A whole life cycle supervision method of livestock machinery
By monitoring livestock machinery throughout its entire lifecycle and optimizing feed dispensing speed and angle, the problem of large feed dispensing deviations has been solved, enabling precise feed delivery and ensuring the health of dairy cows.
Patent Information
- Application Number
- CN202510114348.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-24
- Publication Date
- 2025-11-28
- Estimated Expiration
- 2045-01-24
AI Technical Summary
The lack of a dynamic, full-lifecycle monitoring mechanism for existing livestock machinery leads to large deviations in the automatic feeding process, affecting feeding efficiency.
By acquiring the regional characteristics and current feed ratios of dairy farms, feeding analysis is conducted to determine feed dispensing targets. Combined with the transportation path of automated feed dispensing vehicles, the dispensing speed and angle are optimized to generate the optimal dispensing plan for dynamic monitoring.
It enables precise and efficient feed delivery under different usage periods and breeding environments, improving feed utilization and ensuring the health of dairy cows.
Smart Images
Figure CN120069595B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of control systems, and particularly relates to a full life cycle supervision method of livestock machinery. BACKGROUND
[0002] With the continuous expansion of dairy farming scale, the traditional manual feeding method is increasingly insufficient in accuracy and efficiency. On the one hand, manual operation is difficult to accurately control the weight and distribution of feed, which can easily lead to feed waste or nutritional imbalance. On the other hand, factors such as feed ratio, cow breed and growth stage change frequently, and simple reliance on experience or extensive management often cannot adjust and optimize in time. To solve the above problems, more and more farms begin to introduce automatic feeders to improve feeding efficiency through mechanization and intelligent means. However, in the existing technology, the speed or angle is mostly controlled in the automatic feeding process, and there is a lack of dynamic full life cycle supervision mechanism. Especially after the increase of the use time of the equipment, the wear and tear and performance degradation of the mechanical parts often lead to the continuous accumulation of feeding deviation, which cannot be automatically compensated and corrected, ultimately affecting the feeding accuracy, feed utilization and cow health level.
[0003] Therefore, in the prior art, the livestock machinery lacks a dynamic full life cycle supervision mechanism, resulting in large feeding deviation in the automatic feeding process, which affects the feeding effect. SUMMARY
[0004] The present application provides a full life cycle supervision method of livestock machinery, which solves the technical problem that the livestock machinery lacks a dynamic full life cycle supervision mechanism in the prior art, resulting in large feeding deviation in the automatic feeding process, which affects the feeding effect. It can accurately and efficiently complete the feeding of feed in different use periods and breeding environments, significantly improve the feed utilization and ensure the health of cows.
[0005] The application provides a whole life cycle management method of livestock machinery, which comprises the following steps: acquiring a plurality of dairy cow breeding characteristics of a plurality of regions in a dairy cow farm; performing feeding analysis according to a current feed ratio and the plurality of dairy cow breeding characteristics; determining a plurality of region discharge targets; and mapping and sorting the plurality of region discharge targets according to a transportation path of an automatic discharge vehicle to generate a region discharge target sequence.
[0006] In the implementation, the acquiring of the plurality of dairy cow breeding characteristics of the plurality of regions in the dairy cow farm and the performing of the feeding analysis according to the current feed ratio and the plurality of dairy cow breeding characteristics to determine the plurality of region discharge targets comprises: acquiring the plurality of dairy cow breeding characteristics of the plurality of regions in the dairy cow farm, wherein the dairy cow breeding characteristics at least include a dairy cow breed, a quantity and a growth stage, and the growth stage at least includes a growth period, a mature period, a lactation period, a pregnancy period and a senile period; acquiring a current feed ratio of the automatic discharge vehicle, wherein the current feed ratio includes a material type and a proportion; performing feeding standard retrieval according to the current feed ratio and the plurality of dairy cow breeding characteristics based on a dairy cow breeding specification table to acquire a plurality of region discharge weights, which are set as the plurality of region discharge targets.
[0007] In the implementation, the selecting of a first discharge target in the region discharge target sequence to have a minimum deviation from the first discharge target as an expectation, the performing of discharge speed optimization analysis in combination with a vehicle discharge duration and the current feed ratio, and the acquiring of a first optimized discharge speed comprise: selecting a first region discharge target in the region discharge target sequence as the first discharge target; performing simulation modeling on the automatic discharge vehicle in a three-dimensional simulation platform to generate a discharge simulation model; acquiring a discharge speed adjustment threshold of the automatic discharge vehicle; performing discharge speed optimization analysis in combination with the vehicle discharge duration and the current feed ratio based on the discharge speed adjustment threshold and the discharge simulation model to have a minimum deviation from the first discharge target as an expectation, and outputting the first optimized discharge speed.
