CN115576203B - Embedded electric precipitation intelligent control method and system based on neural network - Google Patents
Embedded electric precipitation intelligent control method and system based on neural network Download PDFInfo
- Publication number
- CN115576203B CN115576203B CN202211278780.7A CN202211278780A CN115576203B CN 115576203 B CN115576203 B CN 115576203B CN 202211278780 A CN202211278780 A CN 202211278780A CN 115576203 B CN115576203 B CN 115576203B
- Authority
- CN
- China
- Prior art keywords
- neural network
- parameters
- dust removing
- dust
- concentration
- 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
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 153
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000001556 precipitation Methods 0.000 title claims description 6
- 239000000428 dust Substances 0.000 claims abstract description 267
- 238000004891 communication Methods 0.000 claims abstract description 39
- 238000005457 optimization Methods 0.000 claims abstract description 37
- 238000007781 pre-processing Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 13
- 238000013507 mapping Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 11
- 238000013500 data storage Methods 0.000 description 10
- 230000005611 electricity Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000003546 flue gas Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Automation & Control Theory (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Feedback Control In General (AREA)
Abstract
The application provides an embedded electric dust removal intelligent control method and system based on a neural network, wherein the system comprises an upper computer and a plurality of lower computers which are connected in a communication way and are arranged on a plurality of dust removal devices, and the method comprises the following steps: the upper computer is used for constructing a first neural network for predicting the concentration of the outlet dust; the upper computer performs optimization based on the first neural network by taking the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of a plurality of dust removing devices is minimized, obtains a first operation parameter and sends the first operation parameter to the lower computer; the lower computer obtains a first control parameter of the dust removing equipment by optimizing with the aim that the actual operation parameter is matched with the first operation parameter and the estimated power consumption of the dust removing equipment is minimized based on the second neural network, and controls the operation of the dust removing equipment according to the first control parameter. According to the scheme, the neural network is utilized, the control parameter optimization is carried out according to the fact that the concentration of the outlet dust reaches the standard and the power consumption is minimized, the optimal first control parameter is obtained, and the power consumption of the dust removing equipment is effectively reduced.
Description
Technical Field
The invention relates to the technical field of flue gas treatment, in particular to an embedded electric dust removal intelligent control method and system based on a neural network.
Background
With the increase of industrialization level, the pollution of natural environment is also more serious. As an important technology for treating dust pollution in the environment, the electric dust removal technology is widely applied to industries such as environmental protection, power generation and the like.
However, in order to ensure that the dust removal effect can meet the related requirements, the conventional electric dust removal device often consumes a large amount of electric energy, and the operation cost is too high.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an embedded electric dust removal intelligent control method and system based on a neural network, so as to provide an electric dust removal scheme with low power consumption.
The application provides an intelligent control method for embedded electric dust removal based on a neural network, which is applied to an intelligent control system for embedded electric dust removal based on the neural network, wherein the system comprises an upper computer and a plurality of embedded lower computers arranged on a plurality of dust removal devices, and the upper computer is in communication connection with each lower computer, and the method comprises the following steps:
The upper computer constructs a first neural network for predicting the concentration of the outlet dust according to the global parameter in a preset time period and the operation parameters of the plurality of dust removing devices; wherein the global parameter comprises at least one of boiler load and outlet dust concentration; the operating parameters include at least one of a voltage average, a voltage peak, and a current value;
The upper computer operates a group intelligent optimization algorithm based on the first neural network with the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of the plurality of dust removing devices is minimized, so as to obtain a first operation parameter;
The upper computer sends the first operation parameters to each lower computer;
The lower computer operates a group intelligent optimization algorithm based on a second neural network constructed in advance, and with the aim that actual operation parameters are matched with the first operation parameters and the estimated power consumption of dust removing equipment to which the lower computer belongs is minimized, so as to obtain first control parameters of the dust removing equipment to which the lower computer belongs; wherein the first control parameter includes at least one of an intermittent off time, a current limit, and a voltage limit; the second neural network is constructed according to control parameters and operation parameters of dust removing equipment to which the lower computer belongs in the preset time period;
And the lower computer controls the operation of the dust removing equipment according to the first control parameter of the dust removing equipment.
Optionally, before the upper computer constructs the first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices, the method further includes:
the upper computer monitors the working state of each dust removing device in real time to obtain the original data in the preset time period;
The upper computer performs data preprocessing on the original data to obtain global parameters in the preset time period and operation parameters of the plurality of dust removing devices; wherein the data preprocessing includes at least one of data cleansing, object mapping, and data assembling.
Optionally, the sending, by the upper computer, the first operation parameter to each lower computer includes:
The upper computer analyzes and packages the first operation parameters to obtain control instructions carrying the first operation parameters;
And the upper computer sends a control instruction carrying the first operation parameter to each lower computer.
Optionally, the communication connection between the upper computer and the lower computer is direct communication connection or indirect communication connection.
Optionally, the upper computer constructs a first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices, and the method includes:
The upper computer obtains a neural network to be trained;
The upper computer inputs the operation parameters and the global parameters in the preset time period into the neural network to be trained to obtain the estimated outlet dust concentration output by the neural network to be trained, and iteratively updates the neural network to be trained by taking the outlet dust concentration of which the estimated outlet dust concentration is matched with the outlet dust concentration of the global parameters as a target until the deviation of the estimated outlet dust concentration and the outlet dust concentration of the global parameters meets a preset first convergence condition to obtain a first neural network.
