Research On Dairy Farm Precision Management Based On Multiscale Environment Analysis

Environment management in Large-scale dairy farms is important to promote animal welfare and ensure stable productivity.In order to find the way to promote the environment management efficiency and reduce the cost from environment control,this study takes Hulunbuir Xiertala dairy farm as the research object and the precision management of large-scaled farm based on multi-scale environmental analysis is carried out.Aiming at finding the different relationship between feeding management and climate,the interaction model of internal and external environment factors based on multy-scales data and the computer fluid dynamics model are studied to find the changing regulation of micro temperature field in the cow barn and constructed the optimized environmental control methods.Aiming at the different degree of impact on economic benefits in large-scale dairy farms by different environmental control method,a comprehensive evaluation index of economic benefits of cow barn based on environmental regulation and optimization was constructed.Based on different DEA analysis models,an economic benefit of cow barns under different environmental management conditions is conducted.In order to solve the difficulty of multi-source data sharing in dairy farms,data inheritance method and resource sharing technology framework are studied and used to construct the dairy farm information management system.The following studies have been carried out:（1）Inside and outside environment factors interaction model based on multiscale data intergration.This study analyzes multiscale environment data inside and outside the barn by correlation analysis and regression model,including temperature,relative humidity,gas content,wind speed and so on.Through the research on the climate data for many years,it is found that the weather suitable for dairy farming in this area has reached nearly 40%.According to temperature change rules the hulunbuir climate can be divided into two seasons,the warm and the cold season.The warm season usually starts from late April and ends in late September,and the highest temperature can reach to 40℃.The cold season is from late September to April with the lowest temperature reaching to-42℃.According to the climate rescouces,the cooling mesures are needed in warm season to prevent cows from heat stress and the keeping warm measures are needed in coldseason to prevent cows from cold stress.Through the analysis of the relationship between the internal and external environmental factors and the predictive regression,it is found that the change of the internal temperature of the barn usually lags behind the change of the outside temperature.The cause is that the barn management measures adopted are different in diferent seasons.The internal and external environmental relations are obviously different in the warm and cold seasons.The influence of the external environment on the inside environment of the barn in the warm season is significant.There is a significant linear correlation with temperature and humidity inside and outside the barn in warm season,and R-Squared of the regression module is 0.9711and 0.9083.There is also a significant linear correlation between the CO₂ and NH₃ inside the barn in warm season,and regression modules with good fitting degrees are moduled.Above all,the correlation between gas concentration and environmental factors such as temperature and humidity in cold season barn is weak,and the prediction regression model can not be well established through temperature and humidity.In cold season,the indoor environment is less affected by the outdoor environment,and the correlation between the indoor gas and environmental factors such as temperature and humidity is weak,and also the prediction regression model can not be well established through temperature and humidity.（2）Regulation mechanism and management of barn environment in warm season based on micro-oenvironment simulation.The energy exchange between the interior structures of livestock barn in warm season was studied,and the grid model of the barn was constructed.The RNG k-εturbulence model was used to simulate and analyze the effects of different wind directions on the temperature field inside the barn.It is found that the boundary conditions of temperature field simulation based on measured data can be replaced by the predicted temperature by regression models,and the simulation results are consistent.When the outside temperature is higher than the inside temperature,the inside temperature rises rapidly within 30minutes,but different wind directions have different effects on the inside area of the barn.The temperature difference in the barn is large.The temperature heterogeneity in the south wind group is the largest,while that in the north wind group is the smallest.The temperature in the South cowbed is generally higher than that in the North cowbed.When the outside temperature is lower than the inside temperature,the inside temperature drops to the outside temperature within 30 minutes.The difference of the total temperature in the barn is not significant,the maximum temperature difference is less than1℃.The temperature changes in different areas of the barn caused by different wind directions are different,and the cooling efficiency of the south wind group is better than that of other groups.According to the analysis,CFD simulation can be used to reduce the number of environmental monitoring sensors deployed inside and outside the house,and to improve the overall temperature of the house;at the same time,when the outside temperature is higher than 25℃,measures such as reducing the air temperature flowing into the building and reducing the influence of solar radiation should be taken to slow down the rapid rise of the inside temperature;when the outside temperature drops rapidly,natural ventilation can be used to cool down the inside temperature.（3）Economic efficiency analysis according to environmental regulation and optimization based on barns.Quantitative research and analysis of the relationship between environmental factors of barn and production performance and health status are made in this study.Daily average milk production and cows eliminating amount represent the production performance and health level of barn respectively.Through forecasting results and empirical analysis,the ideal input-output benefit of barn is constructed.The economic benefit index between actual barn and simulated barn is compared,and the economic benefit of barn unit is evaluated and scored based on environmental control cost for barn.The results show that the use of large cooling equipment only in warm season is not conducive to the economic efficiency of barn,but should be combined with warm season insulation measures.Through the analysis of CCR-DEA model and super-efficiency DEA model,it is found that the economic benefits of No.3and No.4 barn,which almost did not use mechanical temperature control facilities,do not reach the optimal input-output situation.In terms of the input of environmental control,the barn without environmental control has the worst economic efficiency.But it is not that the greater the environmental input,the higher the economic efficiency.When considering the environmental control,we should reduce the cost as much as possible,and improve the management measures of dairy farms in order to achieve the optimization of economic efficiency.According to the climatic characteristics of Hulunbuir,the application of heating facilities in cold season should be given priority to barn,and the application of cooling facilities in warm season should be increased on the basis of this,so as to maximize the benefit of investment.（4）Technology and application of dairy farm multi-source data integration based on resource sharingAiming at the characteristics of heterogeneity,distribution and time series of perceived data in the environment of livestock barns,this paper studies source data layer,intermediate processing layer and database layer from the bottom to the top of the information system application.Inclusion of data protocols from different sources such as sensors and information systems is achieved through Middleware Technology.The ontology model of animal husbandry resources is established by five-element analysis method.Data distributed storage service is builded by SWIFT Technology.Unified resource operation interface is developed using Restful Web API technology.These technologies provide a technical framework for data sharing and utilization.Based on the technology framework of resource sharing,a dairy farm information precision management system is developed by ASP.NET MVC and JQuery,which realizes the functions of dairy farm environment monitoring and dairy cow resume query and so on.