Breeding feed mixing method and device

Abstract

The invention relates to the technical field of feed mixing and stirring, in particular to a breeding feed mixing method and device. The device comprises a mixing control system, the mixing control system analyzes mixing data generated by known historical mixing batches by using a historical feed mixing data analysis module, and further constructs a mixing difficulty matrix of each historical batch class cluster on the basis of the historical data. According to the method, initial stirring parameters can be obtained by determining a matched historical batch class cluster according to the initial stirring parameters, then a mixing difficulty sequence corresponding to the initial stirring parameters is determined by utilizing the mixing difficulty matrix, the stirring speed is controlled by utilizing the mixing difficulty in the mixing difficulty sequence, and the breeding feed is mixed by adjusting the stirring speed on the basis of the final stirring time. Manual intervention is reduced, and the breeding feed is automatically and efficiently mixed within a fixed time by controlling the speed of the stirring roller.

Description

Technical Field

[0001] The invention relates to the technical field of feed mixing and stirring, and in particular to a breeding feed mixing method and device. Background Art

[0002] In order to feed the farmed animals scientifically, a variety of feed raw materials are mixed and fed to the farm, such as grass, puffed feed, nutritional supplements, etc. In order to mix the feed, the farm will use a feed mixing device to mix it. The common feed mixing device contains intermeshing stirring rollers. Through the rotation of the stirring rollers, the feed can be mixed and crushed at the same time.

[0003] In the prior art, when using a feed mixing device to mix feed, the mixing speed and mixing time are manually set for mixing. However, some mixing parameters are often only applicable to mixed batches with specific raw material ratios. If the mixing parameters are not applicable to mixed batches with new raw material ratios, the mixing effect will be poor, the feed will not be fully mixed, the feed quality will be reduced, and the mixing efficiency will be affected. If the optimal mixing parameters are determined for mixed batches with new raw material ratios through more experiments, the cost will increase, and the optimal mixing parameters cannot be determined quickly and effectively. Summary of the invention

[0004] In order to solve the technical problem that the prior art cannot quickly and effectively determine the optimal stirring parameters and control the stirring for the batch to be mixed when mixing feed, thereby resulting in poor feed mixing effect, the purpose of the present invention is to provide a method and device for mixing aquaculture feed, and the technical scheme adopted is as follows: The present invention provides a breeding feed mixing device, comprising a mixing bin, a stirring roller, a material outlet, and a mixing control system; the mixing control system comprises: The historical feed mixing data analysis module is used to divide a number of historical batch clusters according to the raw material distribution in each historical mixing batch; within the historical batch cluster, the mixing quality scores of all historical mixing batches are arranged according to the mixing speed and mixing time of each historical mixing batch, and the mixing data feature matrix of the historical batch cluster is obtained; the growth change trend of each element in the mixing data feature matrix under the mixing time dimension is obtained; according to the change of the growth change trend of each element in the mixing speed dimension, the mixing difficulty corresponding to each element is obtained to form a mixing difficulty matrix; An initial mixing parameter setting module is used to obtain a matching historical batch cluster of a batch to be mixed; the stirring speed and stirring time of the matching historical batch cluster are counted to obtain an initial stirring speed and a final stirring time of the batch to be mixed; The speed control module is used to extract a mixing difficulty sequence from the mixing difficulty matrix matching the historical batch cluster according to the initial mixing speed and the final mixing time; the time range of the mixing difficulty sequence corresponds to the final mixing time; during the mixing process, the initial mixing speed is updated according to the mixing difficulty corresponding to the mixing time in the mixing difficulty sequence; The discharge control module is used to stop stirring after the final stirring time is reached, control the opening of the discharge port, and output the mixed feed in the mixing bin.

[0005] Furthermore, the historical feed mixing data analysis module also includes a historical data classification unit; the historical data classification unit is used to obtain the feature similarity of the raw material distribution characteristics between historical mixing batches, and cluster the historical mixing batches based on the feature similarity using the Kmeans clustering algorithm to obtain a number of historical batch clusters.

