Prediction algorithm of operation abnormality for automatic trouble avoidance system of electric tractor

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K06319
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 41040:Agricultural environmental engineering and agricultural information engineering-related
• Research Institution: Ehime University
• Principal Investigator: Ueka Yuko 愛媛大学, 農学研究科, 准教授 (00527103)
• Co-Investigator(Kenkyū-buntansha): 松井 正実 宇都宮大学, 農学部, 教授 (10603425)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 電動農機 / 異常検知 / 自動化

Outline of Final Research Achievements

In this study, in order to construct an automatic trouble avoidance system for agricultural machinery, an electric tractor that can directly measure power consumption by operating each configuration unit with independent motors was developed.
The changes in travel load and workload can be detected from the changes in power consumption data, and the possibility of predicting from power consumption data of underload conditions such as tire slippage has been confirmed. In addition, the work condition by analyzing the power consumption data of wind winnowing unit of the combine harvester using the analysis method of anomaly detection and change detection was found. In the future, this technology for predicting accidents and monitoring the working conditions of various types of electric agricultural machinery is going to be developed.

Free Research Field

農業環境工学および農業情報工学

Academic Significance and Societal Importance of the Research Achievements

電動化の特性を発揮し，モータ消費電力データから機械負荷異常領域への変化点を抽出し，外乱の多い農業機械の自動制御システムへ応用利用するためのアルゴリズムを構築しようとする点にある．また，コンバインのような複数の作業機能を有する農業機械に展開することで，作業精度が向上するとともに，対象作物多様化のニーズにも資するものと考えている．
電動農業機械については，国際会議などでも近年議論されるようにはなったが，実用化に至っているのは小型機械ばかりである．トラクタ・コンバインなどの主力農業機械の電動化に向けて，本研究成果は，効率的な動力分配可能な新しい機構や制御システムの創造に寄与するものとなる．