Modeling and simulation of a multi source microwave heating of soil based on PSO-BPNN

•Using a self-made multi-source microwave heating cavity as an experimental device.•Test results validate the models, then models validate the neural network models.•Adjustable position of multi-source microwave input unit.•The position of the input unit can significantly improve the heating effect. Due to rapid industrialization, accumulation of organic pollutants in the soil has been an increasing menace, having detrimental effect on crop yields, toxicity to flora and fauna, and environment in total. Microwave soil remediation technology, has garnered a significant attention in the field of thermal degradation of organic-polluted soil. However, the high-temperature remediation often involve the use of microwave devices with multiple sources, leading to competing and coupling effects, significantly impacting heating efficiency. Therefore, an attempt was made to develop a multiple source microwave device with adjustable inputs for soil heating, to design adjustment schemes for multiple input units. A multi-physics field model consistent with the testing apparatus was developed and utilized to generate the temperature for 625 different adjustment schemes. To further refine and to optimize the adjustment schemes a PSO-BPNN model was integrated and utilized. The simulation results were utilized to predict the most effective adjustment scheme, and was tested. The model potentially serves to estimate the optimal adjustment scheme for multiple source microwave inputs for effective heating of soil.