[0006] Furthermore, the historical feed mixing data analysis module also includes a raw material distribution feature extraction module, and the raw material distribution feature extraction module is used to: For each mixed batch, the particle volume of each raw material in the mixed batch and the content ratio in the mixed batch are obtained; the cutting probability of the raw material in the mixed batch is obtained according to the particle volume and the content ratio; the raw material combination characteristic curve of the mixed batch is constructed with the raw material type number as the horizontal axis and the raw material cutting probability as the vertical axis, and the raw material combination characteristic curve is used as the raw material distribution characteristic.

[0007] Furthermore, the horizontal dimension of the mixed data feature matrix is ​​the stirring time, and the vertical dimension is the stirring speed; the specific method for obtaining the mixed data feature matrix includes: At the same stirring speed, the mixing quality scores are sorted from small to large according to the stirring time to obtain a first sequence for each stirring speed; the first sequence is arranged vertically from large to small according to the stirring speed to obtain the mixing data feature matrix.

[0008] Furthermore, the method for acquiring the growth change trend includes: Any element in the mixed data feature matrix is ​​recorded as a target element, and all elements at the corresponding position of the target element in the stirring time dimension direction constitute a time series trend analysis sequence of the target element; the range of the time series trend analysis sequence is obtained; the mixed quality score difference between the target element and the next element in the time series trend analysis sequence is obtained; and the product of the mixed quality score difference and the range is taken as the growth change trend of the target element.

[0009] Furthermore, the method for obtaining the mixed difficulty includes: Any element in the mixed data feature matrix is ​​recorded as a target element, and the neighborhood range of the target element is obtained according to a preset size in the direction of increasing stirring speed with the target element as the starting point; the growth change trends of the elements in the neighborhood range are sorted from small to large according to the stirring speed to obtain a growth change trend sequence; A differential sequence of the growth change trend sequence is obtained, the elements in the differential sequence are accumulated and negatively correlated and normalized to obtain the mixed difficulty of the target element.

[0010] Furthermore, the method for obtaining the matching historical batch clusters includes: The similarity of raw material distribution characteristics between the batch to be mixed and the cluster center of each historical batch cluster is used as the matching degree; and the historical batch cluster with the largest matching degree is used as the matching historical batch cluster.

[0011] Further, obtaining the initial stirring speed and final stirring time of the batch to be mixed includes: Obtaining the average particle volume and average density of all raw materials in the batch to be mixed; taking the product of the average particle volume and the average density as the morphological characteristic of the raw materials in the batch to be mixed; The average stirring speed of the historical mixing batches whose mixing quality scores in the matching historical batch cluster are greater than a preset score threshold is used as the initial stirring speed; The ratio between the volume of the mixing bin and the morphological characteristics of the raw material is used as the cutting effectiveness; The final stirring time is obtained according to the initial stirring speed and the cutting effectiveness.

[0012] Furthermore, the speed control module includes a stirring speed update module, which is used to obtain a speed adjustment coefficient according to the corresponding mixing difficulty and the matching degree of the matching historical batch cluster when the stirring process reaches the stirring time corresponding to each element in the mixing difficulty sequence, and update the initial stirring speed according to the speed adjustment coefficient.

[0013] The present invention provides a method for mixing aquaculture feed, wherein the method utilizes any one of the aquaculture feed mixing devices to mix batches of aquaculture feed to be mixed.

[0014] The present invention has the following beneficial effects: The livestock feed mixing device proposed in the embodiment of the present invention includes a mixing control system, which uses a historical feed mixing data analysis module to analyze the mixing data generated by known historical mixing batches, and then constructs a mixing difficulty matrix for each historical batch cluster based on the historical data. The mixing difficulty can be effectively quantified by statistically analyzing the changes in the mixing quality score in the historical data under the two dimensions of stirring speed and stirring time. For a batch to be mixed with a new raw material ratio, it is only necessary to determine the matching historical batch cluster to obtain the initial stirring parameters, and then use the mixing difficulty matrix to determine the mixing difficulty sequence corresponding to the initial stirring parameters, and then use the mixing difficulty in the mixing difficulty sequence to control the stirring speed, ensuring that the livestock feed is effectively mixed by adjusting the stirring speed based on the final stirring time. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 creative work.

[0016] Figure 1 A schematic diagram of a breeding feed mixing device provided by one embodiment of the present invention; Figure 2 A structural block diagram of a hybrid control system provided by an embodiment of the present invention.

