Breeding management method based on Hu sheep breeding data

Abstract

The invention relates to the technical field of Hu sheep breeding management, in particular to a breeding management method based on Hu sheep breeding data. The method comprises the following steps: acquiring breeding data and breeding indexes of Hu sheep; analyzing the breeding data based on the breeding data index to obtain a Hu sheep breeding result; the method comprises the following steps: converting breeding data, Hu sheep growth rate influence factors and a corresponding time sequence into a matrix X, carrying out multi-path delay embedding transformation on the matrix X by adopting a multi-path delay embedding transformation technology to obtain a high-order Hankel tensor, mapping the high-order Hankel tensor into a core tensor by adopting a tensor decomposition technology, and taking the core tensor as an input feature to obtain a Hu sheep growth rate vector; inputting the input features into an LSSVM model to predict the growth rate of the Hu sheep, and analyzing the growth rate of the Hu sheep based on growth indexes to obtain a growth result of the Hu sheep; and generating a Hu-sheep breeding report based on the Hu-sheep breeding result and the Hu-sheep growth result. The breeding data can be deeply mined and analyzed, and the Hu sheep growth prediction precision and the management decision scientificity are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Hu sheep breeding management, and in particular to a breeding management method based on Hu sheep breeding data. Background Art

[0002] As a unique and excellent sheep breed in my country, Huyang occupies an important position in animal husbandry with its delicious meat, strong reproductive ability and good adaptability. However, traditional Huyang breeding methods often rely on experience and judgment, lack of scientific data support, resulting in low breeding efficiency and insufficient disease prevention and control capabilities. With the continuous development of technologies such as the Internet of Things and big data, intelligent management based on breeding data has become an important way to improve the level of Huyang breeding.

[0003] At present, although there are some intelligent management systems for animal husbandry on the market, there are few systems that are deeply optimized for the specific needs of Huyang breeding.

[0004] Incomplete data collection: Under the traditional breeding model, breeding data mainly relies on manual records, which have problems such as missing data and large errors, making it difficult to fully reflect the growth status and breeding environment of Hu sheep.

[0005] Inaccurate data analysis: Existing systems often lack in-depth mining and analysis of Huyang breeding data, and are unable to accurately predict disease risks, optimize feed ratios, etc., resulting in a lack of scientific basis for management decisions.

[0006] Unscientific management decisions: Due to insufficient data support, breeding managers often rely on experience and judgment when formulating breeding plans and adjusting feed formulas, making it difficult to achieve refined management.

[0007] Therefore, there is an urgent need for a breeding management method based on Huyang breeding data to solve the above problems. Summary of the invention

[0008] The purpose of the present invention is to provide a breeding management method based on lake sheep breeding data: to solve the technical problems that the existing solutions have incomplete collection of lake sheep breeding data, inaccurate data analysis, unscientific management decisions, and are difficult to meet the refined management needs of lake sheep breeding.

[0009] The purpose of the present invention can be achieved through the following technical solutions:

[0010] A breeding management method based on Huyang breeding data, the method comprising:

[0011] Obtain the breeding data and breeding indicators of Hu sheep, where the breeding data includes breeding environment data and non-breeding environment data, and the breeding indicators include breeding data indicators and growth indicators;

[0012] Analyze the breeding data based on the breeding data indicators and obtain the results of Huyang breeding;

[0013] Based on the growth index, the factors affecting the growth rate of Hu sheep and the time corresponding to the factors affecting the growth rate of Hu sheep are obtained, the time corresponding to the factors affecting the growth rate of Hu sheep is converted into a time series, and the factors affecting the growth rate of Hu sheep and the corresponding time series are converted into a matrix X. The matrix X is subjected to a multi-way delayed embedding transformation using a multi-way delayed embedding transformation technique to obtain a high-order Hankel tensor, and then the high-order Hankel tensor is mapped into a core tensor using a tensor decomposition technique. The core tensor is used as an input feature, and the input feature is input into the LSSVM model to predict the growth rate of Hu sheep. The growth rate of Hu sheep is analyzed based on the growth index to obtain the growth result of Hu sheep.

[0014] Generate a Hu Yang breeding report based on the Hu Yang breeding results and Hu Yang growth results.

