Published January 27, 2019

Temperature Monitoring Device Using Bolt IoT Module

Temperature Monitoring device which notifies you via Email for any sudden change in temperature or if it goes out of the defined threshold.

IntermediateFull instructions provided1 hour1,434

Temperature Monitoring Device Using Bolt IoT Module

Things used in this project

Hardware components

Bolt IoT Bolt WiFi Module

Temperature Sensor

Male/Female Jumper Wires

Software apps and online services

Bolt IoT Bolt Cloud

Story

I always wanted to work on Internet of Things as it changes our day-to-day life and literally has endless possibilities of use cases, but I didn't get some reputed and interesting tutorials for the same. When I come across the Internshala IoT training cum competition I couldn't resist myself from doing it at here is the fruit of the training.

This project is about building a Temperature Monitoring Device using Bolt IoT Module. The device sends an e-mail to the user whenever the temperature of the surrounding (wherever it is used) crosses a certain threshold which is previously defined.

Here is the project in action :

Working of the temperature sensor device

I am assuming that the Bolt IoT module is already connected to your system and with a stable internet connection.Step 1 : HardwareSetup1. The three pins of LM35 sensor are connected to the female end of the 3 wires.

2. The blue wire is connected with the sensor pin, green wire with the output pin and yellow wire with the ground pin. The male ends of the respective wires are connected to the 5V input, A0 and ground (GND) of the Bolt IoT module respectively.

1 / 3

3. The final connection will look like this.

4. The Bolt module then needs to be connected with a portable power source (because we need to move the device to a particular place for temperature readings).

Slowly blinking blue light

5. The device then needs to be connected with a stable internet connection.

Stable blue and green light

Hardware connections are complete.

Step 2 : ConfiguringandDeployingdeviceforpredictions

1. Login to https://cloud.boltiot.com/, click on products tab and add a new product. The product will be an output device with collection data as GPIO.

Adding new product

2. After adding the product click on "Configure this product" on the top right corner to configure the product. Configuration screen will open.

Hardware configuration

3. In hardware configuration, we have connected the LM35 sensor with A0 pin of the module so mark check the A0 pin and write a variable name for it.

Hardware configured

4. Now we need to write the code for the predictions in the code section. The code is written in JavaScript. The code first import the 'google-chart' library so that we can use it's graphs. Them we write a title for our graph. Them we choose a chart which we need to use (we used predictionGraph). Then we named the axes of the graph (x, y). We then set a multiplication factor. And at-last we plotted the points on the graph.

JS code.

5. We then save the code after which we will link our bolt device with the product by clicking "Link device to this product" button present on the top right of the screen and then choosing the Bolt Device.

Linking our bolt device

6. Once we linked the device, we then deploy the configurations to our device so that it work according to what we have configured in the product. After deploying the configuration it's time for some action. We place our bolt device to a place of which we want to check the temperature (I have place my bolt device in my refrigerator to get the readings) and the reading will be plotted in the graph.

7. Let the device take readings for sometime (I left my bolt device for 3-4 hrs to take ample of reading for future predictions). Remember, the more readings you take the better predictions your device will show.

Temperature Readings (the initial value is greater because it is the reading at room temperature)

8. Once you are satisfied with the amount of the readings you took, download the data as.CSV file for further use and click on the predict button below to predict the future temperature. The 3 variables predicted points, no. polynomial coefficients and frame size are used to fit our predicted graph and help us to get better and more accurate results. Predicted points defines the number of future points which we want to predict. Number Polynomial Coefficients define the polynomial coefficients for the prediction. Frame size defines the number of previous readings used for the prediction.

The final prediction graph

Temperature prediction is complete.

Step 3 : TemperatureMonitoringbyprocessingtherecordeddataandZ-scoreAnalysis.

1. First get the bolt API key by clicking on the API Tab and under the section for Generate Key, click on Enable.

2. Then build a mailgun account by going to https://app.mailgun.com/sessions/new this will be used to send email notifications for temperature management. After creating the account. Get the mail ID and mail API key.

Click the domain name to get complete details

3. We will now write the configurations for our device in python. The configurations will include the Bolt device ID, Bolt API key, Mailgun sandbox URL (sandbox_url), Mailgun API key, sender email (test@sandbox_url) and receiver email (your email). Apart from these the configuration file will also have frame_size and multiplication factor for Z-score analysis. A sample configuration file is also provided.

4. After writing the configuration file, we write the python code for temperature sensor.4.1. We downloaded the data which we recorded previously and read it in our code using the pandas library. We remove the first reading as it is a reading of room temperature (you may need to adjust the data accordingly). After the data is read we find a minimum and maximum range of the data which will act as a threshold for our device (thresholds can also be defined according to the needs of the user).

Data pre-processing

4.2. We then build a function for Z-Score analysis. The function will return "None" if there are less number of reading and will return an array of upper and lower bounds if the Z-score analysis is complete.

Z-score analysis

4.3. We then configure our code with bolt device and e-mail alert API (mailgun) and also initialize an empty array which will store the data.

Configuring code for device

4.4. Now we write the monitoring and mailing code in a while loop which will run endlessly because the temperature should be monitored every time. In the while loop we have first printed the history data, minimum and maximum threshold values. The output from the Bolt device is then stored in a variable named 'response'. We then fetch the sensor value from the response. We then run the function for Z-score analysis using the history data, and frame size and multiplication factor from the configuration file. If we get the bounds from Z-score analysis then we will move further, otherwise it will print "Not enough data to compute Z-score. Need more data points". Once the upper and lower bounds are calculated, we will then check the sensor value, if it is between the minimum and maximum threshold, then no mail will be sent otherwise the device will mail the user alerting him about the temperature. And using Z-score analysis, the device will mail the user if there is any sudden change in the temperature (or sensor value).

