In order to solve the problem that the traditional heating device can not meet the somatosensory temperature requirements of piglets of different days of age and high energy consumption, the intelligent heat preservation control system for piglets in the delivery house was designed. According to the environmental factors such as temperature, relative humidity and wind speed in the piggery, the intelligent insulation control system was designed for piglets to calculate the appropriate range of somatosensory temperature and adjust the output power of the insulation equipment intellectively and in cluster to meet the demand of somatosensory temperature for piglets of different stages and days of age. The experiment of the piglet insulation control system in the delivery mother pig house was carried out, the experimental results indicated that the system could control the somatosensory temperature of piglets within the set temperature range, and the somatosensory temperature fluctuation range was within ±0.75℃. The coefficient of variation calculated from the somatosensory temperature data at different ages was less than 0.1, which showed weak variability, indicating that the dispersion degree of somatosensory temperature was small, and the effect of systematic control of somatosensory temperature was relatively stable. This study can provide reference for the environmental regulation of delivery sows.

 To achieve intelligent feeding, userfriendly humancomputer interaction, and easy maintenance for lactating sows, a set of intelligent and precise feeding systems for lactating sows was designed. This system includes an intelligent feeder, a handheld personal digital assistant terminal, a central controller, and a CANBus bus. It enables convenient onsite management of pig entry and exit management, dynamic monitoring of pig status, and realtime alarm functions for abnormal conditions. In a largescale sow farm in Yunnan Province, 280 intelligent feeders for lactating sows were installed, and a production performance comparison test was carried out with the production line equipped with traditional feeding. The test results showed that intelligent feeding facilitates realized userfriendly humancomputer interaction, touchsensitive feeding, intelligent wet mixing, noncontact operation, and easy maintenance. Compared with traditional feeding, intelligent feeding increased the average daily feed intake of sows by 0.673kg and reduced the daily feed waste by 0.345kg, indicating significant differences (P<0.01) in the average daily feed intake and daily feed waste at different ages between intelligent feeding and traditional feeding. The number of healthy piglets per month increased by 10.44%, the number of healthy piglets per litter increased by 0.69, and the average weight of weaned piglets at 21 days increased by 0.370kg. The backfat loss of lactating sows decreased by 34.43%, and the 7-day weaning estrus rate increased by 4%. The return on investment analysis of the system demonstrated that intelligent feeding effectively reduces feed waste, improves production performance, and has a high return on investment, which can effectively help largescale pig farms reduce costs. This study provides a reference for the development and application of intelligent feeding systems for sows.

Aiming to solve the problems of high mortality rate of nursery pigs,  high feed waste,  and high labor intensity,  a prototype testing platform for a porridge machine was built,  and an intelligent porridge feeding system for nursery pigs was designed. The system was composed of four parts: a mechanical body,  a handheld terminal,  a central controller,  and a cloud platform. Orthogonal experiment was carried out on the installation height of screw conveyor, the inner diameter of discharge channel and the structure of broken arch in the mechanical body. The optimal parameters combination is obtained as follows: screw conveyor installation height was 60 mm,  discharging channel diameter was 42 mm,  the broken arch structure adapted the combination form of upper and lower broken arch structure,  the filling efficiency was optimal,  the value of λ was 14.2 g/r,  and the water supply error of the system was 143%. When the ratio of water to feed was equal and greater than 1.5∶1,  the liquid level monitoring device can satisfy the requirements of the system. The research can provide a reference for the research and development of feeding equipment for piglets.

In order to improve the efficiency of waste gas purification in largescale pig farming, this paper studied the horizontal exhaust gas purification system using the acid spray scrubbing method in mechanically ventilated pig houses. Combined with the orthogonal test plan and using the range analysis and variance analysis methods, this paper analyzed the influence mechanism of key process parameters such as the ventilation speed, the frequency of the pickling pump, and the spray direction in the acid spray scrubbing method on the efficiency of ammonia removal. The test results showed that the spray direction has a significant effect on the ammonia removal efficiency. The average ammonia removal efficiency was 6.87% higher than the downstream efficiency and 18.88% higher than the countercurrent efficiency. The reduction of the ventilation wind speed can improve the ammonia removal efficiency. When ensuring sufficient acid cleaning pump pressure, the increase of the acid cleaning pump pressure had little effect on the ammonia removal efficiency. Through range analysis and variance analysis, it can be concluded that in horizontal exhaust gas purification system, the spraying direction was crossflow, picking frequency was 30 Hz, and ventilation wind speed was 1 m/s, the ammonia removal effect was theovetically the best. Finally, according to the ammonia removal efficiency and power consumption of each combination, the optimal scheme was listed to provide reference for product design.

In order to obtain the location information of animal husbandry autonomous mobile robots in largescale breeding farms, a map building and location algorithm based on TWODIMENSIONAL code (AprilTag) was proposed. A monocular camera on the robot was used to observe the AprilTag in the environment. The state of the robot was estimated based on extended Kalman Filter (EKF) by integrating the measurement of the robot motion with the wheel odometer. On this basis, the pose of each AprilTag in the global coordinate system was calculated. In order to restrain the increase of cumulative errors in the process of building the graph, the estimated results were taken as the initial values before the loop, and the leastsquares problem with the projection relation was constructed, and the LevenbergMarquardt (LM) method was used to solve the leastsquares problem. The global optimization was used to eliminate the cumulative error when the loop was detected. Experimental results showed that the proposed algorithm can build the AprilTag map in a large range, and the average error and maximum error of robot localization using the map were 0.075 86 m and 0.303 93 m, respectively.

In order to understand the change rule of the influencing factors of the environmental regulation of the reserve sow houses in summer, the temperature, relative humidity and ammonia in the reserve sow house were tested by changing the number of fans, fan combination, and wet curtain. The results showed that the more fans were turned on, the greater the range of temperature decrease (the maximum range was 2.66 ℃) and the greater the range of relative humidity increase (the maximum range was 9.76%), the more uniform the distribution of temperature and humidity in the ventilation direction in the reserve sow house. The maximum cooling range was 4.27 ℃, and the maximum increase of relative humidity was 24.17%. When the left and right fans were turned on, the distribution of temperature and humidity in the shed was more uniform than that of the left or right middle fans. The average THI index in the reserve female sheds was 79.5, with a decrease rate of 3.86%, which could effectively alleviate the heat stress of pigs. The uneven coefficient of temperature field and relative humidity field in the reserve piggery were 0.42 and 0.5, respectively, indicating that the distribution of temperature and humidity in the reserve piggery was relatively uniform.