Research on optimization of crane fault predictive control system based on data mining

Abstract

To ensure the safe functioning of lifting equipment, a data mining-based optimization study of a crane failure predictive control system is provided. To diagnose lifting machinery faults, the system employs decision tree categorization. Using association rules, a correlation study of hoisting machinery defect and failure was performed. When the minimal confidence and support degree are entered, a total of 244 instances of 18 frequent itemset A9 (safety protection device) may be obtained, indicating that lifting machinery does not perform well in this category. A6 (main parts) and A9 appeared a total of 98 times, with support and confidence of 29.4 and 35.6, respectively, indicating that the main parts can detect that the safety protection device is also having problems. A7 (electrical control system) and A9 appeared a total of 67 times, with support and confidence of 20.1 and 27.3, respectively, indicating that the electrical control system can detect that the safety protection device is also having problems; the correlation between them was also quite large. The system’s feasibility and efficacy shows that it has some application value.

Keywords

• lifting machinery
• data mining
• decision tree classification