CN116651617B - Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof - Google Patents

Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof Download PDF

Info

Publication number
CN116651617B
CN116651617B CN202310503249.3A CN202310503249A CN116651617B CN 116651617 B CN116651617 B CN 116651617B CN 202310503249 A CN202310503249 A CN 202310503249A CN 116651617 B CN116651617 B CN 116651617B
Authority
CN
China
Prior art keywords
dust concentration
time sequence
scale
vector
input vector
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310503249.3A
Other languages
Chinese (zh)
Other versions
CN116651617A (en
Inventor
方华华
朱征涛
庄正康
杜弘
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Jiahuan Electronic Co ltd
Original Assignee
Zhejiang Jiahuan Electronic Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang Jiahuan Electronic Co ltd filed Critical Zhejiang Jiahuan Electronic Co ltd
Priority to CN202310503249.3A priority Critical patent/CN116651617B/en
Publication of CN116651617A publication Critical patent/CN116651617A/en
Application granted granted Critical
Publication of CN116651617B publication Critical patent/CN116651617B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B03—SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
    • B03C—MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
    • B03C3/00—Separating dispersed particles from gases or vapour, e.g. air, by electrostatic effect
    • B03C3/34—Constructional details or accessories or operation thereof
    • B03C3/66—Applications of electricity supply techniques
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B03—SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
    • B03C—MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
    • B03C3/00—Separating dispersed particles from gases or vapour, e.g. air, by electrostatic effect
    • B03C3/34—Constructional details or accessories or operation thereof
    • B03C3/66—Applications of electricity supply techniques
    • B03C3/68—Control systems therefor
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B03—SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
    • B03C—MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
    • B03C2201/00—Details of magnetic or electrostatic separation
    • B03C2201/24—Details of magnetic or electrostatic separation for measuring or calculating parameters, efficiency, etc.
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
    • Y02A50/20—Air quality improvement or preservation, e.g. vehicle emission control or emission reduction by using catalytic converters
    • Y02A50/2351—Atmospheric particulate matter [PM], e.g. carbon smoke microparticles, smog, aerosol particles, dust

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Electrostatic Separation (AREA)

Abstract

An electric dust removal frequency conversion amplitude modulation high-voltage power supply and a high-voltage output control method thereof acquire dust concentration values at a plurality of preset time points in a preset time period acquired by a dust concentration monitoring sensor; the artificial intelligence technology based on deep learning is adopted to excavate the real-time state characteristic information of the dust concentration value, and the frequency value of the electric dust collection variable-frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristic of the dust concentration, so that the power supply has better adaptability and regulation and control capability, and the efficiency of the electric dust collection system and the ultra-clean emission level of flue gas emission are effectively improved.