[0015] Furthermore, the breeding data is analyzed based on the breeding data indicators to obtain the results of Huyang breeding, which specifically includes the following processes:

[0016] Based on the breeding data, the average temperature of the breeding environment of the Hu sheep in the management period is obtained, and based on the breeding data indicators, the preset temperature of the Hu sheep breeding in the management period is obtained, and the difference WD between the average temperature and the preset temperature of the Hu sheep breeding is calculated, wherein the management period is the preset breeding period;

[0017] Based on the breeding data, the average humidity of the breeding environment of Huyang in the management period is obtained; based on the breeding data indicators, the preset humidity of Huyang breeding in the management period is obtained; and the difference SD between the average humidity and the preset humidity of Huyang breeding is calculated;

[0018] Based on the breeding data, the number of Hu sheep vaccination types within the management period is obtained; based on the breeding data indicators, the number of Hu sheep vaccination types that should be vaccinated within the management period is obtained; and the proportion of the number of Hu sheep vaccination types to the number of Hu sheep vaccination types that should be vaccinated is calculated.

[0019] Based on the breeding data, the feed intake of Hu sheep during the management period is obtained; based on the breeding data indicators, the feed intake that Hu sheep should take during the management period is obtained; and the difference CL between the feed intake of Hu sheep and the feed intake that Hu sheep should take is calculated;

[0020] Substitute WD, SD, YM and CL into the calculation formula of Huyang breeding index to obtain Huyang breeding index LMS, which is as follows:

[0021]

[0022] Among them, γ, α, and θ are weight coefficients, which are 0.2, 0.6, and 0.2 respectively. F is a constant coefficient determined by the season. Different seasons correspond to different F values, which are user-defined. C is the average injection volume of the vaccine.

[0023] The Lake Sheep Breeding Index LMS is recorded as the Lake Sheep Breeding result.

[0024] Furthermore, the conversion of the factors affecting the growth rate of Hu sheep and the corresponding time series into matrix X specifically includes the following processes:

[0025] The factors affecting the growth rate of Hu sheep and the time series are formed into a matrix Where X is a multidimensional array, R represents the real number domain, each element of the matrix X is a real number, and the size, size or dimension of each dimension of X is I1, I2, ..., I N and T, with a total of N+1 dimensions; T is the length of the time series, N is the number of factors affecting the growth rate of Hu sheep, and each row of the matrix X (I1×I2×...×I N ) represents the time vector of the factors affecting the growth rate of Hu sheep. Each component of the vector represents the value of the factors affecting the growth rate of Hu sheep at different times. Each column of the matrix X (X1×X2×...×X N ) represents the growth rate factor vector of Hu sheep at different times, and each component of the vector represents the value of the factor affecting the growth rate of Hu sheep on that day.

[0026] Furthermore, the multi-path delay embedding transformation technology is used to perform multi-path delay embedding transformation on the matrix X to obtain a high-order Hankel tensor, which specifically includes the following process:

[0027] Hankel(X)=FOLD (I,τ) (X×S1×S2×...×S N );

[0028] Among them, FOLD (I,τ) is the folding operator, FOLD (I,τ) : To copy the matrix.

[0029] Furthermore, mapping the high-order Hankel tensor to the core tensor through tensor decomposition technology specifically includes the following process:

[0030] Based on Tucker decomposition, high-order Hankel tensors are mapped to core tensors:

[0031] Hankel(X)=G×U 1 ×U 2 ×...×U N ;

[0032] in, is the core tensor, is the principal component of the tensor at each order, called the projection matrix, and each projection matrix is ​​orthogonal to each other.

[0033] Furthermore, the input features are input into the LSSVM model to predict the growth rate of Hu sheep, which specifically includes the following process:

[0034] Build the LSSVM model:

[0035] Given a training set D = {(x i ,y i )|(i=1,2...n)}, where x i ,y i are the input and output of the model respectively, and n is the size of the training set;

[0036] The LSSVM regression prediction function is: f(x) = ω Y +b;

[0037] Among them, b is the constant of the regression prediction function, ω is the regression matrix of the regression prediction function; ω Y is the device matrix of the regression prediction function;

[0038] The weight vector of the regression function is obtained by optimizing the penalty function:

[0039]

[0040] C1 is the penalty coefficient, γ is the regularization parameter, e i is the training error;

[0041] The regression prediction model formula is:

[0042] Among them, K(x, z) is the kernel function, α i is the Lagrange multiplier;

[0043] Prediction of Hu sheep growth rate based on LSSVM model.

