Study On Abnormal Detection Of Piggery Environment Based On Multivariate Data

China’s pig industry has experienced more than 40 years of development.Advanced breeding technologies at home and abroad have been widely popularized and applied.The technical level of domestic pig farms has improved compared with the past.However,compared with western developed countries,China’s pig industry still has a very big gap in automation,informatization,high efficiency and intelligence.The breeding of pig farms is affected by many factors,of which the growth environment is a very important factor.A comfortable growth environment is an important guarantee to reduce diseases,improve defense capability and enhance the slaughter rate.If data mining can find the parameters that affect the living environment comfort of pigs,it will create a comfortable environment for pig farm breeding through automatic adjustment.Therefore,this paper organically combines the data of pig breeding environment and pig activity through the main technologies used in the intelligent management process of pig farms,through data analysis and collection,through scientific processing of pig farm data,establishes a certain connection between video information and data information inside the pig farm,obtains a fitting curve of the data,and finds a comfortable living environment for the pig farm.Based on the research topic of pig activity,this paper aims to provide directions for pig farms to explore breeding methods and improve breeding environment through data analysis.The research content of this paper mainly includes the following aspects: First,based on the real-time tracking video of pig farm,according to the activity law of pigs,the time interval of research data is selected,and the minimum set of corresponding environmental monitoring data is extracted according to this time interval.Secondly,according to the data characteristics of different data sources,data preprocessing is carried out on each data set,and a data fusion algorithm is designed to realize the data fusion of the preprocessed data sources.Thirdly,based on curve interpolation and curved surface interpolation,the activity characteristics of pig behavior under the comprehensive influence of single environmental variables and multiple environmental variables are analyzedrespectively,and the change rule of pig behavior with environment is extracted.With the participation of pig farm breeding technicians,the regulation model of pig growth environment index is determined,which provides support for pig farm breeding ambient intelligence monitoring.Through research,the analysis results can be used to better guide the production of pig farms,further ensure the benign development of production,control the production process of pig farms,effectively reduce production costs,improve production efficiency and improve production capacity.From the point of view of disease control,by analyzing the data collected from pig farms,the occurrence of diseases can be avoided as early as possible and the losses caused by diseases in pig farms can be reduced.Therefore,it is of great theoretical and practical significance to deeply mine and analyze the data collected from pig farms.