基于多维度数据融合的电网机组故障自动诊断系统

设计基于多维度数据融合的电网机组故障自动诊断系统，降低电网异常运维次数、提升电网安全性。系统数据采集层使用采集器和智能移动终端采集电网机组数据后，利用网络通信层将电网机组数据传输至数据存储层的SQL数据库存储，服务层调用数据存储层内的电网机组数据完成消息推送、故障查询等功能，并将电力机组数据推送至逻辑层，逻辑层利用多维度电网机组数据，构建基于深度置信网络的多维度数据决策融合模型，结合推理机诊断电网机组故障，并通过人机交互层将接收到的电网机组故障诊断信息以可视化方式呈现给用户。实验表明:该系统通信损失值较低，可保障电网机组数据完整，具备较好的数据融合性能，且故障诊断精度高，可降低电网异常运维次数，应用性较好。

Automatic fault diagnosis system of power grid unit based on multi-dimensional data fusion

An automatic fault diagnosis system for power grid units based on multi-dimensional data fusion is designed to reduce the number of abnormal operation and maintenance of power grid and improve the security of power grid.The system data acquisition layer uses the collector and intelligent mobile terminal to collect the grid unit data，and then uses the network communication layer to transmit the grid unit data to the SQL database of the data storage layer for storage.The service layer calls the grid unit data in the data storage layer to complete the functions of message push and fault query，and pushes the power unit data to the logic layer.The logic layer uses the multi-dimensional grid unit data，a multi-dimensional data decision fusion model based on deep confidence network is constructed，combined with inference engine to diagnose power grid unit fault，and the received power grid unit fault diagnosis information is presented to users in a visual way through human-computer interaction layer.Experiments show that the system has low communication loss value，which can ensure the integrity of power grid unit data，has good data fusion performance，has high fault diagnosis accuracy，can reduce the number of abnormal operation and maintenance of power grid，and has good applicability.