A method and system for solid-liquid separation of cow dung based on convolutional neural network

Abstract

The present invention discloses a cow dung solid-liquid separation method and system based on a convolutional neural network, which belongs to the field of feces treatment technology, and includes: S10, constructing a cow dung solid-liquid separation network model based on a convolutional neural network; S20, using training samples to train the cow dung solid-liquid separation network model; S30, inputting the real-time parameters of the operation of the cow dung solid-liquid separation device into the trained cow dung solid-liquid separation network model, and the trained cow dung solid-liquid separation network model gives a parameter matching scheme, which is the basis for the operation of the cow dung solid-liquid separation device. The present invention provides a method that can provide the optimal parameters for the operation of the cow dung solid-liquid separation device according to the real-time parameters of the operation of the cow dung solid-liquid separation device, so that the separation efficiency of the cow dung solid-liquid separation device is optimized to meet the storage and transportation requirements, saving a lot of manpower and material resources.

Description

Technical Field

[0001] The present invention belongs to the technical field of feces treatment, and in particular relates to a cow dung solid-liquid separation method and system based on convolutional neural network. Background Art

[0002] Dairy cows produce a lot of cow dung during breeding. If the moist cow dung is not separated into solid and liquid, it is easy for bacteria to breed in the cow dung, thus polluting the environment. In order to solve this problem, people invented the cow dung solid-liquid separator. The cow dung treated by the cow dung solid-liquid separator can enter the biogas digester to make biogas, and can also be directly used as fertilizer for fruit trees and forests and as a raw material for organic fertilizer.

[0003] However, in the process of processing cow dung, the extrusion time and drying time of the cow dung solid-liquid separator of the prior art remain unchanged regardless of the water content of the cow dung, resulting in that the processing capacity of the cow dung solid-liquid separator cannot be fully utilized. When the water content of the cow dung is high, the dryness of the cow dung residue produced by the cow dung solid-liquid separator cannot meet the storage and transportation requirements, and two drying operations are required, resulting in a large waste of manpower and material resources.

[0004] Therefore, there is an urgent need for a method and system for separating cow dung solid from liquid that can automatically adjust the extrusion and drying time according to the water content of cow dung to meet the dryness requirements of cow dung residue storage and transportation. Summary of the invention

[0005] The object of the present invention is to provide a method and system for separating cow dung solid from liquid, which can automatically adjust the extrusion and drying time according to the water content of cow dung to meet the dryness requirements of cow dung residue storage and transportation. The present invention adopts the following technical solutions:

[0006] A method for solid-liquid separation of cow dung based on convolutional neural network, comprising:

[0007] S10. Construct a network model for solid-liquid separation of cow dung based on convolutional neural network;

[0008] S20, training the cow dung solid-liquid separation network model using training samples;

[0009] S30, inputting the real-time operating parameters of the cow dung solid-liquid separation device into the trained cow dung solid-liquid separation network model, and providing a parameter matching scheme by the trained cow dung solid-liquid separation network model, and the parameter matching scheme is the basis for the operation of the cow dung solid-liquid separation device.

[0010] Furthermore, the cow dung solid-liquid separation device is provided with a feeding area, an extrusion area, a drying area, a slag discharge area and a drainage area; the training samples include the following parameters: the mass, moisture content and temperature of the cow dung entering the feeding area; the extrusion time and motor speed of the extrusion area; the moisture content and temperature, drying time and drying temperature of the cow dung entering the drying area; the slag discharge amount, slag discharge temperature and moisture content of the slag discharge area; the drainage amount of the drainage area; the real-time parameters of the operation of the cow dung solid-liquid separation device are consistent with the training samples.

[0011] Furthermore, the structure of the cow dung solid-liquid separation network model is as follows: Figure 2 shown.

[0012] Furthermore, the training process of the cow dung solid-liquid separation network model in step S20 is as follows:

[0013] S21, obtaining parameters in the training sample, wherein the parameters in the training sample are the basis for generating the parameter matching scheme of the cow dung solid-liquid separation network model;

[0014] S22, inputting each parameter in the training sample into the cow dung solid-liquid separation network model;

[0015] S23, the cow dung solid-liquid separation network model selects the best matching parameters between various parameters through a genetic algorithm to form a parameter matching scheme;

[0016] S24. The cow dung solid-liquid separation network model sends the parameter matching solution to the cow dung solid-liquid separation device.

[0017] Furthermore, the feeding area is provided with a first quality detection device for measuring the quality of cow dung entering, a first moisture content tester for measuring the moisture content of cow dung, and a first temperature sensor for measuring the temperature of cow dung; the drying area is provided with a second moisture content tester for measuring the moisture content of cow dung to be dried, a second temperature sensor for measuring the temperature of cow dung to be dried, and a third temperature sensor for measuring the drying temperature; the slag discharge area is provided with a second quality detection device for measuring the slag discharge amount, a fourth temperature sensor for measuring the slag discharge temperature, and a third moisture content tester for measuring the moisture content of the slag discharge; the drainage area is provided with a third quality detection device for measuring the drainage amount.