Description

Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof
Technical Field
The application relates to the technical field of intelligent control, in particular to an electric dust removal variable frequency amplitude modulation high-voltage power supply and a high-voltage output control method thereof.
Background
At present, domestic electric dust removing power supplies mainly comprise a single-phase power frequency power supply, a single-phase constant current power supply, a three-phase power frequency power supply, a high-frequency power supply, a pulse power supply and the like. The foreign electric dust removing power supply product mainly comprises a single-phase power frequency power supply, a high-frequency power supply and a pulse power supply.
Currently, high-frequency power supplies are most widely used due to the highest average output voltage and the flexible pulse gap control function and other advanced technologies. The pulse power supply has the best charging effect on fine dust with high specific resistance, and is well applied to electric fields at the tail of electric dust removal. However, the characteristic requirement and the dust removal effect of the power supply on the first electric field of the electric dust removal are not ideal, so that the high-requirement smoke emission of the electric dust removal is greatly influenced.
Those skilled in the art should know that the first electric field dust removal effect is good, can alleviate the dust removal requirement of follow-up electric field greatly, provides basic guarantee for ultra clean emission. The first electric field has high dust concentration, is easy to generate corona sealing, and requires high average voltage and high average current. The average voltage of the high-frequency power supply is high, but the pulse width of the high-frequency current is too narrow, so that the current is smaller (lower than the power frequency power supply); the power frequency power supply (load matching, when the conduction angle is large) has larger current, but the output voltage pulsation is large, and the average voltage is low.
Therefore, an optimized electric dust removal variable frequency amplitude modulation high-voltage power supply is expected.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an electric dust removal frequency conversion amplitude modulation high-voltage power supply and a high-voltage output control method thereof, which are used for acquiring dust concentration values at a plurality of preset time points in a preset time period acquired by a dust concentration monitoring sensor; the artificial intelligence technology based on deep learning is adopted to excavate the real-time state characteristic information of the dust concentration value, and the frequency value of the electric dust collection variable-frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristic of the dust concentration, so that the power supply has better adaptability and regulation and control capability, and the efficiency of the electric dust collection system and the ultra-clean emission level of flue gas emission are effectively improved.
In a first aspect, there is provided an electric dust removal variable frequency amplitude modulation high voltage power supply, comprising:
the data acquisition module is used for acquiring dust concentration values at a plurality of preset time points in a preset time period acquired by the dust concentration monitoring sensor;
the vector arrangement module is used for arranging the dust concentration values of the plurality of preset time points into dust concentration absolute quantity time sequence input vectors according to a time dimension;
the difference value calculation module is used for calculating the difference value between dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain a dust concentration relative value time sequence input vector;
the cascading module is used for cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector;
the double-pipeline feature extraction module is used for enabling the dust concentration multi-dimensional time sequence input vector to pass through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model so as to obtain a multi-scale dust concentration time sequence feature vector;
the optimizing module is used for carrying out dimension distinguishing degree strengthening on the multi-scale dust concentration time sequence feature vector so as to obtain a decoding feature vector; and
And the decoding module is used for carrying out decoding regression on the decoding characteristic vector through a decoder to obtain a decoding value, wherein the decoding value is used for representing the frequency value of the recommended high-voltage power supply at the current time point.
In the above electric precipitation variable frequency amplitude modulation high voltage power supply, the cascade module is configured to: cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector by using the following cascading formula to obtain a dust concentration multidimensional time sequence input vector; wherein, the cascade formula is:
V c ＝Concat[V 1 ，V 2 ]
wherein V is 1 ，V 2 Representing the dust concentration absolute amount timing input vector and the dust concentration relative amount timing input vector,representing a cascade function, V c Representing the dust concentration multi-dimensional time sequence input vector.
In the electric dust removal variable frequency amplitude modulation high-voltage power supply, the first convolution neural network model and the second convolution neural network model respectively use one-dimensional convolution kernels with different scales.
In the above-mentioned electric precipitation frequency conversion amplitude modulation high voltage power supply, the double-pipeline characteristic extraction module includes: the first scale feature extraction unit is used for encoding the first convolutional neural network model to obtain a first scale dust concentration time sequence feature vector;
The second scale feature extraction unit is used for encoding the second convolutional neural network model to obtain a second scale dust concentration time sequence feature vector; and a fusion unit for fusing the first-scale dust concentration time sequence feature vector and the second-scale dust concentration time sequence feature vector to obtain the multi-scale dust concentration time sequence feature vector.
In the above electric precipitation variable frequency amplitude modulation high voltage power supply, the first scale feature extraction unit is configured to: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the first convolution neural network model to take the output of the last layer of the first convolution neural network model as a first-scale dust concentration time sequence characteristic vector, wherein the first convolution neural network model has a one-dimensional convolution kernel with a first scale.
In the above electric precipitation variable frequency amplitude modulation high voltage power supply, the second scale feature extraction unit is configured to: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the second convolution neural network model to obtain a dust concentration time sequence characteristic vector with a second scale from the output of the last layer of the second convolution neural network model, wherein the second convolution neural network model has a one-dimensional convolution kernel with a second scale, and the first scale is different from the second scale.
In the above electric precipitation variable frequency amplitude modulation high voltage power supply, the optimizing module is configured to: performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector by using the following optimization formula to obtain a decoding feature vector; wherein, the optimization formula is:
wherein μ and σ are the mean and standard deviation, v, of the feature value sets at each position in the multi-scale dust concentration time sequence feature vector i Is the characteristic value of the ith position of the multi-scale dust concentration time sequence characteristic vector, and v′ i Is the eigenvalue of the i-th position of the decoded eigenvector.
In the above electric precipitation variable frequency amplitude modulation high voltage power supply, the decoding module is configured to: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value; wherein, the decoding formula is:wherein V is d Representing the decoding eigenvector, Y representing the decoded value, W representing the weight matrix, B representing the bias vector, +.>Representing a matrix multiplication.
In a second aspect, a method for controlling electric precipitation variable frequency amplitude modulation high-voltage output is provided, which comprises the following steps:
acquiring dust concentration values at a plurality of predetermined time points within a predetermined period of time acquired by a dust concentration monitoring sensor;
Arranging dust concentration values of the plurality of preset time points into dust concentration absolute quantity time sequence input vectors according to a time dimension;
calculating the difference value between dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain a dust concentration relative value time sequence input vector;
cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector;
the dust concentration multi-dimensional time sequence input vector is processed through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector;
performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoding feature vector; and
and carrying out decoding regression on the decoding characteristic vector through a decoder to obtain a decoding value, wherein the decoding value is used for representing the frequency value of the recommended high-voltage power supply at the current time point.
In the above-mentioned electric precipitation frequency conversion amplitude modulation high voltage output control method, cascade the said dust concentration absolute quantity time sequence input vector and said dust concentration relative quantity time sequence input vector to obtain dust concentration multidimensional time sequence input vector, including: cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector by using the following cascading formula to obtain a dust concentration multidimensional time sequence input vector; wherein, the cascade formula is:
V c -Cancat[V 1 ，V 2 ]
Wherein V is 1 ，V 2 Representing the dust concentration absolute amount timing input vector and the dust concentration relative amount timing input vector,representing a cascade function, V c Representing the dust concentration multi-dimensional time sequence input vector.
Compared with the prior art, the electric dust collection frequency conversion amplitude modulation high-voltage power supply and the high-voltage output control method thereof provided by the application acquire dust concentration values at a plurality of preset time points in a preset time period acquired by the dust concentration monitoring sensor; the artificial intelligence technology based on deep learning is adopted to excavate the real-time state characteristic information of the dust concentration value, and the frequency value of the electric dust collection variable-frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristic of the dust concentration, so that the power supply has better adaptability and regulation and control capability, and the efficiency of the electric dust collection system and the ultra-clean emission level of flue gas emission are effectively improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is an application scenario diagram of an electric dust removal variable frequency amplitude modulation high voltage power supply according to an embodiment of the application.
