Bio-fermentation feed distribution method

Abstract

An embodiment of the present invention provides a method for distributing bio-fermented feed, comprising: obtaining historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs, the historical feeding data including the historical feeding amount of the bio-fermented feed; calculating the growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data; generating a feeding strategy for the feeding target based on the growth curve, and distributing feeding to the feeding target based on the feeding strategy. By using the historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs, the growth situation of the population to which the target belongs is analyzed from multiple dimensions to obtain a growth curve of the population to which the feeding target belongs, and a feeding strategy for the feeding target is generated through the growth curve. Adopting the feeding strategy of the feeding target can improve the accuracy of the feeding amount in the distribution of bio-fermented feed.

Description

Technical Field

[0001] The present invention relates to the technical field of feed breeding, in particular to a method for distributing biological fermentation feed. Background Art

[0002] Bio-fermented feed is feed made from roughage fermented by microorganisms. It uses microorganisms and complex enzymes as the biological feed starter strains to convert feed ingredients into microbial protein, bioactive small peptide amino acids, microbial active probiotics, and complex enzyme preparations. Roughage is rich in crude fiber and protein, such as cellulose, hemicellulose, pectin, and lignin, but is difficult for animals to directly digest and absorb. Consuming it can increase intestinal burden and cause intestinal diseases. Bio-fermented feed not only compensates for the amino acid deficiencies found in conventional feeds, but also rapidly converts other nutrients in the roughage, enhancing digestion, absorption, and utilization. Traditional bio-fermentation feed distribution is generally fed at regular intervals and in fixed quantities. However, since the growth conditions of farmed animals at different growth stages are highly correlated with the feed fed, providing appropriate feeding strategies at different growth stages can be beneficial to the growth of farmed animals. However, existing feeding strategies generally increase the amount of bio-fermentation feed distribution based on a single dimension such as the size (weight) or breeding time of farmed animals, which can easily result in excessive or insufficient bio-fermentation feed being fed. Therefore, existing bio-fermentation feed distribution has the problem of poor feeding accuracy. Summary of the Invention

[0003] The embodiment of the present invention provides a method for distributing bio-fermentation feed, which aims to solve the problem that the existing feeding strategy is generally based on a single dimension such as the size (weight) or breeding time of the farmed animals to increase the amount of bio-fermentation feed distributed, which easily causes excessive or insufficient feeding of bio-fermentation feed. Therefore, the existing bio-fermentation feed distribution has the problem of poor feeding amount accuracy. By analyzing the historical growth data, historical environmental data and historical feeding data of the population to which the feeding target belongs, the growth situation of the population to which the target belongs is analyzed from multiple dimensions, and the growth curve of the population to which the feeding target belongs is obtained. The feeding strategy of the feeding target is generated through the growth curve. The feeding strategy of the feeding target can improve the accuracy of the amount of bio-fermentation feed distribution.

[0004] In a first aspect, an embodiment of the present invention provides a method for distributing a bio-fermented feed, the method comprising the following steps:

[0005] Acquiring historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs, wherein the historical feeding data includes the historical feeding amount of the bio-fermented feed;

[0006] Calculating a growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data;

[0007] Based on the growth curve, a feeding strategy for the feeding target is generated, and feeding allocation is performed on the feeding target based on the feeding strategy.

[0008] Optionally, the historical feeding data includes a historical feeding amount sequence a n,i , where n represents the nth historical feeding moment, and i represents the ith feeding data type. The growth curve of the population to which the feeding target belongs is calculated based on the historical growth data, the historical environmental data, and the historical feeding data, including:

[0009] According to the historical feeding amount sequence a n,i The historical growth data and the historical environment data are sampled at the time granularity to obtain the historical growth sequence b n,j and historical environmental sequence c n,k , where j represents the jth growth data type and k represents the kth environment data type;

[0010] According to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs.

[0011] Optionally, the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs, including:

[0012] The historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k Perform alignment and splicing to obtain the spliced ​​sequence d n,m , wherein the spliced ​​sequence d n,m Including the splicing features corresponding to each historical moment, m = i + j + k;

[0013] According to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs.

[0014] Optionally, the sequence d after splicing n,m , calculate the growth curve of the population to which the feeding target belongs, including:

[0015] According to the spliced ​​sequence d n,m , construct a multidimensional space, the dimension of the multidimensional space is the same as the concatenated sequence d n,m The dimensions are the same;

[0016] The spliced ​​sequence d n,m Mapping the splicing features corresponding to each historical moment in the multidimensional space to obtain the feature point distribution in the multidimensional space;

[0017] According to the distribution of characteristic points in the multidimensional space, a growth curve of the group to which the feeding target belongs is calculated.