[0018] Furthermore, the convolutional neural network includes an input layer, a convolution layer, and a pooling layer, wherein the number of convolutional layers is 16, the convolutional layer with a convolution kernel size of 3*3, and the number of maximum pooling layers is 5.

[0019] Furthermore, the training samples include 80% of training samples and 20% of testing training samples.

[0020] A cow dung solid-liquid separation system based on convolutional neural network, comprising:

[0021] A data acquisition module is used to collect real-time parameters of the operation of the cow dung solid-liquid separation device;

[0022] A data storage module, connected to the data acquisition module, and used to store the data collected by the data acquisition module;

[0023] A data processing module connected to the data storage module; the data processing module is a model based on a convolutional neural network, and is used to provide a parameter matching scheme according to the real-time parameters of the cow dung solid-liquid separation device stored in the data storage module;

[0024] A training module, connected to the data processing module and the data storage module, for selecting the best matching parameters among the real-time parameters collected by the data collection module through a genetic algorithm, forming a parameter matching scheme and uploading it to the data processing module for training;

[0025] A control module, connected to the data processing module, is used to download the trained data processing module to achieve acquisition, training and identification of real-time parameters of the cow dung solid-liquid separation device;

[0026] An input module, connected to the control module and the data storage module, for inputting the real-time parameters of the cow dung solid-liquid separation device into the trained data processing module;

[0027] A display module, connected to the control module, for displaying the operating parameters given by the trained data processing module;

[0028] Cow dung solid-liquid separation device is used for solid-liquid separation of cow dung.

[0029] Furthermore, the cow dung solid-liquid separation device is provided with a feeding area, an extrusion area, a drying area, a slag discharge area and a drainage area; the feeding area is provided with a first quality detection device for measuring the quality of the cow dung entering, a first moisture content tester for measuring the moisture content of the cow dung and a first temperature sensor for measuring the temperature of the cow dung; the drying area is provided with a second moisture content tester for measuring the moisture content of the cow dung to be dried, a second temperature sensor for measuring the temperature of the cow dung to be dried and a third temperature sensor for measuring the drying temperature; the slag discharge area is provided with a second quality detection device for measuring the slag discharge amount, a fourth temperature sensor for measuring the slag discharge temperature and a third moisture content tester for measuring the slag discharge moisture content; the drainage area is provided with a third quality detection device for measuring the drainage amount.

[0030] Beneficial effects:

[0031] The present invention provides a cow dung solid-liquid separation method and system based on a convolutional neural network, which can automatically adjust the extrusion and drying time, drying temperature, etc. according to the real-time parameters of the cow dung solid-liquid separation device, i.e., the water content of the cow dung, the temperature of the cow dung, etc., to meet the dryness requirements of the cow dung residue. At the same time, the cow dung solid-liquid separation network model gives the optimal parameters for the operation of the cow dung solid-liquid separation device according to the real-time parameters of the operation of the cow dung solid-liquid separation device, so that the separation efficiency of the cow dung solid-liquid separation device is optimized to meet the storage and transportation requirements, saving a lot of manpower and material resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Flowchart of a cow dung solid-liquid separation system based on convolutional neural network

[0033] Figure 2 Schematic diagram of the network model for solid-liquid separation of cow dung

[0034] Figure 3 A diagram of the structure of a cow dung solid-liquid separation system based on convolutional neural network

[0035] Among them, 1. Cow dung solid-liquid separation device; 11. Feeding area; 12. Extrusion area; 13. Drying area; 14. Slag discharge area; 15; Liquid discharge area; 2. Data acquisition module; 3. Data storage module; 4. Data processing module; 5. Training module; 6. Input module; 7. Control module; 8. Display module. DETAILED DESCRIPTION

[0036] Example 1

[0037] A method for solid-liquid separation of cow dung based on convolutional neural network, comprising:

[0038] S10. Construct a cow dung solid-liquid separation network model based on a convolutional neural network. The structure of the cow dung solid-liquid separation network model is as follows: Figure 2 shown.

[0039] The convolutional neural network includes an input layer, a convolution layer, and a pooling layer. The number of convolutional layers is 16, the convolution kernel size is 3*3, and the number of maximum pooling layers is 5.

[0040] S20. Use training samples to train the cow dung solid-liquid separation network model; the training samples include 80% training samples and 20% test training samples.