Fig. 2 is a block diagram of an electric dust removal variable frequency amplitude modulation high voltage power supply according to an embodiment of the application.
Fig. 3 is a block diagram of the dual-line feature extraction module in the electric dust removing variable frequency amplitude modulation high voltage power supply according to the embodiment of the application.
Fig. 4 is a flowchart of a method for controlling electric dust removal variable frequency amplitude modulation high voltage output according to an embodiment of the present application.
Fig. 5 is a schematic diagram of a system architecture of an electric dust removing variable frequency amplitude modulation high voltage output control method according to an embodiment of the application.
Detailed Description
The following description of the technical solutions according to the embodiments of the present application will be given with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.
It should be noted that, the term "first\second\third" related to the embodiment of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.
With the continuous improvement of the national requirements for environmental protection, the requirements for smoke emission of coal-fired power plants are also improved. The newly issued pollutant emission standard (GB 13223-2011) of the thermal power plant is formally implemented in 1 st 2012 and 1 st, the smoke emission requirement of the coal-fired unit is greatly improved, and the smoke emission concentration in the common area is less than or equal to 30mg/Nm3; the smoke emission concentration of the key control area (a type of area) is less than or equal to 20mg/Nm3. Some developed areas require ultra-clean emission, and the smoke emission concentration is less than or equal to 10mg/Nm3. This places higher demands on the dust removal device and the electric dust removal power supply.
In order to meet the national smoke emission requirements and ultra-clean emission with higher requirements, electric dust removal power supplies are also being innovated. The current electric dust removing power supply mainly comprises a single-phase power frequency power supply, a single-phase constant current power supply (an equivalent single-phase power frequency amplitude modulation power supply), a three-phase power frequency power supply, a high-frequency power supply, a pulse power supply and the like, and a single-phase power frequency power supply, a high-frequency power supply and a pulse power supply. Currently, high-frequency power supplies are most widely used due to the highest average output voltage and the flexible pulse gap control function and other advanced technologies. The pulse power supply is the latest generation of products, has the best charging effect on fine dust with high specific resistance, and is well applied to electric fields at the tail of electric dust removal.
However, the characteristic requirement and the dust removal effect of the power supply on the first electric field of the electric dust removal are not ideal, so that the high-requirement smoke emission of the electric dust removal is greatly influenced. It can be understood that the first electric field has good dust removal effect, can greatly lighten the dust removal requirement of the subsequent electric field, and provides basic guarantee for ultra-clean emission. The first electric field has high dust concentration, is easy to generate corona sealing, and requires high average voltage and high average current. The average voltage of the high-frequency power supply is high, but the pulse width of the high-frequency current is too narrow, so that the current is smaller (lower than the power frequency power supply); the power frequency power supply (load matching, when the conduction angle is large) has larger current, but the output voltage pulsation is large, and the average voltage is low. Therefore, the power supply cannot well meet the dust removal requirement of the first electric field. The application provides an electric dust removal variable frequency (50-500 Hz) amplitude modulation high-voltage power supply, which can take the advantages of a power frequency power supply and a high-frequency power supply into account, adopts intelligent variable frequency regulation control to optimally output high average voltage and current, is used for a first electric field of electric dust removal, can ensure that corona is prevented from being sealed, and effectively improves dust removal efficiency. The power supply can also be used for a post-stage electric field, and the performance is better than that of a power frequency power supply.
In one embodiment of the application, the electric dust removal variable frequency amplitude modulation high-voltage power supply is composed of an electric control cabinet and a high-voltage rectification transformer, wherein the electric control cabinet mainly comprises the following research and development contents: an electric main loop, a control loop, a core controller (containing embedded control software), a cabinet structure and the like; the main research and development content of the high-voltage rectification transformer is as follows: the device is applicable to variable frequency (50-500 Hz) amplitude-modulated high-voltage transformers, high-voltage rectifier bridge stacks, high-voltage rectifying rheological box structures and the like.
The main key technology of the electric dust removal frequency conversion amplitude modulation high-voltage power supply comprises the following steps: (1) the electric main loop adopts an AC-DC-AC mode, and the hardware adopts a high-power IGBT to complete DC/AC frequency conversion amplitude modulation inversion; (2) designing a high-power pulse filter reactor suitable for frequency conversion and amplitude modulation; (3) the design of a high-voltage transformer meeting the frequency conversion (50-500 Hz) amplitude modulation is realized; (4) the core controller adopts SPWM sinusoidal pulse width modulation (software) to complete variable frequency amplitude modulation control based on high-speed DSP control.
The main innovation points of the application are as follows: (1) the frequency conversion amplitude modulation self-adaptive synchronous intelligent control technology can automatically adjust the frequency and amplitude along with the change of the load, so that the output voltage and current of the equipment can be maximized, and the dust removal efficiency of a first electric field (high-concentration dust) of electric dust removal can be maximized. (2) The innovative design of the high-voltage transformer adapting to dynamic changes of voltage frequency and amplitude.
In the technical scheme of the application, the main technical indexes comprise:
a) Output direct current average voltage regulation: 10-100 kV
b) Output direct current average current regulation range: 10-2000 mA
c) Primary voltage inversion frequency adjustment range: 50-500 Hz
d) Primary voltage ac amplitude adjustment range: 50-380V
e) SPWM fixed frequency control and automatic control
f) SPWM constant amplitude control and automatic control
g) Flashover frequency control range: 5-50 times/min
In one embodiment of the application, the technical route adopted is as follows: (1) And (2) designing a system scheme for technical data collection, and determining an overall scheme. (3) subsystem design: the design of a main loop and a control loop of an electrical control cabinet, the design of a hardware circuit and software of a core controller, the design of a filter reactor and the design of a cabinet body structure; the design of the variable frequency high-voltage transformer, the design of the high-voltage rectifier bridge and the design of the box body. (4) And (3) finishing the test of the key process, and establishing a project simulation debugging platform and a test platform. (5) The design is perfected on the basis of experiments, and an industrial test prototype is manufactured in a trial mode. (6) And on the basis of an industrial test prototype, the design and shaping are perfected again.
Aiming at the technical problems, the technical conception of the application is as follows: the frequency value of the electric dust collection variable frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristics of dust concentration, so that the power supply has better adaptability and regulation capacity, and the efficiency of an electric dust collection system and the ultra-clean emission level of flue gas emission can be effectively improved.
Specifically, in the technical scheme of the application, during the working process of the electric dust removal variable frequency amplitude modulation high-voltage power supply, dust concentration values at a plurality of preset time points in a preset time period acquired by a dust concentration monitoring sensor are firstly obtained. Here, in an electric dust removal system, dust concentration is a very important parameter, which directly affects the efficiency of electric dust removal and the cleanliness of the emissions. Accordingly, by collecting the dust concentration values at the plurality of preset time points, the distribution characteristics of the dust concentration values in the time dimension can be obtained, so that the frequency of the high-voltage power supply can be more accurately adjusted, and better dust removal effect and ultra-clean emission level of flue gas emission are realized.
Next, the dust concentration values at the plurality of predetermined time points are arranged in a time dimension as a dust concentration absolute amount timing input vector. Here, the dust concentration values at the plurality of predetermined time points are arranged into the dust concentration absolute amount time sequence input vector according to the time dimension because the electric dust removal system needs to consider the change of the dust concentration with time, and then adjust and optimize according to the change rule thereof. In an electric dust removal system, time is a very important parameter, since the dust concentration is not only related to the current point in time, but also to the time relation between the historical data. Accordingly, by arranging the dust concentration values at the plurality of predetermined time points in time dimension as the dust concentration absolute amount time-series input vector, a time-series model can be established so as to understand the trend of the dust concentration over time. Therefore, the dynamic change rule of the dust concentration can be mastered better, and more accurate adjustment and optimization are realized.
Then, the difference value between the dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector is calculated to obtain a dust concentration relative value time sequence input vector. Here, calculating the difference between the dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain the dust concentration relative value time sequence input vector is to describe the change trend of the dust concentration more accurately, so as to improve the prediction and recommendation precision of the electric dust removal system.
Accordingly, in the electric dust removal system, the change of the dust concentration along with time has certain regularity, and the change trend of the dust concentration can be effectively grasped by calculating the difference value between the dust concentration values of two adjacent positions. In addition, the dust concentration absolute time sequence input vector is converted into the dust concentration relative time sequence input vector, so that influence factors among different time points can be eliminated, and the relative change condition of the dust concentration can be described more accurately.