[0018] Optionally, calculating the growth curve of the population to which the feeding target belongs based on the distribution of characteristic points in the multidimensional space includes:

[0019] According to the distribution of characteristic points in the multidimensional space, marginal utility analysis is performed in the time dimension with the dimension of the growth data type as the analysis target to obtain a marginal utility curve that meets the preset conditions;

[0020] According to the marginal utility curve that meets the preset conditions, the growth curve of the population to which the feeding target belongs is determined.

[0021] Optionally, the step of analyzing the marginal utility of the dimension in which the growth data type is located in the time dimension based on the distribution of the feature points in the multidimensional space to obtain a candidate marginal utility curve that meets preset conditions includes:

[0022] According to the distribution of characteristic points in the multidimensional space, analyzing the marginal utility of the dimension where the environmental data type is located in the time dimension to obtain a candidate marginal utility curve;

[0023] The candidate marginal utility curves are screened within a preset marginal utility decreasing rate range to obtain the marginal utility curve that meets the preset conditions.

[0024] Optionally, determining the growth curve of the population to which the feeding target belongs based on the marginal utility curve that meets the preset conditions includes:

[0025] Performing curve fitting on the marginal utility curve that meets the preset conditions to obtain the marginal utility curve of the group to which the feeding target belongs;

[0026] The growth curve of the population to which the feeding target belongs is determined according to the marginal utility curve of the population to which the feeding target belongs.

[0027] In a second aspect, an embodiment of the present invention provides a bio-fermentation feed distribution device, the device comprising:

[0028] An acquisition module is used to acquire historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs, wherein the historical feeding data includes the historical feeding amount of the bio-fermentation feed;

[0029] a calculation module, configured to calculate a growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data;

[0030] A generation module is used to generate a feeding strategy for the feeding target based on the growth curve, and to distribute feeding to the feeding target based on the feeding strategy.

[0031] Optionally, the historical feeding data includes a historical feeding amount sequence a n,i , wherein n represents the nth historical feeding time, i represents the ith feeding data type, and the calculation module is further used to calculate the feeding amount sequence a according to the historical feeding amount sequence a n,i The historical growth data and the historical environment data are sampled at the time granularity to obtain the historical growth sequence b n,j and historical environmental sequence c n,k , where j represents the jth growth data type, and k represents the kth environmental data type; according to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs.

[0032] Optionally, the calculation module is further used to convert the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k Perform alignment and splicing to obtain the spliced ​​sequence d n,m , wherein the spliced ​​sequence d n,m Including the splicing features corresponding to each historical moment, m = i + j + k; according to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs.

[0033] In an embodiment of the present invention, historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs are obtained, and the historical feeding data include the historical feeding amount of the bio-fermented feed; based on the historical growth data, the historical environmental data, and the historical feeding data, a growth curve of the population to which the feeding target belongs is calculated; based on the growth curve, a feeding strategy for the feeding target is generated, and feeding distribution is performed on the feeding target based on the feeding strategy. By using the historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs, the growth situation of the population to which the target belongs is analyzed from multiple dimensions to obtain a growth curve of the population to which the feeding target belongs, and a feeding strategy for the feeding target is generated through the growth curve. By using the feeding strategy of the feeding target, the accuracy of feeding amount distribution of the bio-fermented feed can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 This is a flow chart of a method for distributing bio-fermented feed provided by an embodiment of the present invention;

[0036] Figure 2 This is a schematic structural diagram of a bio-fermentation feed distribution device provided by an embodiment of the present invention;

[0037] Figure 3 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0039] See Figure 1 , Figure 1 This is a flow chart of a method for distributing bio-fermented feed provided by an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0040] 101. Obtain historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs.

[0041] In an embodiment of the present invention, the feeding target may be arable animals, such as chickens, ducks, pigs, cattle, sheep, and other arable animals. Specifically, the feeding target refers to the current batch of arable animals, and the historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs refer to data collected and recorded during the breeding process of previous batches of arable animals. The historical feeding data includes the historical feeding amount and feeding time of bio-fermented feed. The historical feeding data may include the type of feed fed, the feeding amount of each type of feed, and the feeding time. The feed type may include bio-fermented feed. There may be one or more feed types. When there is only one feed type, the feed type is bio-fermented feed. When there are multiple feed types, the feed type includes bio-fermented feed. The historical environmental data includes environmental data such as temperature, humidity, and light intensity. This historical environmental data may be collected in real time via sensors. The historical growth data may include data such as body shape, weight, and color. This historical growth data may be recorded at regular intervals and may be the average growth data of the population.

[0042] 102. Based on historical growth data, historical environmental data, and historical feeding data, the growth curve of the population to which the feeding target belongs is calculated.