[0044] Furthermore, the growth rate of Hu sheep was analyzed based on growth indicators, and the growth results of Hu sheep were obtained, which specifically included the following processes:

[0045] Based on the growth index, the preset Hu sheep growth rate threshold is obtained to determine whether the Hu sheep growth rate exceeds the preset Hu sheep growth rate threshold. If so, a signal that the Hu sheep growth condition meets the standard is generated; if not, a signal that the Hu sheep growth condition does not meet the standard is generated. The signal that the Hu sheep growth condition meets the standard or the signal that the Hu sheep growth condition does not meet the standard is recorded as the Hu sheep growth result.

[0046] Furthermore, before analyzing the breeding data based on the breeding data indicators, the method also includes:

[0047] For breeding data with small missing data and short time intervals, linear interpolation is used to fill in the missing data. The linear interpolation method is expressed as:

[0048]

[0049] Among them, the values ​​corresponding to the breeding data at time k and time k+j are Q k and Q k+j , Q k+i is the missing data at the k+ith moment.

[0050] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0051] The present invention obtains the breeding data and breeding indicators of Huyang, wherein the breeding data includes breeding environment data and non-breeding environment data, and the breeding indicators include breeding data indicators and growth indicators; the breeding data is analyzed based on the breeding data indicators to obtain the Huyang breeding results; the breeding data and the time series corresponding to the factors affecting the growth rate of Huyang are converted into a matrix X, and the matrix X is subjected to a multi-path delayed embedding transformation technology to perform a multi-path delayed embedding transformation to obtain a high-order Hankel tensor, and then the high-order Hankel tensor is mapped to a core tensor by a tensor decomposition technology, and the core tensor is used as an input feature, and the input feature is input into the LSSVM model to predict the growth rate of Huyang, and the growth rate of Huyang is analyzed based on the growth indicator to obtain the growth result of Huyang; and a Huyang breeding report is generated based on the Huyang breeding results and the Huyang growth results. The present invention can perform in-depth mining and analysis on the breeding data, and improve the prediction accuracy of Huyang growth and the scientific nature of management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0053] Figure 1 It is a workflow diagram of a breeding management method based on Huyang breeding data in an embodiment of the present invention;

[0054] Figure 2 It is a workflow diagram of another breeding management method based on Huyang breeding data in an embodiment of the present invention;

[0055] Figure 3 It is a workflow diagram of another breeding management method based on lake sheep breeding data in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0058] This embodiment provides a breeding management method based on Huyang breeding data. Figure 1 is a workflow diagram of a breeding management method based on Huyang breeding data according to an embodiment of the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0059] Step S101: Acquire the breeding data and breeding indicators of Hu sheep, wherein the breeding data includes breeding environment data and non-breeding environment data, and the breeding indicators include breeding data indicators and growth indicators;

[0060] Step S102: Analyze the breeding data based on the breeding data indicators to obtain the results of Hu sheep breeding;

[0061] Step S103: Based on the growth index, the factors affecting the growth rate of Hu sheep and the time corresponding to the factors affecting the growth rate of Hu sheep are obtained, the time corresponding to the factors affecting the growth rate of Hu sheep is converted into a time series, and the factors affecting the growth rate of Hu sheep and the corresponding time series are converted into a matrix X, and the matrix X is subjected to a multi-path delayed embedding transformation using a multi-path delayed embedding transformation technology to obtain a high-order Hankel tensor, and then the high-order Hankel tensor is mapped into a core tensor using a tensor decomposition technology, and the core tensor is used as an input feature;

[0062] Step S104: inputting the input features into the LSSVM model to predict the growth rate of the Hu sheep, analyzing the growth rate of the Hu sheep based on the growth index, and obtaining the growth result of the Hu sheep;

[0063] Step S105: Generate a Hu sheep breeding report based on the Hu sheep breeding results and the Hu sheep growth results.

