Construction of a press machine for extracting sesame oil and evaluating it using artificial intelligence technique

Document Type : Complete scientific research article

Author

Machine design and Mechatronics Department, Institute of Mechanics, Iranian Research Organization for Science and Technology, Tehran. Iran

10.22069/fppj.2025.22825.1839

Abstract

1. Background and purpose
To optimize the mechanical and thermal properties affecting the quality and quantity of sesame oil, the effects of spiral shaft speed and sesame seed moisture content were investigated at three different levels each. A support vector machine and response surface methodology were used to evaluate and model the results. Sesame contains 44% oil and 30% protein. Various mechanical and chemical technologies, along with different devices, are available for oil extraction [20]. It is recommended to use the mechanical method (press) with pressure on the texture of these seeds, which leads to the oil coming out of their texture. The goal of the extraction process is to produce healthy and pure oil [11].
2. Materials and methods
The spiral mechanism of the device is designed to exert maximum pressure on the seeds. To determine the shell thickness, both radial and tangential stresses on the cylinder wall were calculated. The design of the spiral is crucial for oil extraction, feed rate, and internal pressure within the machine. In this research, spiral speeds of 0.4, 0.6, and 0.8 Hz, chamber temperatures of 30°C, 45°C, and 60°C, and seed moisture content of 3.5% and 7% were considered. A neural network was employed for data analysis and modeling [7].
3. Findings
The main factors influencing the neural network's behavior include the input parameters and the number of hidden layers. A multi-layer perceptron (MLP) with hidden layers was used in the artificial neural network structure. The maximum accuracy and lowest error in measuring the acidity of sesame oil were 0.9930 and 0.0023, respectively, while the corresponding values for peroxide measurement were 0.9991 and 0.0006. The RMSE values for acidity and peroxide were 0.118 and 0.106, respectively, and the R² values were 0.87 and 0.98.
4. Conclusion
The extruder's effective length was 390 mm, the grain drive was 30 mm, the depth of the spiral tooth was 10 mm, the outer diameter was 80 mm, and the spiral pitch angle was 82.5 degrees. An increase in temperature reduces the moisture content of sesame seeds in the machine, which in turn affects the quality and quantity of the oil to some extent. Increasing the spiral speed raises the pressure and friction within the chamber, which in turn increases the temperature, potentially leading to the combustion of oxygen in the air with the oil. As a result, the levels of peroxide and oil acidity increase. As a result, the levels of peroxide and oil acidity increase.

Keywords

Main Subjects