[0043] In an embodiment of the present invention, the implicit relationship among historical growth data, historical environmental data, and historical feeding data may be analyzed, and the implicit relationship may be extracted as a growth curve of the group to which the feeding target belongs.

[0044] Specifically, we can analyze the impact of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension, and then extract the law of the impact of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension based on the impact of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension, and obtain the growth curve of the population to which the feeding target belongs based on the law of the impact of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension.

[0045] Since the growth curve of the population to which the feeding target belongs is obtained with the population to which the feeding target belongs as the target, the growth curve is a growth curve adapted to the feeding target, and the environmental data and feeding data are taken into consideration to obtain growth curves in multiple dimensions, so that the growth curve can more accurately describe the influence of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension.

[0046] 103. Based on the growth curve, a feeding strategy for the feeding target is generated, and feeding is distributed to the feeding target based on the feeding strategy.

[0047] In this embodiment of the present invention, a growth curve describes how environmental factors and feeding amounts influence the growth of a population to which a feeding target belongs over time. Therefore, a feeding strategy for the feeding target can be generated based on the growth target of the feeding target. The feeding strategy includes feeding data and environmental data. After the feeding strategy is determined, feeding allocation is performed for the feeding target based on the feeding strategy.

[0048] The growth target can be growth time and growth weight or body shape. The growth target can be pre-entered by the user. For example, the growth target can be "grow to 2 kg in two months", and the feeding strategy is based on "grow to 2 kg in two months" and the growth curve. Specifically, the growth target can be used as a variable of the growth curve, and the growth curve can be obtained by y=f(x) t Represented by, where f() t This is a time-dependent function. x represents feeding and environmental data, and y represents growth data. Entering a growth target is equivalent to entering target y and a time constraint t. Solving for x yields the feeding and environmental data corresponding to the growth target. For example, if target y is 2 kg and time constraint t is less than two months, the output parameter x will be the feeding and environmental data corresponding to "growing to 2 kg in two months."

[0049] In an embodiment of the present invention, historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs are obtained, wherein the historical feeding data includes the historical feeding amount of the bio-fermented feed; based on the historical growth data, the historical environmental data, and the historical feeding data, a growth curve of the population to which the feeding target belongs is calculated; based on the growth curve, a feeding strategy for the feeding target is generated, and feeding distribution is performed on the feeding target based on the feeding strategy. By using the historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs, the growth situation of the population to which the target belongs is analyzed from multiple dimensions to obtain a growth curve of the population to which the feeding target belongs, and a feeding strategy for the feeding target is generated through the growth curve. By using the feeding strategy of the feeding target, the accuracy of the distribution amount of the bio-fermented feed can be improved.

[0050] Optionally, historical feeding data includes historical feeding amount sequence a n,i , where n represents the nth historical feeding moment, i represents the ith feeding data type, and in the step of calculating the growth curve of the population to which the feeding target belongs based on the historical growth data, historical environmental data, and historical feeding data, the historical feeding amount sequence a can be used to calculate the growth curve of the population to which the feeding target belongs. n,i The time granularity is used to sample the historical growth data and historical environment data to obtain the historical growth sequence b n,j and historical environmental sequence c n,k, where j represents the jth growth data type, and k represents the kth environmental data type; according to the historical feeding amount sequence a n,i 、Historical growth sequence b n,j and historical environmental sequence c n,k , calculate the growth curve of the population to which the feeding target belongs.

[0051] In an embodiment of the present invention, the above-mentioned feeding data type can be the type of feed fed, and the above-mentioned historical feeding data can include the type of feed fed, the feeding amount of each type of feed, and the feeding time. The above-mentioned feed type can include at least one of the types of biological fermentation feed, concentrated feed, fresh feed, etc. When there is only one type of feed fed, the feed type fed is biological fermentation feed. The above-mentioned feeding time is a fixed feeding time, such as feeding at a fixed time every day. Historical feeding amount sequence a n,i Indicates the feeding amount of the i-th feed type at the n-th historical feeding time. The above historical feeding amount sequence a n,i The time granularity can be understood as the distribution interval of feeding time.

[0052] Specifically, the distribution intervals of feeding time are different every day. For example, if the feeding time distribution is 7 am, 11 am, 4 pm, 8 pm, and 11 pm, the distribution intervals of feeding time are 4, 5, 4, and 3 respectively. Therefore, according to the historical feeding amount sequence a n,i The time granularity is used to sample the historical growth data and historical environment data to obtain the historical growth sequence b n,j and historical environmental sequence c n,k Since growth data needs to be collected regularly, historical growth data is also a discrete data with a time interval. During the sampling process, sampling can be performed according to the left principle. It can be understood as the historical feeding amount sequence a. n,i Between two historical moments of historical growth data at a certain historical moment, the growth data corresponding to the earlier of the two historical moments is used as the sampling data.

