Coal-fired unit electrostatic dust collection energy-saving optimization method, system and equipment and storage medium

Abstract

The invention relates to a coal-fired unit electrostatic dust collection energy-saving optimization method, system and equipment and a storage medium, and the method comprises the steps: obtaining historically collected system operation data of a coal-fired unit electrostatic dust collection system, and constructing an outlet NOx concentration prediction model and an outlet dust concentration dynamic model according to the system operation data; power optimization control strategies of high-voltage power supplies of all levels of electric fields in the dust removal system are obtained, a model is built based on the outlet NOx concentration prediction model, the outlet dust concentration dynamic model and the optimization control strategies, and an electrostatic dust removal energy-saving optimization model is obtained; acquiring a preset environment-friendly standard concentration, and analyzing the environment-friendly standard concentration through the electrostatic dust collection energy-saving optimization model to obtain an optimal dust collection parameter; and adjusting the current or voltage of each level of electric field in the dust removal system according to the optimal dust removal parameters. Through the constructed energy-saving optimization model, the electrostatic dust collection control system can reduce the dust emission concentration and meet the pollutant emission standard, and meanwhile, the energy consumption of the electric dust collector can be greatly reduced.

Description

Technical Field

[0001] The present application relates to the field of electrostatic precipitators in power plants, and particularly to an energy-saving optimization method, system, equipment, and storage medium for electrostatic precipitators in coal-fired units. Background Art

[0002] Most dust removal systems in coal-fired power plants use electrostatic precipitators. This type of precipitator has the characteristics of high dust removal efficiency, large air volume handling capacity, low pressure loss, low operating cost, and the ability to handle high-temperature flue gas. It is widely used in the field of dust control in thermal power units. With the continuous improvement of environmental protection standards, the required emission concentration of electrostatic precipitators is getting lower and lower, and the energy consumption of electrostatic precipitators is also increasing continuously. In actual operation, due to the flexible operation of the unit load and the variability of coal types, the operating conditions of the electrostatic precipitator change frequently. Therefore, the parameters of the electrostatic precipitator are mostly adjusted manually by operators. However, manual adjustment is prone to errors, making it difficult to ensure that the dust concentration at the outlet of the electrostatic precipitator reaches the emission standard under flexible operation modes through manual operation. Therefore, the phenomenon of excessive dust concentration at the outlet often occurs.

[0003] Currently, in order to avoid excessive dust concentration emissions at the outlet, relatively high electrostatic precipitator power supply parameters (secondary current or voltage) are usually adopted, but this will lead to high energy consumption of the electrostatic precipitator, and the dust removal efficiency and energy consumption are not proportional; if the parameters are not increased, the emission concentration of the electrostatic precipitator will not meet the standard either. Therefore, the existing electrostatic dust removal control system cannot achieve a significant reduction in the energy consumption of the electrostatic precipitator while reducing the dust emission concentration and meeting the pollutant emission standards. Summary of the Invention

[0004] The embodiments of the present application provide an energy-saving optimization method, system, equipment, and storage medium for electrostatic precipitators in coal-fired units, so as to at least solve the problem in the related art that the existing electrostatic dust removal control system cannot achieve a significant reduction in the energy consumption of the electrostatic precipitator while reducing the dust emission concentration and meeting the pollutant emission standards.

[0005] In a first aspect, the embodiments of the present application provide an energy-saving optimization method for electrostatic precipitators in coal-fired units. The method is applied to the electrostatic precipitator system of a coal-fired unit, and the method includes:

[0006] Obtain the system operation data of the electrostatic precipitator system of the coal-fired unit collected historically, and construct an outlet NOx concentration prediction model and an outlet dust concentration dynamic model based on the system operation data;

[0007] Obtain the power optimization control strategy of the high-voltage power supply for each stage of the electric field in the dust removal system, and construct a model based on the outlet NOx concentration prediction model, the outlet dust concentration dynamic model, and the optimization control strategy to obtain an electrostatic precipitator energy-saving optimization model;

[0008] Obtain the preset environmental protection compliance concentration, analyze the environmental protection compliance concentration through the electrostatic precipitation energy-saving optimization model, and obtain the optimal dust removal parameters;

[0009] Adjust the current or voltage of each stage of the electric field in the dust removal system according to the optimal dust removal parameters.

[0010] In one embodiment, the system operation data includes boiler operation data, outlet flue gas data, and electrostatic precipitation system parameters. An outlet NOx concentration prediction model is constructed. The construction of the outlet NOx concentration prediction model according to the system operation data includes:

[0011] Preprocess the boiler operation data, outlet flue gas data, and electrostatic precipitation system parameters. The preprocessing process includes data cleaning, standardization / normalization, missing value filling, and outlier removal;

[0012] Extract feature data related to the outlet dust concentration from the preprocessed data, and perform feature selection through correlation analysis or principal component analysis to obtain high-precision feature data;

[0013] Select a deep learning model according to the sequence characteristics of historical dust concentration data, and train, validate, and optimize the deep learning model with the high-precision feature data to obtain the outlet NOx concentration prediction model.