The application provides an embedded electric dust collection intelligent control system based on a neural network, which comprises an upper computer and a plurality of embedded lower computers arranged on a plurality of dust collection devices, wherein the upper computer is in communication connection with each lower computer;
the upper computer comprises a decision module and an output control module;
The decision module is used for:
Constructing a first neural network for predicting the concentration of the outlet dust according to the global parameter in a preset time period and the operation parameters of the plurality of dust removing devices; wherein the global parameter comprises at least one of boiler load and outlet dust concentration; the operating parameters include at least one of a voltage average, a voltage peak, and a current value;
based on the first neural network, operating a group intelligent optimization algorithm with the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of the plurality of dust removing devices is minimized, so as to obtain a first operation parameter;
the output control module is used for sending the first operation parameters to each lower computer;
the lower computer comprises an energy-saving decision module and an execution module;
The energy-saving decision module is used for:
Based on a pre-constructed second neural network, running a group intelligent optimization algorithm with the aim that actual running parameters are matched with the first running parameters and the expected power consumption of dust removing equipment to which the lower computer belongs is minimized, so as to obtain first control parameters of the dust removing equipment to which the lower computer belongs; wherein the first control parameter includes at least one of an intermittent off time, a current limit, and a voltage limit; the second neural network is constructed according to control parameters and operation parameters of dust removing equipment to which the lower computer belongs in the preset time period;
the execution module is used for: and controlling the operation of the dust removing equipment according to the first control parameter of the dust removing equipment.
Optionally, the upper computer further comprises an acquisition module and a data processing module;
The acquisition module is used for monitoring the working state of each dust removing device in real time and obtaining the original data in the preset time period;
The acquisition module is used for carrying out data preprocessing on the original data to obtain global parameters in the preset time period and operation parameters of the plurality of dust removing devices; wherein the data preprocessing includes at least one of data cleansing, object mapping, and data assembling.
Optionally, when the output control module sends the first operation parameter to each lower computer, the output control module is specifically configured to:
analyzing and packaging the first operation parameters to obtain control instructions carrying the first operation parameters;
and sending a control instruction carrying the first operation parameters to each lower computer.
Optionally, the communication connection between the upper computer and the lower computer is direct communication connection or indirect communication connection.
Optionally, the decision module is configured to, when constructing the first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices, specifically:
obtaining a neural network to be trained;
Inputting the operation parameters and the global parameters in the preset time period into the neural network to be trained to obtain the estimated outlet dust concentration output by the neural network to be trained, and iteratively updating the neural network to be trained by taking the outlet dust concentration of which the estimated outlet dust concentration is matched with the outlet dust concentration of the global parameters as a target until the deviation of the estimated outlet dust concentration and the outlet dust concentration of the global parameters meets a preset first convergence condition to obtain a first neural network.
The embodiment of the application provides an embedded electric dust removal intelligent control method and system based on a neural network, which are applied to the embedded electric dust removal intelligent control system based on the neural network, wherein the system comprises an upper computer and a plurality of embedded lower computers arranged on a plurality of dust removal devices, and the upper computer is in communication connection with each lower computer, and the method comprises the following steps: the upper computer constructs a first neural network for predicting the concentration of the outlet dust according to the global parameter in a preset time period and the operation parameters of a plurality of dust removing devices; wherein the global parameter comprises at least one of a boiler load and an outlet dust concentration; the operating parameters include at least one of a voltage average, a voltage peak, and a current value; the upper computer operates a group intelligent optimization algorithm based on the first neural network with the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of a plurality of dust removing devices is minimized, so as to obtain a first operation parameter; the upper computer sends first operation parameters to each lower computer; the lower computer operates a group intelligent optimization algorithm based on a pre-constructed second neural network and with the aim that actual operation parameters are matched with first operation parameters and the estimated power consumption of dust removing equipment to which the lower computer belongs is minimized, so as to obtain first control parameters of the dust removing equipment to which the lower computer belongs; wherein the first control parameter comprises at least one of an intermittent off time, a current limit, and a voltage limit; the second neural network is constructed according to control parameters and operation parameters of dust removing equipment to which the lower computer belongs in a preset time period; the lower computer controls the operation of the dust removing equipment according to the first control parameter of the dust removing equipment. According to the scheme, the neural network is utilized, the control parameter optimization is carried out according to the fact that the concentration of the outlet dust reaches the standard and the power consumption is minimized, the optimal first control parameter is obtained, and the power consumption of the dust removing equipment is effectively reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic structural diagram of an embedded electric dust removal intelligent control system based on a neural network according to an embodiment of the present application;
Fig. 2 is a schematic structural diagram of an upper computer according to an embodiment of the present application;
Fig. 3 is a schematic diagram of a working principle of a data processing module of an upper computer according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a first neural network according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a lower computer according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a second neural network according to an embodiment of the present application;
Fig. 7 is a flowchart of an intelligent control method for embedded electric dust removal based on a neural network according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order to solve the problem of high power consumption of the existing electric dust removing equipment, the embodiment of the application provides an embedded intelligent electric dust removing control system based on a neural network, which is used for simultaneously controlling a plurality of dust removing equipment to operate according to certain control parameters.
Please refer to fig. 1, which is a schematic diagram of the architecture of the system. The system may include an upper computer 101 and a plurality of lower computers 102, wherein each lower computer 102 is installed in a controlled dust removal device. The upper computer 101 may be a computer device with a strong data processing capability, such as a desktop computer for industrial or personal use, or a server, and the lower computer may be an embedded device, such as an integrated circuit chip, or a programmable logic gate array.
Communication connection is established between the upper computer and each lower computer, and communication interaction can be performed based on a specific network protocol. The communication connection may be a direct communication connection or an indirect communication connection.
The direct communication connection means that a simplex (i.e. only supporting the upper computer to send instruction data to the lower computer) long connection is established between the upper computer and the lower computer, i.e. the upper computer and the lower computer are always connected as long as they are in a starting-up state.
Indirect communication means that the lower computer gathers data to the upper computer at regular time through some intermediate means. The lower computer sends the data to a certain transfer device in real time, and the transfer device reports the received data to the upper computer periodically.
Alternatively, the upper computer may be in direct communication connection with one portion of the lower computers and in indirect communication connection with another portion of the lower computers.
The specific functions and working principles of the upper computer are described below.
