Jinpin KU 2204-SE Edge Computing Server" Helps Intelligent Inspection of Ultra High Voltage Substations

Background

Compared to conventional substations, ultra-high voltage AC substations have high voltage levels, large land areas, complex equipment structures, heavy inspection tasks, and high operational risks. Existing video surveillance systems are mainly used for security, but they lack sufficient coverage and clarity. Robot deployment is also inadequate, with short battery life and inability to cover all inspection points.

Ultra-high voltage AC substations urgently need to develop a remote intelligent inspection system to assist maintenance personnel in their daily inspection work, reduce their workload, and improve emergency response speed, ensuring on-site work safety. To improve the equipment monitoring quality of ultra-high voltage AC substations and reduce the labor intensity and operational risks of maintenance personnel.

Challenge

The remote intelligent inspection system of ultra high voltage AC substations is deployed at the substation end, mainly composed of inspection hosts, robots, high-definition videos, etc. The inspection host issues control and inspection tasks, and the robot host and video host respectively control the robot and camera to carry out joint indoor and outdoor device inspection operations, and upload inspection data, collection files, etc. to the inspection host. The inspection host intelligently analyzes the collected data to form inspection results and reports, and sends timely alarms. At the same time, it has real-time monitoring and intelligent linkage with the main and auxiliary monitoring systems. The inspection host plays a crucial role in the overall architecture, and has high requirements for both hardware and software, with a maximum depth of 450mm, powerful CPU processing capability, GPU collaborative processing capability, support for large data throughput, processing and storage, and support for 220V AC and high-voltage DC power supply.

Solution

Jinpin recommended the KU 2204-SE edge computing server for the customer based on their needs. It provides up to 130 TFLOPS of video and image processing power in a limited space, and provides up to 64TB of storage space. It can quickly process data collected by inspection robots, generate reports, and provide fault warnings.

Jinpin KU 2204-SE AI Edge Computing Server

Features

2 Intel® Xeon® Cascade Lake processors, 2TB memory

Support for 2 NVIDIA Tesla P4/T4 accelerators for accelerated computing

2U 450mm depth, suitable for power network cabinets

Unique GPU cooling design to improve system reliability

Impact

The use of advanced technologies such as high-definition video and robots for remote intelligent inspections has initially realized functions such as automatic meter reading, automatic data analysis and comparison, and automatic recognition of appearance anomalies, which plays an important auxiliary role in improving the quality of maintenance personnel's inspections. It reduces the workload and risks of maintenance personnel and improves the speed and quality of inspections.