[0014] In one embodiment, the power optimization control strategy for the high-voltage power supply of each stage of the electric field in the dust removal system is as follows:

[0015] When the unit load of the dust removal system increases, increase the voltage and / or current of the high-voltage power supply in preset steps; when the unit load of the dust removal system decreases, decrease the voltage and / or current of the high-voltage power supply in preset steps;

[0016] When the outlet dust concentration of the dust removal system increases, increase the voltage and / or current of the high-voltage power supply in preset steps; when the outlet dust concentration of the dust removal system decreases, decrease the voltage and / or current of the high-voltage power supply in preset steps;

[0017] When the dust resistivity of the dust removal system becomes higher, decrease the voltage of the high-voltage power supply in preset steps; when the dust resistivity of the dust removal system becomes lower, increase the voltage of the high-voltage power supply of the electric field in preset steps.

[0018] In one embodiment, the dust removal system includes a pre-stage electric field and a post-stage electric field; wherein, based on the outlet NOx concentration prediction model and the optimization control strategy, a model is constructed to obtain an electrostatic precipitation energy-saving optimization model, including:

[0019] Construct a pre-stage power optimization function for the high-voltage power supply of the pre-stage electric field according to the relationship between the unit load, the inlet dust concentration, the dust resistivity and the high-voltage power supply voltage and current, and use the output values of the outlet NOx concentration prediction model and the outlet dust concentration dynamic model as the feed-forward control of the pre-stage power optimization function to obtain a pre-stage electric field energy-saving optimization model;

[0020] The post-stage electric field adopts PID negative feedback closed-loop regulation, and an interference observer is introduced to establish a post-stage power optimization function for the high-voltage power supply of the post-stage electric field. The post-stage power optimization function is combined with the fuzzy weighting method to obtain a post-stage electric field energy-saving optimization model;

[0021] Combine the pre-stage electric field energy-saving optimization model and the post-stage electric field energy-saving optimization model to obtain the electrostatic precipitator energy-saving optimization model.

[0022] In one embodiment, obtain the preset environmental protection compliance concentration, analyze the environmental protection compliance concentration through the electrostatic precipitator energy-saving optimization model, and obtain the optimal dust removal parameters, including:

[0023] Obtain the current unit load of the dust removal system, and obtain the current dust concentration and current dust resistivity of the dust removal system through a pre-built on-line dust concentration measurement system;

[0024] Input the current unit load, the current dust concentration and the current dust resistivity into the pre-stage electric field energy-saving optimization model to obtain the pre-stage electric field regulation voltage and the pre-stage electric field regulation current.

[0025] In one embodiment, obtain the preset environmental protection compliance concentration, analyze the environmental protection compliance concentration through the electrostatic precipitator energy-saving optimization model, and obtain the optimal dust removal parameters, including:

[0026] Predict the predicted value of the outlet NOx concentration and the predicted value of the dust concentration of the dust removal system according to the outlet NOx concentration prediction model and the outlet dust concentration dynamic model;

[0027] Input the environmental protection compliance concentration, the predicted value of the outlet NOx concentration and the predicted value of the dust concentration into the post-stage electric field energy-saving optimization model to obtain the post-stage electric field regulation voltage and the post-stage electric field regulation current.

[0028] In one embodiment, the method further includes:

[0029] Deploy the optimization model to an artificial intelligence server, perform two-way data reading and writing with a process controller (DCS) through an OPC station, and adjust the operating parameters of the electrostatic precipitator in real time to make the dust emission concentration meet the environmental protection standards and the energy consumption is the lowest.

[0030] Second aspect, an electrostatic precipitator energy-saving optimization system for a coal-fired unit according to an embodiment of the present application is used to execute the electrostatic precipitator energy-saving optimization method for a coal-fired unit described in any of the above embodiments. The system includes:

[0031] A data acquisition module for acquiring the system operation data of the electrostatic precipitator system of the coal-fired unit collected historically;

[0032] A prediction model construction module for constructing an outlet NOx concentration prediction model and an outlet dust concentration dynamic model according to the system operation data;

[0033] An optimization model construction module for obtaining the power optimization control strategy of the high-voltage power supply of each stage of the electric field in the dust removal system, and constructing a model based on the outlet NOx concentration prediction model, the outlet dust concentration dynamic model and the optimization control strategy to obtain an electrostatic precipitator energy-saving optimization model;

[0034] An optimization analysis module for obtaining the current outlet dust concentration, analyzing the current outlet dust concentration and the preset environmental protection compliance concentration through the electrostatic precipitator energy-saving optimization model to obtain the optimal dust removal parameters; and adjusting the current or voltage of each stage of the electric field in the dust removal system according to the optimal dust removal parameters.

[0035] Third aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the electrostatic precipitator energy-saving optimization method for a coal-fired unit described in the first aspect above.

[0036] Fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the electrostatic precipitator energy-saving optimization method for a coal-fired unit described in the first aspect above.