[0064] In summary, the present invention obtains the breeding data and breeding indicators of Hu sheep, wherein the breeding data includes breeding environment data and non-breeding environment data, and the breeding indicators include breeding data indicators and growth indicators; the breeding data is analyzed based on the breeding data indicators to obtain the Hu sheep breeding results; the breeding data and the time series corresponding to the factors affecting the growth rate of Hu sheep are converted into a matrix X, and the matrix X is subjected to a multi-way delayed embedding transformation technology to perform a multi-way delayed embedding transformation to obtain a high-order Hankel tensor, and then the high-order Hankel tensor is mapped to a core tensor by a tensor decomposition technology, and the core tensor is used as an input feature, and the input feature is input into the LSSVM model to predict the growth rate of Hu sheep, and the growth rate of Hu sheep is analyzed based on the growth indicator to obtain the growth results of Hu sheep; a Hu sheep breeding report is generated based on the Hu sheep breeding results and the Hu sheep growth results. The present invention can perform in-depth mining and analysis on breeding data, improve the prediction accuracy of Hu sheep growth and the scientific nature of management decisions.

[0065] In some embodiments, Figure 2 is another workflow diagram of a breeding management method based on lake sheep breeding data according to an embodiment of the present invention, such as Figure 2 As shown, the following steps are specifically included to analyze the breeding data based on the breeding data indicators and obtain the results of Huyang breeding:

[0066] Step S201: obtaining the average temperature of the breeding environment of the Hu sheep in the management period based on the breeding data, obtaining the preset temperature of the Hu sheep breeding in the management period based on the breeding data indicators, and calculating the difference WD between the average temperature and the preset temperature of the Hu sheep breeding, wherein the management period is the preset breeding period;

[0067] Step S202: obtaining the average humidity of the breeding environment of the Hu sheep in the management period based on the breeding data, obtaining the preset humidity for Hu sheep breeding in the management period based on the breeding data indicators, and calculating the difference SD between the average humidity and the preset humidity for Hu sheep breeding;

[0068] Step S203: obtaining the number of Hu sheep vaccine types in the management period based on the breeding data, obtaining the number of Hu sheep vaccine types that should be vaccinated within the management period based on the breeding data indicators, and calculating the ratio YM of the number of Hu sheep vaccine types to the number of Hu sheep vaccine types that should be vaccinated;

[0069] Step S204: obtaining the feed intake of the Hu sheep in the management period based on the breeding data, obtaining the required feed intake of the Hu sheep in the management period based on the breeding data indicators, and calculating the difference CL between the feed intake of the Hu sheep and the required feed intake of the Hu sheep;

[0070] Step S205: Substitute WD, SD, YM and CL into the calculation formula of the Huyang breeding index to obtain the Huyang breeding index LMS, which is as follows:

[0071]

[0072] Among them, γ, α, and θ are weight coefficients, which are 0.2, 0.6, and 0.2 respectively. F is a constant coefficient determined by the season. Different seasons correspond to different F values, which are customized by the user. C is the average vaccine injection volume. It is worth noting that the constant coefficient F corresponds to different values ​​in different seasons, and the constant coefficient F in spring is larger than that in the other three seasons.

[0073] Step S206: Record the Lake Sheep Breeding Index LMS as the Lake Sheep Breeding Result.

[0074] In some embodiments, converting the factors affecting the growth rate of Hu sheep and the corresponding time series into a matrix X specifically includes the following process:

[0075] The factors affecting the growth rate of Hu sheep and the time series are formed into a matrix Where X is a multidimensional array, R represents the real number domain, each element of the matrix X is a real number, and the size, size or dimension of each dimension of X is I1, I2, ..., I N and T, with a total of N+1 dimensions; T is the length of the time series, N is the number of factors affecting the growth rate of Hu sheep, and each row of the matrix X (I1×I2×...×I N The time vector representing the factors affecting the growth rate of Hu sheep. Each component of the vector represents the value of the factors affecting the growth rate of Hu sheep at different times. Each column of the matrix X (X1×X2×...×X N ) represents the growth rate factor vector of Hu sheep at different times, and each component of the vector represents the value of the factor affecting the growth rate of Hu sheep on that day.