[0017] The figure includes: 1 a discharging port blocking door; 2 a stirring roller; 3 a mixing control system; 4 a discharging port; 5 a conveyor belt; 101 a historical feed mixing data analysis module; 102 an initial mixing parameter setting module; 103 a speed control module; and 104 a discharging control module. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a method and device for mixing aquaculture feed according to the present invention, its specific implementation, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0019] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0020] The embodiment of the present invention provides a feed mixing device for breeding, which includes a feed mixing device body, such as a mixing bin, a stirring roller, and a discharge port. The feed mixing device body also includes a mixing control system, which is used to control the stirring speed of the stirring roller and the discharge of the discharge port.

[0021] The specific scheme of a method and device for mixing aquaculture feed provided by the present invention is described in detail below in conjunction with the accompanying drawings.

[0022] See also Figure 1 , which shows a schematic diagram of a breeding feed mixing device provided by an embodiment of the present invention, Figure 1 The breeding feed mixing device comprises a discharge port blocking door 1; a stirring roller 2; a mixing control system 3; a discharge port 4; and a conveyor belt 5. Figure 1 When the breeding feed mixing device is used, after the raw material information of the batch to be mixed is counted, the information is input into the mixing control system 2, and then the raw materials are put into the mixing bin, and the mixing roller 2 starts to rotate after the mixing program is started. Figure 1 The stirring rollers 2 are two mutually meshing stirring rollers, which can not only play the role of feed mixing but also crush large-particle feed such as grass through the rotation of the stirring rollers. During the stirring process, the conveyor belt 5 also starts to rotate, and the conveyor belt 5 rotates in the direction of the discharge port 4. At the same time, the conveyor belt 5 is also equipped with an air bag, which can increase the contact resistance between the conveyor belt 5 and the feed during the rotation. Under the action of the air bag blocking, the feed at the bottom of the mixing bin is driven to accumulate from the through port to the side of the discharge port 4, so that the feed on the side of the discharge port surges upward, and then continues to contact the stirring roller 2, forming a continuous stirring cycle. During the stirring process, the stirring speed is controlled by the mixing control system 3, and the stirring is stopped after the mixing time is reached. The discharge port blocking door 1 at the position of the discharge port 4 is controlled to open. Under the transmission of the conveyor belt 5, the mixed feed can be driven by the expanded air bag to be discharged from the discharge port 4.

[0023] In one embodiment of the present invention, a transparent rectangular observation window is also installed on the mixing bin for observing the mixing state of the feed inside the device.

[0024] It should be noted that, in one embodiment of the present invention, the data calculation process in the hybrid control system is mainly processed by a cloud server, and the corresponding information is collected and transmitted to the cloud using material network technology. After the cloud performs data processing, the corresponding data is fed back to the hybrid control system, and the hybrid control system then feeds back the corresponding control signal to each device.

[0025] See also Figure 2, which shows a structural block diagram of a hybrid control system provided in one embodiment of the present invention, the system includes a hybrid control system including a historical feed mixing data analysis module 101, an initial mixing parameter setting module 102, a speed control module 103 and a discharging control module 104.

[0026] The historical feed mixing data analysis module 101 first divides a number of historical batch clusters according to the raw material distribution in each historical mixing batch. Because the historical mixing batch contains multiple mixing conditions as historical data, for a raw material ratio, multiple historical mixing batches can be included through different stirring speeds and different stirring times. For the batch to be mixed, the historical mixing batch corresponding to the raw material ratio similar to itself is the information that can be used as a reference, so the historical mixing batches are first classified, and then each historical batch cluster is analyzed.

[0027] Because the historical mixing batch is a process that has been mixed, it contains known stirring speed, stirring time and mixing quality score. It should be noted that the mixing quality score is used to evaluate the quantitative mixing effect. Because the historical mixing batch is a process of stirring using existing fixed stirring parameters, there is a fixed stirring time for each batch and a constant stirring speed during this period of time.

[0028] It should be noted that the mixing quality score can be evaluated manually, because the mixing chamber of the mixing device in the embodiment of the present invention includes an observation window, and the staff can observe the internal mixing state through the observation window, and can be scored by multiple dimensions such as color, shape, and fineness, and finally the mixing quality score is obtained by statistics. The specific scoring method is a technical means well known to those skilled in the art and will not be described in detail here.