[0053] According to the historical feeding amount sequence a n,i and historical environmental sequence c n,k In the time dimension, for the historical growth sequence b n,j An analysis is conducted to determine the effects of environmental factors and feeding amounts on the growth of the target population over time. Based on the effects of environmental factors and feeding amounts on the growth of the target population over time, the patterns of their effects are extracted. Based on the patterns of their effects over time, a growth curve for the target population is obtained.

[0054] More specifically, the historical feeding amount sequence a n,i and historical environmental sequence c n,k In the time dimension, for the historical growth sequence b n,j The analysis process can be achieved through b n,j =G(a n,i , c n,k ) t To express, G() t It is a functional reflection of the influence of environmental factors and feeding amount on the growth of the feeding target population in the time dimension. In the known historical growth sequence b n,j , historical feeding amount sequence a n,i and historical environmental sequence c n,k Under this condition, we can calculate the function G() that affects the growth law. t , and then we can use the environmental factors and feeding amount to influence the growth of the feeding target group in the time dimension G() t Get the growth curve y=f(x) of the population to which the feeding target belongs t .

[0055] Since the growth curve of the population to which the feeding target belongs is obtained with the population to which the feeding target belongs as the target, the growth curve is a growth curve adapted to the feeding target, and the environmental data and feeding data are taken into consideration to obtain growth curves in multiple dimensions, so that the growth curve can more accurately describe the influence of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension.

[0056] Optionally, according to the historical feeding amount sequence a n,i 、Historical growth sequence b n,j and historical environmental sequence c n,k In the step of calculating the growth curve of the population to which the feeding target belongs, the historical feeding amount sequence a n,i 、Historical growth sequence b n,j and historical environmental sequence c n,k Perform alignment and splicing to obtain the spliced ​​sequence d n,m , where the spliced ​​sequence d n,m Including the splicing features corresponding to each historical moment, m = i + j + k; according to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs.

[0057] In the embodiment of the present invention, the historical feeding amount sequence a n,i 、Historical growth sequence b n,j and historical environmental sequence c n,kAlign them in chronological order, and splice the feeding data, growth data, and environmental data corresponding to each historical moment to obtain the splicing features corresponding to each historical moment, and then obtain the spliced ​​sequence d n,m It can be seen that the spliced ​​sequence d n,m With the historical feeding amount sequence a n,i 、Historical growth sequence b n,j and historical environmental sequence c n,k The splicing order can be feeding data first, environmental data in the middle, and growth data last. It should be noted that the dimension of feeding data is i, indicating that there are i feed types, the dimension of environmental data is k, indicating that there are k environmental parameters, and the dimension of growth data is j, indicating that there are j types of growth data.

[0058] The spliced ​​sequence d n,m As a sample, extract the spliced ​​sequence d n,m Historical feeding amount sequence a n,i and historical environmental sequence c n,k In the time dimension, for the historical growth sequence b n,j The implicit influence of the historical feeding amount sequence a is parameterized to achieve n,i and historical environmental sequence c n,k In the time dimension, for the historical growth sequence b n,j Through analysis, we can determine the impact of environmental factors and feeding amount on the growth of the feeding target population in the time dimension. Then, based on the impact of environmental factors and feeding amount on the growth of the feeding target population in the time dimension, we can extract the law of the impact of environmental factors and feeding amount on the growth of the feeding target population in the time dimension, and obtain the growth curve of the feeding target population based on the law of the impact of environmental factors and feeding amount on the growth of the feeding target population in the time dimension.

[0059] In a possible embodiment, the above implicit influence can be achieved by b n,j =G(a n,i , c n,k ) t The implicit influence parameterization can be expressed as b n,j =W(a n,i , c n,k ) t +B, where W is the weight parameter and B is the bias parameter. n,j , historical feeding amount sequence a n,i and historical environmental sequence c n,k Under this condition, the weight parameter W and bias parameter B can be calculated, and the function G() of the growth influence law can be obtained according to the calculation.t , according to the influence of environmental factors and feeding amount on the growth of the feeding target group in the time dimension G() t Get the growth curve y=f(x) of the population to which the feeding target belongs t .

[0060] Optionally, according to the spliced ​​sequence d n,m In the step of calculating the growth curve of the population to which the feeding target belongs, the spliced ​​sequence d n,m , construct a multidimensional space, the dimension of the multidimensional space is the same as the sequence d after splicing n,m The dimensions of the concatenated sequence d n,m The splicing features corresponding to each historical moment are mapped to the multidimensional space to obtain the distribution of feature points in the multidimensional space; according to the distribution of feature points in the multidimensional space, the growth curve of the population to which the feeding target belongs is calculated.

