CAJ | 학술논문

　　针对畜禽舍饲养环境多变量耦合的特点，提出了一种多测点传感器监测方案，以适应其复杂的饲养环境。该方案采用最优加权算法对同类传感器数据进行融合，并通过改进的 D-S （ Dempster-Shafer ）证据理论算法将不同传感器的多个参数整合，并根据整合结果对舍内环境状况做出准确判断。该方案克服了传统监测手段的单一性，避免了分别处理各环境因子信息数据的不确定性，增强了畜禽舍环境监控系统的鲁棒性。实验表明，将该算法应用于畜禽舍环境监测系统后，监测结果的可靠性与准确度均有明显提高。
　　침대축금사사양배경다변량우합적특점，제출료일충다측점전감기감측방안，이괄응기복잡적사양배경。해방안채용최우가권산법대동류전감기수거진행융합，병통과개진적 D-S （ Dempster-Shafer ）증거이론산법장불동전감기적다개삼수정합，병근거정합결과대사내배경상황주출준학판단。해방안극복료전통감측수단적단일성，피면료분별처리각배경인자신식수거적불학정성，증강료축금사배경감공계통적로봉성。실험표명，장해산법응용우축금사배경감측계통후，감측결과적가고성여준학도균유명현제고。
This thesis suggests a kind of monitoring plan with multi-measure points and multi-sensors that can be applied to the detection of the animal building environment which is multivariable coupling .This method combines opti-mal weighting algorithm and D-S( Dempster-Shafer ) Theory algorithm to fuse different sensors data in order to make an accurate judgment on the animal building environment .Thus , it enhances the robustness of environmental monitoring and control system .The experimental result shows that this method improves the reliability and accuracy of the animal build -ing's environmental monitoring system .