数字赋能·助力中国·影响世界

On August 10, 2018, Professor Jiang Haiyan and Professor Hu Bin from the School of Information Technology at Nanjing Agricultural University visited Jiangsu Three&Three to guide the deep application of artificial intelligence and big data in agriculture. Jiangsu Three&Three is a smart agriculture IoT solution service provider, focusing on four major fields: smart agriculture, smart industry, smart environmental protection, and smart transportation. It is a pioneer in building smart agriculture.

The "Agricultural Big Data and Intelligent System Research Team" at Nanjing Agricultural University is supported by the theories, methods, and technologies of intelligent science, data science, computational science, and software engineering. Its goal is to analyze, model, simulate, and control the relationship between the growth and development, cultivation management, fertilizer management, and disease and pest diagnosis of field crops, facility crops, aquatic products, and other animals and plants, and to carry out key technology innovation and integrated platform research and development for smart crop systems. At present, the team has 1 professor, 4 associate professors, and 2 lecturers. 10 graduate students.

Direction 1: Intelligent prediction, management, and risk assessment platform technology for crop systems.

Research on the principles, methods, and technologies of time-series data analysis, domain knowledge graph construction, and mechanism model uncertainty optimization control for crop systems. Develop a crop production system analysis, management, push, and intelligent Q&A system driven by domain knowledge graphs, machine learning models, and mechanism models to promote the integration of modeling knowledge relationships, mechanism models, machine learning models, and optimization control models for crop systems. Provide analysis methods and visual analysis tools for agricultural meteorological data analysis, production process control, and agricultural operation push under climate change conditions.

research findings:

1. An intelligent system for dynamically pushing rice growth prediction, warning, and management plans based on knowledge graph and growth model

Integrating the high-yield cultivation model of rice at the agricultural scale and the leading RiceGrow rice growth model in China, using deep learning, evolutionary computing, and SOA technology, we provide intelligent analysis of meteorological conditions during key phenological periods of rice for modern agricultural producers, managers, machinery, and agronomists. We adaptively simulate and predict the growth status of rice, intelligently push meteorological disasters, high-yield cultivation models, precise nitrogen fertilizer management plans for sowing dates, and disease diagnosis information. include:

(1) Method and analysis evaluation tool for predicting meteorological sequences of key phenological periods in rice based on deep networks.

(2) Intelligent calibration methods and tools for crop models.

(3) Rice growth simulation evaluation web service system.

(4) Intelligent push system for high-yield rice cultivation management based on knowledge graph.

Direction 2: Automatic detection and diagnosis technology for phenotype and growth characteristics of animals and plants.

By utilizing technologies such as image and spectral analysis, combined with IoT technology, we aim to study the methods and techniques for automatically measuring, processing, and evaluating the morphology and growth information of animals and plants under abiotic stress (such as high and low temperatures, CO2, nutrients, water, dissolved oxygen, ammonium nitrogen, etc.) and biotic stress (such as diseases and pests), and develop a crop growth data acquisition device and an intelligent health diagnosis system.