1 / 3 • Code in while loop

4.5 Also remember that the sensor value are from the range for 0-1023, so the readings are in this range only. To get the actual value of temperature, divide the sensor values by 10.24

5. The code is now complete and we run the code and hence the device using the python command on terminal. The complete python file for the code is provided.

1 / 2 • Initial output

Now you are up for action.

Thismarksthecompletionoftheproject.

Now monitor you room temperature and show-off your device. All the code is provided for your help. I sure had fun working on this project hope you guys feel the same.

ThankYou!

Code

import pandas as pd
import conf, json, time, math, statistics
from boltiot import Email, Bolt

# read the CSV file of the readings
df = pd.read_csv("file:///home/suraj333/Documents/IOT/temp_recording.csv")

#print the data of the csv files (temperature readings)
print("DATA SET \n",df)

# removing the 1st data row as it was the room temperature
print("\nDATA SET without 1st reading(room temp)\n", df[1:])

# describing the mean, median, mode, max and min from the data
print("\nData Set Description\n", df[1:].describe())

# saving max and min sensor readings in a variable to define the range of temperature
minimum_val = (df[1:].describe()['sensor_reading']['min'])
maximum_val = (df[1:].describe()['sensor_reading']['max'])

# Function to set upper and lower bound by Z-score analysis
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]


# configuring bolt device
mybolt = Bolt(conf.API_KEY, conf.DEVICE_ID)

# configuring e-mail alert requirement
mailer = Email(conf.MAILGUN_API_KEY, conf.SANDBOX_URL, conf.SENDER_EMAIL, conf.RECIPIENT_EMAIL)

# setting up empty list to save the temperature (sensor) readings
history_data=[]

# system will keep on running infinitely
while True:
    # prints the history data
    print("\n\nHistory", history_data)
    
    # prints the minimum temperatre
    print("minimum = ", minimum_val)

    # prints the maximum temperature
    print("maximum = ", maximum_val)

    # saves bolt module response in variable
    response = mybolt.analogRead('A0')

    # saves the response data in variable
    data = json.loads(response)

    # print the response data
    print("complete data ", data)

    # prints current sensor value
    print("data_val ", data['value'])

    # saving the upper and lower bounds by Z-score analysis in a variable named bound
    bound = compute_bounds(history_data,conf.FRAME_SIZE,conf.MUL_FACTOR)

    # prints the bounds measured via Z-score analysis
    print("Bounds", bound)

    # checks if bound is empty or not
    # if it is empty, then it will save the current sensor value in the history_data list otherwise, will directly go to try.
    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']))

    try:
        # saves current sensor value in variable
        sensor_value = int(data['value'])
        print(" sensor_val ", sensor_value)

        # condition for when sensor value crosses the maximum threshold
        if sensor_value > maximum_val:
            temp = sensor_value/10.24
            response = mailer.send_email("Alert", "Fridge door is open.\nTemperature is beyond threshold!!!\nThe current temp val is " +str(temp))
            print(response)

        # condition for when sensor value crosses the minimum threshold
        elif sensor_value < minimum_val:
            temp = sensor_value/10.24
            response = mailer.send_email("Alert", "Fridge is cooling much more than required.\nTemperature is beyond threshold!!!\nThe current temp val is " +str(temp))
            print(response)
                 
        elif bound != None:
            # condition for when current sensor_value is greater than the upper bount calculated by Z-score analysis
            if sensor_value > bound[0]:
                response = mailer.send_email("Warning!!!", "Someone has opened the fridge door!!!")
                print ("Someone has opened the fridge door.")

            # condition for when current sensor_value is lowerer than the lower bound calculated by Z-score analysis           
            elif sensor_value < bound[1]:
                print("Someone decreased the temperature.")
            else:
                history_data.append(sensor_value)

    # in case of any error
    except Exception as e:
        print("Error", e)
    time.sleep(10)

# The URL, device ID and API keys are to be filled by you and is not to be shared with anyone

MAILGUN_API_KEY = "410669354d735497XXXXXXXXXXXXXXXX-3939XXXX-7349XXXX"
SANDBOX_URL = "sandboxc54681880a064db0XXXXXXXXXXXXXXXX.mailgun.org"
SENDER_EMAIL = "test@sandboxc54681880a0XXXXXXXXXXXXXXXXXXXX.mailgun.org"
RECIPIENT_EMAIL = "surajsingla333@gmail.com"
API_KEY = "2fbeXXXX-95XX-44XX-b0XX-b73fXXXXXXXX"
DEVICE_ID = "BOLTXXXXXXXX"
FRAME_SIZE = 4
MUL_FACTOR = 3

Credits

suraj singla

1 project • 3 followers

Temperature Monitoring Device Using Bolt IoT Module

Things used in this project

Hardware components

Software apps and online services

Story

Schematics

Hardware devices used

Connecting devices

Connecting devices

Connecting devices

Connecting devices

Connecting with Power Source

Connecting with WiFi

Code

Temperature_sensor

Temperature_sensor_configurations

Credits

suraj singla

Comments

Embed the widget on your own site

Temperature Monitoring Device Using Bolt IoT Module

Temperature Monitoring Device Using Bolt IoT Module

Things used in this project

Hardware components

Software apps and online services

Story

Schematics

Hardware devices used

Connecting devices

Connecting devices

Connecting devices

Connecting devices

Connecting with Power Source

Connecting with WiFi

Code

Temperature_sensor

Temperature_sensor_configurations

Credits

suraj singla

Comments

Related channels and tags