In today's connected world, the key of success is "the right information in the right time".
In industry, we understand that a machine that’s out of service is a machine that’s losing money. Wouldn’t it be great if you could know how all of your equipment is functioning, and whether it will fail before it actually does break down?
By knowing which equipment needs maintenance, maintenance work can be better planned (spare parts, people, etc.) and what would have been "unplanned stops" are transformed to shorter and fewer "planned stops", thus increasing plant availability. Other potential advantages include increased equipment lifetime, increased plant safety, fewer accidents with negative impact on environment, and optimized spare parts handling. This is what "Predictive Maintenance" is all about.
Step 1: Usage of Different Sensors
Here, we are using Proximity sensor (IR distance interpreter), temperature sensor, piezo-vibration sensor, sound sensor and Light sensors. Along with this sensors, we are using LEDs and buzzer for showing output in the real time.
We are using IR distance interpreter for detecting presence near the machine. So, if any person is near the machine we can turn on the red LED. For sound condition monitoring, it makes possible to “hear” friction and stress in rotating machinery, which can predict deterioration earlier than conventional techniques. So, we are transmitting the current sound value to Intel IoT Cloud, where the values of sound can be plot on the graph. In the industry, it is possible that sometimes due to various conditions, sound value can increase sometimes. So, we have developed this system to look for continuous increase in the sound and if the sound is high for particular detection, then it will raise an alert to the user application.
Similarly, we have used temperature sensor to continuously monitor temperature of the device. On the Intel Cloud, we can set the rules for sending the alert if temperature goes high above certain value. When any person goes in the plant, it is possible that there is not enough light for him to see. So whenever light value goes below certain value, lights will be turn on automatically (Here, blue LED will turn on).
Vibration analysis is most productive on high-speed rotating equipment and can be the most expensive component of a Predictive Maintenance to get up and running. Vibration analysis, when properly done, allows the user to evaluate the condition of equipment and avoid failures. Whenever value of vibration sensor will go above the particular value, it will turn on buzzer.
Step 2: Connect Sensors to Edison
To use the code as it is, connect sensors to Edison as mentioned below.
Red LED- Digital 2
Green LED - Digital 3
Blue LED - Digital 4
IR Distance interpreter - Digital 5
Buzzer - Digital 6
Reset Button - Digital 7
Temperature Sensor - Analog 0
Vibration sensor - Analog 1
Light Sensor - Analog 2
Sound Sensor - Analog 3
Step 3: Setup Intel IoT Cloud Dashboard
Setup enableIoT cloud as mentioned in Intel IoT platforms. Once you have setup account, all the parameters which you will be receiving from Edison board.
After you have registered components to your Edison board, do not forget to restart IoT agent installed on Edison.
systemctl stop iotkit-agent<br>systemctl start iotkit-agent
Use above code to restart IoT agent. Also, after every hiur the activation code will expire, so you need to update the activation code.
iotkit-admin activate activation_code
Add different rules to trigger email, actuation or HTTP endpoint by going to rules tab in the dashboard.
Step 4: Code Your Edison
Upload the code available at the Github on the Edison.
We have developed this hack using Intel XDK. If you don't want to use XDK, directly upload main.js on the Edison. and you can run the application by
With the Intel XDK, you will also be able to view the application in the mobile. We have also developed mobile application.
We are using both the processor available on Edison. We are using 32 bit MCU for reading piezo-vibration sensor value and actuate the buzzer.
Step 5: Run the Application
Once you have uploaded code on the edison, you are all set to sit back and get all the results at any place.
By using predictive maintenance, you’ll notice improved productivity, a reduction of maintenance costs, and longer machinery life. Recent studies on the topic have shown results of a 20% reduction in maintenance budgets among companies who adopt a Predictive Maintenance.
Traditional maintenance methods are labor intensive and often require a halt in production. They can’t determine any issues or hot spots that may develop between a routine inspection, and therefore are not cost effective. Time based or preventive maintenance theories require servicing to be done on machines, whether they need it or not. With the Internet of Things (or Internet of Everything), becoming "The Next Thing" of 21st century, "Predictive Maintenance" is going to become one of the most used/followed practice in the industry.
If you have any queries regarding, drop a comment.