After the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector are obtained, the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector are cascaded to obtain a dust concentration multi-dimensional time sequence input vector, and the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector are fused in such a way to fully consider the influence of different factors on dust concentration change, so that more comprehensive and accurate dust concentration time sequence data are obtained.
And further, the dust concentration multi-dimensional time sequence input vector is processed through a double-pipeline characteristic extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence characteristic vector. In particular, in the technical solution of the present application, the first convolutional neural network model and the second convolutional neural network model use one-dimensional convolutional kernels having different scales, respectively. In this way, in the process of performing multi-scale one-dimensional convolutional encoding on the dust concentration multi-dimensional time sequence input vector by using the dual-pipeline characteristic extraction structure comprising the first convolutional neural network model and the second convolutional neural network model, the dual-pipeline characteristic extraction structure comprising the first convolutional neural network model and the second convolutional neural network model can capture local correlation characteristics of dust concentration absolute values and variation values in the dust concentration multi-dimensional time sequence input vector along different scales under time sequence.
In particular, in an electric dust removal system, the dust concentration changes over time with different dimensions and frequencies. By using one-dimensional convolution kernels with different scales, the influence of characteristic information with different scales can be fully considered, and the modeling capability of the model on dust concentration time sequence data is improved. Meanwhile, the double-pipeline feature extraction structure is adopted, so that the model is deeper and more complicated, and the feature extraction capacity and expression capacity of the model are further improved.
And then, carrying out decoding regression by a decoder by taking the multi-scale dust concentration time sequence characteristic vector as a decoding characteristic vector to obtain a decoding value, wherein the decoding value is used for representing the frequency value of the recommended high-voltage power supply at the current time point. That is, the decoder is used to perform a decoding regression on the multi-scale dust concentration timing feature vector to construct a mapping relationship between dust concentration multi-dimensional timing features and frequency values of a high voltage power supply. Therefore, the frequency value of the electric dust removal variable frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristics of the dust concentration, so that the power supply has better adaptability and regulation capacity, and the efficiency of an electric dust removal system and the ultra-clean emission level of smoke emission can be effectively improved.
In particular, in the technical scheme of the application, when the dust concentration multi-dimensional time sequence input vector passes through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector, the dust concentration multi-dimensional time sequence input vector is fused with the first-scale dust concentration time sequence feature vector and the second-scale dust concentration time sequence feature vector which are respectively obtained by the first convolutional neural network model and the second convolutional neural network model to obtain the multi-scale dust concentration time sequence feature vector. Here, although the first-scale dust concentration time series feature vector and the second-scale dust concentration time series feature vector can express local correlation features of different scales under the time series of the dust concentration absolute value and the variation value, feature correlation still exists, so that the probability density representation of the multi-scale dust concentration time series feature vector in a high-dimensional feature space is insufficient in dimensional distinction, regression probability expression of the multi-scale dust concentration time series feature vector is affected, and accuracy of a decoding value obtained by a decoder of the multi-scale dust concentration time series feature vector is reduced.
Therefore, the applicant of the present application orthogonalizes the multi-scale dust concentration time sequence feature vector, for example, a manifold curved surface dimension expressed as a gaussian probability density, specifically expressed as:
wherein μ and σ are the feature value set v i E means and standard deviation of V, and V i Is the characteristic value of the ith position of the optimized multi-scale dust concentration time sequence characteristic vector.
Here, by characterizing the curved surface unit tangent vector modulo length and the unit normal vector modulo length by the square root of the mean value and standard deviation of the high-dimensional feature set expressing the manifold curved surface, the manifold curved surface of the high-dimensional feature manifold of the multi-scale dust concentration time sequence feature vector V can be subjected to orthogonal projection based on the unit modulo length on the tangent plane and the normal plane, so that the dimensional reconstruction of the probability density of the high-dimensional feature is performed based on the basic structure of the gaussian feature manifold geometry, and the accuracy of the regression probability expression of the optimized multi-scale dust concentration time sequence feature vector is improved by improving the accuracy of the decoding value of the optimized multi-scale dust concentration time sequence feature vector obtained by a decoder.
Fig. 1 is an application scenario diagram of an electric dust removal variable frequency amplitude modulation high voltage power supply according to an embodiment of the application. As shown in fig. 1, in this application scenario, first, dust concentration values (e.g., C as illustrated in fig. 1) at a plurality of predetermined time points within a predetermined period of time acquired by a dust concentration monitoring sensor are acquired; the obtained dust concentration value is then input into a server (e.g., S as illustrated in fig. 1) deployed with an electro-dusting variable frequency amplitude modulated high voltage power supply algorithm, wherein the server is capable of processing the dust concentration value based on the electro-dusting variable frequency amplitude modulated high voltage power supply algorithm to generate a decoded value representative of the frequency value of the recommended high voltage power supply at the current point in time.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
In one embodiment of the application, fig. 2 is a block diagram of an electric dust removal variable frequency amplitude modulation high voltage power supply according to an embodiment of the application. As shown in fig. 2, an electric dust removing variable frequency amplitude modulation high voltage power supply 100 according to an embodiment of the present application includes: a data acquisition module 110 for acquiring dust concentration values at a plurality of predetermined time points within a predetermined period of time acquired by the dust concentration monitoring sensor; a vector arrangement module 120, configured to arrange the dust concentration values at the plurality of predetermined time points into a dust concentration absolute amount time sequence input vector according to a time dimension; a difference calculating module 130, configured to calculate a difference between dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain a dust concentration relative value time sequence input vector;
the cascade module 140 is configured to cascade the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector; the dual-pipeline feature extraction module 150 is configured to pass the dust concentration multi-dimensional time sequence input vector through a dual-pipeline feature extraction structure including a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector; the optimizing module 160 is configured to perform dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoded feature vector; and a decoding module 170, configured to perform decoding regression on the decoded feature vector through a decoder to obtain a decoded value, where the decoded value is used to represent a frequency value of the recommended high-voltage power supply at the current time point.
Specifically, in the embodiment of the present application, the data acquisition module 110 is configured to acquire dust concentration values at a plurality of predetermined time points within a predetermined period of time acquired by the dust concentration monitoring sensor. Aiming at the technical problems, the technical conception of the application is as follows: the frequency value of the electric dust collection variable frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristics of dust concentration, so that the power supply has better adaptability and regulation capacity, and the efficiency of an electric dust collection system and the ultra-clean emission level of flue gas emission can be effectively improved.
Specifically, in the technical scheme of the application, during the working process of the electric dust removal variable frequency amplitude modulation high-voltage power supply, dust concentration values at a plurality of preset time points in a preset time period acquired by a dust concentration monitoring sensor are firstly obtained. Here, in an electric dust removal system, dust concentration is a very important parameter, which directly affects the efficiency of electric dust removal and the cleanliness of the emissions. Accordingly, by collecting the dust concentration values at the plurality of preset time points, the distribution characteristics of the dust concentration values in the time dimension can be obtained, so that the frequency of the high-voltage power supply can be more accurately adjusted, and better dust removal effect and ultra-clean emission level of flue gas emission are realized.
Specifically, in the embodiment of the present application, the vector arrangement module 120 is configured to arrange the dust concentration values at the plurality of predetermined time points into the dust concentration absolute amount timing input vector according to a time dimension. Next, the dust concentration values at the plurality of predetermined time points are arranged in a time dimension as a dust concentration absolute amount timing input vector. Here, the dust concentration values at the plurality of predetermined time points are arranged into the dust concentration absolute amount time sequence input vector according to the time dimension because the electric dust removal system needs to consider the change of the dust concentration with time, and then adjust and optimize according to the change rule thereof. In an electric dust removal system, time is a very important parameter, since the dust concentration is not only related to the current point in time, but also to the time relation between the historical data.
Accordingly, by arranging the dust concentration values at the plurality of predetermined time points in time dimension as the dust concentration absolute amount time-series input vector, a time-series model can be established so as to understand the trend of the dust concentration over time. Therefore, the dynamic change rule of the dust concentration can be mastered better, and more accurate adjustment and optimization are realized.