[0041] The training process of the cow dung solid-liquid separation network model is as follows:

[0042] S21, obtaining parameters in the training sample, wherein the parameters in the training sample are the basis for generating a parameter matching solution for a cow dung solid-liquid separation network model;

[0043] S22, inputting each parameter in the training sample into the cow dung solid-liquid separation network model;

[0044] S23, the cow dung solid-liquid separation network model selects the best matching parameters between various parameters through a genetic algorithm to form a parameter matching scheme;

[0045] S24. The cow dung solid-liquid separation network model sends the parameter matching solution to the cow dung solid-liquid separation device.

[0046] S30, inputting the real-time operating parameters of the cow dung solid-liquid separation device into the trained cow dung solid-liquid separation network model, and the trained cow dung solid-liquid separation network model provides a parameter matching solution, which is the basis for the operation of the cow dung solid-liquid separation device.

[0047] Among them, the cow dung solid-liquid separation device is provided with a feeding area, an extrusion area, a drying area, a slag discharge area and a drainage area; the training samples include the following parameters: the mass, moisture content and temperature of the cow dung entering the feeding area; the extrusion time and motor speed of the extrusion area; the moisture content and temperature of the cow dung entering the drying area, the drying time and the drying temperature; the slag discharge amount, slag discharge temperature and moisture content of the slag discharge area; the drainage amount of the drainage area; the real-time parameters of the cow dung solid-liquid separation device are consistent with the training samples.

[0048] The feeding area is provided with a first quality detection device for measuring the quality of cow dung entering, a first moisture content tester for measuring the moisture content of cow dung, and a first temperature sensor for measuring the temperature of cow dung; the drying area is provided with a second moisture content tester for measuring the moisture content of cow dung to be dried, a second temperature sensor for measuring the temperature of cow dung to be dried, and a third temperature sensor for measuring the drying temperature; the slag discharge area is provided with a second quality detection device for measuring the slag discharge amount, a fourth temperature sensor for measuring the slag discharge temperature, and a third moisture content tester for measuring the moisture content of the slag discharge; the drainage area is provided with a third quality detection device for measuring the drainage amount.

[0049] Example 2

[0050] A cow dung solid-liquid separation system based on convolutional neural network, comprising:

[0051] Data acquisition module 2, used to collect real-time parameters of the operation of the cow dung solid-liquid separation device;

[0052] The data storage module 3 is connected to the data acquisition module 2 and is used to store the data collected by the data acquisition module 2;

[0053] The data processing module 4 is connected to the data storage module 3; the data processing module 4 is a model based on a convolutional neural network, which is used to provide a parameter matching scheme according to the real-time parameters of the cow dung solid-liquid separation device;

[0054] The training module 5 is connected to the data processing module 4 and the data storage module 3, and is used to select the best matching parameters between the real-time parameters collected by the data collection module 2 through a genetic algorithm, and form a parameter matching scheme and upload it to the data processing module 4 for training;

[0055] The control module 7 is connected to the data processing module 4 and is used to download the trained data processing module 4 to achieve the acquisition, training and identification of real-time parameters of the cow dung solid-liquid separation device;

[0056] An input module 6, connected to the control module 7 and the data storage module 3, is used to input the real-time parameters of the cow dung solid-liquid separation device into the trained data processing module 4;

[0057] A display module 8, connected to the control module 7, for displaying the operating parameters given by the training-completed data processing module 4;

[0058] The cow dung solid-liquid separation device 1 is used for solid-liquid separation of cow dung.

[0059] In this embodiment, the real-time parameters of the operation of the cow dung solid-liquid separation device 1 include the following parameters: the mass, water content and temperature of the cow dung entering the feeding area; the extrusion time and motor speed in the extrusion zone; the water content and temperature, drying time and drying temperature of the cow dung entering the drying area; the slag discharge amount, slag discharge temperature and water content in the slag discharge area; and the discharge amount in the discharge area.

[0060] Among them, the cow dung solid-liquid separation device is provided with a feeding area 11, an extrusion area 12, a drying area 13, a slag discharge area 14 and a drainage area 15; the feeding area 11 is provided with a first quality detection device for measuring the quality of cow dung entering, a first moisture content tester for measuring the moisture content of cow dung and a first temperature sensor for measuring the temperature of cow dung; the drying area 13 is provided with a second moisture content tester for measuring the moisture content of cow dung to be dried, a second temperature sensor for measuring the temperature of cow dung to be dried and a third temperature sensor for measuring the drying temperature; the slag discharge area 14 is provided with a second quality detection device for measuring the slag discharge amount, a fourth temperature sensor for measuring the slag discharge temperature and a third moisture content tester for measuring the slag discharge moisture content; the drainage area 15 is provided with a third quality detection device for measuring the discharge amount.

