Accurate control method and system for water content of feed, electronic equipment and storage medium

Abstract

The invention discloses a feed moisture content accurate control method and system, electronic equipment and a storage medium, and the method comprises the steps: uploading a feed formula, online collected data related to the feed moisture content and the like to a cloud end, and taking the data as historical data for training a machine learning model; the machine learning model is trained and verified through historical data, and a rule model used for predicting the water adding amount of the mixing machine is output; and deploying the trained rule model to a central control system, predicting the water adding amount of the mixer based on online monitoring data, and controlling a water adding system to add water according to a predicted value by the central control system. According to the invention, accurate control of the feed moisture content of the granulator production line is successfully realized, and the moisture content deviation of the feed finished product is reduced to + / -0.5%.

Description

Technical Field

[0001] The present invention belongs to the technical field of feed preparation, and in particular relates to a method, system, electronic equipment and storage medium for accurately controlling the moisture content of feed. Background Art

[0002] The moisture content of feed is an important indicator of feed. It is not only related to the quality of finished feed, but also has an important impact on the processing of feed. In order to avoid mold and deterioration during transportation and storage, feed raw materials are generally dried, so the moisture content is low. It is necessary to increase the moisture content of the feed mixture by adding water to the mixer and steam to the conditioner to meet the process requirements. Taking the feed pelletizing production line as an example, the moisture content of the mixture entering the pelletizing bin of the pelletizing machine after modulation is more reasonable between 15% and 17% (different feed formulas will have slightly different reasonable moisture content ranges, but generally they are distributed in a range of two percentage points). At this time, the pellet feed produced has good processing quality, uniform finish, low powdering rate, and low energy consumption. When the moisture content is lower than 15%, the surface of the feed pellets will be rough, resulting in an increase in the powdering rate. When the moisture content is higher than 17%, the mold will slip, resulting in blockage of the mold hole, and the machine needs to be stopped for clearing. According to statistics, the pelletizer has an average of about one shutdown failure caused by mold blockage every 8 hours, and each time the production is delayed by about 1 hour. The moisture content of finished feed also needs to be controlled within 12.5% ​​(there are slight differences between different types of feed), but too low a moisture content will also cause pellet feed to be easily powdered during transportation, reducing the palatability of the feed and directly affecting the animal's eating and nutritional intake. In order to ensure that the feed will not mold and deteriorate, feed mills generally control the moisture content of the feed to a lower level, which inevitably leads to waste of feed raw materials and excessive energy consumption.

[0003] The feed industry generally uses sampling method to detect the moisture content of raw materials, and then sprays water in the mixer to compensate. The moisture content deviation of the finished feed is about ±2%. The moisture content deviation of the finished feed is one percentage point higher than that of the raw materials, resulting in a raw material loss of 1% for the feed company. Therefore, whether from the perspective of the quality of the finished feed or the saving of raw materials, it is necessary to accurately control the moisture content of the feed. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method, system, electronic device and storage medium for accurately controlling the moisture content of feed.

[0005] The technical solution provided by the present invention is specifically as follows:

[0006] A method for accurately controlling the moisture content of feed, comprising the steps of:

[0007] Acquire the test data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed to build a historical database; the test data includes the moisture content of the mixed material in the mixer, the amount of water added in the mixer, the modulation temperature and steam flow of the mixer, and the air volume, humidity and temperature of the cooler;

[0008] Train, validate, and optimize machine learning models through historical databases;

[0009] Convert the trained machine learning model into a rule model and deploy it into the feed production central control system;

[0010] In the feed production process, the rule model outputs the predicted water addition amount of the mixer based on the online detection data, and controls the mixer to add water according to the predicted water addition amount;

[0011] Update the historical database regularly and use the updated historical database to continuously optimize and update the machine learning model.

[0012] Preferably, a near infrared sensor is installed on the side wall of the mixer to obtain the moisture content of the mixed material in the mixer.

[0013] Preferably, a material taking port is provided at the discharge port of the cooler to sample and obtain the moisture content of the finished feed.

[0014] Preferably, the historical database is stored in the cloud, where the machine learning model is trained and the rule model is output.

[0015] A precise control system for feed moisture content based on the above method comprises:

[0016] The data detection module is used to obtain historical detection data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed, and is also used to obtain online detection data related to the moisture content of feed in each link in real time;

[0017] The cloud database module is used to receive and store historical test data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed;

[0018] The machine learning module is used to train the machine learning model based on the historical data stored in the cloud database module, and convert the trained machine learning model into a rule model and output it to the feed production central control system;

[0019] The feed production central control system is used to output the predicted water addition amount of the mixer based on the online detection data obtained by the data detection module according to the deployed rule model, and is also used to control the mixer to add water according to the predicted water addition amount.

[0020] Furthermore, the machine learning module is also used to regularly optimize and update the machine learning model based on the historical data updated by the cloud database module, and reconvert it into a rule model and deploy it into the feed production central control system.

[0021] An electronic device comprises a memory, a processor and a computer program stored and running on the memory, wherein when the processor executes the program, the steps of the method for accurately controlling the moisture content of feed as described above are implemented.