[0076] In some embodiments, the multi-path delay embedding transformation technique is used to perform multi-path delay embedding transformation on the matrix X to obtain a high-order Hankel tensor, which specifically includes the following process:

[0077] Hankel(X)=FOLD (I,τ) (X×S1×S2×...×S N );

[0078] Among them, FOLD (I,τ) is the folding operator, FOLD (I,τ) : To copy the matrix.

[0079] In some embodiments, mapping a high-order Hankel tensor to a core tensor by using a tensor decomposition technique specifically includes the following process:

[0080] Based on Tucker decomposition, high-order Hankel tensors are mapped to core tensors:

[0081] Hankel(X)=G×U 1 ×U 2 ×...×U N ;

[0082] in, is the core tensor, is the principal component of the tensor at each order, called the projection matrix, and each projection matrix is ​​orthogonal to each other.

[0083] In some embodiments, inputting the input features into the LSSVM model to predict the growth rate of Hu sheep specifically includes the following process:

[0084] Build the LSSVM model:

[0085] Given a training set D = {(x i ,y i )|(i=1,2...n)}, where x i ,y i are the input and output of the model respectively, and n is the size of the training set;

[0086] The LSSVM regression prediction function is: f(x) = ω Y +b;

[0087] Among them, b is the constant of the regression prediction function, ω is the regression matrix of the regression prediction function; ω Y is the device matrix of the regression prediction function;

[0088] The weight vector of the regression function is obtained by optimizing the penalty function:

[0089]

[0090] C1 is the penalty coefficient, γ is the regularization parameter, e i is the training error;

[0091] The regression prediction model formula is:

[0092] Among them, K(x, z) is the kernel function, α i is the Lagrange multiplier;

[0093] Prediction of Hu sheep growth rate based on LSSVM model.

[0094] In some embodiments, Figure 3 is another workflow diagram of a breeding management method based on lake sheep breeding data according to an embodiment of the present invention, such as Figure 3 As shown in the figure, the growth rate of Hu sheep is analyzed based on growth indicators, and the growth results of Hu sheep are obtained, which specifically include the following processes:

[0095] Step S301: obtaining a preset Hu sheep growth rate threshold based on the growth index;

[0096] Step S302: determine whether the growth rate of Hu sheep exceeds the preset Hu sheep growth rate threshold, if yes, proceed to step S303, if no, proceed to step S304;

[0097] Step S303: generating a signal indicating that the growth status of the Hu sheep has reached the standard;

[0098] Step S304: generating a signal indicating that the growth condition of the Hu sheep does not meet the standard;

[0099] Step S305: Recording the signal indicating that the growth condition of the Hu sheep meets the standard or the signal indicating that the growth condition of the Hu sheep does not meet the standard as the growth result of the Hu sheep.

[0100] Furthermore, before analyzing the breeding data based on the breeding data indicators, the method also includes:

[0101] For breeding data with small missing data and short time intervals, linear interpolation is used to fill in the missing data. The linear interpolation method is expressed as:

[0102]

[0103] Among them, the values ​​corresponding to the breeding data at time k and time k+j are Q k and Q k+j , Q k+i is the missing data at the k+ith moment.

[0104] In some embodiments, targeted management decisions are made based on the lake sheep breeding report:

[0105] 1. Optimize feed ratio

[0106] Scientifically formulated feed:

[0107] The feed ratio should be adjusted according to the needs of Hu sheep at different growth stages. For example, the proportion of energy feed, such as corn meal and barley, can be increased during the fattening period, while protein feed, such as soybean meal and rapeseed cake, needs to be increased during the lactation and pregnancy period.

[0108] Introducing advanced feed formulation technologies, such as using total mixed ration technology, to ensure that the feed is nutritionally balanced and easy to digest and absorb.

[0109] Add necessary feed additives, such as vitamins, minerals and probiotics, to improve feed utilization and the resistance of Hu sheep.

[0110] Cost Control:

[0111] Make full use of local resources, explore high-quality forage and feed raw materials, and reduce feed costs.

[0112] Manage feed inventory in a refined manner to avoid waste and losses from expiration.

