Leo Zhang
Back to Projects

Smart Power Management System

Chunxiao Technology · 2019 – 2022

Role: Java Backend & Android Developer

Cover

Cover

Smart power platform for factories and buildings: onboard gateways, switches, and meters; monitor per-load electrical parameters; remote control and alarm policies. 15% energy saving, sub-second alarm, 99.9% data collection rate.

15% energy saving, <1s alarm response, 99.9% collection rate, 3+ parks unified management

Problem

Factories and buildings need granular power monitoring and control to reduce energy waste and respond to electrical anomalies quickly.

Solution

IoT platform: Modbus/RS485 gateways and smart meters for telemetry, Spring Cloud + InfluxDB for time-series storage, Apache Flink for stream processing, Vue.js/ECharts dashboards, Android app for mobile monitoring.

Architecture

Smart meters/breakers → Modbus/RS485 → Gateway → MQTT → Spring Cloud → InfluxDB/Flink → Vue.js dashboards + Android app

Key Highlights

  • Ran Java services for multi-site power monitoring under long-lived production load
  • Shipped Android app for live energy readouts and sub-second alarm escalation
  • Wired Modbus/RS485 gateways and smart meters into the telemetry path
  • Achieved 15% average energy consumption reduction

Tech Stack

JavaSpring CloudModbus RTU/TCPDL/T645InfluxDBRabbitMQApache FlinkVue.jsEChartsMQTTRS485

What I Learned

Time-series databases (InfluxDB) are essential for power telemetry; stream processing (Flink) enables real-time anomaly detection; edge gateways reduce cloud bandwidth.