WHITE PAPER
MQTT Platform for AI: Empowering AI with Real-Time Data →

Overview

One of the largest agricultural enterprises in China manages over 53,600 square kilometers of land and cultivates more than 48.74 million acres. As a key national grain production base, the company has maintained stable grain production levels exceeding 40 billion kilograms annually for over 13 years, contributing significantly to China's agriculture industry.

With its vision of becoming a global leader in agriculture, the enterprise has continually invested in smart agriculture and digital transformation, leveraging cutting-edge technology to stay ahead in the competitive agricultural sector.

The Challenge

Managing such vast agricultural operations comes with its own set of challenges. In the seed breeding process, key environmental factors like soil moisture, temperature, and sunlight must be carefully monitored and controlled to ensure optimal seed development. However, this process often involves managing vast, distributed agricultural fields, making data collection and real-time monitoring difficult.

The enterprise faced several challenges in its digitization project, including:

  • Complex Data Collection Conditions: Gathering accurate data on environmental factors in vast and diverse agricultural regions proved challenging.
  • Data Integrity and Real-time Issues: With a variety of sensors and monitoring devices spread across wide areas, the consistency and timeliness of data collection were often compromised.
  • Integration Difficulties: The company's legacy systems struggled with compatibility between equipment and platforms, leading to inefficiencies in data interaction and high maintenance costs.
  • Unreliable Communications: Weak network signals in remote agricultural areas made real-time data transmission difficult.
  • High Concurrency and Data Volume: The breeding process required processing a large volume of real-time data, posing a challenge to ensure platform stability under high message loads.

The Solution

To tackle these challenges, the company initially selected the open-source version of EMQX as the MQTT broker to establish a unified platform for comprehensive data collection and analysis, which is crucial in monitoring and optimizing the seed breeding process.

However, as the company’s operational demands expanded, they recognized the need for enhanced performance and capabilities. Consequently, they transitioned to the enterprise version of EMQX. This upgrade provided them with advanced features and improved reliability, making it the ideal foundation for their ambitious digital transformation project in seed breeding.

The EMQX Platform not only streamlined data collection and processing but also facilitated effective management of the entire seed breeding cycle, from inception to harvest. By leveraging both edge computing and cloud-edge collaboration, the solution allowed for real-time data processing and analysis, thereby enhancing decision-making and operational efficiency.

Architecture diagram

Key capabilities of the solution included:

  • Multi-protocol Data Ingestion: EMQX’s support for various protocols like MQTT, HTTP, QUIC, and WebSocket allowed seamless integration of different sensors and devices. Additionally, it enabled the connection of NB-IoT sensors using CoAP and LwM2M protocols, ensuring reliable data transmission even in regions with weak signals.
  • High Availability Clustering: EMQX’s distributed cluster architecture ensured high concurrency and real-time data processing, enabling the system to handle large message volumes and support future growth.
  • Flexible Data Integration and Rule Engine: With EMQX’s built-in rule engine, the company could define custom data processing rules, efficiently integrating data into back-end systems like Kafka, TimescaleDB, and InfluxDB.
  • Reliable Communication in Weak Networks: EMQX improved data transmission efficiency with the innovative MQTT over QUIC, reducing network latency and ensuring reliable real-time communication in challenging environments.
  • Comprehensive Monitoring: The integrated monitoring platform provided real-time oversight of the system’s performance, helping the company efficiently manage equipment and monitor environmental conditions.

Results

The deployment of EMQ’s IoT solution resulted in several key outcomes:

  • Enhanced Data Stability: By directly forwarding sensor data to Kafka, the new architecture provided stable data ingestion and eliminated previous issues with data loss during high device activity or network fluctuations.
  • Optimized System Architecture: EMQX’s rule engine simplified data processing and reduced the need for complex web service clusters, resulting in improved system capacity and performance.
  • Lower Operational Costs: With advanced logging, issue tracking, and hot configuration updates, EMQX significantly reduced maintenance costs and improved the efficiency of the company’s IoT system operations.

Conclusion

Through EMQ’s innovative cloud-edge IoT solution, the agricultural enterprise successfully achieved its digital and intelligent transformation in seed breeding, reducing operational costs while ensuring high reliability and stability in production. Looking ahead, as the company further analyzes and applies its seed data, it plans to expand its intelligent production capabilities and enhance its competitive edge in agricultural technology.