[0008] In an implementation, based on the discharge speed adjustment threshold and the discharge simulation model, a discharge speed optimization analysis is performed in combination with the vehicle discharge duration and the current feed ratio to minimize deviation from the first discharge target, and the first optimized discharge speed is output, including: rendering the vehicle discharge duration and the current feed ratio to the discharge simulation model to generate a real-time discharge simulation space; randomly selecting a first discharge speed in the discharge speed adjustment threshold, and using the real-time discharge simulation space to simulate a first simulated discharge weight according to the first discharge speed; calculating the deviation of the first simulated discharge weight from the first discharge target to obtain a first weight difference; continuing to randomly select discharge speeds in the discharge speed adjustment threshold for simulation until a predetermined speed optimization number is reached, outputting multiple discharge speeds and multiple weight differences, and selecting the discharge speed corresponding to the minimum weight difference as the first optimized discharge speed.
[0009] In an implementation, based on the current feed ratio, the vehicle discharge duration and the first optimized discharge speed, a discharge angle optimization analysis is performed to expect that the feed falls uniformly into a predetermined area, and the first optimized discharge angle is output, including: collecting a sample feed ratio set, a sample discharge duration set, a sample discharge speed set and a sample discharge angle set according to the breeding discharge log in the historical time zone, and labeling the feed distribution under different sample feed ratios, sample discharge durations, sample discharge speeds and sample discharge angles to obtain a sample feed distribution set, wherein the sample feed distribution includes sample distribution uniformity and sample overflow area ratio, and the overflow area ratio is the ratio of the area of the feed exceeding the predetermined area to the area of the predetermined area; using the sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set as training data to supervise the learning of the random forest to obtain a feed discharge prediction plug-in; using the feed discharge prediction plug-in, based on the current feed ratio, the vehicle discharge duration and the first optimized discharge speed, a discharge angle optimization analysis is performed to expect that the feed falls uniformly into a predetermined area, and the first optimized discharge angle is output.
[0010] In an implementation, the sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set are used as training data to supervise the learning of the random forest to obtain a feed discharge prediction plug-in, including: using the sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set as training data, and dividing the training data into P parts, randomly selecting P times to obtain a first training set, and continuing to select P times to obtain P training sets, wherein P is an integer greater than 10; taking the sample feed ratio, the sample discharge duration, the sample discharge speed and the sample discharge angle as input, and taking the sample distribution uniformity and the sample overflow area ratio as supervision, using the P training sets to supervise the learning of the random forest until convergence, and harvesting P feed discharge prediction units; based on the P feed discharge prediction units, the feed discharge prediction plug-in is integrated to build, wherein the output of the feed discharge prediction plug-in is the mean of the outputs of the P feed discharge prediction units.
[0011] In an implementation, the feed discharge prediction plug-in is used to perform discharge angle optimization analysis according to the current feed ratio, vehicle discharge duration and first optimized discharge speed, with uniform feed falling into a predetermined area as the expectation, and the first optimized discharge angle is output, including: obtaining a discharge angle adjustment threshold of the automatic discharge vehicle, and randomly selecting a first discharge angle in the discharge angle adjustment threshold; inputting the current feed ratio, vehicle discharge duration, first optimized discharge speed and first discharge angle into the feed discharge prediction plug-in to output first distribution uniformity and first overflow area ratio; obtaining a first discharge fitness according to the first distribution uniformity and the first overflow area ratio, wherein the discharge fitness is positively correlated with the distribution uniformity and negatively correlated with the overflow area ratio; continuing to perform discharge angle optimization analysis according to the discharge angle adjustment threshold until a predetermined number of angle optimization times are reached, obtaining a plurality of discharge angles and a plurality of discharge fitnesses, and selecting the discharge angle with the maximum discharge fitness as the first optimized discharge angle.
[0012] In an implementation, the discharge error of the optimized discharge parameter sequence is corrected to obtain an optimal discharge scheme, including: obtaining the cumulative working duration of the automatic discharge vehicle, performing discharge error analysis according to the cumulative working duration to determine the discharge speed compensation coefficient and the discharge angle compensation coefficient, wherein the error analysis channel is constructed based on the BP neural network to perform discharge error analysis, and the sample working duration set, the sample discharge speed compensation coefficient set and the sample discharge angle compensation coefficient set are collected to train the error analysis channel to convergence; according to the discharge speed compensation coefficient and the discharge angle compensation coefficient, the discharge error of the optimized discharge parameter sequence is iteratively corrected to obtain an optimal discharge parameter sequence, and the optimal discharge scheme is generated in combination with the transportation path.