Fig. 2 is a schematic structural diagram of an upper computer. It can be seen that the upper computer 101 specifically includes an acquisition module 1011, a data processing module 1012, a data storage module 1013, a decision module 1014, an output control module 1015 and a database 1016.
The collection module 1011 is connected with external hardware (such as a controlled dust removing device), so as to monitor the operation state of the hardware in real time, so as to obtain data required by the upper computer.
The output control module 1015 is connected to the lower computer through an external communication layer (corresponding to a communication channel), and the upper computer communicates with the lower computer through the output control module 1015.
The specific functions of the respective modules of the upper computer are described below.
The acquisition module is used for monitoring the working state of each dust removing device in real time and obtaining the original data in a preset time period.
Specifically, as shown in fig. 3, a schematic diagram of a working principle of a data processing module according to an embodiment of the present application is shown. The acquisition module comprises a unified reading interface and an acquisition device, wherein the unified reading interface is directly connected with an external data serial port and a network, and the acquisition device reads original data required by an upper computer from the external data serial port and the network in real time through the unified reading interface, and specifically comprises original global parameters and original operation parameters of each electric dust removal device (namely an electric dust removal pipeline).
The preset time period may be any time period set according to actual needs. For example, if the present embodiment is set to control the operation of the dust removing devices according to the operation conditions of the respective dust removing devices within the last 1 day, the preset period may be the last 1 day, or the last 24 hours.
As shown in fig. 3, after the acquisition module acquires the original data, the original data may be preprocessed through an internal data pipeline, so as to obtain global parameters and operation parameters in a preset time period.
The operation parameters herein include the operation parameters of each dust removing device controlled by the system, for example, the voltage average value, the voltage peak value, and the current value of each dust removing device at each moment in the preset time period.
That is, the acquisition module is used for carrying out data preprocessing on the original data to obtain global parameters in a preset time period and operation parameters of a plurality of dust removing devices.
Wherein the data preprocessing includes at least one of data cleansing, object mapping, and data assembling.
The data assembly refers to assembling the collected original data into new data according to a certain format. The object mapping refers to mapping original data into corresponding objects in the upper computer. The data cleaning refers to deleting, interpolating and replacing abnormal or error data in original data, wherein the data which is out of range or obviously does not accord with the current operation condition can be judged as the abnormal or error data.
Optionally, the upper computer further comprises a data processing module, and the data processing module is used for carrying out structuring and normalization processing on the global parameter and the operation parameter output by the acquisition module, so that the global parameter and the operation parameter are converted into a unified data structure in the system, and subsequent analysis and processing are facilitated.
In addition, the data processing module can calculate some necessary parameters which cannot be directly collected according to the collected global parameters and the operation parameters, for example, calculate the power average value of each dust removing device in a preset time period, the heating value of the dust removing device and the like according to the voltage average value and the current value in the operation parameters.
The upper computer also comprises a data storage module and a database, wherein the data storage module is used for storing the processed data into the database and is used for acting on the operation of the database, so that equipment outside the upper computer provides functions of adding, deleting, modifying, searching and the like of the data and provides more customized services.
The data storage module may be encapsulated by the underlying data storage product, shielding the underlying storage by providing a stable interface, thereby enabling the system to retain the ability to replace the storage product. The data storage module provides a series of functions of I/O thread pool maintenance, cache library reading and writing, database reading and writing, cache and database synchronization, metadata maintenance and the like and provides corresponding interfaces.
A decision module for:
and constructing a first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices.
Wherein the global parameter comprises at least one of a boiler load and an outlet dust concentration; the operating parameter includes at least one of a voltage average, a voltage peak, and a current value.
And based on the first neural network, operating a group intelligent optimization algorithm with the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of a plurality of dust removing devices is minimized, so as to obtain a first operation parameter.
In other words, the decision module of the upper computer has the function of analyzing by utilizing the collected operation parameters and the global parameters in the preset time period, and outputting the optimized first operation parameters, wherein the first operation parameters are used for guiding each lower computer to control the operation of the dust removing equipment in a future period of time.
The objective of the upper computer optimization process is to minimize the sum of the power consumption of all the dust removing devices, namely the predicted power consumption, under the limiting condition that the predicted outlet dust concentration is not out of standard (namely not greater than the concentration threshold value) by utilizing the parameter relation established by the first neural network. Preferably, in the process, the variable parameter quantity in the optimization result is limited, so that the variable parameter quantity meets the actual production requirement.
Fig. 4 is an input/output schematic diagram of a first neural network according to the present embodiment. It can be seen that the inputs to the first neural network include global parameters and operational parameters. The global parameters input to the first neural network may be global parameters in a preset time period acquired by the previous acquisition module, that is, the outlet dust concentration and the boiler load in the preset time period. The input of the operating parameters of the first neural network may be a set of operating parameters intended for subsequent use in controlling the operation of the dust removing device.
The estimated outlet dust concentration input by the first neural network can be understood as the predicted outlet dust concentration after each dust removing device has been operated for a period of time, as controlled according to the input set of operation parameters.
For example, an estimated outlet dust concentration of 50% would indicate that the outlet dust concentration would be predicted to reach 50% after a period of time following an input set of operating parameters.
It will be appreciated that the above-described step of constructing the first neural network may be performed only once, or may be performed at regular intervals, for example, at intervals of one week. That is, when the first neural network has not been constructed yet, the decision module constructs the first neural network, and then each time the dust removing device is controlled to operate, the decision module may directly perform the subsequent steps based on the first neural network without constructing a new first neural network each time.
The first operating parameter may be a set of operating parameters, i.e. a set of operating parameters, in particular the first operating parameter may comprise a voltage mean value, a voltage peak value and a current value. The first operation parameter means that when each dust removing device controlled by the system of the embodiment operates according to the first operation parameter, each dust removing device can reach the lowest total power consumption under the premise that the concentration of the outlet dust reaches the standard.
Alternatively, the first neural network of the upper computer in this embodiment may be a data-driven neural network model, and of course, may be other types of neural network models, which is not limited.
