CN116688967A

CN116688967A - Multi-degree-of-freedom activated carbon regenerator based on DBD plasma and application method

Info

Publication number: CN116688967A
Application number: CN202310913847.8A
Authority: CN
Inventors: 王安康; 陈华良; 王晓宇; 刘瑞龙; 李勋臣
Original assignee: Qingxu Hongbo Sewage Treatment Co ltd
Current assignee: Qingxu Hongbo Sewage Treatment Co ltd
Priority date: 2023-07-25
Filing date: 2023-07-25
Publication date: 2023-09-05
Anticipated expiration: 2043-07-25
Also published as: CN116688967B

Abstract

The invention discloses a multi-degree-of-freedom activated carbon regenerator based on DBD plasma and a use method thereof, wherein the outside of an adjusting mechanism comprises at least ten first linear degrees of freedom which are arranged along a coaxial line annular array but have staggered travel directions and axes; each first linear degree of freedom is adaptively connected with a DBD array generating assembly for generating BDB plasmas; 1. high-efficiency regeneration: while the traditional activated carbon regeneration technology has low efficiency, the multi-degree-of-freedom activated carbon regeneration technology based on DBD plasma can remove pollutants adsorbed on activated carbon more efficiently, and the regeneration efficiency of the activated carbon is improved. 2. And (3) multi-degree-of-freedom adjustment: the traditional technology can only realize single parameter adjustment generally, while the multi-degree-of-freedom activated carbon regeneration technology based on DBD plasma can realize multi-degree-of-freedom adjustment, such as reducing adjustment, ionization generation of a gas jet pump of DBD plasma gas, visual detection of a CCD industrial visual camera and the like, and finer control is realized while the regeneration effect is improved.

Description

Multi-degree-of-freedom activated carbon regenerator based on DBD plasma and application method

Technical Field

The invention relates to the technical field of activated carbon regeneration, in particular to a multi-degree-of-freedom activated carbon regenerator based on DBD plasma and a use method thereof.

Background

Activated carbon is a material widely used for air and water purification, and has a large number of micropores and pores, and can adsorb and remove various organic and inorganic pollutants. However, due to its limited adsorption capacity, frequent regeneration is required in order to ensure a stable purification effect.

The activated carbon regeneration technology has become an important research field, and the main purpose of the technology is to recover the adsorption capacity and efficiency of the activated carbon by removing the pollutants adsorbed on the activated carbon, thereby realizing low-cost and high-efficiency regeneration and reuse.

The background technology of activated carbon regeneration mainly comprises the technologies of physical regeneration, chemical regeneration, biological regeneration and the like.

(1) Physical regeneration refers to removing adsorbates on the activated carbon by a physical method, and comprises high-temperature steam regeneration, nitrogen purging regeneration, inert gas purging regeneration and the like. Among them, high temperature steam regeneration is the most common method, and has an advantage of being able to rapidly and effectively remove contaminants adsorbed on activated carbon.

(2) Chemical regeneration refers to the removal of adsorbates on activated carbon by chemical means, including oxidizer regeneration, acid wash regeneration, alkaline wash regeneration, and the like. Among them, the regeneration of an oxidizing agent is a common method for restoring the adsorption performance of activated carbon by oxidatively decomposing contaminants adsorbed on the activated carbon using an oxidizing agent such as hydrogen peroxide, ozone, etc.

(3) Biological regeneration refers to the removal of adsorbates on activated carbon by biological methods, including microbial degradation regeneration, biological adsorption regeneration, and the like. Among them, microbial degradation regeneration is an emerging regeneration technology that restores adsorption performance of contaminants adsorbed on activated carbon by degrading them using microorganisms such as bacteria, fungi, etc. However, this method requires a long time and a complicated operation, and is affected by biological factors and environmental factors.

In summary, the activated carbon regeneration technology is an important research field, and various regeneration methods have advantages and disadvantages and need to be selected and optimized according to specific situations. In the future, with the deep research and the development of technology, the activated carbon regeneration technology is expected to realize more efficient and low-cost regeneration and reuse, and makes a greater contribution to environmental protection and sustainable development.

However, the inventor has long been working and studied to find that the above-mentioned traditional physical, chemical and biological regeneration technologies have the following technical disadvantages in general:

(1) The reaction speed is slow: the traditional technology needs longer reaction time to realize the regeneration effect and has lower efficiency;

(2) the regeneration effect is unstable, namely the traditional technology can be influenced by environmental factors such as temperature, humidity and the like, so that the regeneration effect is unstable;

(3) The active carbon material is limited: the traditional technology can realize the regeneration effect only by using specific activated carbon materials, and is not flexible and has wide applicability;

(4) The manual operation is needed: the conventional technology requires manual operations to adjust reaction parameters and monitor regeneration effects, and may have errors and inconveniences.

Therefore, a multi-degree-of-freedom activated carbon regenerator based on DBD plasma and a use method thereof are provided.

Disclosure of Invention

In view of the foregoing, embodiments of the present invention wish to provide a multi-degree-of-freedom activated carbon regenerator based on DBD plasma and a method of use thereof, so as to solve or alleviate the technical problems existing in the prior art, and at least provide a beneficial choice;

the technical scheme of the embodiment of the invention is realized as follows:

First aspect

A multi-degree-of-freedom activated carbon regenerator based on DBD plasma comprises an adjusting mechanism;

the bottom of the adjusting mechanism is provided with a conventional workbench which is used for supporting the whole device as a whole product and is arranged in a relevant working environment, and a control console for controlling all the following electrical elements is arranged beside the adjusting mechanism;

the outer part of the adjusting mechanism comprises at least ten first linear degrees of freedom which are arranged along a coaxial line annular array and have the travel direction staggered with the axis; each first linear degree of freedom is adaptively connected with a DBD array generating assembly for generating BDB plasmas;

at least four storage components for clamping the active carbon which needs to be regenerated are arranged in the adjusting mechanism in an annular array mode by taking the axis as a reference;

the storage component comprises a second linear degree of freedom along the axis direction, wherein the second linear degree of freedom is used for adjusting the lifting of the activated carbon.

When the device is used, clamping and lifting adjustment are carried out on the device by the object placing component according to the size and the specification of different active carbon, and then the adjusting mechanism synchronously adjusts all DBD array generating components to uniformly output DBD plasmas in an annular mode to carry out omnibearing regeneration on the active carbon.

In the above embodiment, the following embodiments are described: the multi-degree-of-freedom activated carbon regenerator based on DBD plasma mainly comprises a workbench, a control console, an adjusting mechanism, a storage component, a DBD array generating component and the like. The adjusting mechanism comprises an outer ten first linear degrees of freedom arranged along a coaxial line annular array and at least four inner storage components. According to the size and the specification of different activated carbons, clamping and lifting of the activated carbon are realized through lifting adjustment of the storage component, and simultaneously, the adjusting mechanism synchronously adjusts all DBD array generating components to uniformly output DBD plasmas in an annular form to the activated carbon for omnibearing regeneration.

Wherein in one embodiment: the adjusting mechanism comprises a first cylinder body with a central axis as an axis and a second cylinder body which is in sliding fit with the outer part of the first cylinder body;

the second cylinder is driven to lift by the first linear piece;

ten groups of vector components for outputting the first linear degree of freedom are uniformly arranged outside the second cylinder body in the form of an annular array;

the vector component is adapted to be connected with the DBD array generating component.

The first linear member is preferably a first servo cylinder, and the cylinder body and the piston rod of the first servo cylinder are respectively connected with the first cylinder body and the second cylinder body. When the linear motion vector device is used, the first servo electric cylinder drives the second cylinder to slide on the first cylinder to drive all vector components to output the first linear degree of freedom.

In the above embodiment, the following embodiments are described: in this embodiment, the adjustment mechanism includes a first barrel having a central axis as an axis and a second barrel slidably fitted to the outside of the first barrel. The second cylinder is driven to lift by the first linear piece, ten groups of vector components for outputting the first linear degree of freedom are uniformly arranged outside the second cylinder in the form of an annular array, and the vector components are connected with the DBD array generating component in an adaptive mode. The first linear member is preferably a first servo cylinder, and the cylinder body and the piston rod of the first servo cylinder are respectively connected with the first cylinder body and the second cylinder body.

Wherein in one embodiment: the vector assembly comprises a first plate body and an articulated arm, wherein the bottom of the first plate body is hinged to the top of the first cylinder body and the top of the second cylinder body respectively;

the top of the first plate body is hinged with a second plate body;

two ends of the hinge rod are respectively hinged to the outer surface of the second plate body and the top of the hinge arm;

the second board body is provided with a DBD array generating assembly on the surface facing the axis.

In the above embodiment, the following embodiments are described: in this embodiment, the vector assembly includes a first plate body and a hinge arm, the bottoms of which are hinged to the tops of the first cylinder body and the second cylinder body respectively, the top of the first plate body is hinged to the second plate body, two ends of the hinge rod are hinged to the outer surface of the second plate body and the top of the hinge arm respectively, and the second plate body faces to one surface of the axis and is provided with the DBD array generating assembly.