[0013] This application proposes a method for full life-cycle monitoring of livestock machinery. This method involves acquiring several dairy farming characteristics across several areas within a dairy farm, performing feeding analysis based on the current feed formulation and these characteristics, determining feed dispensing targets for several areas, and mapping and sorting these targets according to the transport path of an automated feed dispensing vehicle to generate a regional feed dispensing target sequence. A first feed dispensing target is selected from this sequence, with the minimum deviation from the first target as the expected outcome. Feed dispensing is then performed in conjunction with the vehicle's dispensing time and the current feed formulation. Speed optimization analysis is performed to obtain a first optimized feeding speed. Based on the current feed ratio, vehicle feeding time, and the first optimized feeding speed, and with the expectation that feed will fall evenly into a predetermined area, feeding angle optimization analysis is conducted to output a first optimized feeding angle. This, combined with the first optimized feeding speed, generates a first optimized feeding parameter. Further feeding parameter optimization analysis is performed on other feeding targets in the feeding target sequence for the region, generating an optimized feeding parameter sequence. Feeding error correction is applied to the optimized feeding parameter sequence to obtain the optimal feeding scheme for automatic feeding control in dairy farms. This solves the technical problem in existing livestock machinery where the lack of a dynamic, full-lifecycle monitoring mechanism leads to large feeding deviations during automatic feeding, affecting feeding efficiency. It achieves accurate and efficient feed delivery under different usage periods and farming environments, significantly improving feed utilization and ensuring dairy cow health. Attached Figure Description
[0014] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.
[0015] Figure 1 A schematic diagram of a method for monitoring the entire life cycle of livestock machinery provided in this application embodiment;
[0016] Figure 2 This is a flowchart illustrating a method for monitoring the entire life cycle of livestock machinery to obtain feed discharge targets in several areas, as provided in an embodiment of this application. Detailed Implementation
[0017] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.
[0018] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. The described embodiments should not be regarded as limitations to the present application. All other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.
[0019] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms first\second are only used to distinguish similar objects and do not represent a specific order of the objects. The terms include and have and any variants are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.
[0020] The embodiments of the present application provide a whole life cycle supervision method of livestock machinery, as shown in Figure 1 The method comprises the following steps:
[0021] Obtaining a plurality of dairy farming characteristics of a plurality of regions in a dairy farm, performing feeding analysis according to the current feed ratio and the plurality of dairy farming characteristics, determining a plurality of region discharge targets, and mapping and sorting the plurality of region discharge targets according to the transportation path of the automatic discharge vehicle to generate a region discharge target sequence;
[0022] The first discharge target is selected in the area discharge target sequence, and the minimum deviation from the first discharge target is expected, combined with the vehicle discharge duration and the current feed ratio to perform discharge speed optimization analysis to obtain a first optimized discharge speed; a plurality of dairy farming characteristics of a plurality of areas in the dairy farm are obtained, the dairy farming characteristics including dairy breed, quantity and growth stage, and the growth stage at least including growth period, mature period, lactation period, pregnancy period and old age period. Feeding analysis is performed according to the current feed ratio and the plurality of dairy farming characteristics to determine a plurality of area discharge targets, wherein the area areas of the areas are all the same. According to the transportation path of the automatic discharge vehicle, the plurality of area discharge targets are mapped and sorted, that is, the feeding order and the corresponding area discharge targets are mapped to generate an area discharge target sequence. Further, the first discharge target is selected in the area discharge target sequence, and the first discharge target is the first sorted discharge target in the area discharge target sequence. The minimum deviation from the first discharge target is expected, combined with the vehicle discharge duration and the current feed ratio to perform discharge speed optimization analysis to obtain a first optimized discharge speed.
[0023] As shown in Figure 2 The method provided by the embodiment of the present application further includes: obtaining a plurality of dairy farming characteristics of a plurality of areas in the dairy farm, wherein the dairy farming characteristics at least include dairy breed, quantity and growth stage, and the growth stage at least includes growth period, mature period, lactation period, pregnancy period and old age period; obtaining a current feed ratio of the automatic discharge vehicle, wherein the current feed ratio includes material type and proportion; based on a dairy farming specification table, feeding standard retrieval is performed according to the current feed ratio and the plurality of dairy farming characteristics respectively to obtain a plurality of area discharge weights, which are set as the plurality of area discharge targets.
[0024] The method comprises the following steps: acquiring a plurality of dairy farming characteristics of a plurality of regions in a dairy farm, and performing feeding analysis according to a current feed ratio and the plurality of dairy farming characteristics to determine a plurality of region discharge targets, wherein the dairy farming characteristics at least include dairy cow breeds, quantity and growth stages, and the growth stages at least include growth period, adult period, lactation period, pregnancy period and old age period. Then, a current feed ratio of an automatic discharge vehicle is acquired, wherein the current feed ratio includes material types and their proportions, and the material types include feed materials such as forage, potatoes, corn, etc., and the proportions are actual proportions of each material. A dairy farming specification table is acquired, which contains the nutritional requirements and daily recommended nutrient intake for different dairy cow breeds and different growth stages. Based on the dairy farming specification table, feeding standard retrieval is performed according to the current feed ratio and the plurality of dairy farming characteristics to acquire the quality of the current feed ratio meeting the daily nutritional standard requirements of a single dairy cow, to acquire the standard requirements of a single dairy cow in each region, and to acquire a plurality of region discharge weights according to the quantity of each region and the standard requirements of a single dairy cow, which are set as the plurality of region discharge targets.
[0025] The method provided by the embodiment of the present application further comprises the following steps: selecting a first region discharge target as a first discharge target in the sequence of region discharge targets; performing simulation modeling on the automatic discharge vehicle in a three-dimensional simulation platform to generate a discharge simulation model; acquiring a discharge speed adjustment threshold of the automatic discharge vehicle, and performing discharge speed optimization analysis based on the discharge speed adjustment threshold and the discharge simulation model, with the minimum deviation from the first discharge target as the expectation, in combination with the vehicle discharge duration and the current feed ratio, to output the first optimized discharge speed.
[0026] The first discharge target is selected from the regional discharge target sequence, and a discharge speed optimization analysis is performed in combination with the vehicle discharge duration and the current feed ratio to obtain a first optimized discharge speed, including: selecting a first regional discharge target as the first discharge target from the regional discharge target sequence. The automatic discharge vehicle delivery process is simulated and visualized in a three-dimensional simulation platform, which is constructed by professional software to generate a discharge simulation model. When constructing, the vehicle structure model is constructed to input the basic parameters of the automatic discharge vehicle, such as: vehicle size, load capacity; discharge port size and shape; speed range and adjustment method of the feed conveyor belt. The discharge model simulates the rate, throwing angle, particle distribution and the like of the feed falling from the silo port or the conveyor belt, and adjusts the feed density, viscosity, moisture content and the like according to the feed ratio. Environmental modeling: according to the information of the road surface terrain, temperature and humidity, cowshed structure and the like on the spot, a simple scene model can be made, but the most important thing in this step is the virtual simulation of the discharge process. The above models are assembled and integrated in the three-dimensional simulation platform to form a runnable simulation model, which is used to simulate the output weight, time and feed distribution under different "discharge speed" inputs. The discharge speed adjustment threshold of the automatic discharge vehicle, i.e. the discharge speed adjustable range of the automatic discharge vehicle, is obtained, and based on the discharge speed adjustment threshold and the discharge simulation model, a discharge speed optimization analysis is performed in combination with the first discharge target deviation minimum expectation and the vehicle discharge duration and the current feed ratio, and the first optimized discharge speed is output.
[0027] The method provided by the embodiment of the application further includes: rendering the vehicle discharge duration and the current feed ratio to the discharge simulation model to generate a real-time discharge simulation space; randomly selecting a first discharge speed in the discharge speed adjustment threshold, and simulating a first simulated discharge weight according to the first discharge speed by using the real-time discharge simulation space; calculating the deviation of the first simulated discharge weight from the first discharge target to obtain a first weight difference; continuing to randomly select a discharge speed in the discharge speed adjustment threshold for simulation until a predetermined speed optimization number is reached, outputting a plurality of discharge speeds and a plurality of weight differences, and selecting a discharge speed corresponding to the minimum weight difference as the first optimized discharge speed.
[0028] The first optimization discharge speed is output by adjusting the discharge speed threshold and the discharge simulation model, aiming at the minimum deviation of the first discharge target, combining the vehicle discharge duration and the current feed ratio for discharge speed optimization analysis, including rendering the vehicle discharge duration and the current feed ratio to the discharge simulation model to generate a real-time discharge simulation space. The vehicle discharge duration is a pre-set discharge duration, and since the area of each region is the same, the vehicle discharge duration is also the same. The "vehicle discharge duration" and "current feed ratio" are input into the discharge simulation model in the three-dimensional simulation platform to simulate the material flow and distribution in the actual discharge process, thereby obtaining a real-time discharge simulation space for evaluating the feed delivery effect under different discharge speeds. Further, a first discharge speed is randomly selected in the discharge speed adjustment threshold, and the first simulation discharge weight is obtained by simulating the first discharge speed using the real-time discharge simulation space. The deviation of the first simulation discharge weight from the first discharge target is calculated to obtain the first weight difference. Further, continue to randomly select discharge speeds in the discharge speed adjustment threshold for simulation, until a predetermined speed optimization number is reached, output multiple discharge speeds and multiple weight differences, and select the discharge speed corresponding to the minimum weight difference as the first optimization discharge speed. Further, the optimization discharge speed of the remaining regions is obtained in the same way.
[0029] Based on the current feed ratio, the vehicle discharge duration and the first optimization discharge speed, the discharge angle optimization analysis is performed to aim at the uniform falling of the feed into the predetermined region, to output the first optimization discharge angle, and to generate the first optimization discharge parameter in combination with the first optimization discharge speed; continue to perform discharge parameter optimization analysis on other discharge targets in the sequence of region discharge targets to generate a sequence of optimization discharge parameters, and perform discharge error correction on the sequence of optimization discharge parameters to obtain an optimal discharge scheme for automatic discharge control of the dairy farm.