[0022] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the method for accurately controlling the moisture content of feed as described above.

[0023] Compared with the prior art, the present invention has at least the following beneficial effects:

[0024] The present invention uploads feed formula, online data collected related to feed moisture content, etc. to the cloud as historical data for training machine learning models; trains and verifies machine learning models through historical data, outputs rule models for predicting the amount of water added to the mixer; deploys the trained rule models to the central control system, predicts the amount of water added to the mixer based on online monitoring data, and the central control system controls the water adding system to add water according to the predicted value. The precise control of feed moisture content in the pelletizing machine production line was successfully achieved, and the moisture content deviation of the finished feed was reduced to ±0.5%. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0026] Figure 1 It is a schematic diagram of the framework of a method for accurately controlling moisture content of feed provided by an embodiment of the present invention;

[0027] Figure 2 It is the deviation between the moisture content of a feed product of a feed factory and the target value provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0029] Embodiment 1

[0030] Feed pelleting production generally includes raw material crushing, mixing, conditioning, pelleting, cooling, and finished products. Each link has a different degree of influence on the moisture content of the feed. Crushing process: When the raw materials are crushed by the crushing equipment (such as a crusher or grinder), the bound water in the raw materials is converted into free water due to the crushing of the raw material cells and high-temperature evaporation, resulting in a decrease in moisture content; excessive moisture content of the raw materials will lead to a decrease in crushing efficiency and increase energy consumption; in addition, wet raw materials are easy to adhere to the inside of the crusher, increasing the risk of blockage and equipment wear. Mixing stage: During the mixing stage, the various raw materials are evenly stirred, and the moisture content of the raw materials is required to be uniform. If necessary, additional water or liquid may be added to coordinate the overall moisture content (the amount of water added cannot exceed 2% of the weight of the material, and the water retention rate is generally 40% to 50%). The goal is to ensure that the moisture content of the mixture is moderate, usually between 10% and 13%. Conditioning: In the conditioning stage, the mixed raw materials are steamed to ripen the starch and denature the protein. The steam added in this process is converted into water and integrated into the feed raw materials. The moisture content of the conditioned feed should be controlled at 15% to 17%. The steam added in this stage is used to control the conditioning temperature, and the actual amount of water added is passive. Granulation: In the granulation stage, the moisture content plays a decisive role in the granulation effect. The moisture content of the raw materials should generally be controlled between 10% and 13%, and after conditioning, it is increased to between 14% and 17% to ensure the formation and strength of the pellets. Too high a moisture content may cause the pelletizing die to be blocked (on average, a shutdown caused by mold blockage occurs every 6 to 8 hours, delaying production by about 1 hour each time), while too low a moisture content may cause the pellets to be fragile and brittle. Cooling stage: The pellets just out of the pellet mill are usually at a high temperature and have an increased moisture content. The pellets need to be cooled to room temperature through a cooler, while the moisture content is reduced to the appropriate standard for the finished product.

[0031] In summary, the mixing stage is the only link that can actively change the moisture content of the feed. To this end, this embodiment provides a method for accurately controlling the moisture content of the feed, which controls the moisture content of the finished feed by controlling the amount of water added in the mixing stage.

[0032] like Figure 1 The specific implementation route is as follows:

[0033] (1) Sensors are placed at key links in feed production to collect data related to feed moisture content online, such as the moisture content of the mixed material and the amount of water added during the mixing stage, the modulation temperature and steam flow rate during the modulation stage, the energy consumption during the pelleting stage, the air volume, humidity and temperature during the cooling stage, and the moisture content of the finished feed obtained through sampling testing during the finished product stage.

[0034] (2) The feed production line continues for a period of time until enough historical data (including feed formula, processing parameters, sensor detection data, moisture content of sampled feed products, etc.) is accumulated.

[0035] (3) Train, verify, and optimize machine learning models using historical data.

[0036] (4) Convert the trained machine learning model into a rule model that can be deployed by the feed factory central control system.

[0037] (5) During the feed production process, the rule model predicts and calculates the amount of water added to the mixer based on the online detection data, and controls the mixer to add water according to the predicted amount of water added.

[0038] (6) The central control system regularly uploads relevant data such as feed formula, processing parameters, sensor detection data, and moisture content of sampled feed products to the remote end to update the historical database;

[0039] (7) Use the updated historical database to continuously optimize and update the machine learning model.

[0040] In order to obtain the moisture content of the material in the mixing stage, this embodiment installs a near infrared sensor on the side wall of the mixer (the sensor is installed in the area swept by the mixer blades to ensure that the material at the detection position is always in motion). A material extraction port is set at the discharge port of the cooler to sample and detect the moisture content of the finished feed and upload it to the central control system.

[0041] In order to obtain a more accurate rule model, this embodiment stores the acquired historical data in the cloud, trains the machine learning model in the cloud, and outputs the rule model; the trained model is downloaded to the local system, and the control targets such as the moisture content of the feed are predicted through the data collected by the online detection system, and the local automatic control system performs closed-loop control based on the predicted value and the target value; the local system uploads the online collected data and results (labels) to the cloud, and trains the machine learning model as historical data. Machine learning is a branch of artificial intelligence, and it is a core technology for realizing artificial intelligence. In theory, as long as enough data is collected and the computer has the ability to process this data, an accurate transfer function (black box) between the controllable parameters of the production line (here is the amount of water added in the mixer) and the quality of the feed can be constructed.

