How to Improve Relay Protection Technology (Part 2)

    Mastering and understanding the basic types and approaches of relay protection fault and accident handling are important conditions for improving the level of relay protection fault and accident handling. At the same time, the following issues should be strengthened.

  1. Master sufficient necessary theoretical knowledge

  1.1 Electronic technology knowledge: With the increasing use of microcomputer protection in the power grid, it is an urgent task for relay protection workers to learn electronic technology and microcomputer protection knowledge well.

  1.2 Principles and composition of microcomputer protection: In order to analyze the causes of faults or accidents based on the phenomena generated by protection and automatic devices, and quickly determine the fault location, staff must possess basic knowledge of microcomputer protection, fully master and understand the basic principles and performance of protection, memorize the logic block diagrams of microcomputer protection, and be familiar with circuit principles and component functions.

  2. Possess relevant technical data: To smoothly handle relay protection accidents, materials such as maintenance regulations, device usage and technical manuals, commissioning outlines and records, setting value notices, whole-group commissioning records, and secondary circuit wiring diagrams are indispensable.

  3. Use correct inspection methods: Generally, relay protection accidents can often be identified through simple inspections. If no faulty component is found after some routine checks, it indicates that the fault is relatively hidden and should be given full attention. At this time, the step-by-step reverse inspection method can be adopted, that is, starting from the point where the fault phenomenon is exposed to analyze the cause, and determining the fault scope based on the cause. If the fault cause still cannot be determined, use the sequential inspection method to conduct a comprehensive inspection of the device.



  4. Master skills in handling microcomputer protection accidents: In the accident handling of microcomputer protection, past experience is very valuable; it can help staff quickly eliminate recurring faults, but skills are more important. The following is a summary based on the characteristics of microcomputer protection.

  4.1 Substitution method: This method refers to replacing suspected plug-ins or components that are inconvenient to measure with plug-ins or components of the same specification, same function, and good performance.

  4.2 Comparison method: This method involves comparing various parameters of the faulty device or previous inspection reports; the part with a significant difference is the fault point.

  4.3 Simulation inspection method: This method refers to performing desoldering, open-circuiting, or changing corresponding component parameters on a good device according to the schematic diagram (usually with the cooperation of the manufacturer) to observe whether the same fault phenomenon appears on the device. If the same fault phenomenon appears, the fault location or4. Master skills in handling microcomputer protection accidents: In the accident handling of microcomputer protection, past experience is very valuable; it can help staff quickly eliminate recurring faults, but skills are more important. The following is a summary based on the characteristics of microcomputer protection.

  4.1 Substitution method: This method refers to replacing suspected plug-ins or components that are inconvenient to measure with plug-ins or components of the same specification, same function, and good performance.

  4.2 Comparison method: This method involves comparing various parameters of the faulty device or previous inspection reports; the part with a significant difference is the fault point.

  4.3 Simulation inspection method: This method refers to performing desoldering, open-circuiting, or changing corresponding component parameters on a good device according to the schematic diagram (usually with the cooperation of the manufacturer) to observe whether the same fault phenomenon appears on the device. If the same fault phenomenon appears, the fault location or damagedcomponent is confirmed.