[0061] In the embodiment of the present invention, after multiple splicing, the sequence d n,m In the example, a splicing feature at each historical moment corresponds to a feature point in the multidimensional space. The splicing feature is mapped to the multidimensional space to obtain the feature point distribution in the multidimensional space. The feature point distribution in the multidimensional space can represent the sequence d after multiple splicing. n,m The spatiotemporal distribution of each splicing feature in .

[0062] In multidimensional space, in the time dimension, the characteristic curve b in the remaining dimensions can be constructed with different spatial distributions n,j =W m (a n,i , c n,k ) t +B m , W m is the weight parameter under the mth dimension, B m Bias parameter in the mth dimension. Characteristic curve b in the remaining dimensions n,j =W m (a n,i , c n,k ) t +B m In the known feature point data, the historical growth sequence b n,j , historical feeding amount sequence a n,i and historical environmental sequence c n,k In the case of , the weight parameter W can be calculated m and bias parameter B m , and the random forest weight parameter W m and bias parameter B m Optimize and get the function G() that affects the growth law tThe weight parameter W and bias parameter B in the equation are the influence of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension G() t Get the growth curve y=f(x) of the population to which the feeding target belongs t .

[0063] Optionally, in the step of calculating the growth curve of the population to which the feeding target belongs based on the distribution of characteristic points in the multidimensional space, a marginal utility analysis can be performed in the time dimension with the dimension where the growth data type is located as the analysis target based on the distribution of characteristic points in the multidimensional space to obtain a marginal utility curve that meets preset conditions; based on the marginal utility curve that meets the preset conditions, the growth curve of the population to which the feeding target belongs is determined.

[0064] In an embodiment of the present invention, the above marginal utility analysis can be understood as that as the feeding amount increases, the benefit of growth decreases. Specifically, under fixed environmental data, an increase in the feeding amount in the feeding data corresponds to a decrease in the increment in the growth data. For example, when the feeding amount is 500g, the body weight increases by 10g; when the feeding amount is 1000g, the body weight increases by 14g; when the feeding amount is 1500g, the body weight increases by 15g.

[0065] According to marginal utility analysis, analyze the marginal utility curve b that meets the preset conditions under the dimensions of each environmental data n,j =W r (a n,i , c n,k ) t +B r , r is less than or equal to m. The marginal utility curve b in the remaining dimensions n,j =W r (a n,i , c n,k ) t +B r In the known feature point data, the historical growth sequence b n,j , historical feeding amount sequence a n,i and historical environmental sequence c n,k In the case of , the weight parameter W can be calculated r and bias parameter B r , and the random forest weight parameter W r and bias parameter B r Optimize and get the function G() that affects the growth law t The weight parameter W and bias parameter B in the equation are the influence of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension G() t Get the growth curve y=f(x) of the population to which the feeding target belongs t .

[0066] Optionally, in the step of analyzing the marginal utility of the dimension where the growth data type is located in the time dimension based on the distribution of characteristic points in the multidimensional space to obtain a candidate marginal utility curve that meets the preset conditions, the marginal utility of the dimension where the environmental data type is located in the time dimension can be analyzed based on the distribution of characteristic points in the multidimensional space to obtain a candidate marginal utility curve; the candidate marginal utility curves can be screened within a preset marginal utility deceleration rate range to obtain a marginal utility curve that meets the preset conditions.

[0067] In the embodiment of the present invention, the marginal utility curve b that meets the preset conditions can be obtained under the dimension of each environmental data according to the distribution of characteristic points in the multidimensional space. n,j =W k (a n,i , c n,k ) t +B k , the above-mentioned precondition can be the marginal utility decreasing rate range (W min , W max ), W min is the minimum deceleration rate, W max The minimum deceleration rate and the maximum deceleration rate can be set by the user. k Greater than W min , W k Less than W max The candidate marginal utility curves are screened based on the conditions to obtain the marginal utility curve b that meets the preset conditions. n,j =W r (a n,i , c n,k ) t +B r , r is less than or equal to k. In the marginal utility curve b that meets the preset conditions n,j =W r (a n,i , c n,k ) t +B r In the known feature point data, the historical growth sequence b n,j , historical feeding amount sequence a n,i and historical environmental sequence c n,k In the case of , the weight parameter W can be calculated r and bias parameter B r , and the random forest weight parameter W r and bias parameter B r Optimize and get the function G() that affects the growth law tThe weight parameter W and bias parameter B in the equation are the influence of environmental factors and feeding amount on the growth of the population to which the feeding target belongs in the time dimension G() t Get the growth curve y=f(x) of the population to which the feeding target belongs t .