Wherein in one embodiment: and the top of the second plate body is provided with an air jet pump for jetting DBD plasma gas generated by ionization of the DBD array generating assembly and a CCD industrial vision camera for visually detecting external visual characteristics of the activated carbon.

In the above embodiment, the following embodiments are described: in the embodiment, the CCD industrial vision camera is integrated into the multi-degree-of-freedom activated carbon regenerator, so that the real-time monitoring of the reaction process of the activated carbon and the plasma can be realized. The position of the CCD industrial vision camera is arranged at the top of the second plate body, so that the detection of the external vision characteristics of the activated carbon can be realized, and the condition of plasma reaction can be monitored.

Wherein in one embodiment: the DBD array generating assembly comprises at least two electrodes which are arranged in parallel and a dielectric substance which is matched with the electrodes, and a plane insulating layer is formed on the surface of the dielectric substance;

the electrodes have a space therebetween. So as to form a suitable electric field distribution such that a plasma can be generated on both sides of the insulating layer.

The electrode is electrically connected with a discharge protector for discharge protection.

In the above embodiment, the following embodiments are described: the DBD array generating assembly is composed of at least two electrodes and a dielectric substance, wherein the electrodes are arranged in parallel, a space is reserved between the electrodes to form proper electric field distribution, and a plane insulating layer is formed on the surface of the dielectric substance. Meanwhile, the electrode is electrically connected with the discharge protector to protect the electrode and other components from being damaged by excessive discharge voltage and current.

Wherein in one embodiment: the bottom of the first cylinder body is provided with a storage component; the object placing component comprises a second linear piece for outputting a second linear degree of freedom, the second linear piece is arranged at the bottom of the first cylinder, and the second linear piece is connected with and lifted by a rotary executing piece; the rotary executing piece is connected with an electric chuck for clamping the activated carbon. The second linear member is preferably a second servo cylinder, and the rotary actuator is preferably a servo motor; the cylinder body and the piston rod of the second servo electric cylinder are respectively and fixedly connected to the bottom of the first cylinder body and the outer shell of the servo motor, and the output shaft of the servo motor is fixedly connected with the outer shell of the electric chuck.

In the above embodiment, the following embodiments are described: the bottom of the first barrel is provided with a storage component, wherein the storage component comprises a second linear piece for outputting a second linear degree of freedom, a rotary executing piece and an electric chuck for clamping activated carbon.

Second aspect

The multi-degree-of-freedom activated carbon regeneration using method based on the DBD plasma adopts the multi-degree-of-freedom activated carbon regenerator based on the DBD plasma and comprises the following using steps:

s1, preparing a data set: image data of the activated carbon and plasma reactions are collected while each image is assigned a corresponding label, such as the state of the activated carbon, the plasma reaction conditions, etc.

In the above embodiment, the following embodiments are described: the collection and labeling of the data sets is the basis for constructing the monitoring model, and the data needs to be collected from multiple dimensions, such as the state of the adsorbed substances on the surface of the activated carbon, the reaction area of the plasma, the reaction time and other parameters, and corresponding labels are allocated to each image.

S2, data preprocessing: and carrying out normalization and enhancement processing on the image data. To improve the learning effect of the model.

In the above embodiment, the following embodiments are described: for the collected data set, the accuracy of the monitoring model can be improved by preprocessing the data. For example, image enhancement, noise reduction, gradation processing, and the like are performed, and normalization processing of the image is performed at the same time, so that the variability between data is eliminated.

S3, constructing a convolutional neural network: a convolutional neural network is designed.

In the above embodiment, the following embodiments are described: the convolutional neural network is the core of an image monitoring model, and can extract effective characteristics from a large amount of data by designing a proper network structure so as to accurately monitor the reaction conditions of the activated carbon and the plasmas. The design of the convolutional neural network needs to consider parameters such as network depth, convolutional kernel size, pooling layer and the like so as to obtain a better monitoring effect.

S4, training a model: and training a CNN model by using the prepared data set and the corresponding label, and adjusting the optimizer, the loss function and the evaluation index parameters.

In the above embodiment, the following embodiments are described: the convolutional neural network needs to be trained by using the prepared data set and the corresponding labels, and the aim of the training model is to obtain a model with high accuracy, so that the accuracy and the robustness of the model can be improved by adjusting the optimizer, the loss function and the evaluation index parameters.

S5, monitoring in real time: and deploying the trained CNN model into a CCD industrial vision camera, and monitoring the state of the activated carbon and the plasma reaction condition in real time.

In the above embodiment, the following embodiments are described: and deploying the trained monitoring model into a CCD industrial vision camera, and monitoring the reaction conditions of the activated carbon and the plasmas in real time. The monitoring result is passed as a parameter to the next operation.

S6, automatically adjusting parameters: and according to the monitoring result, automatically adjusting parameters of the activated carbon regenerator, such as the working voltage and the working frequency of the plasma generator.

In the above embodiment, the following embodiments are described: and automatically adjusting parameters of the active carbon regenerator, such as a DBD array generating component, according to the monitoring result.

S7, regenerating plasma: and controlling the DBD array generating assembly 5 to output DBD plasmas on the surface of the activated carbon according to the adjusted parameters, so as to realize the regeneration of the activated carbon. In the process, the dynamic adjustment of the plasmas and the activated carbon is realized through the reducing adjustment of the vector component 304 of the multi-degree-of-freedom activated carbon regenerator, so that the regeneration effect is further optimized.

S8, automatically recording data: image data of the activated carbon and plasma reactions were recorded using a CCD industrial vision camera 3046 and the relevant data was automatically saved for subsequent analysis and study.

S9, stopping operation: after the regeneration of the activated carbon is completed, the operation is stopped, and the activated carbon regenerator and related equipment are shut down.

Wherein in one embodiment: the convolutional neural network includes:

input layer: raw image data is received. Let the size of the input image be width height channels, where width and height represent the width and height of the image, respectively, and channels represent the number of channels of the image, e.g. the number of channels of an RGB image is 3.

Convolution layer: and extracting image features. The method comprises the steps of providing k convolution kernels, wherein the size of each convolution kernel is f times channels, and f represents the width and the height of the convolution kernel; the convolution operation is as follows:

c: an output of the convolutional layer;

i: inputting an image;

k: a convolution kernel;

i and j are each: the width and height indexes of the convolution layer output;

m and n are each: the width and height indexes of the convolution kernel;

c: channel index.

An activation layer: for introducing nonlinearities, including the ReLU function:

a: activating the output of the function;

x: the output of the convolutional layer.

Pooling layer: for reducing the current size of the feature map:

p: outputting a pooling layer;

a: activating the output of the layer;

p: step size of pooling operation.

Full tie layer: flattening the output of the pooling layer and performing linear transformation through a weight matrix:

f: outputting a full connection layer;

w: a weight matrix;

b: a bias vector;

flat (P): the pooling layer output P is flattened into a one-dimensional vector.

Output layer: converting the output of the fully connected layer into a probability distribution for classification, including a softmax function:

s: output or probability distribution of the output layer;

f: outputting a full connection layer;

i and j: category index.

Using a cross entropy loss function:

L: a loss value;

y: the one-hot encoding vector of the actual tag;

s: outputting of the output layer;

calculating the gradient of the loss function L with respect to model parameters such as convolution kernels and full-connected layer weight matrices by back propagation, and then updating the model parameters using gradient descent to reduce the loss value;

compared with the prior art, the invention has the beneficial effects that:

1. high-efficiency regeneration: while the traditional activated carbon regeneration technology has low efficiency, the multi-degree-of-freedom activated carbon regeneration technology based on DBD plasma can remove pollutants adsorbed on activated carbon more efficiently, and the regeneration efficiency of the activated carbon is improved.

2. And (3) multi-degree-of-freedom adjustment: the traditional technology can only realize the adjustment of single parameter, while the multi-degree-of-freedom active carbon regeneration technology based on DBD plasma can realize the adjustment of multi-degree of freedom, such as reducing adjustment, an air jet pump for generating DBD plasma gas by ionization, visual detection of CCD industrial visual camera and the like,

the regeneration effect is improved and finer control is realized.

3. The application range is wide: the multi-degree-of-freedom activated carbon regeneration technology based on DBD plasma has wide application range, can treat activated carbon with various sizes, and is suitable for various fields, such as environmental protection, chemical industry and the like.

4. The degree of automation is high: the multi-degree-of-freedom activated carbon regeneration technology based on DBD plasma can realize high-efficiency regeneration, and simultaneously can realize high-automation-degree operation, such as real-time monitoring of the state of the activated carbon, automatic parameter adjustment and the like, so that the complexity of manual operation is reduced, and the working efficiency and safety are improved.

