Using Neural Network for Diagnostics of an Industrial Gas Turbine

Author(s):  
M. Ghoreyshi ◽  
R. Singh

Gas Turbines are being utilized in increasing numbers for industrial applications because of their increasing in efficiency and reliability. However, they become degraded during operation and their associated maintenance costs can become extremely high for the owners. Hence, successful maintenance techniques are those which are able to reduce maintenance costs and down-time. In recent decades industry has started to use predictive maintenance techniques because of their benefits in reducing down-time compared to traditional techniques like breakdown maintenance, as a result different predictive maintenance and diagnostics techniques have been developed during the last fifteen years. This study, in particular, will focus on performance diagnostic techniques based on Neural Networks. The network features and training algorithms will be discussed to develop an appropriate model for gas turbine diagnostics. In addition, it will be shown how training data can affect training performance. This study follows on from previous work carried out at Cranfield University to develop engine health monitoring techniques; however it will attempt to investigate the different abilities of neural networks for use in industrial gas turbine diagnostics, especially in non-standard ambient temperatures and its advantages compared to Gas Path Analysis (GPA) techniques.

1978 ◽  
Author(s):  
A. W. T. Mottram

The industrial gas turbine requires continual improvement in order to increase output and efficiency, to extend its life and to meet fresh environmental requirements. In the compact industrial gas turbine, derived from the aircraft engine, the required improvements are achieved in three ways: (a) new features are incorporated which have been developed to meet the specific requirements of industrial applications, (b) technical improvements developed initially for aircraft engines are applied to existing industrial engines, and (c) new engines developed for aircraft and to a higher technical standard are introduced into industrial service. This paper describes recent improvements to Rolls-Royce compact industrial gas turbines with particular reference to the Olympus C and Olympus 593.


1983 ◽  
Vol 105 (3) ◽  
pp. 417-421 ◽  
Author(s):  
B. Becker ◽  
M. Kwasniewski ◽  
O. von Schwerdtner

With increasing mass flow and constant rotational speed of the single shaft gas turbine, the diameters and tip speeds of compressor and turbine blading have to be enlarged. A significant further increase in mass flow can be achieved with transonic compressor stages, as they have been in service in aero gas turbines for many years. For industrial applications, weight and stage pressure ratio are not nearly as important as efficiency. Therefore, different design criteria had to be applied, which led to a moderate front stage pressure ratio of 1.5 with a rotor tip inlet Mach number of 1.37 and a high solidity blading. In order to simulate the first three stages of a 200-MW gas turbine, a test compressor scaled by 1:5.4 was built and tested. These measurements confirmed the aerodynamic performance in the design point very well. The compressor map showed a satisfactory part speed behavior. These results prove that the single-shaft industrial gas turbine still has a high development potential with respect to power increase. Additionally, with the higher pressure ratio, the cycle efficiency will be improved considerably.


Author(s):  
G. Rocha ◽  
C. J. Etheridge

Solar Turbines Incorporated has extended the size-range of its’ turbomachinery products with development of the Titan™ 130 industrial gas turbine. The 13 megawatt-class, simple-cycle machine is designed to produce 13.3 MW (17,800 hp) output power with a thermal efficiency of 34.5% at ISO inlet conditions with no losses. The larger gas turbine is intended to meet increasing market demands in gas compression and mechanical pump-drive industrial applications. The overall engine design is based on aerodynamic-scale of the 7 MW-class Taurus™ 70 gas turbine with similar operating cycle parameters. The engine configuration consists of hardware that has been scaled from the Taurus 70 and components that are common with the Mars® gas turbine. As with the Taurus 70 and Mars products, the Titan 130 gas turbine features a low emissions combustion system based on Solar’s proven dry, lean-premix, pollution-prevention technology. The enhance system is capable of reducing pollutant emissions over an extended operating range from part-load to full-load conditions. This paper discusses the evolutionary design of the Titan 130 from the Taurus 70 and Mars gas turbines products. Descriptions of the basic configuration, component scaling techniques and modular design construction are presented.