[0037] The electrostatic precipitator energy-saving optimization method, system, device and storage medium provided by the embodiments of the present application have at least the following technical effects:

[0038] Obtain the system operation data of the electrostatic precipitator system of coal-fired units collected historically, and construct an outlet NOx concentration prediction model and an outlet dust concentration dynamic model based on the system operation data; obtain the power optimization control strategy of the high-voltage power supply of each stage of the electric field in the dust removal system, and construct a model based on the outlet NOx concentration prediction model, the outlet dust concentration dynamic model and the optimization control strategy to obtain an electrostatic precipitator energy-saving optimization model; obtain the preset environmental protection compliance concentration, analyze the environmental protection compliance concentration through the electrostatic precipitator energy-saving optimization model to obtain the optimal dust removal parameters; adjust the current or voltage of each stage of the electric field in the dust removal system according to the optimal dust removal parameters. Through the constructed energy-saving optimization model, the electrostatic precipitator control system can achieve reducing the dust emission concentration and meeting the pollutant discharge standards while significantly reducing the energy consumption of the electrostatic precipitator.

[0039] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more concise and understandable. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments and descriptions thereof of the present application are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0041] Figure 1 is a flowchart of an electrostatic precipitator energy-saving optimization method for a coal-fired unit in an embodiment of the present application;

[0042] Figure 2 is a hardware deployment diagram of an electrostatic precipitator energy-saving optimization system for a coal-fired unit in an embodiment of the present application;

[0043] Figure 3 is a structural block diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be described and explained below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments provided in the present application without creative efforts fall within the scope of protection of the present application.

[0045] Obviously, the accompanying drawings in the following description are only some examples or embodiments of the present application. For those of ordinary skill in the art, without creative efforts, the present application can also be applied to other similar scenarios based on these drawings. In addition, it can also be understood that although the efforts made in such a development process may be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some design, manufacturing, or production changes based on the technical content disclosed in the present application are only conventional technical means and should not be understood as insufficient disclosure of the content of the present application.

[0046] When the term "embodiment" is mentioned in the present application, it means that the specific features, structures, or characteristics described in connection with the embodiment may be included in at least one embodiment of the present application. The appearance of this phrase at various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those of ordinary skill in the art explicitly and implicitly understand that the embodiments described in the present application can be combined with other embodiments without conflict.

[0047] Unless otherwise defined, the technical terms or scientific terms involved in the present application should have the ordinary meaning understood by those of ordinary skill in the technical field to which the present application belongs. The terms "a", "one", "kind", "the", and other similar words involved in the present application do not indicate a limitation in quantity and can represent singular or plural. The terms "including", "comprising", "having", and any variations thereof involved in the present application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may further include unlisted steps or units, or may further include other steps or units inherent to these processes, methods, products, or devices. The terms "connected", "coupled", and other similar words involved in the present application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" involved in the present application refers to two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist. For example, "A and / or B" may represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after. The terms "first", "second", "third", etc. involved in the present application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0048] Based on the above situation, the embodiments of the present application provide a method, system, device, and storage medium for energy-saving optimization of electrostatic precipitators in coal-fired power units.

[0049] In a first aspect, an embodiment of the present application provides an energy-saving optimization method for electrostatic precipitation of coal-fired units. Figure 1 This is the flowchart of the method, as Figure 1 shown. The energy-saving optimization method for electrostatic precipitation of coal-fired units includes the following steps.

[0050] Step S1: Obtain the system operation data of the electrostatic precipitation system of the coal-fired unit collected historically, and construct an outlet NOx concentration prediction model and an outlet dust concentration dynamic model based on the system operation data. Specifically, preprocess the boiler operation data, outlet flue gas data, and electrostatic precipitation system parameters. The preprocessing process includes data cleaning, standardization / normalization, missing value filling, and outlier removal; extract feature data related to the outlet dust concentration or NOx concentration from the preprocessed data, and perform feature selection through correlation analysis or principal component analysis to obtain high-precision feature data; select a deep learning model according to the sequence characteristics of historical dust concentration data or historical NOx concentration data, and use the high-precision feature data as training data to train, validate, and optimize the deep learning model to obtain the outlet NOx concentration prediction model or the outlet dust concentration dynamic model.

[0051] In an embodiment of the present application, when constructing a time series prediction model of the NOx concentration at the outlet of the electrostatic precipitator (i.e., the outlet NOx concentration prediction model), it is necessary to combine the combustion mechanism with the data-driven method and perform dynamic modeling with multi-parameter input. First, obtain the input variables, including flue gas flow Q(t), unit load L(t), flue gas temperature T flue (t), coal input M coal (t), high-voltage voltage V(t) and current I(t) of the electrostatic precipitator, variable power consumption P elec (t) of the electrostatic precipitator, inlet NOx concentration C in (t), and the historically collected NOx concentration sequence is {C out (t - 1), C out (t - 2),..., C out (t - k)}. Then, predict the outlet NOx concentration value at a future moment based on the above input variables, that is, the target variable is:

[0052] y(t) = C out (t + Δt)

[0053] where Δt is the prediction step, such as 1 hour.