[0061] The above description is only a preferred embodiment of the present invention and does not limit the technical scope of the present invention. Therefore, any slight modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A method for solid-liquid separation of cow dung based on convolutional neural network, characterized in that: include: S10, constructing a cow dung solid-liquid separation network model based on a convolutional neural network; the convolutional neural network includes an input layer, a convolutional layer, and a pooling layer, the number of convolutional layers is 16, the convolution kernel size is 3*3, and the pooling layer is a maximum pooling layer with 5 layers; S20, training the cow dung solid-liquid separation network model using training samples; S30, inputting the real-time parameters of the operation of the cow dung solid-liquid separation device into the trained cow dung solid-liquid separation network model, and giving a parameter matching scheme by the trained cow dung solid-liquid separation network model, and the parameter matching scheme is the basis for the operation of the cow dung solid-liquid separation device; The training process of the cow dung solid-liquid separation network model in step S20 is as follows: S21, obtaining parameters in the training sample, wherein the parameters in the training sample are the basis for generating the parameter matching scheme of the cow dung solid-liquid separation network model; S22, inputting each parameter in the training sample into the cow dung solid-liquid separation network model; S23, the cow dung solid-liquid separation network model selects the best matching parameters between various parameters through a genetic algorithm to form a parameter matching scheme; S24, the cow dung solid-liquid separation network model sends the parameter matching solution to the cow dung solid-liquid separation device; The cow dung solid-liquid separation device is provided with a feeding area, an extrusion area, a drying area, a slag discharge area and a liquid discharge area; the training sample includes the following parameters: the mass, water content and temperature of the cow dung entering the feeding area; the extrusion time and motor speed of the extrusion area; The moisture content and temperature of the cow dung entering the drying zone, the drying time, and the drying temperature; the slag discharge volume, slag discharge temperature, and moisture content of the slag discharge zone; the liquid discharge volume of the liquid discharge zone; and the real-time parameters of the cow dung solid-liquid separation device are consistent with the training samples; The feeding area is provided with a first quality detection device for measuring the quality of cow dung entering, a first moisture content tester for measuring the moisture content of cow dung, and a first temperature sensor for measuring the temperature of cow dung; the drying area is provided with a second moisture content tester for measuring the moisture content of cow dung to be dried, a second temperature sensor for measuring the temperature of cow dung to be dried, and a third temperature sensor for measuring the drying temperature; the slag discharge area is provided with a second quality detection device for measuring the slag discharge amount, a fourth temperature sensor for measuring the slag discharge temperature, and a third moisture content tester for measuring the moisture content of the slag discharge; the drainage area is provided with a third quality detection device for measuring the drainage amount.

2. The method for solid-liquid separation of cow dung based on convolutional neural network according to claim 1, characterized in that: The training samples include 80% of training samples and 20% of testing training samples.

3. A cow dung solid-liquid separation system based on convolutional neural network, realizing the cow dung solid-liquid separation method based on convolutional neural network as described in any one of claims 1-2, characterized in that: include: A data acquisition module is used to collect real-time parameters of the operation of the cow dung solid-liquid separation device; A data storage module, connected to the data acquisition module, and used to store the data collected by the data acquisition module; A data processing module connected to the data storage module; the data processing module is a model based on a convolutional neural network, and is used to provide a parameter matching scheme according to the real-time parameters of the cow dung solid-liquid separation device stored in the data storage module; A training module, connected to the data processing module and the data storage module, for selecting the best matching parameters among the real-time parameters collected by the data collection module through a genetic algorithm, forming a parameter matching scheme and uploading it to the data processing module for training; A control module, connected to the data processing module, is used to download the trained data processing module to achieve acquisition, training and identification of real-time parameters of the cow dung solid-liquid separation device; An input module, connected to the control module and the data storage module, for inputting the real-time parameters of the cow dung solid-liquid separation device into the trained data processing module; A display module, connected to the control module, for displaying the operating parameters given by the trained data processing module; Cow dung solid-liquid separation device is used for solid-liquid separation of cow dung.

4. The cow dung solid-liquid separation system based on convolutional neural network according to claim 3, characterized in that: The cow dung solid-liquid separation device is provided with a feeding area, an extrusion area, a drying area, a slag discharge area and a liquid discharge area; the feeding area is provided with a first quality detection device for measuring the quality of the cow dung entering, a first moisture content tester for measuring the moisture content of the cow dung, and a first temperature sensor for measuring the temperature of the cow dung; the drying area is provided with a second moisture content tester for measuring the moisture content of the cow dung to be dried, a second temperature sensor for measuring the temperature of the cow dung to be dried, and a third temperature sensor for measuring the drying temperature; the slag discharge area is provided with a second quality detection device for measuring the slag discharge amount, a fourth temperature sensor for measuring the slag discharge temperature, and a third moisture content tester for measuring the slag discharge moisture content; the liquid discharge area is provided with a third quality detection device for measuring the liquid discharge amount.

Images