[0029] In one embodiment of the present invention, in order to more conveniently obtain the mixing information of historical mixing batches, during the execution of historical mixing batches, because it is a fixed stirring time at a fixed stirring speed, an observation interval can be set within the stirring time, so that the scoring results corresponding to different stirring times at a stirring speed can be obtained in one mixing process.

[0030] In another embodiment of the present invention, in order to more conveniently obtain the mixing quality score of historical mixing batches, a decision tree is first constructed based on the stirring and mixing information of some historical mixing batches. For other historical mixing batches, the mixing quality score corresponding to the historical mixing batch can be quickly determined by inputting the corresponding information into the decision tree model.

[0031] The historical feed mixing data analysis module 101 further counts the stirring speed, stirring time and mixing quality score of each historical mixing batch in each historical batch cluster, takes the stirring speed and stirring time as the dimension information in the two-dimensional matrix, and takes the mixing quality score as the element in the matrix, thereby obtaining the mixed data feature matrix, that is, the mixed data feature matrix is ​​obtained by arranging the mixing quality scores of all historical mixing batches according to the stirring speed and stirring time of each historical mixing batch. Through the construction of the mixed data feature matrix, it is possible to perform targeted analysis in the two dimensions of stirring speed and stirring time in a matrix manner, for example, fix a stirring speed, that is, find any row or column in the matrix, and these elements constitute the change of the mixing quality score at the stirring speed as the stirring time changes. For an element in the matrix, it corresponds to a stirring speed and stirring time. By analyzing its growth change trend under the stirring time dimension, it can be obtained that the element has a cutting efficiency for feed raw materials at the current speed and time, that is, the higher the cutting efficiency, the more significant the growth of the mixing quality score over time, and the greater the growth change trend.

[0032] It should be noted that, for the mixing and stirring process of feed, the increase in stirring speed and stirring time will further cut and mix the feed. Therefore, as the stirring speed increases or the stirring time increases, the mixing quality score will only gradually increase or remain unchanged, and will not gradually decrease.

[0033] As the stirring proceeds, the particles of the raw material itself gradually become smaller after cutting, so it will be more difficult for the stirring roller to further cut the raw material, and the cutting efficiency will gradually decrease; if there is an insignificant trend of mixing quality score changes when changing the stirring speed in this state, it means that the mixing difficulty of the raw materials in this state is high. Therefore, the historical feed mixing data analysis module 101 further obtains the mixing difficulty corresponding to each element according to the change in the growth trend of each element in the stirring speed dimension. Because each element can correspond to a mixing difficulty, the mixing difficulty matrix can be obtained by replacing the elements in the mixed data feature matrix with the mixing difficulty. That is, each element in the mixing difficulty matrix represents the mixing difficulty under the stirring speed and stirring time corresponding to the element.

[0034] Preferably, in one embodiment of the present invention, the historical feed mixing data analysis module also includes a historical data classification unit; the historical data classification unit is used to obtain the feature similarity of the raw material distribution characteristics between historical mixing batches, and cluster the historical mixing batches based on the feature similarity using the Kmeans clustering algorithm to obtain a number of historical batch clusters.

[0035] It should be noted that the Kmeans clustering algorithm is a technical means well known to those skilled in the art. The elbow method is used to obtain the optimal K value before clustering. The specific algorithm is not described or limited here.

[0036] Preferably, in one embodiment of the present invention, the historical feed mixing data analysis module further includes a raw material distribution feature extraction module, and the raw material distribution feature extraction module is used to: For each mixed batch, the particle volume of each raw material in the mixed batch and its content ratio in the mixed batch are obtained. It should be noted that the particle volume can be directly obtained by the type of raw material itself, or it can be measured by multiple sampling tests; the content ratio can be directly obtained by the raw material ratio in the batch to be mixed. The specific method for obtaining the particle volume and content ratio is a technical means well known to those skilled in the art, and will not be elaborated or limited here.