[0054] After determining the input variables and output variables, select an appropriate recurrent neural network model according to the user's needs to construct an outlet NOx concentration prediction model. In this embodiment, the LSTM algorithm model (long short-term memory network) is taken as an example for detailed description. First, construct the time series characteristics of the input variables, and the input feature vector is:

[0055] X(t) = [Q(t), L(t), M coal (t), T flue (t), V(t), I(t), P elec (t), C out (t - 1), C out (t - 2),...]

[0056] The key of the LSTM algorithm model lies in the introduction of the cell state and the gating mechanism. These mechanisms enable the network to selectively remember or forget information, thus better processing long sequence data. Among them, the cell state is the core of the LSTM, like a "conveyor belt", which can transmit information throughout the time series. The design of the cell state enables information to remain unchanged over a long time span, thus capturing long-term dependencies. The LSTM controls the flow of information through three gates (forget gate, input gate, output gate). Among them, the forget gate determines which information is discarded from the cell state; the input gate determines which new information is stored in the cell state; the output gate determines which information is output from the cell state.

[0057] According to the above input variables, construct a NOx generation model for the combustion process. The formula is as follows:

[0058]

[0059] Among them, E a is the activation energy, and R is the gas constant. In this application, both the activation energy and the gas constant can be determined by fitting through historical data. NOx is a gaseous pollutant, and the removal ability of the electrostatic precipitator for it is limited. However, the equipment operating state (such as the air leakage rate and the flue gas residence time) will also affect the outlet NOx concentration. The influence of the air leakage rate and other interferences on the outlet NOx concentration is shown in the following formula:

[0060] C out (t) = C in (t)·(1 - η leak (t)) + ΔC other

[0061] Among them, η leak (t) is the air leakage rate of the electrostatic precipitator (related to the equipment sealing performance); ΔC otherFor other interference factors (such as secondary dust, sensor noise, etc.). Then, the input feature vector and the combustion process NOx generation model are used as features and input into the LSTM algorithm model to obtain a data-driven model. The formula is as follows:

[0062]

[0063] Wherein, is the predicted output, and then compensation is performed through an interference observer. Among them, the interference term refers to the prediction residual caused by unmodeled dynamics (such as catalyst failure, sudden change in fuel sulfur content). The formula for the prediction residual is as follows:

[0064]

[0065] Kalman filter interference estimation is performed. The formula for the state space model is as follows:

[0066]

[0067] Wherein, w(t) and v(t) are the process noise and the observation noise respectively.

[0068] The corrected predicted output is:

[0069]

[0070] Finally, the loss function is obtained, and the model is trained according to the loss function. When the loss value output by the loss function meets the requirements, the outlet NOx concentration prediction model is obtained. The outlet NOx concentration prediction model in this embodiment dynamically captures the non-linear coupling effect of parameters such as unit load, coal quantity, and power consumption on the NOx concentration through a time series gating mechanism; and through the embedding of a mechanism model, it constrains the predicted value to conform to the combustion generation law (high temperature promotes the generation of thermal NOx) and the influence of ESP air leakage; at the same time, an interference observer is added to compensate for unmodeled factors (such as changes in fuel characteristics) in real time, improving the robustness of the model. The outlet NOx concentration prediction model can integrate multi-parameter dynamic characteristics, physical mechanisms, and real-time interference compensation to achieve high-precision NOx concentration prediction.

[0071] When constructing a time series prediction model for the dust concentration at the outlet of the electrostatic precipitator (i.e., the dynamic model of the outlet dust concentration), it is necessary to combine the physical mechanism of the electrostatic precipitator with the data-driven method and perform dynamic modeling through multi-parameter input. The following are the core formulas and key components of the model.

[0072] First, a basic physical mechanism model is constructed. In this embodiment, the dust removal efficiency η(t) of the electrostatic precipitator is the core for predicting the outlet concentration, based on the modified Deutsch equation:

[0073]

[0074] Among them, A is the area of the dust collection plate; Q(t) is the flue gas flow rate; ω(t) is the dust drift velocity. The dust drift velocity is related to the electric field strength E(t) and the dust characteristics. The specific formula is:

[0075] ω(t) = k·E(t) m ·f(R(t), C in (t))

[0076] Among them, E(t) is the electric field strength; R(t) is the dust resistivity (this value changes dynamically and needs to be obtained through laboratory data or online estimation); C in (t) is the inlet dust concentration (if it cannot be measured, it can be estimated through combustion parameters); k and m are experimental fitting parameters, which are related to the dust particle size distribution and adhesion. Users can obtain them by fitting historical data. The calculation formula for the outlet dust concentration is:

[0077] C out (t) = C in (t)·(1 - η(t)) + ΔC res (t)

[0078] Among them, ΔC res (t) is the uncollected dust (affected by factors such as secondary dust emission and air leakage).

[0079] Build a data-driven model (taking the LSTM algorithm model as an example). Among them, the input feature vector is:

[0080]

[0081] Among them, k is the lag step (such as k = 6 hours). Similarly, the LSTM algorithm model controls the flow of information through three gates (forget gate, input gate, output gate). Among them, the forget gate determines which information is discarded from the cell state; the input gate determines which new information is stored in the cell state; the output gate determines which information is output from the cell state.

