Smart Greenhouse Environmental and Actuator Data from a 2024 IoT Study
by Mohammed Ismail Lifta and Wisam Abdullah
arff
Available on 1 platform
Sign in to view source links and access this dataset
Description
37,923 rows of sensor and actuator data collected from a smart greenhouse during a master's thesis project at Tikrit University in 2023-2024. The dataset contains 13 features, including temperature, humidity, water level, soil nutrients (N, P, K), and actuator states for fans and pumps. Data was cleaned, duplicates removed, and categorical columns were one-hot encoded for machine learning use.
Use Cases
Predicting optimal greenhouse conditions based on sensor readings for temperature, humidity, and soil nutrients.
Modeling actuator control logic (fan and pump states) based on environmental sensor data.
Analyzing time-series patterns in soil nutrient levels (N, P, K) and water levels.
Developing anomaly detection systems for greenhouse environmental monitoring.
Strengths
37,923 rows provide a substantial time-series record for analysis.
13 features include key environmental variables (temperature, humidity, NPK) and actuator states.
Data has been preprocessed with duplicate removal and one-hot encoding, making it ready for ML modeling.
Limitations
Description metadata is limited; actual data quality requires manual inspection after download.
Last update date is unknown; freshness unverified.
The dataset's geographic and temporal scope is limited to a single greenhouse study.
Provenance
Source
Master's thesis research by Mohammed Ismail Lifta at Tikrit University, supervised by Assistant Professor Wisam Dawood Abdullah.
Collection Method
Collected from sensors in a smartly-equipped greenhouse, with data linked to Google Sheets for remote monitoring.
Time Range
2023-2024
Freshness
Data collection occurred during the 2023-2024 academic year; last platform update is unknown.
Geography
Iraq (likely at or near Tikrit University)
License is CC-BY-ND-4.0, which prohibits distribution of derivative works and requires attribution.