5. Environmental protection and safety: the multi-degree-of-freedom activated carbon regeneration technology based on DBD plasma does not need to use chemical agents and the like, reduces the pollution to the environment, simultaneously does not harm human bodies due to plasma gas used in the operation process, and ensures the safety of the operation process.

Drawings

In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are necessary for the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a perspective view of the present application;

FIG. 2 is a perspective view of another embodiment of the present application;

FIG. 3 is a schematic perspective view of an adjusting mechanism according to the present invention;

FIG. 4 is a schematic top perspective view of the adjustment mechanism and storage assembly of the present invention;

FIG. 5 is a schematic perspective view of a single-unit storage assembly according to the present invention;

FIG. 6 is a schematic perspective view of a vector assembly of the present invention;

fig. 7 is a schematic perspective view of a vector component and a DBD array generating component according to the present invention;

FIG. 8 is an enlarged perspective view of the area A of FIG. 7 according to the present invention;

reference numerals: 1. a work table; 2. a console; 3. an adjusting mechanism; 301. a first cylinder; 302. a second cylinder; 303. a first linear member; 304. a vector component; 3041. an arm hinge; 3042. a first plate body; 3043. a second plate body; 3044. a hinge rod; 3045. an air jet pump; 3046. CCD industrial vision camera; 4. a storage component; 401. an electric chuck; 402. rotating the actuator; 403. a second linear member; 5. a DBD array generation component; 501. an electrode; 502. a discharge protector.

Description of the embodiments

In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit of the invention, whereby the invention is not limited to the specific embodiments disclosed below;

It should be noted that the terms "first," "second," "symmetric," "array," and the like are used merely for distinguishing between description and location descriptions, and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of features indicated. Thus, a feature defining "first," "symmetry," or the like, may explicitly or implicitly include one or more such feature; also, where certain features are not limited in number by words such as "two," "three," etc., it should be noted that the feature likewise pertains to the explicit or implicit inclusion of one or more feature quantities;

in the present invention, unless expressly stated or limited otherwise, a first feature "up" or "down" a second feature may be the first and second features in direct contact, or the first and second features in indirect contact via an intervening medium. Moreover, a first feature being "above," "over" and "on" a second feature may be a first feature being directly above or obliquely above the second feature, or simply indicating that the first feature is level higher than the second feature. The first feature being "under", "below" and "beneath" the second feature may be the first feature being directly under or obliquely below the second feature, or simply indicating that the first feature is less level than the second feature; meanwhile, all axial descriptions such as X-axis, Y-axis, Z-axis, one end of X-axis, the other end of Y-axis, or the other end of Z-axis are based on a cartesian coordinate system.

In the present invention, unless explicitly specified and limited otherwise, terms such as "mounted," "connected," "secured," and the like are to be construed broadly; for example, the connection can be fixed connection, detachable connection or integrated molding; the connection may be mechanical, direct, welded, indirect via an intermediate medium, internal communication between two elements, or interaction between two elements. The specific meaning of the terms described above in the present invention will be understood by those skilled in the art from the specification and drawings in combination with specific cases.

Examples

In the prior art, the traditional physical, chemical and biological regeneration methods have the defects of low reaction speed, unstable regeneration effect, limitation on active carbon materials and limitation of needing manual operation; for this reason, referring to fig. 1-8, the present invention provides a technical solution to solve the above technical problems:

a multi-degree-of-freedom activated carbon regenerator based on DBD plasma and a use method thereof; in this embodiment, the mechanical mechanism and its principle will be described first: firstly, an adjusting mechanism 3 is included;

the bottom of the adjusting mechanism 3 is provided with a conventional workbench 1 which is used for supporting the whole device as a whole product and is arranged in a relevant working environment, and a control console 2 which controls all the following electrical elements is arranged beside the adjusting mechanism;

The outer part of the adjusting mechanism 3 comprises ten first linear degrees of freedom arranged along a coaxial annular array but with the stroke direction staggered with respect to the axis; each first linear degree of freedom is adapted to have connected thereto a DBD array generating assembly 5 for generating BDB plasma;

at least four storage components 4 for clamping the active carbon which needs to be regenerated are arranged in the adjusting mechanism 3 in an annular array mode by taking the axis as a reference;

the storage component 4 comprises a second linear degree of freedom along the axial direction for adjusting the lifting of the activated carbon.

When the device is used, clamping and lifting adjustment is carried out on the device by the storage component 4 according to the size and the specification of different active carbon, and then the adjustment mechanism 3 synchronously adjusts all the DBD array generating components 5 to uniformly output DBD plasmas in an annular form to the active carbon for omnibearing regeneration.

In the scheme, the method comprises the following steps: the multi-degree-of-freedom activated carbon regenerator based on DBD plasma mainly comprises a workbench 1, a console 2, an adjusting mechanism 3, a storage component 4, a DBD array generating component 5 and the like. Wherein the adjusting mechanism 3 comprises an outer ten first linear degrees of freedom arranged along a coaxial annular array and an inner at least four storage assemblies 4. According to the size and the specification of different activated carbons, clamping and lifting of the activated carbon are realized through lifting adjustment of the storage component 4, and simultaneously, the adjusting mechanism 3 synchronously adjusts all the DBD array generating components 5 to uniformly output DBD plasmas in an annular mode to perform omnibearing regeneration on the activated carbons.

Specific: the regenerator regenerates the activated carbon by the plasma generated by the DBD array generating assembly 5 using a DBD plasma based technique. Specifically, the DBD array generating module 5 ionizes air molecules by a high voltage electric field to form plasma, and the generated plasma has a certain energy, so that an adsorption layer on the surface of the activated carbon can be destroyed, and pollutants adsorbed on the activated carbon are decomposed and removed, thereby realizing the regeneration of the activated carbon.

It will be appreciated that in this embodiment, the regenerator has a multiple degree of freedom design, which can accommodate the regeneration requirements of activated carbon of different sizes and specifications. Meanwhile, the DBD plasma technology is adopted for regeneration, so that the method has the advantages of high efficiency, rapidness, low cost and the like, can remove various organic and inorganic pollutants, and can be reused. The regenerator can also control all electrical elements through the control console 2, has the advantages of simplicity and convenience in operation, high intelligent degree and the like, and improves the operation efficiency and accuracy.

It should be noted that, in the present embodiment, the main idea of the present device is to regenerate activated carbon by using a DBD plasma technology, which is a technology for generating gas discharge plasma by corona discharge, and has the characteristics of high efficiency, low energy consumption, easy control, and the like. The DBD plasma technology is applied to the regeneration of the activated carbon, and the following beneficial effects can be brought:

(1) High-efficiency activated carbon regeneration: the DBD plasma technology can quickly oxidize and decompose the adsorption substances in the activated carbon by generating high-energy active substances such as electrons, positive ions, free radicals and the like, thereby realizing efficient activated carbon regeneration.

(2) No secondary pollution: conventional activated carbon regeneration methods, such as pyrolysis and steam regeneration, may produce secondary pollutants, such as NOx and CO 2. However, when the active carbon is regenerated by using the DBD plasma technology, the problem is not existed, because the reaction products of the DBD plasma are harmless substances such as CO2, H2O and the like.

(3) The controllability is strong: the discharge parameters of the DBD plasma technology can be controlled by adjusting parameters such as voltage, current, frequency and the like, so that the control and optimization of the activated carbon regeneration process are realized, and the activated carbon regeneration efficiency is further improved.

(4) Novelty of: the traditional active carbon regeneration method is mostly based on the physical or chemical reaction principle, while the DBD plasma technology is a technology based on the combination of physical and chemical reactions, and has a unique regeneration principle and a unique technology route.

In some embodiments of the present application, please refer to fig. 1-8 in combination: the adjusting mechanism 3 comprises a first cylinder 301 with a central axis as an axis and a second cylinder 302 which is in sliding fit with the outer part of the first cylinder 301;

The second cylinder 302 is driven to lift by the first linear member 303;

the outside of the second cylinder 302 is uniformly provided with ten sets of vector components 304 for outputting the first linear degree of freedom in the form of an annular array;

the vector component 304 is adapted to be connected with the DBD array generating component 5.

The first linear member 303 is preferably a first servo cylinder, and a cylinder body and a piston rod of the first servo cylinder are connected to the first cylinder 301 and the second cylinder 302, respectively. When in use, the first servo cylinder drives the second cylinder 302 to slide on the first cylinder 301, and drives all the vector components 304 to output the first linear degree of freedom.

In the scheme, the method comprises the following steps: in this embodiment, the adjustment mechanism 3 includes a first cylinder 301 having a central axis as an axis and a second cylinder 302 slidably fitted to the outside of the first cylinder 301. The second cylinder 302 is driven to rise and fall by the first linear member 303, and ten sets of vector components 304 for outputting the first linear degree of freedom are uniformly mounted outside the second cylinder in the form of an annular array, and the vector components 304 are adaptively connected with the DBD array generating component 5. The first linear member 303 is preferably a first servo cylinder, and a cylinder body and a piston rod of the first servo cylinder are connected to the first cylinder 301 and the second cylinder 302, respectively.