[0068] Optionally, in the step of determining the growth curve of the population to which the feeding target belongs based on the marginal utility curve that meets the preset conditions, the marginal utility curve that meets the preset conditions can be curve fitted to obtain the marginal utility curve of the population to which the feeding target belongs; and based on the marginal utility curve of the population to which the feeding target belongs, the growth curve of the population to which the feeding target belongs is determined.

[0069] In the embodiment of the present invention, the historical growth sequence b in the known feature point data n,j , historical feeding amount sequence a n,i and historical environmental sequence c n,k In the case of , the weight parameter W can be calculated r and bias parameter B r , the weight parameter W can be r and bias parameter B r Perform curve fitting to convert b n,j =W r (a n,i , c n,k ) t +B r Fitting to b n,j =W(a n,i , c n,k ) t +B, change b n,j =W(a n,i , c n,k ) t +B is the marginal utility curve of the group to which the feeding target belongs. The marginal utility curve b of the group to which the feeding target belongs can be used. n,j =W(a n,i , c n,k ) t +B performs linear transformation to obtain the growth influence law function b n,j =G(a n,i , c n,k ) t , and through linear transformation, the growth influence law function b n,j =G(a n,i , c n,k ) t Converted to the growth curve of the population to which the feeding target belongs y = f (x) t .

[0070] It should be noted that the bio-fermentation feed distribution method provided in the embodiment of the present invention can be applied to intelligent feeding machines, smart phones, computers, servers and other devices.

[0071] Optionally, an embodiment of the present invention provides a bio-fermentation feed distribution device, see Figure 2 , Figure 2 FIG. 1 is a structural diagram of a bio-fermentation feed distribution device provided by an embodiment of the present invention. Figure 2 As shown, the device includes:

[0072] An acquisition module 201 is configured to acquire historical growth data, historical environmental data, and historical feeding data of a population to which a feeding target belongs, wherein the historical feeding data includes a historical feeding amount of the bio-fermented feed;

[0073] A calculation module 202 is configured to calculate a growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data;

[0074] The generating module 203 is configured to generate a feeding strategy for the feeding target based on the growth curve, and perform feeding distribution for the feeding target based on the feeding strategy.

[0075] Optionally, the historical feeding data includes a historical feeding amount sequence a n,i , wherein n represents the nth historical feeding time, i represents the ith feeding data type, and the calculation module 202 is further used to calculate the feeding amount sequence a according to the historical feeding amount sequence a n,i The historical growth data and the historical environment data are sampled at the time granularity to obtain the historical growth sequence b n,j and historical environmental sequence c n,k , where j represents the jth growth data type, and k represents the kth environmental data type; according to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs.

[0076] Optionally, the calculation module 202 is further configured to convert the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k Perform alignment and splicing to obtain the spliced ​​sequence d n,m , wherein the spliced ​​sequence d n,m Including the splicing features corresponding to each historical moment, m = i + j + k; according to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs.

[0077] Optionally, the calculation module 202 is further configured to calculate the sequence d according to the spliced ​​sequence d n,m , construct a multidimensional space, the dimension of the multidimensional space is the same as the concatenated sequence d n,m The dimensions of the concatenated sequence d n,m The splicing features corresponding to each historical moment in the multidimensional space are mapped to the multidimensional space to obtain the feature point distribution of the multidimensional space; and the growth curve of the population to which the feeding target belongs is calculated based on the feature point distribution of the multidimensional space.

[0078] Optionally, the calculation module 202 is also used to perform marginal utility analysis in the time dimension based on the distribution of characteristic points in the multidimensional space, with the dimension of the growth data type as the analysis target, to obtain a marginal utility curve that meets preset conditions; based on the marginal utility curve that meets the preset conditions, determine the growth curve of the population to which the feeding target belongs.

[0079] Optionally, the calculation module 202 is also used to analyze the marginal utility of the dimension in which the environmental data type is located in the time dimension according to the distribution of feature points in the multidimensional space to obtain candidate marginal utility curves; and to filter the candidate marginal utility curves within a preset marginal utility decreasing rate range to obtain the marginal utility curve that meets the preset conditions.

[0080] Optionally, the calculation module 202 is further used to perform curve fitting on the marginal utility curve that meets the preset conditions to obtain the marginal utility curve of the population to which the feeding target belongs; and determine the growth curve of the population to which the feeding target belongs based on the marginal utility curve of the population to which the feeding target belongs.

[0081] It should be noted that the behavior detection device provided in the embodiment of the present invention can be applied to intelligent feeding machines, smart phones, computers, servers and other devices that can distribute bio-fermented feed.