Specifically, in the embodiment of the present application, the difference calculating module 130 is configured to calculate a difference between the dust concentration values of each two adjacent positions in the dust concentration absolute value timing input vector to obtain a dust concentration relative value timing input vector. Then, the difference value between the dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector is calculated to obtain a dust concentration relative value time sequence input vector. Here, calculating the difference between the dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain the dust concentration relative value time sequence input vector is to describe the change trend of the dust concentration more accurately, so as to improve the prediction and recommendation precision of the electric dust removal system.
Accordingly, in the electric dust removal system, the change of the dust concentration along with time has certain regularity, and the change trend of the dust concentration can be effectively grasped by calculating the difference value between the dust concentration values of two adjacent positions. In addition, the dust concentration absolute time sequence input vector is converted into the dust concentration relative time sequence input vector, so that influence factors among different time points can be eliminated, and the relative change condition of the dust concentration can be described more accurately.
Specifically, in the embodiment of the present application, the cascade module 140 is configured to cascade the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector. After the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector are obtained, the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector are cascaded to obtain a dust concentration multi-dimensional time sequence input vector, and the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector are fused in such a way to fully consider the influence of different factors on dust concentration change, so that more comprehensive and accurate dust concentration time sequence data are obtained.
Wherein, the cascade module 140 is configured to: cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector by using the following cascading formula to obtain a dust concentration multidimensional time sequence input vector; wherein, the cascade formula is:
V c -Concat[V 1 ，V 2 ]
wherein V is 1 ，V 2 Representing the dust concentration absolute amount timing input vector and the dust concentration relative amount timing input vector, Representing a cascade function, V c Representing the dust concentration multi-dimensional time sequence input vector.
Specifically, in the embodiment of the present application, the dual-pipeline feature extraction module 150 is configured to pass the multi-dimensional time sequence input vector of dust concentration through a dual-pipeline feature extraction structure including a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale time sequence feature vector of dust concentration. And further, the dust concentration multi-dimensional time sequence input vector is processed through a double-pipeline characteristic extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence characteristic vector.
In particular, in the technical solution of the present application, the first convolutional neural network model and the second convolutional neural network model use one-dimensional convolutional kernels having different scales, respectively. In this way, in the process of performing multi-scale one-dimensional convolutional encoding on the dust concentration multi-dimensional time sequence input vector by using the dual-pipeline characteristic extraction structure comprising the first convolutional neural network model and the second convolutional neural network model, the dual-pipeline characteristic extraction structure comprising the first convolutional neural network model and the second convolutional neural network model can capture local correlation characteristics of dust concentration absolute values and variation values in the dust concentration multi-dimensional time sequence input vector along different scales under time sequence.
In particular, in an electric dust removal system, the dust concentration changes over time with different dimensions and frequencies. By using one-dimensional convolution kernels with different scales, the influence of characteristic information with different scales can be fully considered, and the modeling capability of the model on dust concentration time sequence data is improved. Meanwhile, the double-pipeline feature extraction structure is adopted, so that the model is deeper and more complicated, and the feature extraction capacity and expression capacity of the model are further improved.
Wherein the first convolutional neural network model and the second convolutional neural network model each use one-dimensional convolutional kernels having different scales.
Fig. 3 is a block diagram of the dual-line feature extraction module in the electric dust removing variable frequency amplitude modulation high voltage power supply according to an embodiment of the present application, and as shown in fig. 3, the dual-line feature extraction module 150 includes: a first scale feature extraction unit 151, configured to encode the first convolutional neural network model to obtain a first scale dust concentration time sequence feature vector; the second scale feature extraction unit 152 is configured to encode the second convolutional neural network model to obtain a second scale dust concentration time sequence feature vector; and a fusion unit 153 for fusing the first-scale dust concentration time-series feature vector and the second-scale dust concentration time-series feature vector to obtain the multi-scale dust concentration time-series feature vector.
It should be noted that the multi-scale neighborhood feature extraction module is essentially a deep neural network model based on deep learning, which is capable of fitting any function by a predetermined training strategy and has a higher feature extraction generalization capability compared to the conventional feature engineering.
The multi-scale neighborhood feature extraction module comprises a plurality of parallel one-dimensional convolution layers, wherein in the process of feature extraction by the multi-scale neighborhood feature extraction module, the plurality of parallel one-dimensional convolution layers respectively carry out one-dimensional convolution coding on input data by one-dimensional convolution check with different scales so as to capture local implicit features of a sequence.
The first scale feature extraction unit 151 is configured to: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the first convolution neural network model to take the output of the last layer of the first convolution neural network model as a first-scale dust concentration time sequence characteristic vector, wherein the first convolution neural network model has a one-dimensional convolution kernel with a first scale.
The second scale feature extraction unit 152 is configured to: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the second convolution neural network model to obtain a dust concentration time sequence characteristic vector with a second scale from the output of the last layer of the second convolution neural network model, wherein the second convolution neural network model has a one-dimensional convolution kernel with a second scale, and the first scale is different from the second scale.
The convolutional neural network (Convolutional Neural Network, CNN) is an artificial neural network and has wide application in the fields of image recognition and the like. The convolutional neural network may include an input layer, a hidden layer, and an output layer, where the hidden layer may include a convolutional layer, a pooling layer, an activation layer, a full connection layer, etc., where the previous layer performs a corresponding operation according to input data, outputs an operation result to the next layer, and obtains a final result after the input initial data is subjected to a multi-layer operation.
The convolutional neural network model has excellent performance in the aspect of image local feature extraction by taking a convolutional kernel as a feature filtering factor, and has stronger feature extraction generalization capability and fitting capability compared with the traditional image feature extraction algorithm based on statistics or feature engineering.
Specifically, in the embodiment of the present application, the optimization module 160 is configured to perform dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoded feature vector. In particular, in the technical scheme of the application, when the dust concentration multi-dimensional time sequence input vector passes through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector, the dust concentration multi-dimensional time sequence input vector is fused with the first-scale dust concentration time sequence feature vector and the second-scale dust concentration time sequence feature vector which are respectively obtained by the first convolutional neural network model and the second convolutional neural network model to obtain the multi-scale dust concentration time sequence feature vector. Here, although the first-scale dust concentration time series feature vector and the second-scale dust concentration time series feature vector can express local correlation features of different scales under the time series of the dust concentration absolute value and the variation value, feature correlation still exists, so that the probability density representation of the multi-scale dust concentration time series feature vector in a high-dimensional feature space is insufficient in dimensional distinction, regression probability expression of the multi-scale dust concentration time series feature vector is affected, and accuracy of a decoding value obtained by a decoder of the multi-scale dust concentration time series feature vector is reduced.
Therefore, the applicant of the present application orthogonalizes the multi-scale dust concentration time sequence feature vector, for example, a manifold curved surface dimension expressed as a gaussian probability density, specifically expressed as: performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector by using the following optimization formula to obtain a decoding feature vector; wherein, the optimization formula is:
wherein μ and σ are the mean and standard deviation, v, of the feature value sets at each position in the multi-scale dust concentration time sequence feature vector i Is the eigenvalue of the ith position of the multi-scale dust concentration time sequence eigenvector, and v' i is the eigenvalue of the ith position of the decoding eigenvector.
Here, by characterizing the curved surface unit tangent vector modulo length and the unit normal vector modulo length by the square root of the mean value and standard deviation of the high-dimensional feature set expressing the manifold curved surface, the manifold curved surface of the high-dimensional feature manifold of the multi-scale dust concentration time sequence feature vector V can be subjected to orthogonal projection based on the unit modulo length on the tangent plane and the normal plane, so that the dimensional reconstruction of the probability density of the high-dimensional feature is performed based on the basic structure of the gaussian feature manifold geometry, and the accuracy of the regression probability expression of the optimized multi-scale dust concentration time sequence feature vector is improved by improving the accuracy of the decoding value of the optimized multi-scale dust concentration time sequence feature vector obtained by a decoder.
