Objective In order to improve the detection accuracy of the seeding amount in rice seedling trays and achieve precise regulation of the seeding amount, a variable-speed weighing detection and intelligent regulation device for seeding amount in rice seedling trays was designed.

Method The seedling tray mass variable-speed weighing mechanism composed of front and rear driving wheels was constructed. The front driving wheel created a spacing between the front and rear seedling trays, ensuring that the seedling trays could achieve the static non-contact precise weighing on the variable-speed weighing mechanism. The rear driving wheel enabled the weighed seedling tray to automatically catch up with the front seedling tray to eliminate the spacing, ensuring continuous conveying of the seedling trays without intervals on the seedling-raising production line. A through-beam photoelectric sensor was used to accurately detect whether the seedling tray had reached the weighing position. The weighing signals were processed by the combined Butterworth and Kalman filtering, which effectively suppressed the interfering factors and further improved the detection accuracy of the seeding amount in the seedling trays. With the programmable logic controller (PLC) as the control core, the deviation between the detected value of the seeding amount in the seedling trays and the set value was compared and analyzed. According to the established regulation model for the seeding amount of rice bud seeds in the seedling trays, intelligent and precise regulation was realized.

Result Under the condition of a seedling raising productivity of 500 trays per hour, the average detection accuracy of the seeding amount in the seedling trays was 96.64%. For hybrid rice with a seeding amount in a single seedling tray ranging from 50 to 80 g, the average regulation accuracy and coefficient of variation were 93.69% and 2.01%, respectively. For conventional rice with a seeding amount in a single seedling tray ranging from 81 to 360 g, the average regulation accuracy and coefficient of variation were 96.01% and 3.13%, respectively.

Conclusion The detection accuracy and regulation performance of the device designed in this study are significantly superior to those of existing devices, which has practical application value for improving the quality of seedlings and ensuring the performance of transplantation.