| 摘要 |
The invention relates to a method for the computer-assisted monitoring of the operation of a technical system, particularly of an electrical energy-generating installation. In said method, output variables (y(t)) of the technical system are prognosticated using a data-driven model (NN) based on corresponding input variables (x(t)). In addition, a confidence measurement (C(t)) is determined for the respective input variables (x(t)), using one or more density estimators (DE), this measurement being higher the greater the similarity of the input variables (x(t)) to known input variables from training data with which the data-driven model (NN) and the density estimator (DE) are taught. Based on said confidence measurement and prognosticated output variables, an average weighted deviation (E(t)) is determined between the prognosticated output variables (y(t)) and the output variables (y0(t)) actually occurring during operation of the technical system. If the average weighted deviation (y(t)) exceeds a predetermined threshold (ETh) multiple and successive times, an error in operation is detected in the technical system and, in particular, an alarm is issued. The claimed method allows error operational states to be detected in a technical system in a simple and reliable manner. The method is particularly suitable for monitoring the operation of a gas turbine, in which case the degree of compressor efficiency or the regulated exhaust gas temperature are preferably used as output variables, and the input variables preferably comprise the position of guide vanes in the gas turbine compressor, the ambient temperature, and pressures and temperatures or volumetric flows in said compressor. |
