METHOD AND DEVICE FOR PROCESS STATE RECOGNITION, DATA STORAGE DEVICE, AND LEARNING DEVICE

• Main Authors: UEKUSA MAKOTO, YAEGASHI OSAMU, TERASAKI TAKESHI
• Format: Patent
• Language: English
• Published: 03.12.1996
• Edition: 6
• Subjects: ANALOGUE COMPUTERS; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PHYSICS; REGULATING

PURPOSE: To facilitate the recognition of variation tendency of process data and their corrections, alteration, and qualitative expression without requiring special data analysis or expert's knowledge to monitor the state of a process. CONSTITUTION: The method and devices are provided with a processor 100 which performs a feature extracting processing wherein a feature vector sequence is generated on the basis of time variation tendency of time-series processing data, and vector retrieving processing for retrieving a reference vector similar to a feature vector, a quantitative expression retrieving processing for retrieving corresponding qualitative expression by regarding a sequence of the class corresponding to the reference vector as a feature vector class sequence, a processing for outputting the qualitative expression, etc., an input device 200 which performs input operation accompanying respective processing operations, an output device 300 which is used for an output processing, a reference vector dictionary 510 wherein the reference vector is stored together with the corresponding class, and a quantitative expression dictionary 520 wherein the class sequence of the feature vector is stored together with the corresponding qualitative expression.