[0082] The behavior detection device provided in the embodiment of the present invention can implement each process implemented by the bio-fermentation feed distribution method in the above method embodiment and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0083] See also Figure 3 , Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, it includes: a memory 302, a processor 301 and a computer program of a bio-fermentation feed distribution method stored in the memory 302 and capable of running on the processor 3401, wherein:

[0084] The processor 301 is configured to call the computer program stored in the memory 302 and execute the following steps:

[0085] Acquiring historical growth data, historical environmental data, and historical feeding data of the population to which the feeding target belongs, wherein the historical feeding data includes the historical feeding amount of the bio-fermented feed;

[0086] Calculating a growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data;

[0087] Based on the growth curve, a feeding strategy for the feeding target is generated, and feeding allocation is performed on the feeding target based on the feeding strategy.

[0088] Optionally, the historical feeding data includes a historical feeding amount sequence a n,i , wherein n represents the nth historical feeding moment, and i represents the ith feeding data type. The processor 301 calculates the growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data, including:

[0089] According to the historical feeding amount sequence a n,i The historical growth data and the historical environment data are sampled at the time granularity to obtain the historical growth sequence b n,j and historical environmental sequence c n,k , where j represents the jth growth data type and k represents the kth environment data type;

[0090] According to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs.

[0091] Optionally, the processor 301 executes the process according to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs, including:

[0092] The historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k Perform alignment and splicing to obtain the spliced ​​sequence d n,m , wherein the spliced ​​sequence d n,m Including the splicing features corresponding to each historical moment, m = i + j + k;

[0093] According to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs.

[0094] Optionally, the processor 301 executes the process according to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs, including:

[0095] According to the spliced ​​sequence d n,m , construct a multidimensional space, the dimension of the multidimensional space is the same as the concatenated sequence d n,m The dimensions are the same;

[0096] The spliced ​​sequence d n,m Mapping the splicing features corresponding to each historical moment in the multidimensional space to obtain the feature point distribution in the multidimensional space;

[0097] According to the distribution of characteristic points in the multidimensional space, a growth curve of the group to which the feeding target belongs is calculated.

[0098] Optionally, the processor 301 calculates the growth curve of the population to which the feeding target belongs based on the distribution of characteristic points in the multidimensional space, including:

[0099] According to the distribution of characteristic points in the multidimensional space, marginal utility analysis is performed in the time dimension with the dimension of the growth data type as the analysis target to obtain a marginal utility curve that meets the preset conditions;

[0100] According to the marginal utility curve that meets the preset conditions, the growth curve of the population to which the feeding target belongs is determined.

[0101] Optionally, the processor 301 performs analysis based on the distribution of feature points in the multidimensional space and the marginal utility of the dimension in which the growth data type is located in the time dimension to obtain a candidate marginal utility curve that meets preset conditions, including:

[0102] According to the distribution of characteristic points in the multidimensional space, analyzing the marginal utility of the dimension where the environmental data type is located in the time dimension to obtain a candidate marginal utility curve;

[0103] The candidate marginal utility curves are screened within a preset marginal utility decreasing rate range to obtain the marginal utility curve that meets the preset conditions.

[0104] Optionally, the processor 301 determines the growth curve of the population to which the feeding target belongs based on the marginal utility curve that meets the preset conditions, including:

[0105] Performing curve fitting on the marginal utility curve that meets the preset conditions to obtain the marginal utility curve of the group to which the feeding target belongs;

[0106] The growth curve of the population to which the feeding target belongs is determined according to the marginal utility curve of the population to which the feeding target belongs.

[0107] It should be noted that the electronic device provided in the embodiment of the present invention can be applied to devices such as smart phones, computers, and servers that can distribute bio-fermented feed.