[0082] Then combine the basic physical mechanism model and the data-driven model, that is, use the predicted value of the mechanism model as the feature input to the data-driven model. After joint optimization, the dynamic model of the outlet dust concentration is obtained. The specific formula is:

[0083]

[0084] Among them, η theory (t) is the theoretical efficiency calculated by the Deutsch equation.

[0085] Then, compensation is carried out through a disturbance observer. Among them, the disturbance term refers to the prediction residual caused by unmodeled dynamics (such as rapping failure and sudden change in dust adhesion). The formula for the prediction residual is as follows:

[0086]

[0087] Perform Kalman filter disturbance estimation. The formula for the state space model is as follows:

[0088]

[0089] Among them, w(t) and v(t) are the process noise and measurement noise respectively.

[0090] The corrected predicted output is:

[0091]

[0092] Physical constraints are added to the model output layer to ensure that the predicted values conform to the actual laws, such as non-negativity constraints and efficiency boundary constraints. Finally, the loss function is obtained, and the model is trained according to the loss function. When the loss value output by the loss function meets the requirements, the dynamic model of the outlet dust concentration is obtained. The dynamic model of the outlet dust concentration can integrate the operating parameters of the electrostatic precipitator, physical mechanisms, and real-time disturbance compensation to achieve high-precision dust concentration prediction, providing a theoretical basis for optimizing voltage setting, rapping strategies, and energy conservation and consumption reduction.

[0093] Step S2: Obtain the power optimization control strategy for the high-voltage power supplies of each stage of the dust removal system. Based on the outlet NOx concentration prediction model, the dynamic model of the outlet dust concentration, and the optimization control strategy, a model is constructed to obtain the electrostatic precipitation energy-saving optimization model.

[0094] In this embodiment, the power optimization control strategy for the high-voltage power supplies of each stage of the dust removal system is as follows: when the unit load of the dust removal system increases, the voltage and / or current of the high-voltage power supply are increased in accordance with a preset step; when the unit load of the dust removal system decreases, the voltage and / or current of the high-voltage power supply are decreased in accordance with a preset step; when the outlet dust concentration of the dust removal system increases, the voltage and / or current of the high-voltage power supply are increased in accordance with a preset step; when the outlet dust concentration of the dust removal system decreases, the voltage and / or current of the high-voltage power supply are decreased in accordance with a preset step; when the dust resistivity of the dust removal system becomes higher, the voltage of the high-voltage power supply is decreased in accordance with a preset step; when the dust resistivity of the dust removal system becomes lower, the voltage of the high-voltage power supply of the electric field is increased in accordance with a preset step.

[0095] During the model construction process, based on the relationship between the unit load, inlet dust concentration, dust resistivity and the voltage and current of the high-voltage power supply, a pre-stage power optimization function of the high-voltage power supply for the pre-stage electric field is constructed, and the output value of the outlet NOx concentration prediction model is used as the feed-forward control of the pre-stage power optimization function to obtain the energy-saving optimization model for the pre-stage electric field; the post-stage electric field adopts PID negative feedback closed-loop regulation, and an interference observer is introduced to establish a post-stage power optimization function of the high-voltage power supply for the post-stage electric field, and the post-stage power optimization function is combined with the fuzzy weighting method to obtain the energy-saving optimization model for the post-stage electric field; the energy-saving optimization model for the pre-stage electric field and the energy-saving optimization model for the post-stage electric field are combined to obtain the energy-saving optimization model for the electrostatic precipitator.

[0096] In the energy-saving optimization of the pre-stage (the first, second, and third electric fields) and post-stage (the fourth and fifth electric fields) of the electrostatic precipitator, it is necessary to dynamically adjust the voltage / current of each electric field to minimize the overall energy consumption while ensuring the total dust removal efficiency. The following is the mathematical expression and key formulas of the model.

[0097] First, construct the objective function. The purpose of the objective function is to minimize the total energy consumption, and the formula is as follows:

[0098]

[0099] Among them, P i is the power of the i-th electric field, and V i , I i are the voltage and current of the i-th electric field respectively.

[0100] Set the constraint conditions. In this embodiment, the constraint conditions include the total dust removal efficiency constraint, the electric field operation constraint, the pre-stage and post-stage electric field coordination constraint, etc. Among them, the total dust removal efficiency constraint formula is as follows:

[0101]

[0102] Among them, η i is the dust removal efficiency of the i-th electric field, which is calculated by the modified Deutsch equation. The modified Deutsch equation is:

[0103]

[0104] Among them, A i is the dust collection area of the i-th electric field, ω i is the drift velocity of the i-th electric field (related to the voltage V and dust resistivity R), and Q is the flue gas flow rate.

[0105] The conditions for the electric field operation constraint are as follows:

[0106] 1. Set the upper and lower limits of the voltage / current as V i，min ≤V≤V i，max 、Ii，min ≤I≤I i，max ;

[0107] 2. Anti-corona suppression (for high specific resistance dust), i.e., V i ≤C crit / R i , where C crit is the critical field strength constant;

[0108] 3. Fit the V-I characteristic curve to determine the current-voltage relationship, i.e., I i = f(V i , R i ).