Specific: the adjusting mechanism 3 in this embodiment controls the lifting motion of the second cylinder 302 through the first servo cylinder, so as to drive all the vector components 304 to output the first linear degree of freedom. The vector component 304 and the DBD array generating component 5 are connected in a matching way, so that the function of generating DBD plasmas can be realized, and the regeneration of the active carbon can be realized. Specifically, the activated carbon is regenerated by breaking the adsorption layer on the surface of the activated carbon by generating DBD plasma to decompose and remove contaminants adsorbed on the activated carbon.

It will be appreciated that in this embodiment, the adjusting mechanism 3 adopts the first servo electric cylinder to control the lifting movement of the second cylinder 302, so as to realize lifting adjustment of the activated carbon, thereby adapting to the regeneration requirements of the activated carbon with different sizes and specifications. Meanwhile, the vector component 304 and the DBD array generating component 5 are connected in an adapting mode, so that DBD plasma is generated, the advantages of high efficiency, high speed, low cost and the like are achieved, various organic and inorganic pollutants can be removed, and the DBD plasma can be reused. The regenerator can also control all electrical elements through the control console 2, has the advantages of simplicity and convenience in operation, high intelligent degree and the like, and improves the operation efficiency and accuracy.

In some embodiments of the present application, please refer to fig. 1-8 in combination: the vector assembly 304 includes a first plate 3042 and an articulated arm 3041 with bottoms hinged to the tops of the first cylinder 301 and the second cylinder 302, respectively;

the top of the first plate 3042 is hinged with a second plate 3043;

two ends of the hinge rod 3044 are respectively hinged to the outer surface of the second plate 3043 and the top of the hinge arm 3041;

the second plate 3043 is mounted with the DBD array generating assembly 5 on a surface facing the axis.

When the second cylinder 302 slides, the hinge rod 3044 is lifted or lowered due to the hinge mode, and then pulls the hinge second plate 3043 to displace; the second plate 3043 is hinged with the top of the first plate 3042, and the bottom of the first plate 3042 is hinged with the first cylinder 301 with fixed relative position; when the first servo motor is operated, all vector components 304 are close to or far from the axis; furthermore, when in practical application, one circle formed by all vector components 304 can generate variable diameter adjustment, so that the adjustment of the distance, the direction and the area of each DBD array generating component 5 relative to the plasma output of the regenerated active carbon is realized; when the method is applied, the method can be used for adjusting the activated carbon with different sizes, but the most core application is that the diameter-changing adjustment process is circularly and reciprocally performed, namely, the method is continuously enlarged, reduced, enlarged and reduced, the dynamic adjustment of plasmas relative to the activated carbon is increased in a fluctuation mode, and the contact area between a reaction area and the activated carbon is adjusted to realize the beneficial effect.

Specifically:

(1) The regeneration efficiency is improved: by adjusting the circular size of the vector assembly 304, the contact area of the plasma with the activated carbon can be more matched. This helps to increase the efficiency of activated carbon regeneration because the plasma can be more evenly distributed on the surface of the activated carbon, thereby increasing the effect of desorption of the adsorbed species from the activated carbon.

(2) The method is suitable for activated carbon with different sizes: the multi-degree-of-freedom activated carbon regenerator can adapt to activated carbon with different sizes by adjusting the round size. This allows a wider range of applications for the regenerator to handle activated carbon of various sizes.

(3) Control the regeneration rate: the circular size of the vector assembly 304 may be adjusted to control the regeneration rate as desired. A larger circular area accelerates the regeneration process because the plasma has a larger contact area with the activated carbon. However, a faster regeneration rate may affect the regeneration effect. By adjusting the circular size, a balance of the regeneration speed and the effect can be achieved.

(4) Optimizing energy consumption: on the premise of not influencing the regeneration effect, the energy consumption can be reduced by adjusting the circular size of the vector component. A smaller circular area may require lower plasma generator power, thereby reducing power consumption. Meanwhile, the round size can be adjusted according to actual demands so as to ensure the balance of the regeneration effect and the energy consumption of the activated carbon.

In the scheme, the method comprises the following steps: in this embodiment, the vector assembly 304 includes a first plate body 3042 and an hinge arm 3041, the bottoms of which are hinged to the top of the first cylinder 301 and the second cylinder 302, respectively, the top of the first plate body 3042 is hinged to a second plate body 3043, both ends of the hinge rod 3044 are hinged to the outer surface of the second plate body 3043 and the top of the hinge arm 3041, respectively, and a DBD array generating assembly 5 is mounted on a surface of the second plate body 3043 facing the axis.

Specific: the vector assembly 304 in this embodiment may be raised or lowered by articulation to thereby pull the articulated second plate 3043 into displacement; because the second plate 3043 is hinged to the top of the first plate 3042, and the bottom of the first plate 3042 is hinged to the first cylinder 301 with a fixed relative position, when the first servo cylinder is operated, all the vector components 304 are close to or far from the axis; furthermore, when in practical use, one "circle" formed by all the vector components 304 will generate reducing adjustment, so as to realize adjustment of the distance, orientation and area of each DBD array generating component 5 relative to the plasma output of the regenerated active carbon.

It will be appreciated that in this particular embodiment, dynamic adjustment of plasma output spacing, orientation and area may be achieved through circular size adjustment of the vector assembly 304. Specifically, by adjusting the circular size of the vector assembly 304, the contact area between the plasma and the activated carbon can be more matched, and the efficiency of activated carbon regeneration can be improved. In addition, the variable diameter adjustment of the vector assembly 304 can also adapt to active carbon with different sizes, so that the adaptability and the application range of the regenerator are improved. The regeneration speed can be controlled by adjusting the round size according to the requirement, and the balance of the regeneration speed and the effect is realized. On the premise of not influencing the regeneration effect, the energy consumption can be reduced, and the economical efficiency and the environmental protection of the regenerator are improved.

In summary, the vector component 304 in this embodiment has a high degree of flexibility and intelligence that enables dynamic adjustment and optimization of activated carbon regeneration. The device is characterized in that the output interval, the direction and the area of the plasmas can be automatically adjusted according to the needs, so that the regeneration efficiency is improved, the device is suitable for activated carbon with different sizes, the regeneration speed is controlled, the energy consumption is optimized, and the device has high practical value and economic benefit.

In some embodiments of the present application, please refer to fig. 1-8 in combination:

the top of the second plate body 3043 is provided with an air jet pump 3045 for jetting air for generating DBD plasma gas by ionization of the DBD array generating assembly 5, and a CCD industrial vision camera 3046 for visually detecting external visual characteristics of the activated carbon.

The following functions can be implemented by introducing the visual detection technology of the CCD industrial visual camera 3046:

(1) Monitoring the state of the activated carbon in real time: the CCD industrial vision camera can monitor the state of the activated carbon in real time, including the parameters of color, shape, size and the like, so as to determine the adsorption and regeneration effects of the activated carbon, and correspondingly adjust and optimize the activated carbon.

(2) Monitoring the plasma reaction: the CCD industrial vision camera can monitor the condition of the plasma reaction in real time, including the distribution of the reaction area, the reaction time and other parameters, so as to control the intensity and the time of the plasma reaction.

(3) Automatically adjusting parameters: the CCD industrial vision camera can automatically adjust parameters of the multi-degree-of-freedom active carbon regenerator, including parameters such as working voltage and frequency of the plasma generator, according to real-time monitoring results, so as to ensure that the multi-degree-of-freedom active carbon regenerator operates in an optimal working state.

In the scheme, the method comprises the following steps: in this embodiment, the integration of the CCD industrial vision camera 3046 into the multi-degree of freedom activated carbon regenerator can enable real-time monitoring of the activated carbon and plasma reaction process. The position of the CCD industrial vision camera 3046 is set at the top of the second plate 3043, so that the detection of the external vision characteristics of the activated carbon can be realized, and the condition of the plasma reaction can be monitored.

Specific: the working principle of the CCD industrial vision camera 3046 is to capture and process an object image by using a CCD chip, so as to realize measurement and detection of appearance and parameters of the object. In the multi-degree-of-freedom activated carbon regenerator, a CCD industrial vision camera can capture and process images of the reaction of activated carbon and plasma, so that real-time monitoring is realized.