[0042] To verify the effectiveness of the method in this embodiment, it was deployed in a feed factory. The moisture content distribution of the finished feed samples tested in half a year was as follows: Figure 2 As shown in the figure, the moisture content deviation is reduced to ±0.5%, achieving precise control of the moisture content of the feed in the pellet mill production line.

[0043] Embodiment 2

[0044] Based on the above method, this embodiment provides a feed moisture content accurate control system, which mainly includes:

[0045] The data detection module is used to obtain historical detection data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed, and is also used to obtain online detection data related to the moisture content of feed in each link in real time;

[0046] The cloud database module is used to receive and store historical test data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed;

[0047] The machine learning module is used to train the machine learning model based on the historical data stored in the cloud database module, and convert the trained machine learning model into a rule model and output it to the feed production central control system;

[0048] The feed production central control system is used to output the predicted water addition amount of the mixer based on the online detection data obtained by the data detection module according to the deployed rule model, and is also used to control the mixer to add water according to the predicted water addition amount.

[0049] In some embodiments, the machine learning module is also used to regularly optimize and update the machine learning model based on the historical data updated by the cloud database module, and reconvert it into a rule model and deploy it into the feed production central control system.

[0050] Embodiment 3

[0051] According to an embodiment of the present invention, an electronic device is provided, which may include a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus. The processor may call the logic instructions in the memory to execute a method for accurately controlling the moisture content of feed, the method comprising: obtaining detection data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed to construct a historical database; training, verifying and optimizing the machine learning model through the historical database; converting the trained machine learning model into a rule model and deploying it in the feed production central control system; the rule model outputs the predicted water addition amount of the mixer based on the online detection data, and controls the mixer to add water according to the predicted water addition amount.

[0052] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in several computer-readable storage media when sold or used as independent products. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in the first embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0053] The above-mentioned product can execute the method described in Example 1, and has the corresponding functional modules and beneficial effects of the method. For technical details not fully described in this embodiment, please refer to the detailed contents provided in Example 1 of the present invention.

[0054] Embodiment 4

[0055] According to an embodiment of the present invention, a computer-readable storage medium is provided, the type of which is as described in Example 3, and the computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the method for accurately controlling the moisture content of feed described in Example 1.

[0056] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the relevant technology, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may also be combined, the steps may be implemented in any order, and there are many other changes in different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for accurately controlling the moisture content of feed, characterized in that: Includes steps: Acquire the test data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed to build a historical database; the test data includes the moisture content of the mixed material in the mixer, the amount of water added in the mixer, the modulation temperature and steam flow of the mixer, and the air volume, humidity and temperature of the cooler; Train, validate, and optimize machine learning models through historical databases; Convert the trained machine learning model into a rule model and deploy it into the feed production central control system; In the feed production process, the rule model outputs the predicted water addition amount of the mixer based on the online detection data, and controls the mixer to add water according to the predicted water addition amount; Update the historical database regularly and use the updated historical database to continuously optimize and update the machine learning model.

2. A method for accurately controlling the moisture content of feed according to claim 1, characterized in that: The moisture content of the mixed material in the mixer is obtained by installing a near infrared sensor on the inner wall of the mixer.

3. A method for accurately controlling the moisture content of feed according to claim 1, characterized in that: By setting a feeding port at the discharge port of the cooler, the moisture content of the finished feed can be sampled.

4. A method for accurately controlling the moisture content of feed according to claim 1, characterized in that: The historical database is stored in the cloud, and the machine learning model is trained and the rule model is output in the cloud.

5. A feed moisture content precision control system based on the method according to any one of claims 1 to 4, characterized in that: include: The data detection module is used to obtain historical detection data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed, and is also used to obtain online detection data related to the moisture content of feed in each link in real time; The cloud database module is used to receive and store historical test data related to the moisture content of feed in each link of feed production and the corresponding moisture content of finished feed; The machine learning module is used to train the machine learning model based on the historical data stored in the cloud database module, and convert the trained machine learning model into a rule model and output it to the feed production central control system; The feed production central control system is used to output the predicted water addition amount of the mixer based on the online detection data obtained by the data detection module according to the deployed rule model, and is also used to control the mixer to add water according to the predicted water addition amount.

6. The feed moisture content precision control system according to claim 5, characterized in that: The machine learning module is also used to regularly optimize and update the machine learning model based on the historical data updated by the cloud database module, and reconvert it into a rule model and deploy it into the feed production central control system.

7. An electronic device comprising a memory, a processor and a computer program stored and running on the memory, characterized in that: When the processor executes the program, the steps of the method for accurately controlling the moisture content of feed according to any one of claims 1 to 4 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the method for accurately controlling the moisture content of feed as described in any one of claims 1 to 4.

Images