[0113] 2. Adjust the breeding plan

[0114] Optimizing the breeding season:

[0115] The breeding and lambing time of Hu sheep are adjusted according to climatic conditions and market demand. For example, breeding in warm seasons is conducive to the survival and growth of lambs.

[0116] Implement a breeding system of three parities in two years, and increase the reproduction rate of ewes and the slaughter rate of lambs by reasonably arranging the breeding time.

[0117] Strengthen breeding sheep management:

[0118] Introduce large, fast-growing, high-quality rams to avoid inbreeding and improve the quality of offspring.

[0119] Strengthen the feeding and management of breeding rams to ensure their good physical condition and improve the quality of semen.

[0120] Carry out meticulous management of ewes to ensure that they become pregnant during the optimal breeding period and improve the conception rate.

[0121] 3. Strengthen feeding management

[0122] Environmental Control:

[0123] Ensure that the sheep house environment is dry, ventilated and well-lit to provide a comfortable growth environment for the Hu sheep.

[0124] Take measures to prevent heatstroke and cool down in the summer, and take measures to keep warm and prevent cold in the winter.

[0125] Group feeding:

[0126] The Hu sheep are raised in groups according to their age, gender, weight and physiological condition to facilitate management and care.

[0127] Males, females and lambs are raised in separate pens to prevent premature mating, excessive mating and the spread of disease.

[0128] Health Monitoring:

[0129] Regularly observe the physiological condition of Hu sheep, test individual health indicators, and promptly detect and deal with abnormal situations.

[0130] Do a good job in disease prevention and control, formulate scientific immunization procedures, and carry out vaccination and deworming regularly.

[0131] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0132] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0133] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0134] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0135] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0136] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A breeding management method based on Huyang breeding data, characterized in that: Methods include: Obtain the breeding data and breeding indicators of Hu sheep, where the breeding data includes breeding environment data and non-breeding environment data, and the breeding indicators include breeding data indicators and growth indicators; Analyze the breeding data based on the breeding data indicators and obtain the results of Huyang breeding; Based on the growth index, the factors affecting the growth rate of Hu sheep and the time corresponding to the factors affecting the growth rate of Hu sheep are obtained, the time corresponding to the factors affecting the growth rate of Hu sheep is converted into a time series, and the factors affecting the growth rate of Hu sheep and the corresponding time series are converted into a matrix X. The matrix X is subjected to a multi-way delayed embedding transformation using a multi-way delayed embedding transformation technology to obtain a high-order Hankel tensor, and then the high-order Hankel tensor is mapped into a core tensor using a tensor decomposition technology. The core tensor is used as an input feature, and the input feature is input into the LSSVM model to predict the growth rate of Hu sheep. The growth rate of Hu sheep is analyzed based on the growth index to obtain the growth results of Hu sheep. Generate a Hu Yang breeding report based on the Hu Yang breeding results and Hu Yang growth results.

2. The breeding management method based on Huyang breeding data according to claim 1 is characterized in that: The breeding data is analyzed based on the breeding data indicators to obtain the results of Huyang breeding, which specifically includes the following processes: Based on the breeding data, the average temperature of the breeding environment of the Hu sheep in the management period is obtained, and based on the breeding data indicators, the preset temperature of the Hu sheep breeding in the management period is obtained, and the difference WD between the average temperature and the preset temperature of the Hu sheep breeding is calculated, wherein the management period is the preset breeding period; Based on the breeding data, the average humidity of the breeding environment of Huyang in the management period is obtained; based on the breeding data indicators, the preset humidity of Huyang breeding in the management period is obtained; and the difference SD between the average humidity and the preset humidity of Huyang breeding is calculated; Based on the breeding data, the number of Hu sheep vaccination types within the management period is obtained; based on the breeding data indicators, the number of Hu sheep vaccination types that should be vaccinated within the management period is obtained; and the proportion of the number of Hu sheep vaccination types to the number of Hu sheep vaccination types that should be vaccinated is calculated. Based on the breeding data, the feed intake of Hu sheep during the management period is obtained; based on the breeding data indicators, the feed intake that Hu sheep should take during the management period is obtained; and the difference CL between the feed intake of Hu sheep and the feed intake that Hu sheep should take is calculated; Substitute WD, SD, YM and CL into the calculation formula of Huyang breeding index to obtain Huyang breeding index LMS, which is as follows: Among them, γ, α, and θ are weight coefficients, which are 0.2, 0.6, and 0.2 respectively. F is a constant coefficient determined by the season. Different seasons correspond to different F values, which are user-defined. C is the average injection volume of the vaccine. The Lake Sheep Breeding Index LMS is recorded as the Lake Sheep Breeding result.

