A regulating valve control method and system for production automation

By obtaining and processing production equipment information in real time and generating control instructions using deep learning models, the problem of low control accuracy of existing regulating valves is solved, and automated nonlinear control is realized, which is suitable for a variety of production scenarios.

CN115289265BInactive Publication Date: 2024-11-22XI'AN PETROLEUM UNIVERSITY
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202210970114.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-12
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing control valve control method is affected by human experience and knowledge level, and has low control accuracy, which cannot meet the characteristics of fast opening, linear, and parabolic flow at the same time, making it difficult to achieve automated nonlinear control needs in food production and other fields.

Method used

By obtaining the image, temperature, liquid level and time information of the production equipment in real time, using convolutional neural networks and recurrent neural networks to extract features and generate control instructions, the automated nonlinear control of the regulating valve is realized, and the flow characteristics requirements of different production stages are met. .

Benefits of technology

It realizes automated nonlinear control of the regulating valve, which can meet the characteristics of fast opening, linear, and parabolic flow at the same time, improves the control accuracy and efficiency of the production process, and is suitable for various automated production scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115289265B_ABST
    Figure CN115289265B_ABST
Patent Text Reader

Abstract

The present invention discloses a control method for a regulating valve applied to production automation, belonging to the field of automatic control technology. The present invention obtains monitoring information in a target production device in real time, then extracts image features by extracting image information of the monitoring information, and performs feature fusion of temperature information, liquid level information and time information, and then obtains control information of the regulating valve through a pre-constructed second neural network model. Finally, the control information is converted into a corresponding control instruction of the valve according to a preset rule, and sent to an execution structure of the regulating valve, thereby realizing automatic nonlinear control of the regulating valve. Compared with the existing control method of the regulating valve, the present invention can simultaneously meet the characteristics of fast opening, linear, equal percentage and parabolic flow, thereby efficiently realizing the demand for automatic nonlinear control in different production links, is suitable for various automated production scenarios, and has certain economic and social benefits.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of automatic control technology, and in particular to a regulating valve control method and system applied to production automation. Background Art

[0002] The regulating valve, also known as the control valve, is a final control element that receives the control signal output by the regulating control unit and uses power operation to change the process parameters such as medium flow, pressure, temperature, liquid level, etc. It is generally composed of an actuator and a valve; it is divided into four types according to the flow characteristics: quick opening, linear, equal percentage and parabolic flow characteristics; the quick opening flow characteristic means that there is a large flow when the opening is small, and as the opening increases, the flow can quickly reach a maximum, and then the opening is increased, the flow changes very little; the linear flow characteristic means that the relative flow of the valve is in a linear relationship with the relative opening, that is, the flow change caused by the unit opening change is a constant; the equal percentage characteristic means that the relative flow change caused by the unit opening change is proportional to the relative flow at that point, that is, the amplification factor of the regulating valve is variable, and it increases with the increase of the relative flow; the parabolic characteristic means that the relative flow change caused by the change of the unit relative opening is proportional to the square root of the relative flow value at this point.

[0003] In the human production process, the development of control valves plays a very important role; nowadays, with the continuous development of technologies such as the Internet of Things, the production process has gradually moved towards production automation, so the control of control valves must also meet the needs of production automation. For example, in the food production process, the valve needs to be controlled with fast opening, linear, equal percentage and parabolic flow characteristics to meet the nonlinear control of food production in different links; however, the above control requirements are generally achieved through manual control and adjustment, which is greatly affected by human experience and knowledge level, has low control accuracy, and cannot meet the fast opening, linear, equal percentage and parabolic flow characteristics at the same time, and thus it is difficult to efficiently realize the automated nonlinear control requirements in different production links.

[0004] Therefore, it is urgent to propose a regulating valve control method and system for production automation to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to solve the defects in the prior art and to propose a regulating valve control method and system for production automation.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A regulating valve control method applied to production automation comprises the following steps:

[0008] Acquiring monitoring information of a target production device in real time based on a production monitoring device, wherein the monitoring information includes at least one of image information, temperature information, liquid level information and time information of the target production device;

[0009] Extracting image information from the target production equipment and preprocessing the image information, and extracting image features of the image information based on a pre-built first neural network model;

[0010] Combining the temperature information, liquid level information, time information and the image features to generate a real-time feature vector in the production process of the target production equipment;

[0011] Inputting the real-time feature vector of the target production equipment during the production process into a pre-built second neural network model for prediction to obtain control information of the regulating valve;

[0012] Convert the control information of the regulating valve into a corresponding control instruction of the valve according to a preset rule, and send it to the execution structure of the regulating valve;

[0013] During the production process, the control instructions corresponding to the valve are received in real time, and the valve of the regulating valve is controlled in real time nonlinearly according to the control instructions, so as to meet the valve control requirements of different production stages.