First, a procedure for constructing the first neural network will be described.
Optionally, the decision module is configured to, when constructing the first neural network for predicting the outlet dust concentration according to the global parameter and the operation parameters of the plurality of dust removing devices within the preset time period, specifically:
Obtaining a neural network to be trained; the neural network to be trained may be an initial neural network pre-configured in a decision module.
Inputting the operation parameters and the global parameters in a preset time period into the neural network to be trained to obtain the estimated outlet dust concentration output by the neural network to be trained, and iteratively updating the neural network to be trained by taking the outlet dust concentration of which the estimated outlet dust concentration is matched with the global parameters as a target until the deviation of the estimated outlet dust concentration and the outlet dust concentration of the global parameters meets a preset first convergence condition to obtain the first neural network.
Specifically, when the decision module constructs the first neural network, the operation parameters and the global parameters in the preset time period obtained by the acquisition module can be input into the neural network to be trained, and after the neural network to be trained is processed, the estimated outlet dust concentration is output.
And then, calculating the difference between the estimated outlet dust concentration and the actual outlet dust concentration recorded in the global parameter in the preset time period, and taking the deviation of the estimated outlet dust concentration and the actual outlet dust concentration as the loss value of the neural network to be trained.
After obtaining the loss value, if the loss value does not meet the preset first convergence condition, the estimated outlet dust concentration is considered to be not matched with the real outlet dust concentration in the global parameter, and then the parameter of the neural network to be trained can be updated according to the loss value.
And after the primary parameter updating is completed, repeating the steps of inputting the operation parameters and the global parameters which are obtained by the acquisition module and are in a preset time period into the neural network to be trained, and outputting the estimated outlet dust concentration after the processing of the neural network to be trained.
If the estimated outlet dust concentration obtained for the second time is still not matched with the actual outlet dust concentration, updating again according to the loss value of the second time, inputting the operation parameter and the global parameter in the preset time period for the third time, and the like until the deviation of the estimated outlet dust concentration output by the neural network to be trained and the outlet dust concentration of the global parameter at a certain time meets the first convergence condition.
When the deviation of the estimated outlet dust concentration output by the neural network to be trained and the outlet dust concentration of the global parameter meets the first convergence condition, the neural network to be trained at the moment can be regarded as a constructed first neural network.
The first convergence condition may be that a loss value of the neural network to be trained is smaller than a preset first threshold value.
The decision module operates a group intelligent optimization algorithm based on the first neural network with the aim that the concentration of the outlet dust is not greater than a preset concentration threshold and the estimated power consumption of a plurality of dust removing devices is minimized, and the specific process of obtaining the first operation parameter can include:
in the group intelligent optimization algorithm, each set of operation parameters corresponds to an individual to be optimized, and one set of operation parameters can comprise a voltage average value, a voltage peak value and a current value.
When the intelligent optimization algorithm of the group is operated, a plurality of individuals are initialized first,
And inputting the individual and the global parameters acquired in the previous preset time period into a first neural network for each individual to obtain the expected outlet dust concentration when the individual controls the plurality of dust removing devices to operate, namely obtaining the estimated outlet dust concentration corresponding to the individual, and removing the individual with the estimated outlet dust concentration larger than the preset concentration threshold.
Next, according to the device characteristics of the dust removing devices, for each individual, the total estimated power consumption of the plurality of dust removing devices when the plurality of dust removing devices are controlled to operate by the individual is calculated, and the estimated power consumption corresponding to each individual is regarded as the fitness of the individual.
Subsequently, the current plurality of individuals are updated through a group intelligent optimization algorithm according to the fitness of each individual, and a plurality of new individuals can be generated through updating.
After each update, the first neural network is required to be utilized to evaluate whether the estimated outlet dust concentration corresponding to each updated individual meets the standard (namely, whether the estimated outlet dust concentration is not greater than a concentration threshold value), and after the unqualified individual is removed, the fitness is calculated again and the next update is carried out according to the fitness.
When the fitness of an individual is smaller than a preset first fitness threshold after a certain update, namely the expected power consumption is smaller than the preset expected power consumption threshold, the intelligent optimization algorithm of the group can be stopped, and a group of operation parameters corresponding to the individual with the fitness smaller than the first fitness threshold are determined as the first operation parameters.
Or when the updated frequency accumulation is larger than the preset maximum iteration frequency, stopping running the group intelligent optimization algorithm, and determining a group of running parameters corresponding to the individuals with the minimum current fitness as the first running parameters.
The population intelligent optimization algorithm of the embodiment can be a genetic algorithm, or can be a particle swarm algorithm, or can be other similar algorithms, and is not limited.
In some alternative embodiments, the first operation parameters may respectively correspond to each dust removing device, that is, the upper computer may sequentially output multiple sets of first operation parameters, where each set of first operation parameters corresponds to only one dust removing device controlled by the system.
Alternatively, only one set of first operating parameters may be output, with which each dust removing device is controlled.
The output control module is used for sending the first operation parameters to each lower computer.
Optionally, when the output control module sends the first operation parameter to each lower computer, the output control module is specifically configured to:
Analyzing and packaging the first operation parameters to obtain control instructions carrying the first operation parameters;
and sending a control instruction carrying the first operation parameter to each lower computer.
Specifically, the first operation parameters output by the decision module can be expressed in the form of a decision vector, after the output control module obtains the decision vector, the decision vector is analyzed to obtain each first operation parameter, then the first operation parameters are packaged to form corresponding control instructions, and finally the output control module sends the control instructions to each lower computer through a communication layer.
The method for obtaining the corresponding instruction for the first operation parameter package specifically may include completing mapping from a machine name to an address, that is, determining a communication address of a lower computer according to a name of the lower computer to which the control instruction is to be sent, and splitting and constructing the first operation parameter recorded according to a data structure of an upper computer, so that the first operation parameter is converted from the data structure of the upper computer and adapted to the data structure of the lower computer.
