Authors

Abstract

With the speedy development of information technology, smart agriculture has become the key to the transformation and upgrading of modern agriculture. To improve the precision and practicality of data analysis, a data analysis model for grain harvesters based on a combination of K-means and Apriori is designed. This model collects grain harvester data in real-time, uses the K-means for preliminary clustering, and integrates Apriori algorithms to dynamically adjust clustering centers to improve clustering accuracy. At the same time, the model introduces ResNet to extract image features of grain harvesters, thereby enhancing the comprehensiveness of data analysis. Comparative experiments show that the algorithm has a high adjusted Rand index of 0.92, an F-value of 0.89, a convergence time of only 12.4 seconds, and a clustering accuracy of 95% for agricultural machinery databases. The analysis of actual operational data of grain harvesters shows that their faults are concentrated in the transmission device, accounting for up to 45%. When operating under high load, the work efficiency drops sharply, and when the rated power exceeds 30%, the work efficiency is only 72%. When working in a wet and muddy environment, the failure rate reaches 42.8%. From the above results, the data analysis model for grain harvesters based on the combination of K-means and Apriori algorithm proposed in the study can perform cluster analysis on the data of grain harvesters, laying a solid foundation for the sustainable development of smart agriculture.

Abstract in English

随着信息技术的飞速发展，智慧农业已成为现代农业转型升级的关键。为了提高数据分析的精度和实用性，设计了一种基于K-means和Apriori相结合的粮食收割机数据分析模型。该模型实时采集粮食收割机数据，采用K-means进行初步聚类，并结合Apriori算法动态调整聚类中心，提高聚类精度。同时，该模型引入ResNet提取粮食收割机图像特征，增强了数据分析的全面性。对比实验表明，该算法调整后的Rand指数为0.92，f值为0.89，收敛时间仅为12.4秒，对农机数据库的聚类准确率达到95%。对粮食收割机实际运行数据的分析表明，其故障集中在传动装置上，占比高达45%。在高负荷下运行时，工作效率急剧下降，当额定功率超过30%时，工作效率仅为72%。在潮湿泥泞的环境下工作时，故障率达到42.8%。从以上结果可以看出，本研究提出的基于K-means和Apriori算法相结合的粮食收割机数据分析模型可以对粮食收割机数据进行聚类分析，为智慧农业的可持续发展奠定坚实的基础。