[0014] Furthermore, the production monitoring equipment includes an industrial CCD camera, a temperature sensor, a liquid level meter and a timer.

[0015] Furthermore, the preprocessing includes:

[0016] Extracting the production process image from the image information and performing binarization processing on it;

[0017] The production process image after binarization is subjected to median filtering using a filter window of size 3*3.

[0018] Furthermore, the first neural network model is specifically a convolutional neural network CNN; and the second neural network model is specifically a recurrent neural network RNN.

[0019] Furthermore, the corresponding control instructions are obtained by converting the control information according to preset rules, and the preset rules can be manually set according to different production scenarios; the corresponding control instructions include four corresponding instructions of quick opening, linear, equal percentage and parabolic flow characteristics.

[0020] Furthermore, the regulating valve is specifically a regulating valve having quick-opening, linear, equal percentage and parabolic flow characteristics at the same time, or a combined regulating device having quick-opening flow characteristics, linear flow characteristics, equal percentage flow characteristics and parabolic flow characteristics respectively.

[0021] A regulating valve control system applied to production automation, comprising:

[0022] The data acquisition module is used to acquire image information, temperature information, liquid level information and time information in the target production equipment according to the industrial CCD camera, temperature sensor, liquid level meter and timer;

[0023] The first deep learning module is used to extract image information from the target production equipment and preprocess it, and extract image features of the image information based on a pre-built first neural network model;

[0024] The feature fusion module is used to combine the temperature information, liquid level information, time information and the image features to generate a real-time feature vector in the production process of the target production equipment;

[0025] The second deep learning module is used to input the real-time feature vector in the production process of the target production equipment into a pre-built second neural network model for prediction, so as to obtain the control information of the regulating valve;

[0026] The control processing module is used to convert the control information of the regulating valve into a corresponding control instruction of the valve according to a preset rule, and send it to the execution structure of the regulating valve;

[0027] The execution control module is used to receive the valve corresponding control instructions in real time during the production process, and perform real-time nonlinear control on the valve of the regulating valve according to the control instructions to meet the valve control requirements of different production stages.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] The present application discloses a control method and system for regulating valves applied to production automation, which obtain monitoring information in target production equipment in real time, then extract image features by extracting image information of the monitoring information, and perform feature fusion of temperature information, liquid level information and time information, and then obtain control information of the regulating valve through a pre-constructed second neural network model, and finally convert it into corresponding control instructions of the valve according to preset rules, and send it to the execution structure of the regulating valve, thereby realizing automatic nonlinear control of the regulating valve; compared with the existing control method of regulating valves, the present invention can simultaneously meet the characteristics of fast opening, linear, equal percentage and parabolic flow, thereby efficiently realizing the demand for automatic nonlinear control in different production links, and is suitable for various automated production scenarios, with certain economic and social benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] 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.

[0031] Figure 1 This is an overall flow chart of a regulating valve control method applied to production automation proposed by the present invention;

[0032] Figure 2 This is a schematic diagram of the overall structure of a regulating valve control system applied to production automation proposed by the present invention. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0034] In the description of the present invention, it is necessary to understand that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0035] In one embodiment, referring to Figure 1 , provides a regulating valve control method applied to production automation, comprising the following steps:

[0036] S101: Acquiring monitoring information of a target production device in real time based on a production monitoring device, wherein the monitoring information includes at least one of image information, temperature information, liquid level information and time information of the target production device;

[0037] Specifically, the production monitoring equipment includes an industrial CCD camera, a temperature sensor, a liquid level meter and a timer.

[0038] It should be noted here that the production monitoring equipment is not limited to the equipment listed above. Equipment can be added or deleted according to different production conditions. This embodiment does not impose too many limitations on this. Any equipment that can directly monitor the production process can be used as the production monitoring equipment in this application.

