Search

CN-121986723-A - Accurate feeding device

CN121986723ACN 121986723 ACN121986723 ACN 121986723ACN-121986723-A

Abstract

The invention discloses a precise feeding device, and relates to the technical field of automatic breeding in animal husbandry. The device comprises a biological characteristic acquisition module, a dynamic nutrition reconstruction module, an environment perception feedback assembly and a self-cleaning residue prevention mechanism, wherein the biological characteristic acquisition module is used for constructing a multidimensional biological characteristic model comprising facial textures, body condition three-dimensional point clouds and infrared thermal imaging, the dynamic nutrition reconstruction module comprises a main bin and a plurality of trace element adding bins, milligram-level nutrition allocation is realized through a high-precision spiral metering pump, the environment perception feedback assembly is used for monitoring feeding environment parameters in real time and correcting an energy supply model. The invention solves the technical problems that the prior art only depends on single radio frequency identification, is easy to lose, can not respond to the nutrition demand change of individuals in sub-health state in real time, and causes cross contamination due to drug feed residues. Through a multisource data fusion algorithm, dynamic and accurate feeding of 'one pig one strategy and one meal one preparation' is realized, and the feed conversion rate and the group health level are remarkably improved.

Inventors

  • LI CHANGYING

Assignees

  • 西南大学

Dates

Publication Date
20260508
Application Date
20260124

Claims (10)

  1. 1. An accurate feeding device, characterized by comprising: a main body frame (1), the main body frame (1) defining a feeding channel (11) for receiving the feeding of a single animal; the feeding execution mechanism (2) is arranged at the front end of the feeding channel (11), and the feeding execution mechanism (2) comprises a discharging mixing cavity (21) and a terminal feeding trough (22) which are communicated with the feeding execution mechanism; The biological characteristic acquisition module (3) is arranged at the inlet and above the feeding channel (11), and the biological characteristic acquisition module (3) comprises a binocular vision depth camera (31) and an infrared thermal imaging sensor (32) and is used for acquiring identity information, three-dimensional body condition data and body surface temperature distribution data of livestock in a non-contact manner; The nutrition supply system (4) comprises at least one base stock bin (41) and a plurality of micro-additive bins (42), and the base stock bin (41) and the micro-additive bins (42) are communicated with the discharging mixing cavity (21) through independent precise metering conveying mechanisms; The central control unit (5) is electrically connected with the biological characteristic acquisition module (3), the feeding execution mechanism (2) and the nutrition supply system (4) respectively; The central control unit (5) is configured to receive multidimensional data acquired by the biological characteristic acquisition module (3), calculate an identity ID, a weight, a Body Condition Score (BCS) and a health state index of livestock through a deep learning algorithm, call a preset dynamic nutrition model based on the health state index, calculate the base stock quantity and the proportion of specific micro-additives required by the current meal of the livestock, and control the nutrition supply system (4) to convey materials with corresponding proportions to the blanking mixing cavity (21) for instantaneous mixing and then throwing.
  2. 2. The precise feeding device according to claim 1, wherein a bidirectional differential mixing auger (211) is arranged in the feeding mixing cavity (21), a nanometer hydrophobic coating is sprayed on the surface of a blade of the mixing auger (211), an electronic gate (212) which is controlled to open and close is arranged at the bottom of the feeding mixing cavity (21), a high-pressure air flow nozzle (213) is arranged below the electronic gate (212), and the angle of the high-pressure air flow nozzle (213) faces the inner surface of the electronic gate (212) and dead angle areas of the terminal feeding trough (22) and is used for executing a pneumatic cleaning program after each feeding is finished so as to prevent medicine-containing feed residues.
  3. 3. The precise feeding device according to claim 1, wherein the precise metering and conveying mechanism of the micro-additive bin (42) is matched with a star-shaped discharge valve (421) by adopting a micro-vibration feeder driven by piezoelectric ceramics, and a photoelectric counting sensor (422) is arranged at a discharge hole of the star-shaped discharge valve (421) and used for calibrating the number or volume of dropping particles of the micro-additive in real time so as to realize milligram-level precise nutrition addition.
  4. 4. The precise feeding device according to claim 1, wherein a suspended weighing platform (12) is arranged at the bottom of the feeding channel (11), at least four high-precision pressure sensors (121) are uniformly distributed below the weighing platform (12), a dynamic filtering algorithm is arranged in the central control unit (5) and used for filtering vibration noise generated by feeding and walking of livestock, obtaining real-time net weight of the livestock, and generating a weight-volume correlation coefficient by combining volume data calculated by the binocular vision depth camera (31) and used for monitoring growth density change of the livestock.
  5. 5. The accurate feeding device according to claim 1 is characterized in that an environment sensing assembly is further integrated on the main body frame (1), the environment sensing assembly comprises an ammonia concentration sensor, a temperature and humidity sensor and an illumination intensity sensor, the central control unit (5) corrects energy supply parameters in the nutrition model according to individual characteristics of livestock and real-time data of the environment sensing assembly, and when the environment temperature is lower than a preset lower limit threshold, the proportioning coefficient of energy feed in a base material is automatically improved.
  6. 6. The precise feeding device according to claim 1, wherein the biological characteristic acquisition module (3) further comprises a lateral light supplementing lamp group (33) and a multi-angle face capturing camera (34) which are arranged on the side wall of the feeding channel (11), the multi-angle face capturing camera (34) is used for capturing facial micro-expression characteristics of livestock, and the central control unit (5) evaluates pain or stress level of the livestock by analyzing gesture characteristics of eyes and ears, and automatically triggers feeding instructions of specific anti-stress additives when the high stress level is detected.
  7. 7. The precise feeding device according to claim 2, wherein the terminal feeding trough (22) is formed by stamping 304 food-grade stainless steel, the inner wall of the terminal feeding trough is designed to be a right-angle-free arc transition structure, an infrared blocking type material level sensor (221) is arranged at the edge of the feeding trough (22) and used for monitoring the situation of residual materials in the feeding trough, the central control unit (5) calculates the feeding speed of livestock according to the residual material reduction rate, and if the feeding speed is lower than 80% of the historical average value of the individual, a health early warning signal is generated and pushed to a cloud management platform.
  8. 8. The precise feeding device according to claim 1, wherein an intelligent shunt door (7) is arranged at the outlet of the feeding channel (11), the shunt door (7) is provided with three working positions, namely a normally open position, a locked feeding position and a shunt channel position, and when the central control unit (5) judges that livestock are in a diseased or oestrus state, the shunt door (7) is controlled to be switched to the shunt channel position after feeding is finished, and the livestock are guided to an isolation area.
  9. 9. The precise feeding device according to claim 1, further comprising a power-off protection and data storage module, wherein a super capacitor group and a nonvolatile memory are integrated in the device, and when external power supply is interrupted accidentally, the super capacitor group provides instantaneous power to drive the electronic gate (212) to close and save the current feeding task state, and data is automatically transmitted after power is recovered, so that the continuity and the integrity of feeding data are ensured.
  10. 10. The precise feeding device according to any one of claims 1 to 9, wherein the central control unit (5) is connected to a remote cloud server through a 5G/NB-IoT communication module, a population evolution algorithm is deployed on the cloud server, historical feeding data and growth results of all networking devices can be collected, weight parameters of the dynamic nutrition model and the biological recognition algorithm are periodically and iteratively updated, and updated firmware OTA is pushed to the central control unit (5).