Optionally, the control instruction may include other necessary information in addition to the above information.
Fig. 5 is a schematic structural diagram of a lower computer according to an embodiment of the present application.
The lower computer may include an instruction module 1021, a power saving decision module 1022, a data storage module 1023, an execution module 1024, a data reading module 1025, and a database 1026.
The instruction module is connected with the communication layer and is communicated with the upper computer through the communication layer. The instruction module is used for providing a receiving inlet of the control instruction so as to receive the control instruction of the upper computer and transmitting the control instruction to the energy-saving decision module of the lower computer. Specifically, the instruction module may establish a TCP long connection with the host computer to receive the host computer instruction.
In addition, the instruction module can also call a unified control interface in the system to realize the control of the dust removing equipment where the lower computer is located. An interface may also be provided to enable manual control by a user. When the lower computer detects that the connection with the upper computer is disconnected, the instruction module is also responsible for a certain disaster tolerance capacity, such as automatically issuing an emergency instruction for processing the emergency instruction to be adjusted to the maximum or keeping the current state.
The advantage of doing so is that when the upper computer fails, the lower computer can still independently control the dust removal equipment in which the lower computer is positioned to work, so that the reliability of the dust removal system is improved.
The data storage module, the data reading module and the database can periodically collect and store local data through the hardware of the dust removal equipment where the lower computer is located, can also receive and store global parameters, operation parameters and other data provided by the upper computer, and are responsible for the assembly and preprocessing of the data.
Besides, the data storage module, the data reading module and the database of the lower computer can have the same functions as those of the data storage module, the data reading module and the database of the upper computer, and the detailed description is omitted.
The objective of the lower computer optimization process is to make the predicted electric energy consumption (namely the dust removal equipment where the lower computer is located) of the lower computer be small under the limiting condition that the predicted operation parameter (namely the operation parameter output by the second neural network) is close enough to the first optimized operation parameter carried by the upper computer instruction by utilizing the parameter relation established by the second neural network model.
The energy-saving decision module is used for:
And based on a pre-constructed second neural network, running a group intelligent optimization algorithm with the aim that the actual operation parameters are matched with the first operation parameters and the estimated power consumption of the dust removing equipment of the lower computer is minimized, and obtaining the first control parameters of the dust removing equipment.
Wherein the first control parameter comprises at least one of an intermittent off time, a current limit, and a voltage limit; the second neural network is constructed according to control parameters and operation parameters of dust removing equipment of the lower computer in a preset time period.
Fig. 6 is a schematic diagram of input and output of the second neural network of the lower computer.
As shown in fig. 6, the input of the second neural network includes a global parameter within a preset time period, an operation parameter of the dust removing device where the lower computer is located within the preset time period, and a set of control parameters; the output includes the predicted operating parameters of the dust removing apparatus when the dust removing apparatus is controlled in accordance with the input set of control parameters, that is, the predicted operating parameters after the preset period of time (noted as predicted operating parameters), and the predicted electricity consumption of the dust removing apparatus when the dust removing apparatus is controlled in accordance with the set of control parameters.
The construction process of the second neural network of the lower computer can be as follows:
The method comprises the steps of inputting the determined control parameters of the dust removing equipment in a preset time period, inputting the operation parameters and the global parameters in the preset time period into a neural network to be trained to obtain the operation parameters output by the neural network to be trained, comparing the operation parameters output by the neural network to be trained with the operation parameters provided by an upper computer in the preset time period, and taking the deviation of the operation parameters and the operation parameters as loss values of the neural network to be trained.
If the loss value does not meet the preset second convergence condition after the first output, updating the parameters of the neural network to be trained according to the loss value, then performing second input (still inputting the parameters), obtaining second output, if the loss value does not meet the second convergence condition after the second output, updating the parameters of the neural network to be trained again, repeating the third time, and so on until the loss value meets the second convergence condition after a certain output.
When the loss value meets the second convergence condition, the neural network to be trained at the moment can be determined to be a constructed second neural network.
The second convergence condition may be that the loss value is smaller than a preset second threshold.
The neural network to be trained used in the training of the lower computer can be the same as or different from the neural network to be trained used in the training of the upper computer.
The specific process of the energy-saving decision module based on the pre-constructed second neural network, to obtain the first control parameter of the corresponding dust-removing device by operating the group intelligent optimization algorithm with the aim that the actual operation parameter is matched with the first operation parameter and the estimated power consumption of the dust-removing device of the lower computer is minimized, may include:
In the population intelligent optimization algorithm, each individual corresponds to a set of control parameters, which may include an intermittent off time, a current limit, and a voltage limit.
When the group intelligent optimization algorithm is operated, a plurality of individuals are initialized first.
And then, inputting the operation parameters and the global parameters of each individual and the previously acquired preset time period into a second neural network to obtain the operation parameters and the power consumption which the dust removing equipment is expected to reach when the dust removing equipment (referring to the dust removing equipment where the lower computer is) is controlled to operate according to the individual, namely, obtaining the predicted operation parameters and the predicted power consumption which correspond to the individual.
And then, calculating the deviation between the predicted operation parameter of the individual and the first operation parameter provided by the upper computer, recording the deviation as the operation parameter deviation, and calculating the fitness of the individual according to the operation parameter deviation of the individual and the predicted electricity consumption of the individual, wherein the smaller the operation parameter deviation is, the higher the fitness is, and the lower the predicted electricity consumption is, the higher the fitness of the individual is.
Subsequently, the current plurality of individuals are updated through a group intelligent optimization algorithm according to the fitness of each individual, and a plurality of new individuals can be generated through updating.
After each update, the predicted operation parameters and predicted electricity consumption of each new individual need to be reevaluated by using a second neural network so as to obtain the fitness of the new individual, and then the next update is carried out according to the fitness of each individual.