[0037] The cutting probability of the raw materials in the mixed batch is obtained according to the particle volume and content ratio, that is, the larger the particle volume and the greater the content ratio, the greater the probability that the raw material will be cut during the cutting and mixing process. The raw material combination characteristic curve of the mixed batch is constructed with the raw material type number as the horizontal axis and the raw material cutting probability as the vertical axis, and the raw material combination characteristic curve is used as the raw material distribution characteristic. That is, the cutting probability is used as the characteristic of the raw materials in the mixed batch during the mixing process, thereby realizing the quantification of the raw material distribution. It should be noted that in the process of constructing the raw material combination characteristic curve, the types of raw materials need to be quantified. Each raw material can be numbered in the form of a label, and natural numbers starting from 0 are used as the mark of each raw material in turn, thereby realizing the quantification of the raw material types.

[0038] In one embodiment of the present invention, the product of the particle volume and the content ratio is used as the cutting probability.

[0039] It should be noted that, because the raw material distribution characteristic in one embodiment of the present invention is a curve, the DTW algorithm can be used in the clustering process to obtain the dynamic time warping similarity between the raw material combination characteristic curves, and then obtain clusters.

[0040] Preferably, in one embodiment of the present invention, the horizontal dimension of the mixed data feature matrix is ​​the stirring time, and the vertical dimension is the stirring speed; the specific method for obtaining the mixed data feature matrix includes: At the same stirring speed, the mixing quality scores are sorted from small to large according to the stirring time to obtain the first sequence of each stirring speed; the first sequence is arranged vertically from large to small according to the stirring speed to obtain a mixed data feature matrix. It should be noted that in certain embodiments of the present invention, the timing information corresponding to the same position between the first sequences may be different, for example, at speed A, it contains information of 5 minutes, 8 minutes, and 10 minutes, while at speed B, it contains information of 4 minutes, 5 minutes, 7 minutes, and 10 minutes. Because the lengths of the first sequences at the two speeds are different, and the timing information between elements with the same sequence number is also different, it is necessary to standardize the first sequence, and interpolate the data without corresponding timing information in the first sequence through an interpolation data processing method, so that the lengths between the first sequences are the same and the stirring time dimensions correspond.

[0041] Preferably, in one embodiment of the present invention, the method for acquiring the growth change trend includes: Any element in the mixed data feature matrix is ​​recorded as the target element, and all elements at the corresponding position of the target element in the mixing time dimension direction constitute the time series trend analysis sequence of the target element.

[0042] It should be noted that, in one embodiment of the present invention, the horizontal dimension of the mixed data feature matrix is ​​the stirring time, and the vertical dimension is the stirring speed. Then the time series trend analysis sequence of the target element is the entire row where the target element is located in the matrix. Similarly, if in other embodiments of the present invention, the vertical dimension of the mixed data feature matrix is ​​the stirring time, and the horizontal dimension is the stirring speed, then the time series trend analysis sequence is the entire column where the target element is located in the matrix.

[0043] The range of the time series trend analysis sequence is obtained. The range represents the variation of the mixing quality score in the entire mixing time dimension. At the mixing speed where the target element is located, the larger the corresponding range is, the more significant the change in the score is with the change of mixing time at this speed, and the more obvious the growth trend is.

[0044] Obtain the difference in mixed quality scores between the target element and the next element in the time series trend analysis sequence. The greater the difference in mixed quality scores, the greater the growth trend under the target element position.

[0045] Therefore, the product of the mixed quality score difference and the range is taken as the growth trend of the target element.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining the mixed difficulty includes: Any element in the mixed data feature matrix is ​​recorded as a target element, and the neighborhood range of the target element is obtained according to a preset size from the target element as the starting point in the direction of increasing stirring speed. In one embodiment of the present invention, the preset size is selected as 5, that is, from the target element as the starting point, 5 elements are found to constitute the neighborhood range along the direction of increasing stirring speed.

[0047] It should be noted that, in one embodiment of the present invention, because the vertical axis of the mixed data feature matrix is ​​the stirring speed, that is, the direction in which the stirring speed increases is the vertical upward direction of the matrix.

[0048] According to the stirring speed, the growth trend of the elements in the neighborhood range is sorted from small to large to obtain the growth trend sequence. The difference sequence of the growth trend sequence is obtained. The elements in the difference sequence represent the growth change of the growth trend sequence. The larger the element, the more obvious the change of the growth trend, and the lower the mixing difficulty of the current state of the target element. Therefore, the elements in the difference sequence are accumulated and negatively correlated and normalized to obtain the mixing difficulty of the target element.