[0039] S102: extracting image information from the target production equipment and preprocessing the image information, and extracting image features of the image information based on a pre-built first neural network model;

[0040] Specifically, the preprocessing includes:

[0041] Extracting the production process image from the image information, and performing binarization processing on it to convert it into a production process grayscale image;

[0042] The grayscale image of the production process obtained after the binarization process is subjected to median filtering using a filter window of size 3*3 to perform denoising;

[0043] Specifically, the first neural network model is a convolutional neural network CNN.

[0044] S103: combining the temperature information, liquid level information, time information and the image feature to generate a real-time feature vector of the target production equipment during the production process;

[0045] S104: inputting the real-time feature vector of the target production equipment during the production process into a pre-built second neural network model for prediction to obtain control information of the regulating valve;

[0046] Specifically, the second neural network model is a recurrent neural network RNN;

[0047] It should be noted that the control information of the regulating valve specifically refers to the valve opening data obtained after training based on the image information, temperature information, liquid level information and time information;

[0048] An example is given for the control information: assuming that the present invention is applied to a food manufacturing process, in the initial link, the raw materials for preparing a certain food are first input into the boiler through the regulating valve and the pipeline. It is assumed here that the food is a fruit juice drink. Therefore, in the initial stage of production, the image of the production process collected by the CCD camera is empty, and the value of the timer is about the initial time period. Therefore, the valve opening data of the regulating valve at this time is obtained after a deep learning, feature fusion and a second deep learning. At this time, the valve opening data should meet the fast-opening flow characteristics, so the control information at this time is converted into an instruction of the fast-opening flow characteristics according to the preset rules to control the valve opening, thereby realizing the control requirements in the initial stage of production;

[0049] As mentioned above, assuming that the production stage of the juice drink reaches the mixing stage, water needs to be evenly added to the boiler during the mixing stage. The same operation is performed as above. At this time, the control information will be converted into a control instruction of a linear flow characteristic according to a preset rule. At this time, the valve opening is in a linear relationship, thereby realizing linear control of the regulating valve;

[0050] As mentioned above, the control of the equal percentage and parabolic flow characteristics of the regulating valve is also the same, and the nonlinear control of the regulating valve is realized according to different requirements in different production links;

[0051] It should be noted here that the regulating valve is arranged at the processing end of direct production and preparation, and it can be connected to a certain pipeline for conveying a uniform substance, such as water; it can also be connected to multiple pipelines for conveying different substances and realize nonlinear control by monitoring the conditions in the boiler.

[0052] S105: converting the control information of the regulating valve into a corresponding control instruction of the valve according to a preset rule, and sending the instruction to the execution structure of the regulating valve;

[0053] Specifically, the corresponding control instructions are obtained by converting the control information according to preset rules, and the preset rules can be manually set according to different production scenarios; the corresponding control instructions include four corresponding instructions of fast opening, linear, equal percentage and parabolic flow characteristics;

[0054] It should be noted here that the preset rule specifically refers to the conversion relationship between the control information and the corresponding control instruction, which can be the system default conversion rule, or it can be a later self-modification setting made by the user according to the needs of his own production scenario.

[0055] S106: During the production process, receiving the control instruction corresponding to the valve in real time, and performing real-time nonlinear control on the valve of the regulating valve according to the control instruction, so as to meet the valve control requirements of different production stages;

[0056] Specifically, the regulating valve is a regulating valve having quick-opening, linear, equal percentage and parabolic flow characteristics at the same time, or a combined regulating device having quick-opening flow characteristics, linear flow characteristics, equal percentage flow characteristics and parabolic flow characteristics respectively.