Specifically, in the embodiment of the present application, the decoding module 170 is configured to perform decoding regression on the decoded feature vector through a decoder to obtain a decoded value, where the decoded value is used to represent a frequency value of the recommended high-voltage power supply at the current time point. And then, carrying out decoding regression by a decoder by taking the multi-scale dust concentration time sequence characteristic vector as a decoding characteristic vector to obtain a decoding value, wherein the decoding value is used for representing the frequency value of the recommended high-voltage power supply at the current time point. That is, the decoder is used to perform a decoding regression on the multi-scale dust concentration timing feature vector to construct a mapping relationship between dust concentration multi-dimensional timing features and frequency values of a high voltage power supply. Therefore, the frequency value of the electric dust removal variable frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristics of the dust concentration, so that the power supply has better adaptability and regulation capacity, and the efficiency of an electric dust removal system and the ultra-clean emission level of smoke emission can be effectively improved.
Wherein, the decoding module 170 is configured to: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value; wherein, the decoding formula is: Wherein V is d Representing the decoding eigenvector, Y representing the decoded value, W representing the weight matrix, B representing the bias vector, +.>Representing a matrix multiplication.
In summary, an electric dust collection variable frequency amplitude modulation high voltage power supply 100 according to an embodiment of the present application is illustrated, which acquires dust concentration values at a plurality of predetermined time points within a predetermined period of time acquired by a dust concentration monitoring sensor; the artificial intelligence technology based on deep learning is adopted to excavate the real-time state characteristic information of the dust concentration value, and the frequency value of the electric dust collection variable-frequency amplitude modulation high-voltage power supply is adaptively adjusted based on the real-time state characteristic of the dust concentration, so that the power supply has better adaptability and regulation and control capability, and the efficiency of the electric dust collection system and the ultra-clean emission level of flue gas emission are effectively improved.
In one embodiment of the present application, fig. 4 is a flowchart of a method for controlling the electric dust removal variable frequency amplitude modulation high voltage output according to an embodiment of the present application. As shown in fig. 4, the method for controlling the electric dust removal variable frequency amplitude modulation high-voltage output according to the embodiment of the application comprises the following steps: 210, acquiring dust concentration values at a plurality of preset time points in a preset time period acquired by a dust concentration monitoring sensor; 220, arranging dust concentration values of the plurality of preset time points into dust concentration absolute quantity time sequence input vectors according to a time dimension; 230, calculating the difference value between dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain a dust concentration relative value time sequence input vector; 240, cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector; 250, passing the dust concentration multi-dimensional time sequence input vector through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector; 260, performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoding feature vector; and 270, performing decoding regression on the decoding feature vector through a decoder to obtain a decoding value, wherein the decoding value is used for representing the frequency value of the recommended high-voltage power supply at the current time point.
Fig. 5 is a schematic diagram of a system architecture of an electric dust removing variable frequency amplitude modulation high voltage output control method according to an embodiment of the application. As shown in fig. 5, in the system architecture of the electric dust removal variable frequency amplitude modulation high voltage output control method, firstly, dust concentration values at a plurality of predetermined time points in a predetermined time period acquired by a dust concentration monitoring sensor are acquired; then, arranging dust concentration values of the plurality of preset time points into dust concentration absolute quantity time sequence input vectors according to a time dimension; then, calculating the difference value between dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain a dust concentration relative value time sequence input vector; then, cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector; then, the dust concentration multi-dimensional time sequence input vector is processed through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector; then, carrying out dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoding feature vector; and finally, carrying out decoding regression on the decoding characteristic vector through a decoder to obtain a decoding value, wherein the decoding value is used for representing the frequency value of the recommended high-voltage power supply at the current time point.
In a specific example, in the above electric precipitation variable frequency amplitude modulation high voltage output control method, cascading the dust concentration absolute amount time sequence input vector and the dust concentration relative amount time sequence input vector to obtain a dust concentration multi-dimensional time sequence input vector includes: cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector by using the following cascading formula to obtain a dust concentration multidimensional time sequence input vector; wherein, the cascade formula is:
V c ＝Ccncat[V 1 ，V 2 ]
wherein V is 1 ，V 2 Representing the dust concentration absolute amount timing input vector and the dust concentration relative amount timing input vector,representing a cascade function, V c Representing the dust concentration multi-dimensional time sequence input vector.
In a specific example, in the above electric precipitation variable frequency amplitude modulation high voltage output control method, the first convolutional neural network model and the second convolutional neural network model use one-dimensional convolutional kernels with different scales, respectively.
In a specific example, in the above electric precipitation variable frequency amplitude modulation high voltage output control method, passing the dust concentration multi-dimensional time sequence input vector through a dual-pipeline feature extraction structure including a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector includes: encoding the first convolutional neural network model to obtain a first scale dust concentration time sequence feature vector; coding the second convolutional neural network model to obtain a second scale dust concentration time sequence feature vector; and fusing the first-scale dust concentration time sequence feature vector and the second-scale dust concentration time sequence feature vector to obtain the multi-scale dust concentration time sequence feature vector.
In a specific example, in the above method for controlling electric dust removal variable frequency amplitude modulation high voltage output, the encoding of the first convolutional neural network model to obtain the first scale dust concentration time sequence feature vector includes: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the first convolution neural network model to take the output of the last layer of the first convolution neural network model as a first-scale dust concentration time sequence characteristic vector, wherein the first convolution neural network model has a one-dimensional convolution kernel with a first scale.
In a specific example, in the above method for controlling electric dust removal variable frequency amplitude modulation high voltage output, the encoding of the second convolutional neural network model to obtain the second scale dust concentration time sequence feature vector includes: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the second convolution neural network model to obtain a dust concentration time sequence characteristic vector with a second scale from the output of the last layer of the second convolution neural network model, wherein the second convolution neural network model has a one-dimensional convolution kernel with a second scale, and the first scale is different from the second scale.
In a specific example, in the above electric precipitation variable frequency amplitude modulation high voltage output control method, performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoded feature vector, including: performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector by using the following optimization formula to obtain a decoding feature vector; wherein, the optimization formula is:
wherein μ and σ are the multi-scale dustMean and standard deviation of feature value sets of each position in concentration time sequence feature vector, v i Is the characteristic value of the ith position of the multi-scale dust concentration time sequence characteristic vector, and v' i Is the eigenvalue of the i-th position of the decoded eigenvector.
In a specific example, in the above electric precipitation variable frequency amplitude modulation high voltage output control method, the decoding regression is performed on the decoding feature vector by a decoder to obtain a decoding value, where the decoding value is used to represent a frequency value of a recommended high voltage power supply at a current time point, and the method includes: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value; wherein, the decoding formula is: Wherein V is d Representing the decoding eigenvector, Y representing the decoded value, W representing the weight matrix, B representing the bias vector, +.>Representing a matrix multiplication.
It will be appreciated by those skilled in the art that the specific operation of the steps in the above-described electric dust removing variable frequency amplitude modulation high voltage output control method has been described in detail in the above description of the electric dust removing variable frequency amplitude modulation high voltage power supply with reference to fig. 1 to 3, and thus, repetitive descriptions thereof will be omitted.
The present application also provides a computer program product comprising instructions which, when executed, cause an apparatus to perform operations corresponding to the above-described method.
In one embodiment of the present application, there is also provided a computer-readable storage medium storing a computer program for executing the above-described method.
It should be appreciated that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the forms of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects may be utilized. Furthermore, the computer program product may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Methods, systems, and computer program products of embodiments of the present application are described in the flow diagrams and/or block diagrams. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.
The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.