[0030] Based on the current feed ratio, vehicle discharge duration and first optimized discharge speed, a discharge angle is randomly selected in the discharge angle adjustment threshold, and discharge angle optimization analysis is performed to uniformly fall the feed into the predetermined area as the expectation according to the feed discharge prediction plug-in to obtain the discharge angle with the highest discharge fitness, output the first optimized discharge angle, and combine the first optimized discharge speed to generate the first optimized discharge parameter. Continue to use the same analysis method to perform discharge parameter optimization analysis on other discharge targets in the region discharge target sequence to generate an optimized discharge parameter sequence, and perform discharge error correction on the optimized discharge parameter sequence to avoid discharge errors caused by the increase of the running duration of the automatic discharge vehicle, and obtain the optimal discharge scheme for the automatic discharge control of the dairy farm. The technical problem of the prior art that the livestock machinery lacks a dynamic full life cycle supervision mechanism, resulting in large discharge deviation in the automatic discharge process and affecting the feeding effect is solved. The method provided by the embodiment of the application can accurately and efficiently complete the feed delivery under different use periods and breeding environments, significantly improve the feed utilization rate and ensure the health of the dairy cows.
[0031] The method provided by the embodiment of the application further includes: collecting a sample feed ratio set, a sample discharge duration set, a sample discharge speed set and a sample discharge angle set according to the breeding discharge log in the historical time zone, and labeling the feed distribution under different sample feed ratios, sample discharge durations, sample discharge speeds and sample discharge angles to obtain a sample feed distribution set, wherein the sample feed distribution includes sample distribution uniformity and sample overflow area proportion, and the overflow area proportion is the ratio of the area of the feed exceeding the predetermined area to the area of the predetermined area; using the sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set as training data, performing supervised learning on the random forest to obtain a feed discharge prediction plug-in; and using the feed discharge prediction plug-in, performing discharge angle optimization analysis according to the current feed ratio, vehicle discharge duration and first optimized discharge speed to uniformly fall the feed into the predetermined area as the expectation, and outputting the first optimized discharge angle.
[0032] The optimal discharge speed is obtained by the above steps, which can make the weight of the feeding closest to the target value under the given conditions of feed ratio and discharge time. However, only meeting the accuracy of the feeding weight cannot guarantee the uniform distribution of the feed in the predetermined area (such as the feeding trough or the specific area of the cowshed). If the discharge angle is too large or too small, it often causes uneven distribution of the feed or overflow of the predetermined area. Therefore, in order to ensure the uniform distribution of the feed in the predetermined area, the discharge angle needs to be optimized. By obtaining the historical data records formed in the daily feeding of the farm in the historical time interval, including the "feed ratio", "discharge time", "discharge speed", "discharge angle" and the actual distribution of the feed in the cowshed or feeding trough after each discharge, etc., the breeding discharge log in the historical time interval is obtained. According to the breeding discharge log in the historical time interval, a sample feed ratio set, a sample discharge time set, a sample discharge speed set and a sample discharge angle set are collected, and the feed distribution under different sample feed ratios, sample discharge times, sample discharge speeds and sample discharge angles is labeled. The labeled feed distribution of the corresponding sample is obtained, which includes sample distribution uniformity and sample overflow area ratio. The overflow area ratio is the ratio of the area of the feed exceeding the predetermined area to the area of the predetermined area, and the uniformity is an evaluation index reflecting the distribution of the feed, with 1 being completely uniform and lower indicating more uneven. Subsequently, the sample feed ratio set, the sample discharge time set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set are used as training data to supervise the learning of the random forest, and a feed discharge prediction plug-in is constructed. Using the feed discharge prediction plug-in, according to the current feed ratio, vehicle discharge time and first optimized discharge speed, the discharge angle is optimized and analyzed with the expectation of uniform feed falling into the predetermined area, and the first optimized discharge angle is output. The first optimized discharge angle corresponds to the first optimized discharge speed.
[0033] The method provided by the embodiments of the present application further includes: using the sample feed ratio set, the sample discharge time set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set as training data, and dividing the training data into P parts, and selecting P times with replacement to obtain a first training set, and continuing to select P times to obtain P training sets, wherein P is an integer greater than 10; taking the sample feed ratio, the sample discharge time, the sample discharge speed and the sample discharge angle as input, and taking the sample distribution uniformity and the sample overflow area ratio as supervision, using the P training sets to supervise the learning of the random forest until convergence, and harvesting P feed discharge prediction units; and based on the P feed discharge prediction units, the feed discharge prediction plug-in is integrated and constructed, wherein the output of the feed discharge prediction plug-in is the average of the outputs of the P feed discharge prediction units.