When the fitness of an individual is greater than a preset second fitness threshold after a certain update, the predicted electricity consumption of a group of control parameters corresponding to the individual can be considered to be small enough (namely smaller than the electricity consumption threshold), and when the operation of the dust removing equipment is controlled according to the individual, the operation parameters reached by the dust removing equipment are close to the first operation parameters, and at the moment, the intelligent optimization algorithm of the group can be stopped, and the group of control parameters corresponding to the individual with the fitness greater than the second fitness threshold are determined as the first control parameters.
Or when the accumulated number of times of updating is larger than the preset maximum iteration number, stopping running the group intelligent optimization algorithm, and determining a group of control parameters corresponding to the individual with the maximum current fitness as the first control parameters.
It can be understood that the lower computer of each dust removing device can calculate and obtain a group of first control parameters of the dust removing device according to the functions, and the first control parameters can be the same or different among different dust removing devices.
The execution module of the lower computer has the function of realizing the control parameters of the energy-saving decision output to the controller for execution through means such as a hardware interface of the controller. The execution module has a retry mechanism.
That is, the execution module is to: and controlling the operation of the dust removing equipment according to the first control parameter of the dust removing equipment.
The retry mechanism refers to that if the execution module fails to control the dust removing device to operate according to the first control parameter for the first time, the execution module can wait for a period of time and then try to control the dust removing device to operate according to the first control parameter again.
The application provides an embedded electric dust removal intelligent control system based on a neural network, which comprises an upper computer and a plurality of lower computers which are connected in a communication way and are arranged on a plurality of dust removal devices, wherein the method comprises the following steps: the method comprises the steps that a decision module of an upper computer constructs a first neural network for predicting outlet dust concentration, optimizes with the aim that the outlet dust concentration is not more than a preset concentration threshold value and the predicted power consumption of a plurality of dust removing devices is minimized based on the first neural network, obtains a first operation parameter, and then sends the first operation parameter to a lower computer through an output control module; the energy-saving decision module of the lower computer is based on the second neural network, the first control parameter of the dust removing equipment is obtained by optimizing the actual operation parameter to match the first operation parameter and the expected power consumption of the dust removing equipment is minimized, and the execution module controls the operation of the dust removing equipment according to the first control parameter. According to the scheme, the neural network is utilized, the control parameter optimization is carried out according to the fact that the concentration of the outlet dust reaches the standard and the power consumption is minimized, the optimal first control parameter is obtained, and the power consumption of the dust removing equipment is effectively reduced.
In addition, the present embodiment has the following advantages:
In the first aspect, the whole control process is separated into two parts by means of an upper computer and a lower computer, so that the whole decision process has higher interpretability compared with a black box model.
In the second aspect, part of decision functions are configured in a lower computer to be realized, so that the influence of faults of the upper computer on decisions of all dust removing equipment can be minimized.
In a third aspect, a more optimal energy-saving decision is provided through a neural network and a group intelligent optimization algorithm.
According to the embedded electric dust collection intelligent control system based on the neural network provided by the embodiment of the application, the embodiment of the application also provides an embedded electric dust collection intelligent control method based on the neural network, which is applied to the embedded electric dust collection intelligent control system based on the neural network, and the system comprises an upper computer and a plurality of embedded lower computers arranged on a plurality of dust collection devices, wherein the upper computer is in communication connection with each lower computer.
Referring to fig. 7, a flowchart of the method may include the following steps.
S701, the upper computer constructs a first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices.
Wherein the global parameter comprises at least one of a boiler load and an outlet dust concentration; the operating parameter includes at least one of a voltage average, a voltage peak, and a current value.
S702, the upper computer operates a group intelligent optimization algorithm based on the first neural network with the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of a plurality of dust removing devices is minimized, so as to obtain a first operation parameter.
S703, the upper computer sends the first operation parameters to each lower computer.
And S704, the lower computer operates a group intelligent optimization algorithm based on the pre-constructed second neural network with the aim that the actual operation parameters are matched with the first operation parameters and the expected power consumption of the dust removing equipment to which the lower computer belongs is minimized, so as to obtain the first control parameters of the dust removing equipment to which the lower computer belongs.
Wherein the first control parameter comprises at least one of an intermittent off time, a current limit, and a voltage limit; the second neural network is constructed according to control parameters and operation parameters of dust removing equipment of the lower computer in a preset time period.
And S705, the lower computer controls the operation of the dust removing equipment according to the first control parameter of the dust removing equipment.
Optionally, before the upper computer constructs the first neural network for predicting the outlet dust concentration according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices, the method further comprises:
the upper computer monitors the working state of each dust removing device in real time to obtain the original data in a preset time period;
The upper computer performs data preprocessing on the original data to obtain global parameters in a preset time period and operation parameters of a plurality of dust removing devices; wherein the data preprocessing includes at least one of data cleansing, object mapping, and data assembling.
Optionally, the upper computer sends the first operation parameter to each lower computer, including:
the upper computer analyzes and packages the first operation parameters to obtain control instructions carrying the first operation parameters;
The upper computer sends a control instruction carrying a first operation parameter to each lower computer.
Optionally, the communication connection between the upper computer and the lower computer is a direct communication connection or an indirect communication connection.
Optionally, the upper computer constructs a first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices, and the method comprises the following steps:
The upper computer obtains a neural network to be trained;
The upper computer inputs the operation parameters and the global parameters in a preset time period into the neural network to be trained to obtain estimated outlet dust concentration output by the neural network to be trained, and iteratively updates the neural network to be trained by taking the outlet dust concentration of which the estimated outlet dust concentration is matched with the global parameters as a target until the deviation of the estimated outlet dust concentration and the outlet dust concentration of the global parameters meets a preset first convergence condition to obtain the first neural network.
The specific implementation manner of the embedded electric precipitation intelligent control method based on the neural network provided by the embodiment of the application can be referred to the working principle of the corresponding module in the embedded electric precipitation intelligent control system based on the neural network provided by the embodiment of the application, and is not repeated.