[0109] In this embodiment, for the collaborative constraint of the front-stage and rear-stage electric fields, the front-stage electric field is set to preferentially capture dust. The front-stage electric fields (one, two, three) bear the main dust removal load, and the rear-stage electric fields (four, five) optimize the capture of fine particles. It is expressed by the formula as follows:

[0110] η 前级 = 1 - (1 - η1)(1 - η2)(1 - η3) ≥ η 前级，min

[0111] It is also necessary to set the gradient limit of the rear-stage electric field voltage. For example, the rear-stage electric field voltage needs to be higher than the front-stage to avoid re-entrainment of dust. For example, V4 ≥ 0.9V3, V5 ≥ 0.9V4.

[0112] After determining the optimization conditions, a dynamic optimization model is carried out. First, the state space equation is set. The system states in this embodiment include the voltages V i (t), currents I i (t), and inlet concentrations C in，i (t).

[0113] Because the outlet concentration of the front stage is the inlet concentration of the rear stage, the inlet concentration of the rear stage is predicted through the above outlet NOx concentration prediction model and outlet dust concentration dynamic model. Therefore, the state transition equation is as follows:

[0114] C in，i+1 (t) = C in,i (t)·(1 - η i (t))

[0115] Then, according to the current working conditions (such as flue gas volume Q(t), dust concentration C in，1 (t), specific resistance R i (t)) and the above constraints, the optimal voltage distribution is dynamically solved:

[0116]

[0117] During the solution process, the model can be solved by the decomposition and coordination method and the intelligent optimization algorithm.

[0118] Specifically, when the decomposition and coordination method is adopted, the front stage is roughly adjusted, that is, the first three-stage electric fields are preferentially optimized to quickly reduce the inlet concentration, and then the rear stage is finely adjusted. Under the remaining dust removal requirements, the last two-stage electric fields are optimized. The specific formula is as follows:

[0119]

[0120] min(P4 + P5) s.t. η 后级 ≥η min / η 前级

[0121] When the intelligent optimization algorithm is adopted, the genetic algorithm can be used for solution, and the coding variables and fitness function are determined by the user. The particle swarm optimization algorithm can also be used. Particle Swarm Optimization (PSO) is an optimization algorithm based on swarm intelligence, which searches for the optimal solution of the problem by simulating the cooperation and information sharing of individuals (particles) in the swarm.

[0122] Finally, compensation is carried out through the disturbance observer. In this embodiment, the disturbance term is the unmodeled factors (such as the efficiency fluctuation after rapping and ash cleaning, and the sudden change of dust resistivity):

[0123] d i (t) = η i (t) - η model，i (t)

[0124] Compensation strategy: Dynamically adjust the voltage setting value:

[0125]

[0126] (Kp is the proportional gain, adjusted according to experiments)

[0127] Step S3, obtain the preset environmental protection compliance concentration, and analyze the environmental protection compliance concentration through the electrostatic precipitator energy-saving optimization model to obtain the optimal dust removal parameters.

[0128] Specifically, for the front-stage electric field, obtain the current unit load of the dust removal system, and obtain the current dust concentration and current dust resistivity of the dust removal system through the pre-built online dust concentration measurement system; input the current unit load, current dust concentration and current dust resistivity into the front-stage electric field energy-saving optimization model to obtain the front-stage electric field regulation voltage and front-stage electric field regulation current.

[0129] For the post-stage electric field, predict the predicted outlet NOx concentration and the predicted dust concentration of the dust removal system according to the outlet NOx concentration prediction model and the outlet dust concentration dynamic model; input the environmental protection compliance concentration, the predicted outlet NOx concentration and the predicted dust concentration into the post-stage electric field energy-saving optimization model to obtain the post-stage electric field regulation voltage and the post-stage electric field regulation current.

[0130] Step S4, adjust the current or voltage of each stage of the electric field in the dust removal system according to the optimal dust removal parameters.

[0131] The energy-saving optimization solution of this application can include six parts: data collection, equipment research and development, data analysis, model construction, model debugging, and model deployment.

[0132] Before constructing the energy-saving optimization model, equipment research and development is also required, that is, the research and development of on-line monitoring equipment. In this embodiment, the research and development of a dust concentration on-line measurement system can be based on the characteristics of dust charge, construct the entire on-line measurement system, realize the on-line detection of dust specific resistance and concentration, and lay a data support for the establishment of the subsequent dust collector energy-saving optimization model.

[0133] In this embodiment, the research and development of a dust concentration on-line measurement system can be based on the characteristic that friction between dust particles and between dust particles and the flue wall will cause the particles to carry charges. Through the research and development of current signal measurement electrodes, amplifier circuits, and the construction of a dust specific resistance algorithm model, the entire on-line measurement system can be completed. At the same time, the research and development of detection sampling devices and pretreatment devices are carried out to realize the on-line detection of dust specific resistance and concentration, and lay a data support for the establishment of the subsequent dust collector energy-saving optimization model.

