Efficient machinery is required for seedbed leveling in rice trellis seedling transplantation at present. In this study, a self-propelled machine was designed for the rice trellis seedbed precision leveling, in order to combine with the technical requirements of standardized seedling transplantation in the cold region. Electromechanical hydraulics was also integrated with the automatic leveling control system. The hydraulic oil pump and DC stepping motor were taken as the power end in the leveling system. The automatic control strategy was selected to drive the electromechanical roll angle leveling mechanism and hydraulic pitch angle compensation device in using attitude angle data during operation. The better level of the seedbed was obtained to carry out the leveling operation of the precision leveling machine. A Simulink module was used to simulate and then test the performance of roll angle leveling and pitch angle compensation. The results show that better response speed and accuracy were achieved in the roll angle leveling. The minimum action error was observed in the pitch angle compensation of the crushed soil scraper cylinder, the roll angle leveling time was 1.62 s, the overshooting amount was 1.5%, which fully meeting the requirements of stable control. The field test was conducted to verify the simulation. The automatic leveling control system was also embedded in the 1ZJP-2seedbed precision leveling machine. The test showed that the average value of attitude angle, root mean square error (RMSE), and higher stability were found in the operating machine equipped with the automatic leveling system, compared with the control group, the average value of the rolling angle of the operating machine equipped with theautomatic leveling system was −1.88°, the RMSE was 1.95°, and the average absolute error was 1.88°, the average value of the rolling angle of the operating machine in the control group was −2.37°, and the RMSE was 2.43°, and the average absolute error was 2.31°, the average RMSE of the action of pitch angle compensated soil crushing scraper cylinder was 0.145 cm. The performance of seedbed leveling was examined using an optical leveling instrument. The maximum average standard deviation of the seedbed elevation was reduced by 1.05 cm after the application of the automatic leveling control, compared with the pre-operation without leveling. The maximum average standard deviation of seedbed elevation without leveling was only 0.325 cm. The leveling operation was improved by about 229%-500%, compared with the normal operation. The quality of operations was improved by about 268%-384% than before.

High hot-air temperature can improve the quality of paddy in the process of mixed-flow rice drying, while low hot-air temperature can affect the drying rate of paddy. The variable temperature drying is confined to selecting the node of temperature change, the large fluctuation range of temperature, and the long time to reach the specified temperature. In this study, a drying control system was designed to keep the temperature of paddy in the rubbery region of the glass transition curve. The glass transition curve of the paddy was also determined for the control strategy of variable temperature. The opening size of the valve was adjusted in the hot-air mixing device, in order to change the hot-air temperature. A control model was established between the valve opening of the hot air mixing device and the paddy temperature using Logistic regression. Least square identification was used to identify the parameters of the variable temperature control model. Genetic algorithm (GA) was used to optimize the membership degree of fuzzy PID control. The objective function converged to 0.118 in the process of genetic optimization. The amplitude width was then determined as the optimal membership function. Simulink simulation showed that the response time of fuzzy PID control was 66.43 s, and the overshoot was 3.600% when the temperature was set at 42 ℃. The response time of fuzzy PID control was 37.06 s after GA optimization, and the overshoot was reduced to 0.120%. The external signal interference of 5 s was added in the time of 150 s, in order to test the anti-interference performance of the variable temperature control system. The adjustment time of the fuzzy PID control after GA optimization was 4.19 s less than before, and the overshoot was reduced by 0.050%. Once the temperature signal was input at 42 ℃ and the temperature rose to 47 ℃, the adjustment time of the fuzzy PID control after GA optimization was 16.79 s less, and the overshoot was 0.338% less than before. A test system was self-developed for the mixed-flow variable temperature drying, according to the mixed-flow two-way air inlet dryer. Firstly, the temperature-changing test was carried out on the variable-temperature control system under different temperatures. The temperature was set at 37 ℃, 42 ℃, 47 ℃ and 52 ℃. The average response time of the fuzzy PID control was 32.37 s after GA optimization, which was 69.93 s before optimization. In the response time test of gradient temperature rise, the average response time of fuzzy PID control was 27.00 s after GA optimization, which was 45.63 s before that. Compared with the fuzzy PID control after GA optimization, the adjustment time of the target and gradient temperature was shortened by 37.56 s and 18.63 s, respectively. After that, 800 s was divided into eight intervals at the stable temperature, in order to test the stability performance of the variable temperature control system. When the rice temperature reached 42 ℃ in dynamic stability, the paddy temperature varied from 41.9 ℃ to 42.1 ℃, where the average relative error was less than 0.4%, and the coefficient of variation was less than 0.5%. A performance test of the variable temperature control system was carried out on the dryer of Nongjiang Farm in Jiansanjiang National Agricultural High-tech Demonstration Zone, Heilongjiang Province, China. The response time of the fuzzy PID control system after GA optimization was less than 30 s, where the steady-state temperature error was ±0.15 ℃, and the average relative error was less than 0.5%. The simulation and field test showed that the stable temperature control system with glass transition fully met the drying requirements in paddy production.