Author(s):  
Glenn McAndrews

Electric starter development programs have been the subject of ASME technical papers for over two decades. Offering significant advantages over hydraulic or pneumatic starters, electric starters are now poised to be the preferred choice amongst gas turbine customers. That they are not now the dominant starter in the field after decades of investment and experimentation is attributable to many factors. As with any new technology, progress is often unsteady, depending on budgets, market conditions, customer buy-in, etc. Additionally, technological advances in the parent technologies, in this case electric motors, can abruptly and rapidly change, further disturbing the best laid introduction plans. It is therefore not too surprising that only recently, is the industry beginning to see the deployment of electric starters on production gas turbines. The earliest adoption occurred on smaller gas turbine units, generally less than 10 MW in power. More recently, gas turbines greater than 10 MWs are being sold with electric starters. The authors expect that regardless of their size or fuel supply, most all future gas turbine users will opt for electric starters. This may even include the “larger” frame machines with power greater than 100 MW. Starting with some past history, this paper will not only summarize past development efforts, it will attempt to examine the current deployment of electric starters throughout the marine and industrial gas turbine landscapes. The large-scale acceptance of electric start systems for both new production and retrofit will depend on the favorable cost/benefit assessment when weighing both first cost and life cycle cost. The current and intense activity in electric vehicle applications is giving rise to even more power dense motors. The paper will look at some of these exciting applications, the installed products, and the technologies behind the products. To what extent these new products may serve the needs of the gas turbine community will be the central question this paper attempts to answer.


Author(s):  
John Hartranft ◽  
Bruce Thompson ◽  
Dan Groghan

Following the successful development of aircraft jet engines during World War II (WWII), the United States Navy began exploring the advantages of gas turbine engines for ship and boat propulsion. Early development soon focused on aircraft derivative (aero derivative) gas turbines for use in the United States Navy (USN) Fleet rather than engines developed specifically for marine and industrial applications due to poor results from a few of the early marine and industrial developments. Some of the new commercial jet engine powered aircraft that had emerged at the time were the Boeing 707 and the Douglas DC-8. It was from these early aircraft engine successes (both commercial and military) that engine cores such as the JT4-FT4 and others became available for USN ship and boat programs. The task of adapting the jet engine to the marine environment turned out to be a substantial task because USN ships were operated in a completely different environment than that of aircraft which caused different forms of turbine corrosion than that seen in aircraft jet engines. Furthermore, shipboard engines were expected to perform tens of thousands of hours before overhaul compared with a few thousand hours mean time between overhaul usually experienced in aircraft applications. To address the concerns of shipboard applications, standards were created for marine gas turbine shipboard qualification and installation. One of those standards was the development of a USN Standard Day for gas turbines. This paper addresses the topic of a Navy Standard Day as it relates to the introduction of marine gas turbines into the United States Navy Fleet and why it differs from other rating approaches. Lastly, this paper will address examples of issues encountered with early requirements and whether current requirements for the Navy Standard Day should be changed. Concerning other rating approaches, the paper will also address the issue of using an International Organization for Standardization, that is, an International Standard Day. It is important to address an ISO STD DAY because many original equipment manufacturers and commercial operators prefer to rate their aero derivative gas turbines based on an ISO STD DAY with no losses. The argument is that the ISO approach fully utilizes the power capability of the engine. This paper will discuss the advantages and disadvantages of the ISO STD DAY approach and how the USN STD DAY approach has benefitted the USN. For the future, with the advance of engine controllers and electronics, utilizing some of the features of an ISO STD DAY approach may be possible while maintaining the advantages of the USN STD DAY.