[0057] Based on the control method of the regulating valve applied to production automation proposed in the first embodiment, Figure 2 This embodiment provides a regulating valve control system applied to production automation, including:

[0058] The data acquisition module 210 is used to acquire image information, temperature information, liquid level information and time information in the target production equipment according to the industrial CCD camera, temperature sensor, liquid level meter and timer;

[0059] The first deep learning module 220 is used to extract image information from the target production equipment and preprocess it, and extract image features of the image information based on a pre-built first neural network model;

[0060] The feature fusion module 230 is used to combine the temperature information, liquid level information, time information and the image features to generate a real-time feature vector in the production process of the target production equipment;

[0061] The second deep learning module 240 is used to input the real-time feature vector of the target production equipment during the production process into a pre-built second neural network model for prediction, so as to obtain the control information of the regulating valve;

[0062] The control processing module 250 is used to convert the control information of the regulating valve into a valve corresponding control instruction according to a preset rule, and send it to the execution structure of the regulating valve;

[0063] The execution control module 260 is used to receive the valve corresponding control instructions in real time during the production process, and perform real-time nonlinear control on the valve of the regulating valve according to the control instructions to meet the valve control requirements of different production stages.

[0064] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A regulating valve control method applied to production automation, characterized in that: The steps include: Acquiring monitoring information of a target production device in real time based on a production monitoring device, wherein the monitoring information includes image information, temperature information, liquid level information and time information of the target production device; Extracting image information from the target production equipment and preprocessing the image information, and extracting image features of the image information based on a pre-built first neural network model; Combining the temperature information, liquid level information, time information and the image features to generate a real-time feature vector in the production process of the target production equipment; Inputting the real-time feature vector of the target production equipment during the production process into a pre-built second neural network model for prediction to obtain control information of the regulating valve; Convert the control information of the regulating valve into a corresponding control instruction of the valve according to a preset rule, and send it to the execution structure of the regulating valve; During the production process, the control instructions corresponding to the valve are received in real time, and the valve of the regulating valve is controlled in real time nonlinearly according to the control instructions, so as to meet the valve control requirements of different production stages.

2. The regulating valve control method applied to production automation according to claim 1, characterized in that: The production monitoring equipment includes an industrial CCD camera, a temperature sensor, a liquid level meter and a timer.

3. The regulating valve control method applied to production automation according to claim 1, characterized in that: The preprocessing comprises: Extracting the production process image from the image information and performing binarization processing on it; The production process image after binarization is subjected to median filtering using a filter window of size 3*3.

4. The regulating valve control method for production automation according to claim 1, characterized in that: The first neural network model is specifically a convolutional neural network CNN; the second neural network model is specifically a recurrent neural network RNN.

5. The regulating valve control method applied to production automation according to claim 1, characterized in that: The corresponding control instructions are obtained by converting the control information according to preset rules, and the preset rules can be manually set according to different production scenarios; the corresponding control instructions include four corresponding instructions of quick opening, linear, equal percentage and parabolic flow characteristics.

6. The regulating valve control method applied to production automation according to claim 1, characterized in that: The regulating valve is specifically a regulating valve having quick-opening, linear, equal percentage and parabolic flow characteristics at the same time, or a combined regulating device having quick-opening flow characteristics, linear flow characteristics, equal percentage flow characteristics and parabolic flow characteristics respectively.

7. A regulating valve control system used in production automation, characterized in that: include: The data acquisition module is used to acquire image information, temperature information, liquid level information and time information in the target production equipment according to the industrial CCD camera, temperature sensor, liquid level meter and timer; The first deep learning module is used to extract image information from the target production equipment and preprocess it, and extract image features of the image information based on a pre-built first neural network model; The feature fusion module is used to combine the temperature information, liquid level information, time information and the image features to generate a real-time feature vector in the production process of the target production equipment; The second deep learning module is used to input the real-time feature vector in the production process of the target production equipment into a pre-built second neural network model for prediction, so as to obtain the control information of the regulating valve; The control processing module is used to convert the control information of the regulating valve into a corresponding control instruction of the valve according to a preset rule, and send it to the execution structure of the regulating valve; The execution control module is used to receive the valve corresponding control instructions in real time during the production process, and perform real-time nonlinear control on the valve of the regulating valve according to the control instructions to meet the valve control requirements of different production stages.

Citation Information

Patent Citations

  • Multi-point sensing and intelligent control method for temperature of gas heating stove

    CN101464096A

  • Method for controlling automatic chemical dosing of high-turbidity water system

    CN103011356A

  • Heating furnace fuel valve pulse optimal control device and method

    CN109765943A

  • Real-time prediction self-adjusting temperature control method based on BP neural network, computer readable storage medium and air conditioner

    CN110686350A

  • Intelligent gas well valve adjusting system

    CN112943181A