3. The breeding management method based on Huyang breeding data according to claim 1 is characterized in that: The specific process of converting the factors affecting the growth rate of Hu sheep and the corresponding time series into matrix X includes the following steps: The factors affecting the growth rate of Hu sheep and the time series are formed into a matrix Where X is a multidimensional array, R represents the real number domain, each element of the matrix X is a real number, and the size, size or dimension of each dimension of X is I1, I2, ..., I N and T, with a total of N+1 dimensions; T is the length of the time series, N is the number of factors affecting the growth rate of Hu sheep, and each row of the matrix X (I1×I2×...×I N The time vector representing the factors affecting the growth rate of Hu sheep. Each component of the vector represents the value of the factors affecting the growth rate of Hu sheep at different times. Each column of the matrix X (X1×X2×...×X N ) represents the growth rate factor vector of Hu sheep at different times, and each component of the vector represents the value of the factor affecting the growth rate of Hu sheep on that day.

4. A breeding management method based on Huyang breeding data according to claim 3, characterized in that: The multi-path delay embedding transformation technology is used to perform multi-path delay embedding transformation on the matrix X to obtain a high-order Hankel tensor. The specific process includes the following: Hankel(X)=FOLD (I,τ) (X×S1×S2×...×S N ); Among them, FOLD (I,τ) is the folding operator, To copy the matrix.

5. The breeding management method based on Huyang breeding data according to claim 4 is characterized in that: Mapping high-order Hankel tensors to core tensors through tensor decomposition technology specifically includes the following processes: Based on Tucker decomposition, high-order Hankel tensors are mapped to core tensors: Hankel(X)=G×U 1 ×U 2 ×...×U N ; in, is the core tensor, is the principal component of the tensor at each order, called the projection matrix, and each projection matrix is ​​orthogonal to each other.

6. The breeding management method based on Huyang breeding data according to claim 1 is characterized in that: Inputting the input features into the LSSVM model to predict the growth rate of Hu sheep specifically includes the following process: Build the LSSVM model: Given a training set D = {(x i ,y i )|(i=1,2...n)}, where x i ,y i are the input and output of the model respectively, and n is the size of the training set; The LSSVM regression prediction function is: f(x) = ω Y +b; Among them, b is the constant of the regression prediction function, ω is the regression matrix of the regression prediction function; ω Y is the device matrix of the regression prediction function; The weight vector of the regression function is obtained by optimizing the penalty function: C1 is the penalty coefficient, γ is the regularization parameter, e i is the training error; The regression prediction model formula is: Among them, K(x, z) is the kernel function, α i is the Lagrange multiplier; Prediction of Hu sheep growth rate based on LSSVM model.

7. A breeding management method based on Huyang breeding data according to claim 6, characterized in that: The growth rate of Hu sheep was analyzed based on growth indicators, and the growth results of Hu sheep were obtained, which specifically included the following processes: Based on the growth index, the preset Hu sheep growth rate threshold is obtained to determine whether the Hu sheep growth rate exceeds the preset Hu sheep growth rate threshold. If so, a signal that the Hu sheep growth condition meets the standard is generated; if not, a signal that the Hu sheep growth condition does not meet the standard is generated. The signal that the Hu sheep growth condition meets the standard or the signal that the Hu sheep growth condition does not meet the standard is recorded as the Hu sheep growth result.

8. The breeding management method based on Huyang breeding data according to claim 1 is characterized in that: Before analyzing the breeding data based on the breeding data indicators, the method further includes: For breeding data with small missing data and short time intervals, linear interpolation is used to fill in the missing data. The linear interpolation method is expressed as: Among them, the values ​​corresponding to the breeding data at time k and time k+j are Q k and Q k+j , Q k+i is the missing data at the k+ith moment.

Images