A simple dataset of Water pressure and flow rate in a drip irrigation system based on internet of things measuring tools
Description
The dataset contains multiple metrics related to the water distribution system in drip irrigation systems, such as pressure and water flow rate. The dataset contains two types of water pressure data: Pressure 1, which represents the water pressure in the upstream part, and Pressure 2, which represents the water pressure in the downstream segment. Data was collected over a span of 5 hours, specifically on May 24, 2024. The data is offered in two distinct formats: raw and filtered. This data can be utilised by researchers, students, and farmers to do additional analysis and provide novel insights.
Files
Steps to reproduce
The data was acquired by directly extracting it from the drip irrigation system, use the HX710b sensor to quantify water pressure and the YF-S201 sensor to measure water flow rate. The sensor acquires data, which is then analysed and transformed by the ESP32 microcontroller into kilo pascals for water pressure and liters/minute for water flow rate. In the meantime, the data is transmitted to a cloud database (Thingspeak) using an internet connection. In this research, the Thingspeak platform is used as a database storage place . There are two datasets, namely raw data and filtered data with a comma separated values (.csv) extension, both of which have a table structure as shown in Table 1. The raw data is raw_data.csv, and the filtered data is filtered_data .csv. Each table consists of columns: id, created_at,id, pressure_1, pressure_2 and water rate. During the process of data preprocessing, multiple steps are undertaken to transform raw data into filtered and prepared data that is suitable for use. There are three methods employed in the process of data cleansing. The initial step involves rectifying incomplete recorded data. Subsequently, duplicate data that may introduce bias in analysis is identified and eliminated. The last stage is the removal of outlier data. The dataset comprises two categories of data files: raw data and filtered data. Both types of files contain measurements of water pressure, and water flow rate. The raw data of value is derived from real-time recorded measurements of IoT data. The provided data has been filtered based on the indicated optimal values: pressure 1 data ranging from 17.53 to 37.17 KPa, pressure 2 data ranging from 5.07 to 14.86 kPa, water flow rate data ranging from 10.89 to 51.00 L/Minutes. A total of 911 raw data points were generated during the recording process conducted on May 24 from 08.00 WIB to 13.00 WIB. These data include information such as date time, water pressure 1, water pressure 2, and water flow rate. After doing preprocessing, which involved removing incomplete data, duplicates, and outliers, the original dataset of 1329 was reduced to 911 entries. The data obtained from this preprocessing can thereafter be utilised by students, researchers, and anybody requiring to conduct additional analysis employing techniques such as machine learning, deep learning, and other methodologies