Description

Accurate feeding device Technical Field The invention relates to the technical field of modern animal husbandry breeding equipment, in particular to a precise feeding device. Background Along with the popularization of large-scale and intensive culture modes, precise feeding (PrecisionFeeding) has become a key technical path for improving the feed conversion rate, reducing the culture cost, reducing the environmental pollution and guaranteeing animal welfare. Traditional feeding methods generally employ bulk feeding (PHASEFEEDING), i.e., providing a uniform formulation of feed to a population at the same stage of growth. However, this approach ignores significant differences in genetic potential, growth rate, health status, and feeding behavior among animal individuals. Some automated feeding equipment has emerged in the prior art. For example, chinese patent CN112931269a discloses an intelligent accurate feeding system, which recognizes the identity of pigs through RFID electronic ear tags and adjusts the feeding amount according to the body weight. However, the prior art solutions have the following pain points and limitations: The identification dependence is single, namely the identification dependence mainly depends on the RFID earmark. In an actual cultivation environment, the ear tag is easy to fall off or damage due to fight and scratch, so that the identification is failed. In addition, the ear tag mounting process is traumatic to animals, is prone to inflammation, and violates animal welfare principles. The dimension of nutrition regulation is insufficient, the existing 'precision' is limited to 'quantity' precision, namely, the total feeding quantity is controlled, and the 'quality' real-time reconstruction cannot be realized. For example, when a pig is in sub-health, has reduced immunity or is in a highly stressed state, its need for vitamins, electrolytes or specific drugs may increase, while the need for energy may decrease. The existing equipment can not adjust microscopic nutritional ingredients of the feed on line in real time, and can only be manually intervened or administrated in a whole group, so that huge drug waste and drug residue risks are caused. The lack of multidimensional body condition sensing can not fully reflect the growth state of animals only by body weight. The body condition score (BodyConditionScore, BCS) is an important index for measuring the nutrition reserve of animals, and is high in subjectivity and low in efficiency mainly by manual visual inspection at present. The lack of comprehensive perception of body temperature, body surface damage, mental state (e.g., pain recognition through facial expression) results in a feeding decision model that is too simple to achieve true physiological demand matching. Cross-infection risk-when a drug or functional additive needs to be fed to a specific individual, the common blanking channel of the existing equipment is easy to generate residues. When the next healthy animal feeds, the residual medicine may be miseaten, resulting in food safety hazards (such as antibiotic residues) and drug resistance problems. Therefore, there is a need to develop an intelligent and accurate feeding device which can identify the identity and health status of an individual in a non-contact multi-dimensional manner, dynamically reconstruct the nutrition formulation in real time, and has a self-cleaning function so as to meet the high standard requirements of the modern intelligent animal husbandry. Disclosure of Invention The invention aims to provide a precise feeding device so as to solve the problems in the background technology. In order to achieve the above purpose, the present invention provides the following technical solutions: an accurate feeding device comprising: a body frame defining a feed channel for receiving a single animal feed; The feeding execution mechanism is arranged at the front end of the feeding channel and comprises a discharging mixing cavity and a terminal feeding trough which are communicated with the feeding execution mechanism; The biological characteristic acquisition module is arranged at the inlet of the feeding channel and above the feeding channel and comprises a binocular vision depth camera and an infrared thermal imaging sensor, and is used for acquiring identity information, three-dimensional body condition data and body surface temperature distribution data of livestock in a contactless manner; The nutrition supply system comprises at least one base stock bin and a plurality of micro-additive bins, and the base stock bin and the micro-additive bins are communicated with the discharging mixing cavity through independent precise metering conveying mechanisms; the central control unit is respectively and electrically connected with the biological characteristic acquisition module, the feeding execution mechanism and the nutrition supply system; The central control unit is configured to receive multidimensiona