It can be appreciated that in this embodiment, the visual detection technology of the CCD industrial visual camera is introduced, so as to monitor the reaction condition of the activated carbon and the plasma in real time, thereby realizing the following functions: (1) monitoring the state of the activated carbon in real time: the CCD industrial vision camera can monitor the state of the activated carbon in real time, including the parameters of color, shape, size and the like, so as to determine the adsorption and regeneration effects of the activated carbon, and correspondingly adjust and optimize the activated carbon. (2) monitoring the plasma reaction: the CCD industrial vision camera can monitor the condition of the plasma reaction in real time, including the distribution of the reaction area, the reaction time and other parameters, so as to control the intensity and the time of the plasma reaction. (3) automatically adjusting parameters: the CCD industrial vision camera can automatically adjust parameters of the multi-degree-of-freedom active carbon regenerator, including parameters such as working voltage and frequency of the plasma generator, according to real-time monitoring results, so as to ensure that the multi-degree-of-freedom active carbon regenerator operates in an optimal working state. Through real-time monitoring and automatic parameter adjustment, the intelligent control of the multi-degree-of-freedom active carbon regenerator can be realized, and the regeneration efficiency and stability of the multi-degree-of-freedom active carbon regenerator are improved.

the DBD array generating assembly 5 includes at least two parallel electrodes 501 and a dielectric adapted thereto, and a planar insulating layer is formed on the surface of the dielectric;

with a spacing between the electrodes 501. So as to form a suitable electric field distribution such that a plasma can be generated on both sides of the insulating layer.

The electrode 501 is electrically connected to a discharge protector 502 for discharge protection.

In the scheme, the method comprises the following steps: the DBD array generating assembly 5 is composed of at least two parallel electrodes 501 and a dielectric, the electrodes are spaced to form a proper electric field distribution, and a planar insulating layer is formed on the surface of the dielectric. Meanwhile, the electrode 501 is electrically connected with the discharge protector 502 to protect the electrode 501 and other components from excessively high discharge voltage and current.

Specific: the generation of DBD plasma requires a high voltage to excite the gas so that the gas molecules are ionized to form a plasma. The electrode 501 and the dielectric in the DBD array generating assembly 5 may form an insulating layer, and the electric field intensity is concentrated on the surface of the insulating layer under the action of high voltage, so as to form a partial discharge region, thereby generating plasma.

It is understood that in the present embodiment, the DBD array generating assembly 5 is designed to effectively generate DBD plasma, and can be applied to a multi-degree-of-freedom activated carbon regenerator, and interact with activated carbon to regenerate the activated carbon. Meanwhile, the design of the discharge protector 502 can effectively protect electrodes and other components from being damaged by excessively high discharge voltage and current, and prolong the service life of the DBD array generating assembly 5.

Further, the DBD plasma regeneration technology is that by placing active carbon in a DBD plasma reactor, under the action of a high-voltage alternating-current electric field, organic substances on the surface of the active carbon are electrolyzed and oxidized and decomposed into smaller inorganic substances, and the inorganic substances have good adsorption performance. Meanwhile, the plasma reaction can also generate active substances such as ozone, and the active substances can decompose and oxidize organic substances more thoroughly, so that the active carbon has better regeneration effect.

Specifically, in the DBD plasma reactor, a high-voltage alternating current electric field is formed between electrodes, and when the electric field strength reaches a certain value, gas molecules are ionized to form plasma. In the plasma reaction area, ionized oxygen and water vapor react with organic substances to generate oxidants such as ozone, hydrogen peroxide and the like, and the oxidants can decompose the organic substances more thoroughly, so that the activated carbon has better regeneration effect.

In addition, the DBD plasma regeneration technology can also control the intensity and time of the plasma reaction by adjusting the working voltage, frequency and other parameters of the plasma reactor, thereby controlling the regeneration effect of the activated carbon. Meanwhile, the state of the activated carbon and the plasma reaction condition can be monitored in real time by using a visual detection technology, and the parameters of the activated carbon regenerator are automatically adjusted so as to ensure that the activated carbon regenerator runs in an optimal working state.

Specifically, the discharge voltage of the electrode 501 of the DBD array generating assembly 5 is 10-20 kV, and the power frequency is 50Hz; the gas injected by the gas injection pump 3045 is an inert or molecular gas such as air, ammonia, oxygen, gas development, etc., and the flow rate is 0.5-2L/min. The implementation of the above technical parameters is disclosed by the following papers:

(1) The method for regenerating active carbon is compared with the development trend research, the paper has the classification number of X703.1 and the publication number of 1000-8942, and the paper is published in the 11 th stage of 2018 of coal technology.

(2) The research on DBD plasma degradation of organic matters and activated carbon regeneration of activated carbon adsorption is characterized in that the paper has classification number X703.5 and publication number 1000-8713, and is published in the volume 36 and 9 of the environmental science journal 2016.

In some embodiments of the present application, please refer to fig. 1-8 in combination: the bottom of the first cylinder 301 is provided with a device assembly 4;

the storage component 4 comprises a second linear member 403 for outputting a second linear degree of freedom, the second linear member 403 is arranged at the bottom of the first cylinder 301, and the second linear member 403 is connected with and lifted by a rotary executing member 402;

to the rotary actuator 402, an electric chuck 401 for clamping activated carbon is connected.

The second linear member 403 is preferably a second servo cylinder, and the rotary actuator 402 is preferably a servo motor;

the cylinder body and the piston rod of the second servo electric cylinder are fixedly connected to the bottom of the first cylinder 301 and the outer shell of the servo motor respectively, and the output shaft of the servo motor is fixedly connected with the outer shell of the electric chuck 401.

In the scheme, the method comprises the following steps: the bottom of the first cylinder 301 is mounted with a storage assembly 4, wherein the storage assembly 4 comprises a second linear member 403 for outputting a second linear degree of freedom, a rotary actuator 402 and an electric chuck 401 for clamping activated carbon.

Specific: lifting of the second linear member 403 and rotation of the rotary actuator 402 can effect clamping and adjustment of the activated carbon. The electric chuck 401 is designed to firmly fix the activated carbon at the end of the rotary actuator 402, so that the activated carbon cannot displace and oscillate during operation, thereby ensuring the normal operation of the multi-degree-of-freedom activated carbon regenerator.

It will be appreciated that, in this particular embodiment,

(1) Clamping activated carbon: the electric chuck 401 can firmly clamp the activated carbon, thereby avoiding the displacement and vibration of the activated carbon in the regeneration process and ensuring the precision and efficiency of regeneration.

(2) And (3) adjusting the height of the activated carbon: the elevation of the second linear member 403 can adjust the height of the activated carbon to accommodate activated carbon of different sizes and shapes.

(3) Rotating and adjusting the position of the activated carbon: rotation of the rotary actuator 402 can adjust the position of the activated carbon to ensure that it is within the range of plasma output and to improve regeneration efficiency.

(4) Automatically adjusting parameters: the second servo electric cylinder can automatically adjust parameters of the multi-degree-of-freedom type active carbon regenerator so as to ensure that the multi-degree-of-freedom type active carbon regenerator operates in an optimal working state.

Summarizing, the above-mentioned technology addresses the shortcomings of the conventional technology by solving the following principles:

(1) Firstly, a regeneration technology based on DBD plasma is introduced, so that adsorbed substances on the activated carbon can be removed more thoroughly, and compared with a hot air flow regeneration or steam regeneration mode in the traditional technology, the method is more efficient and energy-saving. In addition, as the plasma is generated on the surface of the activated carbon, the contact area between the regeneration area and the activated carbon can be controlled more accurately, so as to achieve more optimal regeneration effect.

(2) Secondly, a multi-degree-of-freedom activated carbon regenerator is adopted, and a reducing adjustment technology and a visual detection technology are introduced into the regenerator. The application of the techniques can realize the accurate positioning and self-adaptive adjustment of the activated carbon, thereby reducing the manual intervention on the activated carbon in the traditional technique and improving the regeneration efficiency and consistency.

(3) Finally, the convolutional neural network is adopted to monitor the image data of the active carbon and the plasma reaction in real time and automatically adjust parameters. The technology can monitor the state of the active carbon and the plasma reaction condition more accurately, and automatically adjust the parameters of the regenerator according to the real-time monitoring result so as to achieve the optimal regeneration effect.

(4) Therefore, the instinct concrete implementation technology can solve some defects existing in the traditional technology from the principle aspect, and bring effective innovation and progress in the aspects of improving the regeneration efficiency, reducing the energy waste and the like.

In the scheme, all electric elements of the whole device are powered by mains supply; specifically, the electric elements of the whole device are in conventional electrical connection with the commercial power output port through the relay, the transformer, the button panel and other devices, so that the energy supply requirements of all the electric elements of the device are met.

In the scheme, all pneumatic elements of the whole device are powered by an external compressed air bottle matched with an air pump; specifically, the pneumatic element of the whole device is in conventional pneumatic connection with the air pump output port of the compressed air cylinder through devices such as an electromagnetic valve, a reversing valve, a pipe body and the like;

preferably, the driving synchronization of the pneumatic elements is controlled by a controller.

The above is disclosure of the mechanical principle aspect of the present embodiment, and a method for using the multi-degree of freedom activated carbon regenerator based on DBD plasma will be further described, which adopts the multi-degree of freedom activated carbon regenerator based on DBD plasma as described above, and includes the following steps:

In the scheme, the method comprises the following steps: the collection and labeling of the data sets is the basis for constructing the monitoring model, and the data needs to be collected from multiple dimensions, such as the state of the adsorbed substances on the surface of the activated carbon, the reaction area of the plasma, the reaction time and other parameters, and corresponding labels are allocated to each image.

