Research On Inverter Monitoring System Based On Cloud Storage

With the growing shortage of global energy, wind and solar power has become the focus of attention. Whether in remote mountainous areas, island or some special environment such as portable air conditioner, many Inverter work in different environments, we need to ensure the security and stability. On the other hand even if a fault has occurred, it must be detected and timely feedback to relevant personnel, so that we can reduce the losses to minimum. In tradition, we use network monitoring and detection of artificial that cost a lot of manpower and resources, but the effect is not good. In recent years, inverter industry uses high-precision sensor technology and the arrival of the era of big data, also the cloud computing technology is gradually penetrated into all walks of life. In this paper, the cloud computing technology is introduced into monitoring system. We find the abnormality information by extracting and analyzing the inverter detection information. Then, provide it to maintenance personnel. So that it saves inverter cost and improves inverter operational efficiency.In this paper, we designed and implemented a complete monitoring system on the cloud platform. The system consists of four parts: inverter design, control and monitoring software, web design and cloud Hadoop cluster. We took advantage of hardware circuit to implement the inverter process, the generated of AC output by control circuit board. We designed many detection circuit on the control circuit to collect the temperature、voltage、current information. We made use of Modbus protocol to transmit these data between monitoring software and control circuit board. Large amounts of data to calculate on the monitoring software, and the result of calculated will be display on the screen. Furthermore, the monitoring software will transmit the received data to network server and Hadoop clusters. Through remote access to the network server, we can easily check the status of the inverter. We made use of K-Means clustering algorithm to find the exception information. Then feedback it to maintenance personnel for further processing. After completion of the overall design and debugging, we had carried on the related performance test and the key algorithms were simulated, and achieved the desired results.