Author(s):  
T. L. Ragland

After industrial gas turbines have been in production for some amount of time, there is often an opportunity to improve or “uprate” the engine’s output power or cycle efficiency or both. In most cases, the manufacturer would like to provide these uprates without compromising the proven reliability and durability of the product. Further, the manufacturer would like the development of this “Uprate” to be low cost, low risk and result in an improvement in “customer value” over that of the original design. This paper describes several options available for enhancing the performance of an existing industrial gas turbine engine and discusses the implications for each option. Advantages and disadvantages of each option are given along with considerations that should be taken into account in selecting one option over another. Specific options discussed include dimensional scaling, improving component efficiencies, increasing massflow, compressor zero staging, increasing firing temperature (thermal uprate), adding a recuperator, increasing cycle pressure ratio, and converting to a single shaft design. The implications on output power, cycle efficiency, off-design performance engine life or time between overhaul (TBO), engine cost, development time and cost, auxiliary requirements and product support issues are discussed. Several examples are provided where these options have been successfully implemented in industrial gas turbine engines.


Author(s):  
Xueyou Wen ◽  
Jiguo Zou ◽  
Zheng Fu ◽  
Shikang Yu ◽  
Lingbo Li

Steam-injected gas turbines have a multitude of advantages, but they suffer from the inability to recover precious demineralized water. The present paper describes the test conditions and results of steam injection along with an attempt to achieve water recovery, which were obtained through a series of tests conducted on a S1A-02 small-sized industrial gas turbine. A water recovery device incorporating a compact finned spiral plate cooling condenser equipped with filter screens has been designed for the said gas turbine and a 100% water recovery (based on the design point) was attained.


Author(s):  
Thomas Wagner ◽  
Robert J. Burke

The desire to maintain power plant profitability, combined with current market fuel gas pricing is forcing power generation companies to constantly look for ways to keep their industrial gas turbine units operating at the highest possible efficiency. Gas Turbines Operation requires the compression of very large quantities of air that is mixed with fuel, ignited and directed into a turbine to produce torque for purposes ranging from power generation to mechanical drive of pumping systems to thrust for air craft propulsion. The compression of the air for this process typically uses 60% of the required base energy. Therefore management of the compression process efficiency is very important to maintain overall cycle efficiency. Since fouling of turbine compressors is almost unavoidable, even with modern air filter treatment, and over time results in lower efficiency and output, compressor cleaning is required to maintain gas turbine efficiency.


1978 ◽  
Vol 100 (4) ◽  
pp. 704-710
Author(s):  
Ch. Just ◽  
C. J. Franklin

The need for a thorough and systematic standard evaluation program for new materials for modern industrial gas turbines is shown by several examples and facts. A complete list of the data required by the designer of an industrial gas turbine is given, together with comments to some of the more important properties. A six-phase evaluation program is described which minimizes evaluation time, cost, and the risk of introducing a new material.


2021 ◽  
Author(s):  
Enzo Losi ◽  
Mauro Venturini ◽  
Lucrezia Manservigi ◽  
Giuseppe Fabio Ceschini ◽  
Giovanni Bechini ◽  
...  

Abstract A gas turbine trip is an unplanned shutdown, of which the most relevant consequences are business interruption and a reduction of equipment remaining useful life. Thus, understanding the underlying causes of gas turbine trip would allow predicting its occurrence in order to maximize gas turbine profitability and improve its availability. In the ever competitive Oil & Gas sector, data mining and machine learning are increasingly being employed to support a deeper insight and improved operation of gas turbines. Among the various machine learning tools, Random Forests are an ensemble learning method consisting of an aggregation of decision tree classifiers. This paper presents a novel methodology aimed at exploiting information embedded in the data and develops Random Forest models, aimed at predicting gas turbine trip based on information gathered during a timeframe of historical data acquired from multiple sensors. The novel approach exploits time series segmentation to increase the amount of training data, thus reducing overfitting. First, data are transformed according to a feature engineering methodology developed in a separate work by the same authors. Then, Random Forest models are trained and tested on unseen observations to demonstrate the benefits of the novel approach. The superiority of the novel approach is proved by considering two real-word case-studies, involving filed data taken during three years of operation of two fleets of Siemens gas turbines located in different regions. The novel methodology allows values of Precision, Recall and Accuracy in the range 75–85 %, thus demonstrating the industrial feasibility of the predictive methodology.


Sign in / Sign up

Export Citation Format

Share Document