Freezer Temperature Control
The aim of the project is to built an Arduino-based controller for the freezer temperature. Arduino collects logs of temperature and humidity. The board also have a light sensor in order to detect if the door of the fridge is open or closed. We assume the sensor is in the feezer, so in a dark enviroment if the door is closed. All the data are sent with a timestamp calculate with a RTC module. The data are sent through mqtt protocol to a online mqtt broker, then the data are saved on Azure Cosmos DB, a noSql database. An application was built in order to visualize the data.
2 Arduino hardware
The hardware part used is composed by:
- Arduino UNO wifi Rev
- DHT22 sensor
- RTC clock module
- Photo resistor sensor
The arduino board is connected to the DHT sensor for humidity and temperature. A photo resistor is connected to detect the light in the freezer and to check if it is open or not. Arduino reads the data from the sensors and sends a json string with a timestamp throught mqtt to the CloudMqtt broker service. The json string is alike:
1 {
2 "humidity" : 63.5,
3 "temperature" : 19.3,
4 "lightSensor" : "10",
5 "year" : "2020"
6 "mounth" : "1",
7 "day" : "22",
8 "hours" : "3",
9 "minutes" : "21",
10 "seconds" : "1"
11 }
The schematics for Arduino is the following:
3 Cloud Infrastructure
The arduino board is connected to CloudMqtt broker services,through mqtt. The mes- sage are sent every 2 seconds. The message are sent with low QoS and without any particular security features because losing some data it’s not a relevant problem for the project. The messages are collected with a small azure virtual machine instance and write on cosmos DB database. A noSQL database was chosen because there’s no need for relation in the DB, but the data are just saved from json.
3.1 MQTT
Mqtt is a lightweight, push-subscribe network protocol that transports messages be- tween devices. The protocol is the first choice to exchange messages between devices because:
- Light enough to run on an embedded device with limited processor or memory resources
- Can work with low and unreliable bandwidht
The message can be sent with 3 different QoS.
- At most once (QoS 0) the message is sent only once and the client and broker take no additional steps to acknowledge delivery (fire and forget)
- At least once (QoS 1) the message is re-tried by the sender multiple times until acknowledgement is received (acknowledged delivery)
- Exactly once (QoS 2) the sender and receiver engage in a two-level handshake to ensure only one copy of the message is received (assured delivery)
In the project a QoS 0 is used because the messages are sent continuously, so a mes- sage lost is irrelevant for the project. CloudMqtt is the online broken choose because the azure broker don’t follow the complete mqtt standard and the ArduinoMqtt library have a low support for the ssl connection required from the Azure IoT hub. The data so are sent to CloudMqtt topic called: /devices/tempcontrol.
3.2 Azure Virtual Machine
The messages sent to the broker are consumed by an Azure Virtual Machine wich is subscribed to the /devices/tempcontrol topic. The performance required is very low, so the virtual machine is the cheapest available, with 1 core and 512MiB of RAM. A nodeJS script is used for the mqtt connection to CloudMqtt provider and the data are clean for an easy usage. The data are saved as json for convenience, alike:
1 {
2 "humidity" : 64.2,
3 "temperature" : 19.2,
4 "doorStatus" : "close",
5 "date" : "2020-01-12T23:41:08Z"
6 }
The nodeJS script manages also the connection to CosmosDB for data persistence. If the freezer door is open an email is sent to inform of that.
3.3 Cosmos DB
The persistence are managed by the Azure noSQL database Cosmos DB. A noSQL database was choosen because the data are collected as a datawarehouse, so there’s no relation between data. The main advantage of a noSQL is the scalability and per- formance, which are astonishingly fast compare to a sql database. The data are saved with a mandatory autogenerated id on the database in a form as the following:
1 {
2 "_id" : ObjectId("5e1ba13907194748c74120c7"),
3 "humidity" : 64.2,
4 "temperature" : 19.2,
5 "doorStatus" : "close",
6 "date" : "2020-01-12T23:41:08Z"
7 }
4 Client Application
The data are visualized by a client application developed with flutter framework. Flutter is a Google framework which aim to develop multi platform application (Android, Ios and web) with a single codebase. The main language of flutter is dart, then the application is transpiled for all the supported platform.
4.1 Flutter Client Application
The application is currently tested on Android devices. The application is divided in two tabs, one for the humidity data and one for the temperature data:
The history of the data is shown in a chart, then we could see the average and the last measurement. The door status is visible in both the tabs. The data are updated in real time with Cosmos DB server.
4.2 IFTTT
IFTTT is a online service to create chains of simple conditional statements, called applets. The server application sent to a IFTTT webhooks a signal if the freezer door is open, then an email with the data is sent to inform the user