In the scheme, the method comprises the following steps: for the collected data set, the accuracy of the monitoring model can be improved by preprocessing the data. For example, image enhancement, noise reduction, gradation processing, and the like are performed, and normalization processing of the image is performed at the same time, so that the variability between data is eliminated.

In the scheme, the method comprises the following steps: the convolutional neural network is the core of an image monitoring model, and can extract effective characteristics from a large amount of data by designing a proper network structure so as to accurately monitor the reaction conditions of the activated carbon and the plasmas. The design of the convolutional neural network needs to consider parameters such as network depth, convolutional kernel size, pooling layer and the like so as to obtain a better monitoring effect.

In the scheme, the method comprises the following steps: the convolutional neural network needs to be trained by using the prepared data set and the corresponding labels, and the aim of the training model is to obtain a model with high accuracy, so that the accuracy and the robustness of the model can be improved by adjusting the optimizer, the loss function and the evaluation index parameters.

S5, monitoring in real time: the trained CNN model is deployed into a CCD industrial vision camera 3046, and the state of the activated carbon and the plasma reaction condition are monitored in real time.

In the scheme, the method comprises the following steps: the trained monitoring model is deployed into a CCD industrial vision camera 3046 to monitor the reaction conditions of the activated carbon and the plasmas in real time. The monitoring result is passed as a parameter to the next operation.

In the scheme, the method comprises the following steps: based on the monitoring result, parameters of the activated carbon regenerator, such as the DBD array generating assembly 5, are automatically adjusted.

The multi-degree-of-freedom activated carbon regeneration using method based on the DBD plasma can automatically adjust regeneration parameters, realize the omnibearing and accurate regeneration of the activated carbon, and improve the regeneration efficiency and quality. Meanwhile, by introducing a visual detection technology of a CCD industrial visual camera, the real-time monitoring and automatic adjustment of the state of the activated carbon and the plasma reaction are realized, and the accuracy and the stability of regeneration are improved.

In some embodiments of the application: the convolutional neural network includes:

c: an output of the convolutional layer;

i: inputting an image;

k: a convolution kernel;

i and j are each: the width and height indexes of the convolution layer output;

m and n are each: the width and height indexes of the convolution kernel;

c: channel index.

A[x] = max(0, x)

a: activating the output of the function;

x: the output of the convolutional layer.

Pooling layer: for reducing the current size of the feature map:

p: outputting a pooling layer;

a: activating the output of the layer;

p: step size of pooling operation.

f: outputting a full connection layer;

w: a weight matrix;

b: a bias vector;

s: output or probability distribution of the output layer;

f: outputting a full connection layer;

i and j: category index.

Using a cross entropy loss function:

l: a loss value;

y: the one-hot encoding vector of the actual tag;

s: outputting of the output layer;

this convolutional neural network-based model includes an input layer, a convolutional layer, an activation layer, a pooling layer, a fully-connected layer, and an output layer. By training the model, the real-time monitoring of the state of the activated carbon and the plasma reaction can be realized. In the training process, the difference between the prediction and the actual label is measured by using a cross entropy loss function, and model parameters are updated through a back propagation algorithm and an optimization algorithm.

Exemplary: the color status of the activated carbon is monitored by a two-dimensional gray scale image. The color states of activated carbon are divided into two categories: class 0 represents non-adsorbed saturation, and class 1 represents adsorbed saturation. The size of the input image is 8x8, i.e., width=8, height=8, channels=1.

Based on the convolutional neural network model described above, the following parameters may be set:

(1) Convolution layer: 1 convolution kernel is used, 3x3 in size. In this case, the output size after convolution is 6×6.

(2) Roll activation layer: a ReLU activation function is used.

(3) Rolling pool layer: using 2x2 max pooling, the pooled output size was 3x3.

(4) Rolling the full connection layer: the 3x3 output is flattened into 9 nodes and then connected to 2 output nodes (corresponding to two categories).

To simplify this exemplary derivation, an 8x8 gray scale image will be predicted to determine the color status of the activated carbon. In practical applications, higher gray scale images are required.

(1) Roll input layer: a gray scale image of 8×8 is directly input.

(2) Roll convolution layer: the convolution kernel of 3*3 is applied for the convolution operation. Resulting in a convolved output of 6*6.

(3) Roll activation layer: a ReLU activation function is applied to the convolved output. The output size was kept 6*6.

(5) Rolling pool layer: 2 x 2 max pooling is applied to the activation output. Resulting in a pooled output of 3*3.

(6) Rolling the full connection layer: the pooled output of 3*3 is flattened into a 9-node vector, which is then linearly transformed using a weight matrix and bias vectors. Resulting in 2 output nodes.

(7) Roll output layer: the softmax function was applied to the full connection layer output resulting in a probability distribution of 2 categories.

By comparing the output probability distribution, the color state of the activated carbon corresponding to the input image can be judged. For example, if the probability of category 0 is greater than the probability of category 1, a conclusion may be drawn: activated carbon represented by the input image is not saturated by adsorption. Otherwise, adsorption saturation is indicated.

Further, the linkage between the model based on the CCD industrial vision camera and the first servo electric cylinder needs to be realized, and vision detection information needs to be converted into a control signal of the first servo electric cylinder:

(1) Preprocessing image data: firstly, image data is obtained from a CCD industrial vision camera, and is preprocessed, such as scaling, clipping, normalization and the like, so that the CCD industrial vision camera is suitable for a trained convolutional neural network model.

(2) Running the model and obtaining a prediction result: and inputting the preprocessed image data into a trained model to obtain a prediction result of the model on the state of the activated carbon. This may include the color, shape, size, etc. characteristics of the activated carbon.

(3) Analyzing the prediction result and generating a control signal: and analyzing whether the state of the activated carbon needs to be adjusted according to the prediction result of the model. And then, generating corresponding control signals according to the degree and the direction which are required to be adjusted. For example, if activated carbon adsorption is saturated, it may be necessary to increase the stroke of the first servo cylinder to increase the regeneration strength.

(4) Transmitting a control signal to the first servo electric cylinder: and sending the generated control signal to the first servo electric cylinder controller. The controller will adjust the operating parameters of the first servo cylinder, such as speed, acceleration, travel, etc., in accordance with the control signal.

(4) Monitoring and adjusting in real time: in the whole process, the state of the activated carbon and the operation condition of the first servo electric cylinder need to be continuously monitored. If the state of the activated carbon is found to change or the first servo electric cylinder runs abnormally, the control signal needs to be adjusted in time to realize the self-adjusting function.

In some embodiments of the present application, the specific implementation program of the above-listed method in this embodiment is described in c++ language, and the principle is that:

a pre-trained TensorFlow model was loaded using OpenCV's dnn module. The input image is then loaded, preprocessed (resized and normalized) and set as input to the network. Next, the model is run and the output is obtained, and then the category with the highest probability is found. Finally, the predicted category is output.

To use this example program, an OpenCV library (guaranteed to include dnn modules) needs to be installed and linked to the item. A pre-trained model file is also needed, which can be trained using frameworks such as TensorFlow or pyrerch.

Specifically, the procedure includes the steps of:

(1) Introducing necessary header files: a header file included in the OpenCV library is required to use the functions of OpenCV. The iostream is used to process input/output.

(2) Program entry point: the main function is the entry point for the c++ program. The argc and argv parameters are used to obtain parameters passed from the command line.

(3) Checking command line parameters: two parameters are required: model files and input images. If the number of parameters is incorrect, the usage information is printed and an error code is returned.

(4) Loading a pre-trained model: the model filename is obtained from the command line parameters and the model is loaded using the readNetFromTensorflow function of OpenCV. This will return a cv:: dnn:: net object, which will be used for reasoning. Merely exemplary illustrations are provided herein, and specific training models will be presented below;

(5) Loading an input image: the input image file name is obtained from the command line parameters and the image is loaded using the imread function of OpenCV. This will return a cv:: mat object, which represents the image matrix. The image is loaded in grayscale mode using the cv: IMREAD_ GRAYSCALE flag.

(6) Preprocessing an input image: the image is first resized to match the input size of the model (8 x8 pixels in this example). The image data type is then converted to a 32-bit floating point number and the pixel values are normalized to the 0-1 range. Finally, the image was converted to a 4D tensor (batch, channel, row, column) for DNN modules using the cv: DNN:blob Frominimage function.

(7) Setting network input: the preprocessed image is set as an input to the network.

(8) Running the model and obtaining an output: the model is run using forward method and the output is obtained. The output will be a cv:Mat object, representing the classification probability.

(9) Find the category with the highest probability: the category with the highest probability is found using the cv:minMaxLoc function. The minMaxLoc function can find the minimum and maximum values in the matrix and their positions. Only the position of the maximum is of interest, so the other parameter is set to null ptr. The location of the maximum (max_loc) will be used to determine the category of prediction.