The application provides an embedded electric dust removal intelligent control method based on a neural network, wherein the system comprises an upper computer and a plurality of lower computers which are connected in a communication way and are arranged on a plurality of dust removal devices, and the method comprises the following steps: the upper computer is used for constructing a first neural network for predicting the concentration of the outlet dust; the upper computer performs optimization based on the first neural network by taking the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of a plurality of dust removing devices is minimized, obtains a first operation parameter and sends the first operation parameter to the lower computer; the lower computer obtains a first control parameter of the dust removing equipment by optimizing with the aim that the actual operation parameter is matched with the first operation parameter and the estimated power consumption of the dust removing equipment is minimized based on the second neural network, and controls the operation of the dust removing equipment according to the first control parameter. According to the scheme, the neural network is utilized, the control parameter optimization is carried out according to the fact that the concentration of the outlet dust reaches the standard and the power consumption is minimized, the optimal first control parameter is obtained, and the power consumption of the dust removing equipment is effectively reduced.
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 apparatus 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 apparatus. 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 apparatus that comprises the element.
It should be noted that the terms "first," "second," and the like herein are merely used for distinguishing between different devices, modules, or units and not for limiting the order or interdependence of the functions performed by such devices, modules, or units.
Those skilled in the art will be able to make or use the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The utility model provides an embedded electric precipitation intelligent control method based on neural network, its characterized in that is applied to the embedded electric precipitation intelligent control system based on neural network, the system includes host computer and installs at a plurality of embedded lower computers of many dust collecting equipment, host computer and each lower computer communication connection, the method includes:
The upper computer constructs a first neural network for predicting the concentration of the outlet dust according to the global parameter in a preset time period and the operation parameters of the plurality of dust removing devices; wherein the global parameter comprises at least one of boiler load and outlet dust concentration; the operating parameters include at least one of a voltage average, a voltage peak, and a current value;
The upper computer operates a group intelligent optimization algorithm based on the first neural network with the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of the plurality of dust removing devices is minimized, so as to obtain a first operation parameter;
The upper computer sends the first operation parameters to each lower computer;
The lower computer operates a group intelligent optimization algorithm based on a second neural network constructed in advance, and with the aim that actual operation parameters are matched with the first operation parameters and the estimated power consumption of dust removing equipment to which the lower computer belongs is minimized, so as to obtain first control parameters of the dust removing equipment to which the lower computer belongs; wherein the first control parameter includes at least one of an intermittent off time, a current limit, and a voltage limit; the second neural network is constructed according to control parameters and operation parameters of dust removing equipment to which the lower computer belongs in the preset time period;
And the lower computer controls the operation of the dust removing equipment according to the first control parameter of the dust removing equipment.
2. The method according to claim 1, wherein before the upper computer constructs the first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices, the method further comprises:
the upper computer monitors the working state of each dust removing device in real time to obtain the original data in the preset time period;
The upper computer performs data preprocessing on the original data to obtain global parameters in the preset time period and operation parameters of the plurality of dust removing devices; wherein the data preprocessing includes at least one of data cleansing, object mapping, and data assembling.
3. The method of claim 1, wherein the sending, by the upper computer, the first operating parameter to each of the lower computers comprises:
The upper computer analyzes and packages the first operation parameters to obtain control instructions carrying the first operation parameters;
And the upper computer sends a control instruction carrying the first operation parameter to each lower computer.
4. The method of claim 1, wherein the communication connection of the upper computer and the lower computer is a direct communication connection or an indirect communication connection.
5. The method of claim 1, wherein the upper computer constructs a first neural network for predicting the concentration of the outlet dust according to the global parameter in the preset time period and the operation parameters of the plurality of dust removing devices, comprising:
The upper computer obtains a neural network to be trained;
The upper computer inputs the operation parameters and the global parameters in the preset time period into the neural network to be trained to obtain the estimated outlet dust concentration output by the neural network to be trained, and iteratively updates the neural network to be trained by taking the outlet dust concentration of which the estimated outlet dust concentration is matched with the outlet dust concentration of the global parameters as a target until the deviation of the estimated outlet dust concentration and the outlet dust concentration of the global parameters meets a preset first convergence condition to obtain a first neural network.
6. The embedded electric dust removal intelligent control system based on the neural network is characterized by comprising an upper computer and a plurality of embedded lower computers arranged on a plurality of dust removal devices, wherein the upper computer is in communication connection with each lower computer;
the upper computer comprises a decision module and an output control module;
The decision module is used for:
Constructing a first neural network for predicting the concentration of the outlet dust according to the global parameter in a preset time period and the operation parameters of the plurality of dust removing devices; wherein the global parameter comprises at least one of boiler load and outlet dust concentration; the operating parameters include at least one of a voltage average, a voltage peak, and a current value;
based on the first neural network, operating a group intelligent optimization algorithm with the aim that the concentration of the outlet dust is not more than a preset concentration threshold value and the estimated power consumption of the plurality of dust removing devices is minimized, so as to obtain a first operation parameter;
the output control module is used for sending the first operation parameters to each lower computer;
the lower computer comprises an energy-saving decision module and an execution module;
The energy-saving decision module is used for:
Based on a pre-constructed second neural network, running a group intelligent optimization algorithm with the aim that actual running parameters are matched with the first running parameters and the expected power consumption of dust removing equipment to which the lower computer belongs is minimized, so as to obtain first control parameters of the dust removing equipment to which the lower computer belongs; wherein the first control parameter includes at least one of an intermittent off time, a current limit, and a voltage limit; the second neural network is constructed according to control parameters and operation parameters of dust removing equipment to which the lower computer belongs in the preset time period;
the execution module is used for: and controlling the operation of the dust removing equipment according to the first control parameter of the dust removing equipment.
7. The system of claim 6, wherein the host computer further comprises an acquisition module and a data processing module;
The acquisition module is used for monitoring the working state of each dust removing device in real time and obtaining the original data in the preset time period;
The acquisition module is used for carrying out data preprocessing on the original data to obtain global parameters in the preset time period and operation parameters of the plurality of dust removing devices; wherein the data preprocessing includes at least one of data cleansing, object mapping, and data assembling.