[0049] In one embodiment of the present invention, the cumulative sum of the elements of the difference sequence is negatively correlated in the form of a reciprocal. In order to avoid the denominator being 0, the reciprocal of the result of adding the positive integer 1 to the cumulative sum of the elements of the difference sequence is used as the mixed difficulty.

[0050] The initial mixing parameter setting module 102 is used to obtain the matching historical batch cluster of the batch to be mixed. That is, the matching historical batch cluster is the reference information of the batch to be mixed. The stirring speed and stirring time of the matching historical batch cluster are further counted to obtain the initial stirring speed and final stirring time of the batch to be mixed. It should be noted that in order to achieve rapid stirring of the aquaculture feed, the final stirring time obtained in the embodiment of the present invention will no longer change in the subsequent modules, and only the initial stirring speed is adjusted to achieve an effective mixing result after the final stirring time.

[0051] Preferably, in one embodiment of the present invention, the method for obtaining matching historical batch clusters includes: The similarity of raw material distribution characteristics between the batch to be mixed and the cluster center of each historical batch cluster is taken as the matching degree; the historical batch cluster with the largest matching degree is taken as the matching historical batch cluster.

[0052] It should be noted that, in one embodiment of the present invention, the raw material distribution characteristic is a curve, so the matching degree can also be obtained through the DTW algorithm, which will not be described in detail.

[0053] Preferably, in one embodiment of the present invention, obtaining the initial stirring speed and the final stirring time of the batch to be mixed comprises: The average particle volume and average density of all raw materials in the batch to be mixed are obtained. The product of the average particle volume and the average density is used as the raw material characteristic of the batch to be mixed. It should be noted that the average particle volume and the average density can be obtained according to the type of raw materials and the raw material ratio of the batch to be mixed, and the specific details are not repeated here.

[0054] Because the mixing effects of historical mixing batches vary, in order to make the initial stirring speed have a certain reference, the average stirring speed of historical mixing batches whose mixing quality scores in the matching historical batch clusters are greater than a preset scoring threshold is used as the initial stirring speed. In one embodiment of the present invention, the scoring threshold is set to match the top 20 percent of the mixing quality scores in the matching historical batch clusters.

[0055] The ratio between the volume of the mixing bin and the morphological characteristics of the raw materials is used as the cutting effectiveness. The final stirring time is obtained according to the initial stirring speed and the cutting effectiveness. The greater the initial stirring speed, the faster the mixing can be completed by adjusting the speed, and the shorter the final stirring time; the greater the cutting effectiveness, the smaller the morphological characteristics of the raw materials, the more difficult it is to mix in the mixing bin, and the longer the final stirring time required. As an example, in one embodiment of the present invention, the final stirring time is expressed by the formula: ;in is the final mixing time, is the preset empirical coefficient, is the volume of the mixing bin, is the initial stirring speed, To evaluate the particle size, is the average density, In one embodiment of the present invention, the empirical coefficient is set to 1.2.

[0056] The speed control module 103 is used to intercept the mixed difficulty sequence in the mixed difficulty matrix matching the historical batch cluster according to the initial mixing speed and the final mixing time; the time series range of the mixed difficulty sequence corresponds to the final mixing time. For example, according to the initial mixing speed, a corresponding row of data or a column of data is selected in the mixed difficulty matrix as the initial mixed difficulty sequence. The time series range of the initial mixed difficulty sequence is the range under the mixing time dimension of the mixed difficulty matrix as a whole, for example, it may include a time series range of 4 minutes, 5 minutes, 6 minutes, 7 minutes, 8 minutes, 9 minutes, and 10 minutes, and the final mixing time is determined to be 8 minutes, then the initial mixed difficulty sequence needs to be intercepted so that it only includes the mixed difficulty information of 4 minutes, 5 minutes, 6 minutes, 7 minutes, and 8 minutes to form a mixed difficulty sequence.

[0057] Furthermore, during the mixing process, since there is a corresponding timing information in the mixing difficulty sequence for each mixing moment, when the mixing moment reaches the timing information in the mixing difficulty sequence, the initial mixing speed can be updated according to the corresponding mixing difficulty, that is, the greater the mixing difficulty, the faster the mixing speed needs to be, so that the feed in the mixing bin can contact the mixing roller more, and effective mixing and cutting can be achieved. The speed control module 103 updates the initial mixing speed and controls the mixing roller.

