Published May 18, 2020

Anomaly Temperature Detector

Want to get notified in your email if your Freezer is not working properly. Then this device will do it so.

IntermediateFull instructions provided5 hours333

Things used in this project

Hardware components

Bolt IoT Bolt WiFi Module

Temperature Sensor

Male/Female Jumper Wires

USB-A to B Cable

Software apps and online services

Bolt IoT Bolt Cloud

Linux subsystem for windows

Mailgun

Story

There are times when you want rest of your work but your work is like ensuring correct temperature of a freezer and if it is out of limits you need to see the matter. So to give you rest this device will help you. Basically the project uses Temperature sensor which sends the temperature value to the bolt cloud and notifies you about anomaly in temperature by sending an E- mail to you. It uses Z- score analysis of ML to find the anomaly and online mail service to send a notification. Also You can know about future temperature using Polynomial Regression technique.

Applications

In Pharmaceutical industries medicines need to be stored in a temperature range. So this device can ensure that the temperature of surrounding in which the medicines are stored is suitable.
In cold storage also temperature needs to be checked at regular intervals.
Anywhere where you need to maintain a temperature and don't want temperature beyond certain range.

Theory

The value sent by sensor need to be divided by 10.24 to get actual temperature value.

Z-score analysis is used for detecting the anomalous behavior in values that are measured. First we calculate mean of the values and then calculate the z score which is added to value to set thresholds at that point.

Frame size tells about number of past values that needs to be taken in account for calculating z-score. And multiplication factor is the value set according to our need of thresholds.

Hardware Setup

Connect the LM35 temperature sensor according to its pin by bolt wifi module. Connect Ground pin to GND, Vout to A0, 5v pin to 5v.

Pin Diagram of LM35

Connect Bolt wifi module via usb charging adapter.

Software Setup

If you are having ubuntu OS nice else setup a subsystem for linux on your windows. You can view how to setup in following link https://www.youtube.com/watch?v=Cvrqmq9A3tA
Install following stuff before moving. Python3 and BoltIOT. Use the following commands.

Now create your account on Mailgun.

Steps

After connecting your device to Bolt cloud, We need to create product in following way.

Select A0 pin and name it as of your choice.(here I have named it as temp)

Now to apply Polynomial regression write the code given below and save the changes.
Now Link This product to your device.

Now you can view the data on a graph and can see the future determined data as well.

Now to use Z-score analysis first we need to create configuration file in which we need to write Api key of bolt device and our mailgun account which you can view in respective dashboards. You can view it in code as shown below. At First It is important to check the boundaries of temperature of your refrigerator so that the frame size and multiplication factor can be set accordingly to detect the anomaly.
Then code of anomaly detection need to be written as shown below.
To see the results run the python code by command 'sudo python3 'file_name''.

Following is the mail received when anomaly is detected.

Conclusion

Hence Our project is complete. Now we can rest during our busy schedule without any worry as if there is a fault we will be notified by our great friend.

MAILGUN_API_KEY= 'c46e376f604c352d*******b064705612-*********-4591ce69' #can be found in your mailgun account

SANDBOX_URL= ' sandboxdba7620ba3d2469ab59fe4**********.mailgun.org' #can be found in your mailgun account

SENDER_EMAIL= 'test@sandboxdba7620ba**********fe44f85f14856.mailgun.org'  #email sent from this id. can be found in the mailgun account.

RECIPIENT_EMAIL= 'aadarsh@*******.com'  #email to which mail is to be sent

API_KEY= 'aadfd682-****-4ca1-****-457****4d6a1' #your bolt api key, can be found from your cloud page.

DEVICE_ID= 'BOLT******'  #device id of your bolt device.
FRAME_SIZE= 4
MUL_FACTOR= 3

   
import conf, json, time, math, statistics
from boltiot import Email, Bolt
def compute_bounds(history_data,frame_size,factor):
    if len(history_data)<frame_size :
        return None

    if len(history_data)>frame_size :
        del history_data[0:len(history_data)-frame_size]
    Mn=statistics.mean(history_data)
    Variance=0
    for data in history_data :
        Variance += math.pow((data-Mn),2)
    Zn = factor * math.sqrt(Variance / frame_size) High_bound = history_data[frame_size-1]+Zn
    Low_bound = history_data[frame_size-1]-Zn
    return [High_bound,Low_bound]

mybolt = Bolt(conf.API_KEY, conf.DEVICE_ID)
mailer = Email(conf.MAILGUN_API_KEY, conf.SANDBOX_URL, conf.SENDER_EMAIL, conf.RECIPIENT_EMAIL)
history_data=[]

while True:
    response = mybolt.analogRead('A0')
    data = json.loads(response)
    if data['success'] != 1:
        print("There was an error while retriving the data.")
        print("This is the error:"+data['value'])
        time.sleep(10)
        continue

    print ("This is the value "+data['value'])
    sensor_value=0
    try:
        sensor_value = int(data['value'])
    except e:
        print("There was an error while parsing the response: ",e)
        continue

    bound = compute_bounds(history_data,conf.FRAME_SIZE,conf.MUL_FACTOR)
                                                                                                                                                                                                                       if not bound:                                                                                                                                                                                                          required_data_count=conf.FRAME_SIZE-len(history_data)
        print("Not enough data to compute Z-score. Need ",required_data_count," more data points")
        history_data.append(int(data['value']))
        time.sleep(10)
        continue

    try:
        if sensor_value > bound[0] :
            print ("Someone opened the fridge. Sending an EMAIL.")
            response = mailer.send_email("Alert","Someone Opened the door of fridge")
            print("This is the response ",response)
        elif sensor_value < bound[1]:
            print ("Someone changed the settings of fridge.. Sending an EMAIL.")
            response = mailer.send_email("Alert","Someone Changed the settings of Fridge")
            print("This is the response ",response)
        history_data.append(sensor_value);
    except Exception as e:
        print ("Error",e)
    time.sleep(10)

Credits

Aadarsh

1 project • 0 followers

Anomaly Temperature Detector

Things used in this project

Hardware components

Software apps and online services

Story

Applications

Theory

Hardware Setup

Software Setup

Steps

Conclusion

Code

Polynomial Regression Product

Configuration File

Anomaly detection

Credits

Aadarsh

Comments

Embed the widget on your own site

Anomaly Temperature Detector

Anomaly Temperature Detector

Things used in this project

Hardware components

Software apps and online services

Story

Applications

Theory

Hardware Setup

Software Setup

Steps

Conclusion

Code

Polynomial Regression Product

Configuration File

Anomaly detection

Credits

Aadarsh

Comments

Related channels and tags