(10) Output predicted category: std:cout is used to output the predicted category to the console.

(11) Ending the program and returning a success code: after the program is executed, a return of 0 indicates success.

The core principle of this example program is to load and run a pre-trained convolutional neural network model using the OpenCV library. The program first loads the model and the input image, and then pre-processes the input image to adapt to the input requirements of the model. Then, the program sets the preprocessed image as network input and performs reasoning. Finally, the program finds the category with the highest probability and outputs it.

The above procedure also enables the linkage with the first servo cylinder, requiring programming according to the communication protocol and hardware interface of the cylinder. The following is a control program of the first servo electric cylinder, which is connected through a serial interface; meanwhile, the specific embodiment uses the asio library of C++:

in this procedure, the necessary header file is first contained, and then a function named control_service_cylinder is defined, which accepts the serial port name and the target location as parameters. The serial port is opened using the asio library and the required baud rate is set.

Next, a command string is constructed to send the target position to the servo cylinder. Here, it is assumed that the electric cylinder accepts a command in a format similar to "MOVE < position > \r\n". Adjustment is required according to the actual communication protocol of the first servo cylinder.

The response is then read from the first servo cylinder and output to the console. In main function, call control_service_cylinder function, and transfer into actual serial port name and target position.

For further disclosure of the present technology, an example of training the convolutional neural network model described above using PyTorch will be presented below. In practical application, it is necessary to classify a set of actual activated carbon images and their corresponding saturation information, and ensure that the pyrerch library is installed:

(1) Importing necessary libraries: a defined convolutional neural network model structure. Here is a simple example, which can be modified according to the requirements:

(2) Adding a full connection layer: a data set is prepared. The CIFAR-10 dataset is used here as an example:

(3) Compiling a model, and setting an optimizer, a loss function and an evaluation index:

(4) Training a model: evaluating performance of the model on the test set:

this procedure demonstrates how a convolutional neural network is constructed and trained using TensorFlow. First, necessary libraries are imported and model structures are defined. The CIFAR-10 dataset is then loaded and the data is preprocessed. Next, the model is compiled, and an optimizer, a loss function, and an evaluation index are set. And finally training the model by using training data, and evaluating the performance of the model on a test set.

The above code examples demonstrate how Convolutional Neural Network (CNN) models are constructed, trained, and evaluated using a TensorFlow or pyrerch framework. Convolutional Neural Networks (CNNs) are a type of deep learning model that is particularly well suited for processing image data. CNNs learn to extract features from input images and classify through a series of convolution, pooling, activation, and fully connected layers. In this case, the following functions can be implemented with the trained CNN model:

(1) Monitoring the state of the activated carbon in real time: by training the CNN, it is possible to learn to extract key features (such as parameters of color, shape, and size) from the activated carbon image in order to determine adsorption and regeneration effects. This allows the effect of the regeneration process to be known and adjusted and optimised accordingly by monitoring the state of the activated carbon.

(2) Monitoring the plasma reaction: similarly, the trained CNN can be used to monitor plasma reaction conditions, including reaction zone distribution, reaction time, etc. This helps to control the intensity and time of the plasma reaction in real time to ensure optimal results.

(3) Automatically adjusting parameters: the trained CNN model can provide feedback for the activated carbon regenerator according to the real-time monitored activated carbon state and plasma reaction condition. This information can then be used to automatically adjust parameters of the multi-degree of freedom activated carbon regenerator, such as the operating voltage, frequency, etc. of the plasma generator to ensure that it operates in an optimal operating state.

In short, the CNN model constructed and trained using TensorFlow or pyrerch can help monitor key parameters in the activated carbon regeneration process in real time and automatically adjust system parameters accordingly to achieve optimal regeneration.

Further:

s1, monitoring an activated carbon regeneration process by using a Convolutional Neural Network (CNN) model and driving a first servo electric cylinder to perform the whole-flow steps:

s2, preparing a data set: image data of the activated carbon and plasma reactions are collected while each image is assigned a corresponding label (e.g., state of activated carbon, plasma reaction conditions, etc.).

S3, data preprocessing: the image data is normalized, enhanced, etc. to improve the learning effect of the model.

S4, constructing a convolutional neural network: a suitable CNN structure is designed, including convolutional layers, pooling layers, activation functions, full-join layers, and the like.

S5, training a model: and training a CNN model by using the prepared data set and the corresponding label, and adjusting parameters such as an optimizer, a loss function, an evaluation index and the like.

S6, evaluating a model: and evaluating the trained CNN model on the verification set and the test set to ensure that the CNN model has enough generalization capability.

S7, monitoring in real time: and deploying the trained CNN model to a monitoring system, and monitoring the state of the activated carbon and the plasma reaction condition in real time.

S8, automatically adjusting parameters: and according to the monitoring result, automatically adjusting parameters of the activated carbon regenerator, such as the working voltage, the frequency and the like of the plasma generator.

S9, driving a first servo electric cylinder: and transmitting the monitoring result and the adjusted parameters to a control system of the first servo electric cylinder, and adjusting the driving force, the stroke and the like of the first servo electric cylinder in real time according to the requirement.

S10, feedback optimization: according to the actual running condition, the CNN model and the first servo electric cylinder control strategy are continuously optimized, and the active carbon regeneration effect is improved.

The technical features of the above-described embodiments may be combined in any manner, and for brevity, all of the possible combinations of the technical features of the above-described embodiments may not be described, however, they should be considered as the scope of the present description as long as there is no contradiction between the combinations of the technical features.

Example two

In order to make the above-described embodiments of the present invention more comprehensible, embodiments accompanied with the present invention are described in detail by way of example. The present invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit of the invention, so that the invention is not limited to the embodiments disclosed below.

The present embodiment is based on the relevant principles described in the above detailed description, where exemplary applications are:

in the current example scenario, a batch of used activated carbon needs to be regenerated, and the multi-degree-of-freedom activated carbon regenerator based on DBD plasma provided in the above embodiment is used. The batch of used activated carbon has lost its adsorption performance after adsorbing organic matters, and needs to be regenerated to recover the adsorption performance.

The using steps are as follows:

s1, preparing activated carbon: the used activated carbon first needs to be collected and ready for regeneration. In practice, activated carbon may require pretreatment, such as removal of impurities and the like.

S2, placing activated carbon: the collected activated carbon is placed in a placement component 4, and the placement component 4 is positioned at the bottom of a first cylinder 301 of the multi-degree-of-freedom activated carbon regenerator based on DBD plasma and is used for storing the activated carbon.

S3, starting equipment: starting the multi-degree-of-freedom activated carbon regenerator based on DBD plasma, and enabling the equipment to start working by operating a computer control system.

S4, ionizing plasma gas: the plasma gas is injected into the DBD array generating assembly 5 by the injection pump 3045, and an ionized plasma is generated.

S5, adjusting plasma parameters: plasma parameters, such as operating voltage and frequency, etc., are adjusted by the console 2 to enable the ionized plasma to efficiently process the activated carbon.

S6, monitoring the state of the activated carbon: the activated carbon is monitored in real time by using a CCD industrial vision camera 3046, including parameters such as color, shape, size and the like, so as to determine the adsorption and regeneration effects thereof, and correspondingly adjust and optimize the adsorption and regeneration effects.

S7, adjusting DBD array parameters: according to the monitoring result, the parameters of the multi-degree-of-freedom active carbon regenerator, including the parameters of electrode spacing, electric field distribution and the like of the DBD array, are automatically adjusted by a computer control system so as to optimize the active carbon regeneration effect.

S8, activating activated carbon: activated carbon is placed in a plasma reaction area, activated by plasma reaction, and allowed to have adsorption capacity again. During the activation process, the position and angle of the activated carbon need to be adjusted according to the real-time monitoring result so as to make the activated carbon fully contact with the plasma reaction area.

And S9, continuously executing the movement track of the first linear member 303, driving the electric chuck 401 to rotate by the rotating executing member 402, and clamping the activated carbon.

S10, closing the air source valve and the vacuum pump, and stopping air entering and extracting.

And S11, lifting the second linear member 403 with the second linear degree of freedom upwards, so that the activated carbon is separated from the first cylinder 301 and falls into a container for receiving the activated carbon, and the regeneration is completed.

S12, repeating the steps S1-S11 to finish the continuous repeated regeneration of the activated carbon.

S13, periodic maintenance: periodic inspections and maintenance of the equipment are performed to ensure long-term stable operation thereof.

The above examples merely illustrate embodiments of the invention that are specific and detailed for the relevant practical applications, but are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Example III

adsorption tower-carbon discharge tank: and the saturated active carbon is discharged from the bottom of the adsorption tower under self-pressure and enters a carbon discharge tank. In this step, the position and movement of the activated carbon may be adjusted using a multi-degree of freedom activated carbon regenerator or (and) activated carbon regenerator in order to accurately discharge the saturated activated carbon into the carbon discharge tank.

