Abstract: This signal processing device comprises: an acquisition unit for acquiring an acoustic signal; a measurement unit for measuring an acoustic level of the acoustic signal for every one of first frequency bands, which are a plurality of frequency bands of a preset first bandwidth; a calculation unit that, on the basis of the plurality of acoustic levels of the first frequency bands, identifies an acoustic feature quantity indicating the separation degree from normal acoustic levels of second frequency bands, which are a plurality of frequency bands of a second bandwidth that is wider than the first bandwidth; a first determination unit for determining whether the acoustic levels measured for every one of the first frequency bands are a first threshold value or greater; and a second determination unit for determining whether the acoustic feature quantity is a second threshold value or greater.

Abstract: An image acquisition unit acquires image data in which an image of a normal monitoring target is captured. An image processing unit generates a plurality of duplicate image data pieces by performing different image processing causing a change in color tone on the image data within a range not exceeding a normal range of the monitoring target. A learning unit trains a model so as to output a value used for determining normality of the monitoring target from the image data, in which the image of the monitoring target is captured, using the plurality of duplicate image data pieces as training data.

Abstract: A state quantity acquiring unit is configured to acquire a state quantity of a turbine. A time calculation unit is configured to calculate an operable time at an over firing operation of the turbine based on a design life of the turbine and a state quantity. A distance calculation unit configured to calculate a Mahalanobis distance based on the state quantity. A determination unit configured to determine whether or not the over firing operation of the turbine is possible by the Mahalanobis distance and the operable time.

Abstract: An image acquisition unit acquires image data in which an image of a normal monitoring target is captured. An image processing unit generates a plurality of duplicate image data pieces by performing different image processing causing a change in color tone on the image data within a range not exceeding a normal range of the monitoring target. A learning unit trains a model so as to output a value used for determining normality of the monitoring target from the image data, in which the image of the monitoring target is captured, using the plurality of duplicate image data pieces as training data.

Abstract: A state quantity acquiring unit is configured to acquire a state quantity of a turbine. A time calculation unit is configured to calculate an operable time at an over firing operation of the turbine based on a design life of the turbine and a state quantity. A distance calculation unit configured to calculate a Mahalanobis distance based on the state quantity. A determination unit configured to determine whether or not the over firing operation of the turbine is possible by the Mahalanobis distance and the operable time.

Abstract: A state quantity acquiring unit is configured to acquire a state quantity of a turbine including a temperature of the turbine. A variable calculation unit is configured to calculate a history variable with respect to a history of the state quantity. A time calculation unit is configured to calculate the operable time at an over firing operation of the turbine based on a history variable corresponding to a design life of the turbine and a calculated history variable.