[0058] Preferably, in one embodiment of the present invention, the speed control module includes a stirring speed update module, which is used to obtain a speed adjustment coefficient according to the corresponding mixing difficulty and the matching degree of the matching historical batch cluster when the stirring process reaches the stirring time corresponding to each element in the mixing difficulty sequence, and update the initial stirring speed according to the speed adjustment coefficient. That is, the embodiment of the present invention takes into account the matching degree with the matching historical batch cluster when obtaining the speed adjustment coefficient, that is, the matching degree can be used as a reference confidence degree, and the greater the matching degree, the more accurate the information of the matching historical batch cluster is, and the greater the confidence degree.

[0059] As an example, the speed adjustment coefficient in the embodiment of the present invention is to multiply the matching degree and the mixing difficulty, and normalize the multiplication result. Because the normalized result is a coefficient between 0 and 1, the normalized result is added to 0.7 to obtain the speed adjustment coefficient. That is, when the normalized result is greater than 0.3, it proves that the mixing difficulty is greater at this time, and the stirring speed needs to be increased, then the speed adjustment coefficient is a result greater than 1, and the speed adjustment coefficient can be multiplied by the initial stirring speed to achieve the update of the initial stirring speed.

[0060] It should be noted that the initial stirring speed is a continuous updating process. After being updated once, if the stirring time continues to reach the time corresponding to the element in the mixing difficulty sequence, it needs to be continuously updated until the final stirring time is reached.

[0061] The discharge control module 104 is used to stop the mixing after the final mixing time is reached, control the opening of the discharge port, and output the mixed feed in the mixing bin.

[0062] The present invention also proposes a method for mixing livestock feed, which uses any one of the above embodiments of a livestock feed mixing device to mix the livestock feed batches to be mixed. When mixing, the staff places the raw materials of the batches to be mixed in the mixing bin, and then inputs the raw material information of the batches to be mixed, such as the raw material ratio, raw material type, etc., through the control panel, and then starts mixing. The mixing control system automatically determines the initial mixing parameters and continuously adjusts the stirring speed according to the input information through the processing of the cloud server. There is no manual participation in the whole process, which improves the mixing quality of the final feed mixture and the mixing efficiency of the whole mixing process.

[0063] In summary, the livestock feed mixing device proposed in the embodiment of the present invention includes a mixing control system, which uses a historical feed mixing data analysis module to analyze the mixing data generated by known historical mixing batches, and then constructs a mixing difficulty matrix for each historical batch cluster based on the historical data. By determining the matching historical batch clusters, the initial mixing parameters can be obtained, and then the mixing difficulty matrix is ​​used to determine the mixing difficulty sequence corresponding to the initial mixing parameters, and then the mixing difficulty in the mixing difficulty sequence is used to control the mixing speed, and the livestock feed is mixed by adjusting the mixing speed based on the final mixing time. The embodiment of the present invention reduces manual intervention and realizes automated and efficient mixing of livestock feed within a fixed time by controlling the speed of the mixing roller.

[0064] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0065] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A breeding feed mixing device, comprising a mixing bin, a stirring roller, and a discharge port, characterized in that: Also includes a hybrid control system; the hybrid control system includes: The historical feed mixing data analysis module is used to divide a number of historical batch clusters according to the raw material distribution in each historical mixing batch; within the historical batch cluster, the mixing quality scores of all historical mixing batches are arranged according to the mixing speed and mixing time of each historical mixing batch, and the mixing data feature matrix of the historical batch cluster is obtained; the growth change trend of each element in the mixing data feature matrix under the mixing time dimension is obtained; according to the change of the growth change trend of each element in the mixing speed dimension, the mixing difficulty corresponding to each element is obtained to form a mixing difficulty matrix; An initial mixing parameter setting module is used to obtain a matching historical batch cluster of a batch to be mixed; the stirring speed and stirring time of the matching historical batch cluster are counted to obtain an initial stirring speed and a final stirring time of the batch to be mixed; The speed control module is used to extract a mixing difficulty sequence from the mixing difficulty matrix matching the historical batch cluster according to the initial mixing speed and the final mixing time; the time range of the mixing difficulty sequence corresponds to the final mixing time; during the mixing process, the initial mixing speed is updated according to the mixing difficulty corresponding to the mixing time in the mixing difficulty sequence; The discharge control module is used to stop stirring after the final stirring time is reached, control the opening of the discharge port, and output the mixed feed in the mixing bin.