Carbon discharge tank-regeneration overhead tank: the carbon after carbon washing needs to be sent into a regeneration overhead tank for regeneration treatment. In this step, a multi-degree of freedom activated carbon regenerator may be used to control the movement of the carbon, and the carbon after the carbon washing is fed into a regeneration overhead tank to be ready for the next regeneration process.

Regeneration overhead tank-multistage furnace: and (5) the carbon after carbon washing enters a multi-stage furnace for regeneration treatment. In this step, a multiple degree of freedom activated carbon regenerator or (and) activated carbon regenerators may be used to assist in delivering the carbon washed carbon to the multi-stage furnace for regeneration.

Regenerator-quench tank: in the regenerating oven, the activated carbon undergoes stages of drying, roasting, activation, and the like. The regenerated active carbon falls into the quenching tank from the feed opening of the regenerating furnace. In this step, the multi-degree-of-freedom activated carbon regenerator or (and) activated carbon regenerator may be used to control the movement and position of the carbon so that it smoothly falls into the quenching tank from the regenerator, and the quenching process of the activated carbon is achieved by controlling the temperature and cooling rate.

Quench tank-new char tank: the activated carbon cooled by the quenching tank needs to be stored in a new carbon tank for reuse of the adsorption tower. In this step, a multi-degree of freedom activated carbon regenerator or (and) an activated carbon regenerator may be used to control the movement and position of the carbon, and the cooled activated carbon is fed into a new carbon tank for storage.

New carbon tank-adsorption tower: the mixture of the new carbon in the new carbon tank and the regenerated active carbon is sent to an adsorption tower to ensure that the adsorption tower has enough active carbon to reach the effluent standard. In this step, a multi-degree-of-freedom activated carbon regenerator or (and) an activated carbon regenerator may be used, which may assist in controlling the delivery and supply of activated carbon during the transition between the new carbon tank and the adsorption tower. By adjusting the first linear degree of freedom, i.e., adjusting the circular size and position of the vector assembly 304, it can be ensured that the proper activated carbon is supplied to the top of the adsorption tower to ensure that the adsorption tower has sufficient activated carbon for the adsorption process of the process water.

The variable diameter adjusting function of the multi-degree-of-freedom activated carbon regenerator is used, so that the output quantity and the position of the activated carbon can be flexibly adjusted according to actual requirements, and the requirements of an adsorption tower can be met. The degree-of-freedom activated carbon regenerator not only can provide accurate control and adjustment, but also can monitor the state of activated carbon in real time and perform feedback and automatic adjustment through a visual detection technology. Therefore, it can ensure that the adsorption tower always has enough active carbon supply in the adsorption process, and improve the adsorption efficiency and the water quality treatment effect.

In summary, the multi-degree-of-freedom activated carbon regenerator based on DBD plasma can be applied to the processes of carbon discharge of an adsorption tower, regeneration of a high-level tank, a multi-stage furnace, a quenching tank, a new carbon tank, an adsorption tower and the like in process description. By controlling the movement, position and supply amount of the activated carbon, the method can realize accurate regeneration and supply of the activated carbon and improve the adsorption efficiency and the water quality treatment effect. Meanwhile, by introducing a visual detection technology, real-time monitoring and automatic adjustment can be realized, and the activated carbon regeneration process is further optimized.

Claims

1. The multi-degree-of-freedom activated carbon regenerator based on DBD plasma is characterized by comprising an adjusting mechanism (3);

the outer part of the adjusting mechanism (3) comprises at least ten first linear degrees of freedom which are arranged along a coaxial line annular array and have the stroke direction staggered with the axis; each first linear degree of freedom is adapted to have connected thereto a DBD array generating assembly (5) for generating BDB plasma;

At least four storage components (4) for clamping active carbon needing regeneration operation are arranged in the adjusting mechanism (3) in an annular array mode by taking the axis as a reference;

the storage component (4) comprises a second linear degree of freedom along the axial direction for adjusting the lifting of the activated carbon.

2. The DBD plasma-based multi-degree of freedom activated carbon regenerator of claim 1, wherein: the adjusting mechanism (3) comprises a first cylinder (301) with a central axis as an axis and a second cylinder (302) which is in sliding fit with the outside of the first cylinder (301);

the second cylinder (302) is driven to lift by the first linear piece (303);

ten groups of vector components (304) for outputting the first linear degree of freedom are uniformly arranged outside the second cylinder (302) in the form of an annular array;

the vector component (304) is adapted to be connected with the DBD array generating component (5).

3. The DBD plasma-based multi-degree of freedom activated carbon regenerator of claim 2, wherein: the vector assembly (304) comprises a first plate body (3042) and an articulated arm (3041), wherein the bottom of the first plate body is hinged to the top of the first cylinder (301) and the top of the second cylinder (302) respectively;

the top of the first plate body (3042) is hinged with a second plate body (3043);

Two ends of the hinge rod (3044) are respectively hinged to the outer surface of the second plate body (3043) and the top of the hinge arm (3041);

a DBD array generating module (5) is mounted on one surface of the second plate body (3043) facing the axis.

4. The DBD plasma-based multi-degree of freedom activated carbon regenerator of claim 3, wherein: the top of the second plate body (3043) is provided with an air jet pump (3045) for jetting DBD plasma gas generated by ionization of the DBD array generating component (5), and a CCD industrial vision camera (3046) for visually detecting external visual characteristics of the active carbon.

5. The DBD plasma-based multi-degree-of-freedom activated carbon regenerator according to any one of claims 2 to 4, wherein: the DBD array generating component (5) comprises at least two electrodes (501) which are arranged in parallel and a dielectric substance which is matched with the electrodes;

the electrodes (501) have a space therebetween.

6. The DBD plasma-based multi-degree of freedom activated carbon regenerator of claim 5, wherein: the electrode (501) is electrically connected with a discharge protector (502) for discharge protection.

7. The DBD plasma-based multi-degree of freedom activated carbon regenerator of claim 5, wherein: the bottom of the first cylinder body (301) is provided with a storage component (4);

The storage component (4) comprises a second linear piece (403) for outputting a second linear degree of freedom, the second linear piece (403) is arranged at the bottom of the first cylinder (301), and the second linear piece (403) is connected with and lifted by a rotary executing piece (402);

an electric chuck (401) for clamping the activated carbon is connected to the rotary actuator (402).

8. A multi-degree-of-freedom activated carbon regeneration using method based on DBD plasma is characterized by comprising the following steps: the multi-degree-of-freedom activated carbon regenerator based on DBD plasma according to any one of claims 1 to 7, comprising the following steps:

s1, preparing a data set: collecting image data of the reaction of the activated carbon and the plasmas, and simultaneously distributing corresponding labels for each image;

s2, data preprocessing: carrying out normalization and enhancement treatment on the image data;

s3, constructing a convolutional neural network: designing a convolutional neural network;

s4, training a model: training a CNN model by using the prepared data set and the corresponding label, and adjusting an optimizer, a loss function and evaluation index parameters;

s5, monitoring in real time: the trained CNN model is deployed into a CCD industrial vision camera (3046), and the state of the activated carbon and the plasma reaction condition are monitored in real time;

S6, automatically adjusting parameters: according to the monitoring result, automatically adjusting parameters of the active carbon regenerator, such as working voltage, frequency and the like of the plasma generator;

s7, driving the first linear member (303): and transmitting the monitoring result and the adjusted parameters to a control system of the first linear piece (303), and adapting the contact area of the current active carbon adjustment reaction area and the active carbon.

9. The DBD plasma-based multi-degree-of-freedom activated carbon regeneration and use method according to claim 8, wherein the method is characterized in that:

the convolutional neural network includes:

input layer: receiving original image data;

convolution layer: extracting image features; the method comprises the steps of providing k convolution kernels, wherein the size of each convolution kernel is f times channels, and f represents the width and the height of the convolution kernel; the convolution operation is as follows:

c: an output of the convolutional layer;

i: inputting an image;

k: a convolution kernel;

i and j are each: the width and height indexes of the convolution layer output;

m and n are each: the width and height indexes of the convolution kernel;

c: channel index;

a: activating the output of the function;

x: an output of the convolutional layer;

pooling layer: for reducing the current size of the feature map:

P: outputting a pooling layer;

a: activating the output of the layer;

: step length of pooling operation;

f: outputting a full connection layer;

w: a weight matrix;

b: a bias vector;

flat (P): flattening the pooling layer output P into a one-dimensional vector;

s: output or probability distribution of the output layer;

i and j: category index.

10. The DBD plasma-based multi-degree-of-freedom activated carbon regeneration and use method according to claim 9, wherein the method is characterized in that: using a cross entropy loss function:

l: a loss value;

y: the one-hot encoding vector of the actual tag;

s: outputting of the output layer;

the gradient of the loss function L with respect to the model parameters is calculated by back propagation, and then the model parameters are updated using gradient descent to reduce the loss value.