to simulate the formation of the leading-edge horseshoe vortex, where similarity compared with that in the turbine cascade is satisfied. The shaped cooling hole is located in the passage. With this specially designed model, the interaction mechanism between the cooling jet and the passage vortex can therefore be separated from the crossflow and the pressure gradient, which also affect the cooling jet. The loss-analysis method based on the entropy generation rate is introduced, which locates where losses of the cooling capacity occur and reveals the underlying mechanism during the mixing process. Results show that the cooling performance is sensitive to the hole location. The injection/passage vortex interaction can help enhance the coolant lateral coverage, thus improving the cooling performance when the hole is located at the downwash region. The coolant is able to conserve its structure in that, during the interaction process, the kidney vortex with the positive rotating direction can survive with the negative-rotating passage vortex, and the mixture is suppressed. However, the larger-scale passage vortex eats the negative leg of the kidney vortices when the cooling hole is at the upwash region. As a result, the coolant is fully entrained into the main flow. Changes in the blowing ratio alter the overall cooling effectiveness but have a negligible effect on the interaction mechanism. The optimum blowing ratio increases when the hole is located at the downwash region.
RESUMO
Generally, industrial gas turbines (IGT) face transient behavior during start-up, load change, shutdown and variations in ambient conditions. These transient conditions shift engine thermal equilibrium from one steady state to another steady state. In turn, various aero-thermal and mechanical stresses are developed that are adverse for engine's reliability, availability, and overall health. The transient behavior needs to be accurately predicted since it is highly related to low cycle fatigue and early failures, especially in the hot regions of the gas turbine. In the present paper, several critical aspects related to transient behavior and its modeling are reviewed and studied from the point of view of identifying potential research gaps within the context of fault detection and diagnostics (FDD) under dynamic conditions. Among the considered topics are, (i) general transient regimes and pertinent model formulation techniques, (ii) control mechanism for part-load operation, (iii) developing a database of variable geometry inlet guide vanes (VIGVs) and variable bleed valves (VBVs) schedules along with selection framework, and (iv) data compilation of shaft's polar moment of inertia for different types of engine's configurations. This comprehensive literature document, considering all the aspects of transient behavior and its associated modeling techniques will serve as an anchor point for the future researchers, gas turbine operators and design engineers for effective prognostics, FDD and predictive condition monitoring for variable geometry IGT.
RESUMO
The creation and exploitation of gas turbine engines (GTE) often involve two mutually exclusive tasks related to ensuring the highest reliability while achieving a good economic and environmental performance of the power plant. The value of the radial clearance between the blade tips of the compressor or turbine and the stator is a parameter that has a significant impact on the efficiency and safety of the GTE. However, the radial displacements that form tip clearances are only one of the components of the displacements made by GTE elements due to the action of power loads and thermal deformations during engines' operation. The impact of loads in conjunction with natural aging is also the reason for the wear of the GTE's structural elements (for example, bearing assemblies) and the loss of their mechanical strength. The article provides an overview of the methods and tools for monitoring the dangerous states of the GTE (blade tips clearances, impellers and shafts displacements, debris detecting in lubrication system) based on the single-coil eddy current sensor, which remains operational at the temperatures above 1200 °C. The examples of practical application of the systems with such sensors in bench tests of the GTE are given.
RESUMO
The main objective of this paper is to present and analyze an innovative configuration of integrated solar combined cycle (ISCC). As novelties, the plant includes a recuperative gas turbine and the conventional bottoming Rankine cycle is replaced by a recently developed double recuperative double expansion (DRDE) cycle. The configuration results in a fuel saving in the combustion chamber at the expense of a decreased exhaust gas temperature, which is just adequate to feed the DRDE cycle that uses propane as the working fluid. The solar contribution comes from a solar field of parabolic trough collectors, with oil as the heat transfer fluid. The optimum integration point for the solar contribution is addressed. The performance of the proposed ISCC-R-DRDE design conditions and off-design operation was assessed (daily and yearly) at two different locations. All results were compared to those obtained under the same conditions by a conventional ISCC, as well as similar configurations without solar integration. The proposed configuration obtains a lower heat rate on a yearly basis in the studied locations and lower levelized cost of energy (LCOE) than that of the ISCC, which indicates that such a configuration could become a promising technology.
RESUMO
Computational Fluid Dynamics (CFD) results are often presented in a deterministic way despite the uncertainties related to boundary conditions, numerical modelling, and discretization error. Uncertainty quantification is the field studying how these phenomena affect the numerical result. With these methods, the results obtained are directly comparable with the experimental ones, for which the uncertainty related to the measurement is always shown. This work presents an uncertainty quantification approach applied to CFD: the test case consists of an industrial prismatic gas turbine vane with standard film cooling shaped holes system on the suction side only. The vane was subject of a previous experimental test campaign which had the objective to evaluate the film cooling effectiveness through pressure-sensitive paint technique. CFD analyses are conducted coherently with the experiments: the analogy between heat and mass transfer is adopted to draw out the adiabatic film effectiveness, solving an additional transport equation to track the concentration of CO2 used as a coolant fluid. Both steady and unsteady simulations are carried out: the first one using a RANS approach with k-ω SST turbulence model the latter using a hybrid LES-RANS approach. Regarding uncertainty quantification, three geometrical input parameters are chosen: the hole dimension, the streamwise inclination angle of the holes, and the inlet fillet radius of the holes. Polynomial-chaos approach in conjunction with the probabilistic collocation method is used for the analysis: a first-order polynomial approximation was adopted which required eight evaluations only. RANS approach is used for the uncertainty quantification analysis in order to reduce the computational cost. Results show the confidence interval for the analysis as well as the probabilistic output. Moreover, a sensitivity analysis through Sobol's indices was carried out which prove how these input parameters contribute to the film cooling effectiveness, in particular, when dealing with the additive manufacturing process.