2. A breeding feed mixing device according to claim 1, characterized in that: The historical feed mixing data analysis module also includes a historical data classification unit; the historical data classification unit is used to obtain the feature similarity of the raw material distribution characteristics between historical mixing batches, and cluster the historical mixing batches based on the feature similarity using the Kmeans clustering algorithm to obtain a number of historical batch clusters.

3. A breeding feed mixing device according to claim 1, characterized in that: The historical feed mixing data analysis module also includes a raw material distribution feature extraction module, and the raw material distribution feature extraction module is used to: For each mixed batch, the particle volume of each raw material in the mixed batch and the content ratio in the mixed batch are obtained; the cutting probability of the raw material in the mixed batch is obtained according to the particle volume and the content ratio; the raw material combination characteristic curve of the mixed batch is constructed with the raw material type number as the horizontal axis and the raw material cutting probability as the vertical axis, and the raw material combination characteristic curve is used as the raw material distribution characteristic.

4. A breeding feed mixing device according to claim 1, characterized in that: The horizontal dimension of the mixed data feature matrix is ​​the stirring time, and the vertical dimension is the stirring speed; the specific method for obtaining the mixed data feature matrix includes: At the same stirring speed, the mixing quality scores are sorted from small to large according to the stirring time to obtain a first sequence for each stirring speed; the first sequence is arranged vertically from large to small according to the stirring speed to obtain the mixing data feature matrix.

5. A breeding feed mixing device according to claim 1, characterized in that: The method for acquiring the growth change trend comprises: Any element in the mixed data feature matrix is ​​recorded as a target element, and all elements at the corresponding position of the target element in the stirring time dimension direction constitute a time series trend analysis sequence of the target element; the range of the time series trend analysis sequence is obtained; the mixed quality score difference between the target element and the next element in the time series trend analysis sequence is obtained; and the product of the mixed quality score difference and the range is taken as the growth change trend of the target element.

6. A breeding feed mixing device according to claim 1, characterized in that: The method for obtaining the mixed difficulty includes: Any element in the mixed data feature matrix is ​​recorded as a target element, and the neighborhood range of the target element is obtained according to a preset size in the direction of increasing stirring speed with the target element as the starting point; the growth change trends of the elements in the neighborhood range are sorted from small to large according to the stirring speed to obtain a growth change trend sequence; A differential sequence of the growth change trend sequence is obtained, the elements in the differential sequence are accumulated and negatively correlated and normalized to obtain the mixed difficulty of the target element.

7. A breeding feed mixing device according to claim 1, characterized in that: The method for obtaining the matching historical batch clusters includes: The similarity of raw material distribution characteristics between the batch to be mixed and the cluster center of each historical batch cluster is used as the matching degree; and the historical batch cluster with the largest matching degree is used as the matching historical batch cluster.

8. The aquaculture feed mixing device according to claim 1, characterized in that: The method of obtaining the initial stirring speed and final stirring time of the batch to be mixed comprises: Obtaining the average particle volume and average density of all raw materials in the batch to be mixed; taking the product of the average particle volume and the average density as the morphological characteristic of the raw materials in the batch to be mixed; The average stirring speed of the historical mixing batches whose mixing quality scores in the matching historical batch cluster are greater than a preset score threshold is used as the initial stirring speed; The ratio between the volume of the mixing bin and the morphological characteristics of the raw material is used as the cutting effectiveness; The final stirring time is obtained according to the initial stirring speed and the cutting effectiveness.

9. A breeding feed mixing device according to claim 7, characterized in that: The speed control module includes a stirring speed updating module, which is used to obtain a speed adjustment coefficient according to the corresponding mixing difficulty and the matching degree of the matching historical batch cluster when the stirring process reaches the stirring time corresponding to each element in the mixing difficulty sequence, and update the initial stirring speed according to the speed adjustment coefficient.

10. A method for mixing aquaculture feed, characterized in that: The method utilizes a livestock feed mixing device as claimed in any one of claims 1 to 9 to mix the livestock feed batches to be mixed.

Images