Diagnosis of defects on peeled veneers by the fuzzy expert system
Grant-in-Aid for Scientific Research (C).
|Research Institution||University of Tsukuba|
TOCHIGI Toshiro Associate Professor Institute of Agricultural and Forest Engineering University of Tsukuba, 農林工学系, 助教授 (00015661)
|Project Fiscal Year
1992 – 1993
Completed(Fiscal Year 1993)
|Budget Amount *help
¥2,100,000 (Direct Cost : ¥2,100,000)
Fiscal Year 1993 : ¥900,000 (Direct Cost : ¥900,000)
Fiscal Year 1992 : ¥1,200,000 (Direct Cost : ¥1,200,000)
|Keywords||Veneer peeling / Diagnosis / Expert system / 単板切削 / 故障診断 / エキスパートシステム / ベニヤレース / 単板品質 / ファジィ理論|
The quality control and the production control in veneer peeling are invitable procedures. During the peeling of veneers, experienced and skilled lathe oprators usually control the operation of veneer lathes by using their empirical knowledges, to produce high qualities of veneers. They also control and adjust the operation of lathe intuitively and instanteneously, when they detect defects on peeling veneers.
Therefore, objectives in the study were to collect emperical knowledges of skilled lathe operators concering with veneer production, and to build pu the diagnostic technique which is applicable, to predict causes of veneer defects in process. We believe that expert systems are powerful and useful tools to solve these kinds of problems.
Empirical knowledges about the veneer peeling, which lathe operators acquired through their experiences, were collected by mailed questionnaires. The questionnaire was consisted of filling probabilities of causes, in percentage representation, affecte
d veneer defects. We sent the questionn aire to 118 veneer and plywood mills affiliated to the Japan Plywood Manuf acturers' Association.
Since the conventional expert system mostly builds its knowledge base by the production rule, IF-THEN-explanation, has some disadvantages, we developed a new expert system, i.e., Fuzzy Expert System. The difinition of qualities of veneers and the decision of the optimum operational condition is formulated easily and realistically if they are represented by the fuzzy expert system on the comprehension of their vaguenesses.
After the construction of the knowledge base, we examined the operational procedures of the fuzzy expert system under supposed veneer defects to indicated causes and their probabilities to them through the simulation.
In the future study, in order to establish the reliability of the system, its verification is inevitable in the real peeling operation.
Research Output (3results)