[0108] The electronic device provided in the embodiment of the present invention can implement each process implemented by the bio-fermentation feed distribution method in the above method embodiment and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0109] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the computer program implements the various processes of the bio-fermentation feed distribution method or the application-end bio-fermentation feed distribution method provided in the embodiment of the present invention, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0110] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0111] The above disclosure is merely a preferred embodiment of the present invention and certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for distributing biological fermentation feed, characterized in that: The method comprises the following steps: Obtain historical growth data, historical environmental data and historical feeding data of the population to which the feeding target belongs, wherein the historical feeding data includes the historical feeding amount of the bio-fermented feed; the historical feeding data includes the historical feeding amount sequence a n,i , where n represents the nth historical feeding moment, and i represents the i-th feeding data type; Calculating a growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data; The calculation of the growth curve of the population to which the feeding target belongs based on the historical growth data, the historical environmental data and the historical feeding data includes: n,i The historical growth data and the historical environment data are sampled at the time granularity to obtain the historical growth sequence b n,j and historical environmental sequence c n,k , where j represents the jth growth data type, and k represents the kth environmental data type; according to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs; According to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs, including: The historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k Perform alignment and splicing to obtain the spliced ​​sequence d n,m , wherein the spliced ​​sequence d n,m Including the splicing features corresponding to each historical moment, m = i + j + k; According to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs; According to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs, including: According to the spliced ​​sequence d n,m , construct a multidimensional space, the dimension of the multidimensional space is the same as the concatenated sequence d n,m The dimensions are the same; The spliced ​​sequence d n,m Mapping the splicing features corresponding to each historical moment in the multidimensional space to obtain the feature point distribution in the multidimensional space; Calculating a growth curve of the population to which the feeding target belongs based on the distribution of characteristic points in the multidimensional space; The step of calculating the growth curve of the population to which the feeding target belongs based on the distribution of characteristic points in the multidimensional space includes: According to the distribution of characteristic points in the multidimensional space, marginal utility analysis is performed in the time dimension with the dimension of the growth data type as the analysis target to obtain a marginal utility curve that meets the preset conditions; Determining the growth curve of the population to which the feeding target belongs based on the marginal utility curve that meets the preset conditions; Based on the growth curve, a feeding strategy for the feeding target is generated, and feeding allocation is performed on the feeding target based on the feeding strategy.

2. The method according to claim 1, wherein The method of analyzing the marginal utility of the dimension where the growth data type is located in the time dimension based on the distribution of the characteristic points in the multidimensional space to obtain a candidate marginal utility curve that meets the preset conditions includes: According to the distribution of characteristic points in the multidimensional space, analyzing the marginal utility of the dimension where the environmental data type is located in the time dimension to obtain a candidate marginal utility curve; The candidate marginal utility curves are screened within a preset marginal utility decreasing rate range to obtain the marginal utility curve that meets the preset conditions.

3. The method according to claim 2, wherein Determining the growth curve of the population to which the feeding target belongs based on the marginal utility curve that meets the preset conditions includes: Performing curve fitting on the marginal utility curve that meets the preset conditions to obtain the marginal utility curve of the group to which the feeding target belongs; The growth curve of the population to which the feeding target belongs is determined according to the marginal utility curve of the population to which the feeding target belongs.

4. A biological fermentation feed distribution device, characterized in that: The device comprises: The acquisition module is used to obtain the historical growth data, historical environmental data and historical feeding data of the group to which the feeding target belongs, wherein the historical feeding data includes the historical feeding amount of the biological fermentation feed; the historical feeding data includes the historical feeding amount sequence a n,i , where n represents the nth historical feeding moment, and i represents the i-th feeding data type; A calculation module is configured to calculate a growth curve of a population to which a feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data; the calculation module is configured to calculate a growth curve of a population to which a feeding target belongs based on the historical growth data, the historical environmental data, and the historical feeding data, including: calculating a growth curve of a population to which a feeding target belongs based on the historical feeding amount sequence a n,i The historical growth data and the historical environment data are sampled at the time granularity to obtain the historical growth sequence b n,j and historical environmental sequence c n,k , where j represents the jth growth data type, and k represents the kth environmental data type; according to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs; According to the historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k , calculate the growth curve of the population to which the feeding target belongs, including: The historical feeding amount sequence a n,i , the historical growth sequence b n,j And the historical environment sequence c n,k Perform alignment and splicing to obtain the spliced ​​sequence d n,m , wherein the spliced ​​sequence d n,m Including the splicing features corresponding to each historical moment, m = i + j + k; According to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs; According to the spliced ​​sequence d n,m , calculate the growth curve of the population to which the feeding target belongs, including: According to the spliced ​​sequence d n,m , construct a multidimensional space, the dimension of the multidimensional space is the same as the concatenated sequence d n,m The dimensions are the same; The spliced ​​sequence d n,m Mapping the splicing features corresponding to each historical moment in the multidimensional space to obtain the feature point distribution in the multidimensional space; Calculating a growth curve of the population to which the feeding target belongs based on the distribution of characteristic points in the multidimensional space; The step of calculating the growth curve of the population to which the feeding target belongs based on the distribution of characteristic points in the multidimensional space includes: According to the distribution of characteristic points in the multidimensional space, marginal utility analysis is performed in the time dimension with the dimension of the growth data type as the analysis target to obtain a marginal utility curve that meets the preset conditions; Determining the growth curve of the population to which the feeding target belongs based on the marginal utility curve that meets the preset conditions; A generation module is used to generate a feeding strategy for the feeding target based on the growth curve, and to distribute feeding to the feeding target based on the feeding strategy.

Images