Claims (9)

1. An electric dust removal variable frequency amplitude modulation high voltage power supply, characterized by comprising:
the data acquisition module is used for acquiring dust concentration values at a plurality of preset time points in a preset time period acquired by the dust concentration monitoring sensor;
the vector arrangement module is used for arranging the dust concentration values of the plurality of preset time points into dust concentration absolute quantity time sequence input vectors according to a time dimension;
the difference value calculation module is used for calculating the difference value between dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain a dust concentration relative value time sequence input vector;
the cascading module is used for cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector;
the double-pipeline feature extraction module is used for enabling the dust concentration multi-dimensional time sequence input vector to pass through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model so as to obtain a multi-scale dust concentration time sequence feature vector;
The optimizing module is used for carrying out dimension distinguishing degree strengthening on the multi-scale dust concentration time sequence feature vector so as to obtain a decoding feature vector; and
the decoding module is used for carrying out decoding regression on the decoding characteristic vector through a decoder to obtain a decoding value, wherein the decoding value is used for representing the frequency value of the recommended high-voltage power supply at the current time point;
wherein, the optimization module is used for: performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector by using the following optimization formula to obtain a decoding feature vector;
wherein, the optimization formula is:
wherein μ and σ are the mean and standard deviation, v, of the feature value sets at each position in the multi-scale dust concentration time sequence feature vector i Is the characteristic value of the ith position of the multi-scale dust concentration time sequence characteristic vector, and v i ， Is the eigenvalue of the i-th position of the decoded eigenvector.
2. The electric dust removal variable frequency amplitude modulation high voltage power supply according to claim 1, wherein the cascade module is configured to: cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector by using the following cascading formula to obtain a dust concentration multidimensional time sequence input vector;
Wherein, the cascade formula is:
wherein V is 1 ，V 2 A timing input vector representing the absolute amount of dust concentration and a timing input vector representing the relative amount of dust concentration, cancat [. Cndot. ]]Representing a cascade function, V c Representing the dust concentration multi-dimensional time sequence input vector.
3. The electro-dedusting variable frequency amplitude modulation high voltage power supply according to claim 2, wherein the first convolutional neural network model and the second convolutional neural network model each use one-dimensional convolutional kernels with different dimensions.
4. The electric dust removal variable frequency amplitude modulation high voltage power supply according to claim 3, wherein the dual-line feature extraction module comprises:
the first scale feature extraction unit is used for encoding the first convolutional neural network model to obtain a first scale dust concentration time sequence feature vector;
the second scale feature extraction unit is used for encoding the second convolutional neural network model to obtain a second scale dust concentration time sequence feature vector; and
and the fusion unit is used for fusing the first-scale dust concentration time sequence feature vector and the second-scale dust concentration time sequence feature vector to obtain the multi-scale dust concentration time sequence feature vector.
5. The electric dust removal variable frequency amplitude modulation high voltage power supply according to claim 4, wherein the first scale feature extraction unit is configured to: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the first convolution neural network model to take the output of the last layer of the first convolution neural network model as a first-scale dust concentration time sequence characteristic vector, wherein the first convolution neural network model has a one-dimensional convolution kernel with a first scale.
6. The electric dust removal variable frequency amplitude modulation high voltage power supply according to claim 5, wherein the second scale feature extraction unit is configured to: and respectively carrying out convolution processing, pooling processing and linear activation processing on the dust concentration multi-dimensional time sequence input vector based on a one-dimensional convolution kernel in forward transfer of layers by using each layer of the second convolution neural network model to obtain a dust concentration time sequence characteristic vector with a second scale from the output of the last layer of the second convolution neural network model, wherein the second convolution neural network model has a one-dimensional convolution kernel with a second scale, and the first scale is different from the second scale.
7. The electric dust removal variable frequency amplitude modulation high voltage power supply according to claim 6, wherein the decoding module is configured to: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value;
wherein, the decoding formula is:wherein V is d Representing the decoding eigenvector, Y representing the decoded value, W representing the weight matrix, B representing the bias vector, +.>Representing a matrix multiplication.
8. The electric precipitation frequency conversion amplitude modulation high-voltage output control method is characterized by comprising the following steps of:
acquiring dust concentration values at a plurality of predetermined time points within a predetermined period of time acquired by a dust concentration monitoring sensor;
arranging dust concentration values of the plurality of preset time points into dust concentration absolute quantity time sequence input vectors according to a time dimension;
calculating the difference value between dust concentration values of every two adjacent positions in the dust concentration absolute value time sequence input vector to obtain a dust concentration relative value time sequence input vector;
cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector;
the dust concentration multi-dimensional time sequence input vector is processed through a double-pipeline feature extraction structure comprising a first convolutional neural network model and a second convolutional neural network model to obtain a multi-scale dust concentration time sequence feature vector;
Performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoding feature vector; and
performing decoding regression on the decoding characteristic vector through a decoder to obtain a decoding value, wherein the decoding value is used for representing a frequency value of a recommended high-voltage power supply at the current time point;
the step of performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector to obtain a decoding feature vector comprises the following steps: performing dimension discrimination enhancement on the multi-scale dust concentration time sequence feature vector by using the following optimization formula to obtain a decoding feature vector;
wherein, the optimization formula is:
wherein μ and σ are the mean and standard deviation, v, of the feature value sets at each position in the multi-scale dust concentration time sequence feature vector i Is the characteristic value of the ith position of the multi-scale dust concentration time sequence characteristic vector, and v i ， Is the eigenvalue of the i-th position of the decoded eigenvector.
9. The method according to claim 8, wherein concatenating the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector to obtain a dust concentration multidimensional time sequence input vector, comprises: cascading the dust concentration absolute time sequence input vector and the dust concentration relative time sequence input vector by using the following cascading formula to obtain a dust concentration multidimensional time sequence input vector;
Wherein, the cascade formula is:
wherein V is 1 ，V 2 A Concat [. Cndot.]Representing a cascade function, V c Representing the dust concentration multi-dimensional time sequence input vector.
CN202310503249.3A 2023-05-06 2023-05-06 Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof Active CN116651617B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310503249.3A CN116651617B (en) 2023-05-06 2023-05-06 Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310503249.3A CN116651617B (en) 2023-05-06 2023-05-06 Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof

Publications (2)

Publication Number Publication Date
CN116651617A CN116651617A (en) 2023-08-29
CN116651617B true CN116651617B (en) 2023-11-14

Family

ID=87721499

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310503249.3A Active CN116651617B (en) 2023-05-06 2023-05-06 Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof

Country Status (1)

Country Link
CN (1) CN116651617B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118136163A (en) * 2024-04-26 2024-06-04 山西国辰建设工程勘察设计有限公司 Intelligent dust removal method and system for coal mining working face

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1997035667A1 (en) * 1996-03-28 1997-10-02 ABB Fläkt Aktiebolag Method for controlling an electrostatic precipitator
WO2010001925A1 (en) * 2008-07-01 2010-01-07 新日本製鐵株式会社 Method for estimating amount of dustfall, device for estimating amount of dustfall, and program for estimating amount of dustfall
CN106321487A (en) * 2016-11-25 2017-01-11 中铁隧道集团有限公司 Three-phase type tunnel construction ventilation control method
CN106861915A (en) * 2017-02-09 2017-06-20 广东松山职业技术学院 Electric cleaner intelligent monitoring diagnostic control system based on GA technologies
CN108940596A (en) * 2018-07-12 2018-12-07 浙江大学 A kind of electric dust removing system optimal control method based on mixed model
CN112099343A (en) * 2020-07-29 2020-12-18 福建龙净环保股份有限公司 Intelligent energy-saving optimization method of electric precipitation system based on neural network
CN112818033A (en) * 2021-01-28 2021-05-18 河北工业大学 Bag breaking intelligent detection method of bag type dust collector based on neural network
KR20220138250A (en) * 2021-04-05 2022-10-12 한국전자통신연구원 A method and an electronic device for inferring occurrence of highly-concentrated fine dust
CN115453945A (en) * 2022-09-26 2022-12-09 北京科技大学 Comprehensive digging face intelligent ventilation control system based on mobile environment detection device
CN115931665A (en) * 2022-12-30 2023-04-07 中国矿业大学 Dust monitoring system and method based on underground WiFi6 communication network

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10245595B2 (en) * 2014-06-13 2019-04-02 Flsmidth A/S Controlling a high voltage power supply for an electrostatic precipitator

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1997035667A1 (en) * 1996-03-28 1997-10-02 ABB Fläkt Aktiebolag Method for controlling an electrostatic precipitator
WO2010001925A1 (en) * 2008-07-01 2010-01-07 新日本製鐵株式会社 Method for estimating amount of dustfall, device for estimating amount of dustfall, and program for estimating amount of dustfall
CN106321487A (en) * 2016-11-25 2017-01-11 中铁隧道集团有限公司 Three-phase type tunnel construction ventilation control method
CN106861915A (en) * 2017-02-09 2017-06-20 广东松山职业技术学院 Electric cleaner intelligent monitoring diagnostic control system based on GA technologies
CN108940596A (en) * 2018-07-12 2018-12-07 浙江大学 A kind of electric dust removing system optimal control method based on mixed model
CN112099343A (en) * 2020-07-29 2020-12-18 福建龙净环保股份有限公司 Intelligent energy-saving optimization method of electric precipitation system based on neural network
CN112818033A (en) * 2021-01-28 2021-05-18 河北工业大学 Bag breaking intelligent detection method of bag type dust collector based on neural network
KR20220138250A (en) * 2021-04-05 2022-10-12 한국전자통신연구원 A method and an electronic device for inferring occurrence of highly-concentrated fine dust
CN115453945A (en) * 2022-09-26 2022-12-09 北京科技大学 Comprehensive digging face intelligent ventilation control system based on mobile environment detection device
CN115931665A (en) * 2022-12-30 2023-04-07 中国矿业大学 Dust monitoring system and method based on underground WiFi6 communication network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于 GA-BP 神经网络的矿井节能通风控制系统 设计;宿敬肖;煤炭技术;第41卷(第12期);全文 *
电除尘器高频电源和脉冲电源应用分析;周建伟;顾立群;;能源研究与信息(第03期);全文 *

Also Published As

Publication number Publication date
CN116651617A (en) 2023-08-29

Similar Documents

Publication Publication Date Title
CN116651617B (en) Electric dust removal variable frequency amplitude modulation high-voltage power supply and high-voltage output control method thereof
Katić et al. Smart detection of voltage dips using voltage harmonics footprint
WO2021258810A1 (en) Converter parameterized constant admittance modeling method based on cross initialization
CN105938578A (en) Large-scale photovoltaic power station equivalent modeling method based on clustering analysis
CN110045610A (en) Inverter modified multistep model predictive control method, equipment and storage equipment
CN204615631U (en) A kind of busbar voltage ripple compensation control circuit of circuit of power factor correction
CN113285481B (en) Grid-connected converter inductance parameter online estimation method, prediction control method and system
CN103365213A (en) Extreme value optimization self-adjusting digital PID control method used for megawatt inverter system
CN104715342A (en) Electrical equipment fault handling and aid decision making method based on case-based reasoning
CN106849705A (en) A kind of current predictive control method based on single-phase Vienna rectifiers
CN103018583A (en) Level number selecting and verifying method based on flexible direct-current transmission system of MMC (Modular Multilevel Converter)
CN111193295A (en) Distribution network flexibility improvement robust optimization scheduling method considering dynamic reconfiguration
Li et al. Fill missing data for wind farms using long short-term memory based recurrent neural network
Cui et al. T-type inverter fault diagnosis based on GASF and improved AlexNet
Kim et al. Intelligent tuning of PID controller with disturbance function using immune algorithm
Wu et al. Classification of power loads based on an improved denoising deconvolutional auto-encoder
CN112861315B (en) One-dimensional descent search method for non-convex single-target optimal power flow global solution of power system
CN110968073B (en) Double-layer tracing identification method for commutation failure reasons of HVDC system
CN104868742B (en) It is a kind of to realize the virtual power control method that full-bridge isolates DC DC converter decompression transformation pattern minimum current stress
CN110707762B (en) Multi-energy complementary alternating current-direct current hybrid micro-grid load flow feasible region calculation method
Hamdy et al. Development of optimal pi controllers of an inverter-based decentralized energy generation system based on equilibrium optimization algorithm
CN113346501A (en) Power distribution network voltage optimization method and system based on brainstorming algorithm
Liu et al. An Integrated Optimization Design Method of Single-Phase PV Inverter Based on Machine Learning
Chen et al. Research on fault diagnosis method of power transformer based on graph neural network
Yu et al. Real-Time Monitoring Method for Thyristor Losses in Ultra High Voltage Converter Station Based on Wavelet Optimized GA-BP Neural Network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant