为了处理大量分布式存储的农业环境数据,实现农业设施智能控制,基于内存计算框架Spark提出一种并行化的Dirichlet过程混合模型聚类方法,对农业环境及设施数据进行训练得到预测模型,执行对温室大棚天窗开度的预测任务。通过对比实验验证了模型预测的可行性,对预测正确率进行统计,并测试了所提出的并行化聚类的执行效率。实验结果表明,提出的方法具有较高的执行效率及准确性。 In order to deal with a large amount of distributed and stored agricultural environment data and realize intelligent control of agricultural facilities, Spark proposes a parallel Dirichlet process hybrid model clustering method based on the memory computing framework. The agricultural environment and facility data are trained to obtain the prediction model. Greenhouse roof opening forecast of the task. The feasibility of the model prediction is verified through comparative experiments, the statistical accuracy of the prediction accuracy is tested, and the execution efficiency of the parallelized clustering is tested. Experimental results show that the proposed method has high efficiency and accuracy.