Energy Capturing IOT Device with ESP8266

At Phoenix Industries, we have a considerable number of injection molding machines, many of which are a bit older. Consequently, not all of them have the capability to connect to the internet. As a response to this challenge, I undertook a project to create a device that can extract data from the energy meters using Modbus communication and transmit this data to the cloud. This enables us to analyze the collected data effectively. 

The energy meters used in our machines are Schneider Electric's EasyLogic PM1200 and PM2100 series. These meters are equipped with an RS-485 communication line, which allows us to extract data from them. In this project, we utilized an ESP8266 microcontroller as the primary controller. Extracting data from Modbus, where the data is 32 bytes long, presented a challenging task. We needed to convert this raw data into meaningful information 

A major concern in this project is ensuring data reliability, particularly because of the unreliable internet connection, which has caused numerous issues. To address this challenge, I devised a solution involving SD card backups, but it's not the conventional SD card backup. The ESP8266, with its limited processing power and memory compared to other microcontrollers, posed a unique challenge. To overcome this, I implemented a strategy where I attempted to upload the energy data to Google Firebase as JSON. However, if the upload was unsuccessful, I saved the data as JSON in a text file on the SD card. When an internet connection became available, the system would automatically transmit the saved data back to Google Firebase. This approach allowed us to maintain system functionality even during extended periods without internet connectivity. 

When sending and backing up data in Google Firebase, I also needed to minimize the amount of data read and written to the database to reduce unnecessary costs. This was crucial because our web dashboard displayed real-time energy meter data, including voltage, current, power, and power factor. Uploading this data every second, even when users weren't actively viewing it, would lead to inefficient database usage. To address this, I implemented a solution where I only uploaded real-time data when users were actively viewing the dashboard. To achieve this, I utilized Google Firebase callback functions to wake up the ESP8266 and send real-time data when requested by the user.