In order to monitor feeding behavior of sows and further attain the sow's precise feeding, an intelligent production performance testing system was designed in this study, which played functions in sows' automatic identification, body weight perception, automatic feeding data acquisition and data analysis simultaneously. The system was composed of electric ear tag identification module, precise feed flow control module, feed trough and boar weighing module, data communication and remote control module. The mechanical device system was constituted of feeding bin, brackets, railing and blocking apron. The mechanical device system was constituted of feeder's vertical wall, weighting platform, flapper, feed loading device, feed bin, control box, switch of discharge and ear tag recognizer. Electronic control systems included microprocessor (LPC1766, ARM Cortex-M3, Working temperature -40-105℃, Operating voltage 2.0-3.6 V, flash 256 K, low power consumption et al.), RS232 reader port, data storage chip (the default storage capacity is 256 KB), circuit of watchdog, weighing circuit, exterior-drivers circuit, JTAG connector circuit and stabilivolt source circuit. Among above, the sensor used for pigs weighing was Delux ADS1232 which had 2 rate options, 10 times per second and 80 times per second, with high precision and large range of features. The performance testing experiment revealed that: 1) the system's precision meets the monitoring requirement of sow production performance. The discharge rate of feeder depended on the level of feed in stock bin, and the average amount of unloading feed was 93±2 g at one time; the range of pig weighing was 0-200 kg with the precision error below 10 g, and the dynamic weighing error was below 0.5% of pig's weight. 2) The feeding behavior monitor for 40 gilts (25-60 kg) showed that the frequency of free feed intake was 10-12 times per day, the average feed time was 78 min, the feed conversion ratio was 2.33:1, and their weight gain was converged to the Gompertz curves (e.g. Wt=172.1exp(-4.0187exp(-0.0122*t)),Wt means body weight, kg;t means day old, day), the predicted decreasing daily weight gain of growing pigs by Gompertz curve occurred at day 111-117, with corresponding inflection point weight in the range of 63-64 kg. The observed and predicted results above could precisely determine the growth performance, indicating that the software systems and hardware devices could satisfy the requirement of growth performance determination in sows. 3) The wiper motor rather than early stepping motor was used in feed discharging control system, which reduced the cost of production. In addition, the combined wiper motor with cylindrical scraper structure decreased the discharge rate of feeder and improved the precision of unloading control system. 4) The core chip in control system was imported, multi-redundant, and protection systems were applied in circuit design. Multiply functional verification was adopted in software writing. The redundancy design in software and hardware eliminated the interference of power, electrical machine and electromagnetic wave, and improved the systems' reliability and stability. 5) The collected data could be saved or transferred, which facilitates the accumulation of pig production, data mining and sow breeding.%为开展种猪生产性能的智能化、自动化测定及开展种猪采食行为学研究,该研究设计了一种集自动识别、体质量感知、采食行为数据自动采集、数据分析与处理于一体的种猪生产性能智能测定系统.该系统主要由猪只耳标识别模块、精准下料控制模块、料槽及猪只个体称质量模块、现场数据通讯模块及远程中央控制模块组成.系统机械部分主要包括饲喂站的竖直侧墙、称质量平台、活动挡板、下料机构、料仓、控制盒、出料口开关及耳标识读器等组成.电路控制系统包括微处理器(LPC1766,内核为ARM Cortex-M3内核的微控制器)、RS232读卡器接口、数据存储芯片(预设存储256 KB数据)、看门狗电路、称质量电路、外围驱动电路、JTAG接口电路及稳压电源电路.系统性能测试结果表明:1)测量精度如下:饲喂下料没有范围限制,取决于喂料仓的储料状态,单次下料量及动态误差为93±2g;猪只体质量秤量程为0~200 kg,计量精度为10 g,称量动态误差占猪只体质量的0.5%以下,符合测定需求;2)对40头种公猪后裔的生长肥育猪饲喂测试结果表明,在25~60 kg体质量范围内,自由采食日均次数10~12次,日均采食时间78 min,测试期间料肉比(FCR)为2.33:1,且生长规律符合Gompertz曲线,通过该模型预测的日增质量下降的拐点发生111~117 d之间,对应的拐点体质量在63~64 kg范围内.上述实际观察及预测结果较好地反映了测定对象的生产性能,开发的软件及硬件系统达到了种猪生产性能测定的要求;3)系统下料控制部分,首次采用雨刷电机取代早期采用的步进电机,不仅成本下降,尤其结合圆柱式刮板下料机构,降低了单次下料量,改善了下料的精度;4)系统核心芯片采用进口器件,电路设计采用多重冗余和保护电路,软件的编写采用了多重功能验证,并通过长期可靠性测试;软件和硬件的冗余设计,提高了控制系统的可靠性,消除来自电源、电机、电磁波干扰,该测定系统具有极高的可靠性;测定的数据通过计算机系统可长期保存或升迁,便于数据量的积累和开展种猪选育的大数据挖掘分析.
展开▼