8. The system of claim 6, wherein the output control module is configured to, when sending the first operating parameter to each of the lower computers:
analyzing and packaging the first operation parameters to obtain control instructions carrying the first operation parameters;
and sending a control instruction carrying the first operation parameters to each lower computer.
9. The system of claim 6, wherein the communication connection of the upper computer and the lower computer is a direct communication connection or an indirect communication connection.
10. The system of claim 6, wherein the decision module is configured to, when constructing the first neural network for predicting the outlet dust concentration according to the global parameter and the operation parameters of the plurality of dust removing devices within the preset time period:
obtaining a neural network to be trained;
Inputting the operation parameters and the global parameters in the preset time period into the neural network to be trained to obtain the estimated outlet dust concentration output by the neural network to be trained, and iteratively updating the neural network to be trained by taking the outlet dust concentration of which the estimated outlet dust concentration is matched with the outlet dust concentration of the global parameters as a target until the deviation of the estimated outlet dust concentration and the outlet dust concentration of the global parameters meets a preset first convergence condition to obtain a first neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211278780.7A CN115576203B (en) | 2022-10-19 | 2022-10-19 | Embedded electric precipitation intelligent control method and system based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211278780.7A CN115576203B (en) | 2022-10-19 | 2022-10-19 | Embedded electric precipitation intelligent control method and system based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115576203A CN115576203A (en) | 2023-01-06 |
| CN115576203B true CN115576203B (en) | 2024-06-11 |
Family
ID=84584134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211278780.7A Active CN115576203B (en) | 2022-10-19 | 2022-10-19 | Embedded electric precipitation intelligent control method and system based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115576203B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117406608B (en) * | 2023-12-14 | 2024-03-12 | 中铁三局集团有限公司 | Tunnel dust removal control method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103186119A (en) * | 2011-12-28 | 2013-07-03 | 陈美玲 | Design method for electric dust remover automatic monitoring system |
| 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 |
| CN114191953A (en) * | 2021-12-07 | 2022-03-18 | 国网河北能源技术服务有限公司 | Flue gas desulfurization and denitrification control method based on convolutional neural network and XGboost |
| CN114492727A (en) * | 2021-12-13 | 2022-05-13 | 中国大唐集团科学技术研究院有限公司西北电力试验研究院 | Intelligent prediction and control method for inlet smoke concentration of dust removal system |
-
2022
- 2022-10-19 CN CN202211278780.7A patent/CN115576203B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103186119A (en) * | 2011-12-28 | 2013-07-03 | 陈美玲 | Design method for electric dust remover automatic monitoring system |
| 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 |
| CN114191953A (en) * | 2021-12-07 | 2022-03-18 | 国网河北能源技术服务有限公司 | Flue gas desulfurization and denitrification control method based on convolutional neural network and XGboost |
| CN114492727A (en) * | 2021-12-13 | 2022-05-13 | 中国大唐集团科学技术研究院有限公司西北电力试验研究院 | Intelligent prediction and control method for inlet smoke concentration of dust removal system |
Non-Patent Citations (3)
| Title |
|---|
| 垃圾焚烧二恶英净化过程建模与优化;唐影;;中外能源;20170115(01);全文 * |
| 基于先进控制的燃煤机组超低排放指标优化控制技术;孙成富;胡炜;胡翔;顾奇凯;沈恒芳;;浙江电力;20170725(07);全文 * |
| 电除尘器自动监控***设计;李依琼;徐;王进;卢泽锋;;计算机测量与控制;20081225(12);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115576203A (en) | 2023-01-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111783846B (en) | Intelligent energy consumption service cooperative control system and method | |
| CN115576203B (en) | Embedded electric precipitation intelligent control method and system based on neural network | |
| CN110046182A (en) | A kind of huge hydroelectric power plant's intelligent alarm threshold setting method and system | |
| CN117713221B (en) | Micro-inversion photovoltaic grid-connected optimization system | |
| CN111723519B (en) | Transformer fault diagnosis device and method based on logistic regression and naive Bayes | |
| CN109886232A (en) | A kind of power grid image identification system neural network based | |
| CN117454093A (en) | Intelligent power module management method and system | |
| CN117933660A (en) | Energy management platform based on Internet of things | |
| CN117346129A (en) | Temperature regulation and control system for low-temperature heating surface of high-temperature superheater of boiler | |
| CN102497135A (en) | Rule-engine-based method for monitoring photovoltaic power station | |
| CN111624874A (en) | Pump station cluster intelligent prediction method and system for urban sewage treatment | |
| WO2024016586A1 (en) | Machine room temperature control method and apparatus, and electronic device and storage medium | |
| CN110889495A (en) | State maintenance analysis method for silk making equipment based on active parameters | |
| CN114756722B (en) | Multi-data centralized management control system and method | |
| CN113837486B (en) | RNN-RBM-based distribution network feeder long-term load prediction method | |
| CN112099438A (en) | Machine tool energy-saving control method and device based on current signals | |
| CN116760033B (en) | Real-time power demand prediction system based on artificial intelligence | |
| CN117505074B (en) | Electric precipitation control method and system based on energy-saving analysis | |
| Qin et al. | Cnn-gru-attention based short-term load forecasting of distribution networks | |
| CN117995374B (en) | Intelligent data platform for medical waste management and equipment monitoring | |
| Mishra et al. | Data mining technology and its applications to power systems | |
| CN118210240A (en) | Multi-device energy consumption monitoring and management method based on Internet of things | |
| CN117520889A (en) | Heat treatment furnace energy consumption mode identification method | |
| CN117635103A (en) | SNN-based intelligent substation equipment state maintenance method | |
| Long et al. | Porting and Optimization of Electric Power Consumers Anomaly Behaviour Monitoring on ARM Platform |
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 |