[0034] The sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set are used as training data to supervise learning of the random forest to obtain a feed discharge prediction plug-in, including: using the sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set and the sample feed distribution set as training data, and equally dividing the training data into P parts, and selecting P times with replacement to obtain a first training set, and iteratively selecting P times in the same way to obtain P training sets, wherein P is an integer greater than 10. Further, taking the sample feed ratio, the sample discharge duration, the sample discharge speed and the sample discharge angle as input data, and taking the sample distribution uniformity and the sample overflow area ratio as supervision data, P random forests are supervised learned by the P training sets respectively until convergence, and P feed discharge prediction units are obtained. The P feed discharge prediction units are integrated to construct the feed discharge prediction plug-in, wherein the output of the feed discharge prediction plug-in is the mean value of the outputs of the P feed discharge prediction units.
[0035] The method provided by the embodiment of the application further includes: obtaining a discharge angle adjustment threshold of the automatic discharge vehicle, and randomly selecting a first discharge angle in the discharge angle adjustment threshold; inputting the current feed ratio, the vehicle discharge duration, the first optimized discharge speed and the first discharge angle into the feed discharge prediction plug-in to output a first distribution uniformity and a first overflow area ratio; evaluating a first discharge fitness according to the first distribution uniformity and the first overflow area ratio, wherein the discharge fitness is positively correlated with the distribution uniformity and negatively correlated with the overflow area ratio; continuing to perform discharge angle optimization analysis according to the discharge angle adjustment threshold until a predetermined angle optimization number is reached, obtaining a plurality of discharge angles and a plurality of discharge fitnesses, and selecting a discharge angle with the maximum discharge fitness as the first optimized discharge angle.
[0036] The outputting the first optimized discharge angle includes: obtaining a discharge angle adjustment threshold of the automatic discharge vehicle, and randomly selecting a first discharge angle in the discharge angle adjustment threshold. Further, the current feed ratio, vehicle discharge duration, first optimized discharge speed and first discharge angle are input into the feed discharge prediction plug-in to output corresponding first distribution uniformity and first overflow area proportion. The first distribution uniformity and the first overflow area are weighted according to actual needs, for example, if the discharge uniformity is better, the corresponding distribution uniformity weight parameter configuration proportion is high. For example, the first distribution uniformity result is A, the first overflow area result is B, the weight parameter allocated to the first distribution uniformity is 0.7, and the weight parameter allocated to the first overflow area is 1-0.7=0.3. The first discharge fitness is obtained according to the evaluation of the first distribution uniformity and the first overflow area proportion, and the first discharge fitness is calculated according to the fitness formula 0.7*A-0.3*B, wherein the discharge fitness is positively correlated with the distribution uniformity and negatively correlated with the overflow area proportion. Random selection is continued in the discharge angle adjustment threshold, and discharge angle optimization analysis is performed according to the selected result until a predetermined angle optimization number is reached, a plurality of discharge angles and a plurality of discharge fitnesses are obtained, and the discharge angle with the maximum discharge fitness is selected as the first optimized discharge angle.
[0037] The method provided by the embodiment of the present application further includes: obtaining the cumulative working duration of the automatic discharge vehicle, performing discharge error analysis according to the cumulative working duration, determining a discharge speed compensation coefficient and a discharge angle compensation coefficient, wherein the discharge error analysis is performed based on a BP neural network to construct an error analysis channel, and the error analysis channel is trained to convergence by collecting a sample working duration set, a sample discharge speed compensation coefficient set and a sample discharge angle compensation coefficient set; and performing discharge error traversal correction on the optimized discharge parameter sequence according to the discharge speed compensation coefficient and the discharge angle compensation coefficient to obtain an optimal discharge parameter sequence, and combining the transport path to generate the optimal discharge scheme.
[0038] In the optimization of the sequence of the parameters of the material discharge error correction, the optimal material discharge scheme is obtained, including: obtaining the cumulative working time of the automatic material discharge vehicle, and performing material discharge error analysis according to the cumulative working time to obtain the material discharge speed compensation coefficient and the material discharge angle compensation coefficient. When performing the material discharge error analysis, the error analysis channel is constructed based on the BP neural network to perform the material discharge error analysis. The error analysis channel is constructed based on the BP neural network model, and the historical working record data of the automatic material discharge vehicle is acquired, the sample working time set, the sample material discharge speed compensation coefficient set and the sample material discharge angle compensation coefficient set are acquired, the sample working time, the sample material discharge speed compensation coefficient and the sample material discharge angle compensation coefficient correspond to each other, and the sample material discharge speed compensation coefficient and the sample material discharge angle compensation coefficient are compensation ratios. The sample working time set is used as the training data, and the sample material discharge speed compensation coefficient set and the sample material discharge angle compensation coefficient set are used as the supervision data to supervise the training of the BP neural network, until the accuracy of the model output meets the requirements, and the error analysis channel is obtained.