[0134] More specifically, the steps of constructing the dust specific resistance algorithm model include: first, determine the physical model, that is, based on the flat electrode method, establish an electric field distribution model to relate voltage, current, dust layer thickness, and specific resistance; then preprocess the signal, such as applying filtering and gain adjustment to process the original current signal to eliminate noise and non-linear effects; when performing parameter calibration, the parameters in the model, such as electrode area, spacing, etc., can be determined through experimental data, and a quantitative relationship between the current signal and the specific resistance is established; then perform error compensation, for example, introduce iR drop compensation and correction of environmental factors such as temperature and humidity; finally, perform algorithm implementation, convert the model into a mathematical formula or computer algorithm, integrate it into the on-line system, and process data in real time and output the specific resistance value.

[0135] The on-line monitoring system for the dust concentration of the flue gas at the inlet of the dust collector provided in this embodiment realizes the on-line monitoring of the dust concentration and the dust specific resistance, and the monitoring results provide data support for the establishment of the dust collector energy-saving optimization model.

[0136] First, data collection is carried out, and data acquisition work is carried out on important parameters such as boiler operation and flue gas, as well as parameters of the entire dust removal system. For the time series prediction model of the NOx concentration at the outlet of the post-stage electrostatic precipitator and the dynamic model of the outlet dust concentration, multi-parameter signals such as the variable power consumption of the electrostatic precipitator, unit load, flue gas volume, coal input volume, flue gas temperature, dust at the outlet of the precipitator, and dust at the chimney outlet are introduced. Combining the real-time operation parameters of the high-voltage and low-voltage equipment of the electrostatic precipitator collected by the system, a research on the time series prediction model of the dust concentration at the outlet of the electrostatic precipitator based on data-driven is carried out.

[0137] In the embodiment of the present application, the unit load is used to reflect the power generation, which can affect the combustion conditions and the amount of flue gas generated; the coal input volume can determine the original dust amount (i.e., the inlet concentration) generated by combustion; the flue gas volume can determine the treatment load of the electrostatic precipitator; the flue gas temperature affects the dust resistivity and the corona discharge efficiency; the variable power consumption is used to reflect the energy consumption; the voltage / current directly affects the electric field strength; the rapping period affects the dust cleaning effect; the dust at the outlet of the precipitator and the dust at the chimney outlet are directly related to the target variable (i.e., the outlet dust concentration), and need to be used as input or supervision signals.

[0138] After the data collection is completed, it is necessary to perform preprocessing analysis on the data. Data analysis can make the operation characteristics of the dust removal equipment clearer and easier to discover and solve data problems that affect the modeling quality. In this embodiment, the data analysis includes data integrity analysis, data anomaly characteristic analysis, and data correlation analysis. Specifically, the data is standardized / normalized. The difference in the dimensions of different parameters is large (such as the unit of flue gas volume is m 3 / h, and the unit of voltage is kV), and standardization (Z-score) or normalization (Min-Max) is required; then the missing values are processed. For the short-term fault data of the sensor, linear interpolation or filling based on the similarity of working conditions (such as mean filling under the same load) is used; finally, the outliers are removed, and the 3σ principle or the isolation forest algorithm is used to identify abnormal working conditions (such as a sudden increase in dust concentration caused by a broken bag of the precipitator).

[0139] Then, according to the preprocessed data, a model is constructed. Specifically, in the present application, the dust removal optimization control is a multi-objective non-linear optimization problem, which requires simultaneously meeting the non-exceedance of dust emissions and the optimal energy consumption. For this reason, an accurate dust prediction model needs to be constructed, and then the best operation parameter combination is obtained through an intelligent optimization algorithm.

[0140] The establishment of the time series prediction model of the NOx concentration at the outlet of the electrostatic precipitator is carried out. Multi-parameter signals such as the variable power consumption of the electrostatic precipitator, unit load, flue gas volume, coal input volume, flue gas temperature, dust at the outlet of the precipitator, and dust at the chimney outlet are introduced. Combining the real-time operation parameters of the high-voltage and low-voltage equipment of the electrostatic precipitator collected by the system, a research on the time series prediction model of the dust concentration at the outlet of the electrostatic precipitator based on data-driven is carried out.

[0141] Specifically, on the basis of predicting the outlet dust concentration, an energy-saving optimization model for the front-stage (the first, second, and third stages) and the rear-stage (the fourth and fifth stages) electric fields of the dust collector is established. Feedforward and feedback methods are respectively used to formulate intelligent energy-saving control strategies for the high-voltage power supplies of each stage of the electric field, and closed-loop control is carried out. Specifically, for the first three stages of the electric field, power optimization of the high-voltage power supply is carried out based on load, dust concentration, and dust resistivity, and automatic adjustment control of the high-voltage power supply is carried out to basically remove the flue gas dust. The last two stages of the power plant are subjected to PID negative feedback closed-loop regulation based on the outlet NOx concentration time series prediction model and the outlet dust concentration dynamic model. An interference observer is introduced to establish a dynamic model of the secondary current of the high-frequency power supply of the last second-stage electric field on the outlet dust concentration. The fuzzy weighting method is used to realize the smooth switching of the control action, and the automatic adjustment of the high-voltage power supply power of the last two stages of the electric field is established to realize the control and adjustment of the secondary dust emission and automatic vibration of the dust collector.

