SumTime-Turbine: A Knowledge-Based System to Communicate Gas Turbine Time-Series Data
نویسندگان
چکیده
SumTime-Turbine produces textual summaries of archived timeseries data from gas turbines. These summaries should help experts understand large data sets that cannot be visually presented in a single graphical display. SumTime-Turbine is based on pattern detection, knowledge-based temporal abstraction (KBTA), and natural language generation (NLG) technology. A prototype version of the system has been implemented and is currently being evaluated.
منابع مشابه
An Approach to Generating Summaries of Time Series Data in the Gas Turbine Domain
In this paper, we propose an approach to generating summaries of time series data in the gas turbine domain using AI techniques. Through the think-aloud method with the aid of visualization of temporal data using time series workbench (TSW), both domain knowledge from experts about how to solve problems in the gas turbine and information about how domain experts analyze the archived temporal da...
متن کاملIGR For GR/M76881/01: Generating Summaries of Time-Series Data (SumTime) Background/Context
Background/Context The goal of the SumTime project was to develop better techniques for generating English summaries of numerical time-series data. The modern world is being flooded with such data. For example, a typical gas-turbine has 250 sensors, each sampling once per minute. This produces 200MB of data per day, which a maintenance engineer may have one hour (per day) to attempt to understa...
متن کاملChoosing the content of textual summaries of large time-series data sets
Natural Language Generation (NLG) can be used to generate textual summaries of numeric data sets. In this paper we develop an architecture for generating short (a few sentences) summaries of large (100KB or more) time-series data sets. The architecture integrates pattern recognition, pattern abstraction, selection of the most significant patterns, microplanning (especially aggregation), and rea...
متن کاملConstrained Model Predictive Control of Low-power Industrial Gas Turbine
Nowadays, extensive research has been conducted for gas turbine engines control due to growing importance of gas turbine engines for different industries and the need to design a suitable control system for a gas turbine as the heart of the industry. In order to design gas turbine control system, various control variables can be used, but in the meantime, fuel flow inserting into combustion cha...
متن کاملPerformance analysis a gas turbine cycle equipped with a double acting type stirling engine in a power generating unit
The aim of this study is to investigate the performance of a gas turbine cycle equipped with a Stirling engine from the thermodynamic point of view. In this system, part of the heat loss from the gas turbine is transmitted to a Stirling engine to generate more power. In the proposed system analysis, the governing equations of the hybrid cycle are modeled in MATLAB software and Schmidt and ideal...
متن کامل