[0039] The cumulative working time of the automatic material discharge vehicle is input into the error analysis channel, and the output material speed compensation coefficient and the material discharge angle compensation coefficient are obtained. According to the material discharge speed compensation coefficient and the material discharge angle compensation coefficient, the material discharge error correction is performed on the sequence of the optimized material discharge parameters, and the material discharge speed compensation is taken as an example. The optimized material discharge speed is the sum of 1 and the material discharge speed compensation coefficient multiplied by the original material discharge speed. After compensation correction of all material discharge parameter sequences, the optimal material discharge parameter sequence is obtained, and the optimal material discharge scheme is generated in combination with the transportation path.
[0040] The technical scheme provided by the embodiment of the present application comprises the following steps: obtaining a plurality of dairy farming characteristics of a plurality of regions in a dairy farm, performing feeding analysis according to a current feed ratio and the plurality of dairy farming characteristics, determining a plurality of region discharge targets, and mapping and sorting the plurality of region discharge targets according to the transportation path of an automatic discharge vehicle to generate a region discharge target sequence; selecting a first discharge target in the region discharge target sequence, taking the minimum deviation from the first discharge target as the expectation, performing discharge speed optimization analysis in combination with the vehicle discharge duration and the current feed ratio to obtain a first optimized discharge speed; performing discharge angle optimization analysis based on the current feed ratio, the vehicle discharge duration and the first optimized discharge speed, taking the uniform falling of feed into a predetermined region as the expectation, outputting a first optimized discharge angle, and combining the first optimized discharge speed to generate a first optimized discharge parameter; continuing to perform discharge parameter optimization analysis on other discharge targets in the region discharge target sequence to generate an optimized discharge parameter sequence, performing discharge error correction on the optimized discharge parameter sequence, and obtaining an optimal discharge scheme for automatic discharge control of the dairy farm. The technical problem of the prior art that the livestock breeding machinery lacks a dynamic full-life-cycle supervision mechanism, resulting in large discharge deviation in the automatic discharge process and affecting the feeding effect is solved. The feeding of the feed can be accurately and efficiently completed under different use periods and breeding environments, the feed utilization rate is significantly improved, and the health of the dairy cows is ensured.
[0041] The above are only the preferred embodiments of the present application and the technical principles applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and more other equivalent embodiments can be included without departing from the concept of the present application.
Claims
1. A method for monitoring the entire life cycle of livestock machinery, characterized in that the method... The method comprises the following steps: acquiring a plurality of dairy farming characteristics of a plurality of regions in a dairy farm, performing feeding analysis according to a current feed ratio and the plurality of dairy farming characteristics, determining a plurality of region discharge targets, and mapping and sorting the plurality of region discharge targets according to a transportation path of an automatic discharge vehicle to generate a region discharge target sequence; selecting a first discharge target in the region discharge target sequence, taking the minimum deviation from the first discharge target as the expectation, and performing discharge speed optimization analysis combined with the vehicle discharge duration and the current feed ratio to obtain a first optimized discharge speed; based on the current feed ratio, the vehicle discharge duration and the first optimized discharge speed, taking the uniform falling of feed into the predetermined region as the expectation, performing discharge angle optimization analysis to output a first optimized discharge angle, and combining the first optimized discharge speed to generate a first optimized discharge parameter; continuing to perform discharge parameter optimization analysis on other discharge targets in the region discharge target sequence to generate an optimized discharge parameter sequence, and performing discharge error correction on the optimized discharge parameter sequence to obtain an optimal discharge scheme for automatic discharge control of the dairy farm; selecting a first discharge target in the region discharge target sequence, taking the minimum deviation from the first discharge target as the expectation, and performing discharge speed optimization analysis combined with the vehicle discharge duration and the current feed ratio to obtain a first optimized discharge speed, which comprises: selecting the first region discharge target in the region discharge target sequence as the first discharge target; performing simulation modeling on the automatic discharge vehicle in a three-dimensional simulation platform to generate a discharge simulation model; acquiring a discharge speed adjustment threshold of the automatic discharge vehicle, and based on the discharge speed adjustment threshold and the discharge simulation model, taking the minimum deviation from the first discharge target as the expectation, combining the vehicle discharge duration and the current feed ratio to perform discharge speed optimization analysis, and outputting the first optimized discharge speed; based on the discharge speed adjustment threshold and the discharge simulation model, taking the minimum deviation from the first discharge target as the expectation, combining the vehicle discharge duration and the current feed ratio to perform discharge speed optimization analysis, and outputting the first optimized discharge speed, which comprises: rendering the vehicle discharge duration and the current feed ratio to the discharge simulation model to generate a real-time discharge simulation space; randomly selecting a first discharge speed in the discharge speed adjustment threshold, and using the real-time discharge simulation space to simulate a first simulated discharge weight according to the first discharge speed; calculating the deviation of the first simulated discharge weight from the first discharge target to obtain a first weight difference; continuing to randomly select discharge speeds in the discharge speed adjustment threshold for simulation until a predetermined speed optimization number is reached, outputting a plurality of discharge speeds and a plurality of weight differences, and selecting the discharge speed corresponding to the minimum weight difference as the first optimized discharge speed; based on the current feed ratio, the vehicle discharge duration and the first optimized discharge speed, taking the uniform falling of feed into the predetermined region as the expectation, performing discharge angle optimization analysis to output a first optimized discharge angle, which comprises: According to the historical time zone, the sample feed ratio set, the sample discharge duration set, the sample discharge speed set, and the sample discharge angle set are collected, and the feed distribution under different sample feed ratios, sample discharge durations, sample discharge speeds, and sample discharge angles is labeled to obtain a sample feed distribution set, wherein the sample feed distribution includes sample distribution uniformity and sample overflow area ratio, and the overflow area ratio is the ratio of the area of the feed beyond the predetermined area to the area of the predetermined area; The sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set, and the sample feed distribution set are used as training data to supervise the learning of the random forest to obtain a feed discharge prediction plug-in; The feed discharge prediction plug-in is used to perform discharge angle optimization analysis according to the current feed ratio, vehicle discharge duration, and first optimized discharge speed, with the expectation that the feed is evenly distributed in the predetermined area, and the first optimized discharge angle is output. The sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set, and the sample feed distribution set are used as training data to supervise the learning of the random forest to obtain a feed discharge prediction plug-in, including: The sample feed ratio set, the sample discharge duration set, the sample discharge speed set, the sample discharge angle set, and the sample feed distribution set are used as training data, and the training data is equally divided into P parts, and P times are selected with replacement to obtain a first training set, and P times are iteratively selected to obtain P training sets, wherein P is an integer greater than 10. The sample feed ratio, sample discharge duration, sample discharge speed, and sample discharge angle are used as input, and the sample distribution uniformity and sample overflow area ratio are used as supervision, and the P training sets are used to supervise the learning of the random forest until convergence, and P feed discharge prediction units are obtained. The P feed discharge prediction units are integrated to construct the feed discharge prediction plug-in, wherein the output of the feed discharge prediction plug-in is the mean of the outputs of the P feed discharge prediction units. The feed discharge prediction plug-in is used to perform discharge angle optimization analysis according to the current feed ratio, vehicle discharge duration, and first optimized discharge speed, with the expectation that the feed is evenly distributed in the predetermined area, and the first optimized discharge angle is output, including: An adjustment threshold for the discharge angle of the automatic discharge vehicle is obtained, and a first discharge angle is randomly selected from the adjustment threshold. The current feed ratio, vehicle discharge duration, first optimized discharge speed, and first discharge angle are input into the feed discharge prediction plug-in to output first distribution uniformity and first overflow area ratio. A first discharge fitness is evaluated according to the first distribution uniformity and first overflow area ratio, wherein the discharge fitness is positively correlated with the distribution uniformity and negatively correlated with the overflow area ratio. The discharge angle optimization analysis is continued according to the discharge angle adjustment threshold until a predetermined number of angle optimization times are reached, a plurality of discharge angles and a plurality of discharge fitnesses are obtained, and the discharge angle with the maximum discharge fitness is selected as the first optimized discharge angle.
2. The livestock machinery lifecycle management method according to claim 1, wherein, Obtaining a plurality of dairy farming characteristics of a plurality of regions in a dairy farm, performing feeding analysis according to the current feed ratio and the plurality of dairy farming characteristics, and determining a plurality of region discharge targets, including: Obtaining a plurality of dairy farming characteristics of a plurality of regions in a dairy farm, wherein the dairy farming characteristics at least include dairy breed, quantity and growth stage, and the growth stage at least includes growth period, mature period, lactation period, pregnancy period and old age period; Obtaining the current feed ratio of the automatic discharge vehicle, wherein the current feed ratio includes material type and proportion; Based on the dairy farming specification table, the feeding standard retrieval is performed according to the current feed ratio and the plurality of dairy farming characteristics, and the region discharge weight is obtained, which is set as the region discharge target.
3. The livestock machinery lifecycle management method according to claim 1, wherein, The optimized discharge parameter sequence is corrected for discharge error to obtain an optimal discharge scheme, including: Obtaining the cumulative working time of the automatic discharge vehicle, performing discharge error analysis according to the cumulative working time, and determining the discharge speed compensation coefficient and the discharge angle compensation coefficient, wherein the discharge error analysis is performed based on the BP neural network to construct an error analysis channel, and the error analysis channel is trained to convergence by collecting sample working time set, sample discharge speed compensation coefficient set and sample discharge angle compensation coefficient set; According to the discharge speed compensation coefficient and the discharge angle compensation coefficient, the optimized discharge parameter sequence is corrected for discharge error to obtain an optimal discharge parameter sequence, and the optimal discharge scheme is generated in combination with the transportation path.
Citation Information
Patent Citations
Mobile environment monitoring automatic feeding system and device
CN111903547A
Cow breeding device
CN217564490U