[0142] In the embodiment of the present application, aiming at the pain point of the excessively high energy consumption of the electrostatic precipitator commonly existing in coal-fired power plants, by utilizing the charge change characteristics, an online monitoring system for the dust concentration of the flue gas at the inlet of the dust collector is developed to measure the dust concentration at the inlet of the dust collector, and an energy-saving optimization model for the front-stage (the first, second, and third stages) and the rear-stage (the fourth and fifth stages) power plants of the dust collector is established based on this. Feedforward and feedback methods are respectively used to formulate intelligent energy-saving control strategies for the high-voltage power supplies of each stage of the electric field, and closed-loop control is carried out to minimize the power consumption of equipment operation to the greatest extent on the premise of ensuring environmental protection compliance.

[0143] In this embodiment, the front-stage electric field specifically refers to that for the first three stages of the electric field, power optimization of the high-voltage power supply is carried out based on load, dust concentration, and dust resistivity, and automatic adjustment control of the high-voltage power supply is carried out to basically remove the flue gas dust; the rear-stage electric field, the last two stages of the electric field are subjected to PID negative feedback closed-loop regulation based on the outlet NOx concentration time series prediction model and the outlet dust concentration dynamic model. An interference observer is introduced to establish a dynamic model of the secondary current of the high-frequency power supply of the last second-stage electric field on the outlet dust concentration. The fuzzy weighting method is used to realize the smooth switching of the control action, and the automatic adjustment of the high-voltage power supply power of the last two stages of the electric field is established to realize the control and adjustment of the secondary dust emission and automatic vibration of the dust collector.

[0144] In the process of optimizing the power of the high-voltage power supply, since the power of the electrostatic precipitator is usually determined by voltage and current, and the settings of voltage and current are affected by dust characteristics. For example, high specific resistance dust may cause back corona, in which case the voltage needs to be reduced or pulse power supply mode needs to be adopted. High dust concentration may require higher current to maintain sufficient charge. High load may require greater corona power to cover a larger gas flow. Therefore, a possible model needs to take dust removal efficiency and power consumption as objectives, take load, medium, dust concentration, and specific resistance as input variables, and adjust voltage and current as control variables. Then, establish a trade-off between these two objectives or transform it into a single-objective optimization problem

[0145] The outlet NOx concentration time series prediction model is a machine learning or statistical model based on time series data, used to predict the future change trend of nitrogen oxides (NOx) concentration at the chimney outlet of coal-fired power plants. By analyzing historical NOx concentration data and related influencing factors (such as process parameters, environmental conditions, equipment status, etc.), this model captures the complex relationships among time dependence, periodicity, trend, and external variables, so as to achieve accurate prediction of outlet NOx

[0146] The steps to introduce a Disturbance Observer (DOB) include: 1) Establish a mathematical model of the system, including the relationships among the power supply, dust removal efficiency, and various variables; 2) Determine the main disturbance sources, such as load changes, coal quality changes, and dust characteristic changes; 3) Design a disturbance observer to estimate these disturbances and use the estimated values to adjust the optimization algorithm, such as correcting the voltage or current setting values to compensate for the influence of disturbances; 4) It may be necessary to combine the DOB with a control strategy (such as PID) or directly consider the disturbance estimated value in the optimization algorithm

[0147] In the process of constructing the dynamic model of the outlet dust concentration, mainly based on the characteristic parameters of the input variables (load changes, coal quality changes, dust characteristic changes, electrostatic precipitator performance changes), by quantifying the influence of different operating variables (such as electrostatic precipitator voltage, differential pressure of bag filter) on the outlet concentration, predict the dynamic changes of the future dust concentration

[0148] After obtaining the energy-saving optimization model, it is necessary to debug the model. The constructed model has passed the theoretical reliability verification. However, to ensure that it can truly generate benefits, on-site commissioning verification of the model is required. The model can be further optimized according to the actual working condition characteristics and feedback data

[0149] Finally, deploy the model to the hardware system, refer to Figure 2, the model is deployed on an artificial intelligence server and then placed in the engineer's station. By connecting to the idle communication card of the DCS through an RJ45 or RS485 communication cable, two-way communication can be achieved. The model and algorithm are deployed in the intelligent server, and two-way data reading and writing are realized between the intelligent server and the process controller (DCS) through the OPC station: the actual operating parameters of the DCS acquisition system are collected, and the results after operation, simulation, and optimization in the server are sent back to the DCS for execution.

[0150] In summary, the electrostatic precipitator energy-saving optimization method for coal-fired units provided in this application avoids the necessity of using a sampling system through in-situ measurement of the online measurement system based on the charged-variation dust concentration. It monitors the operating parameters online in real time, warns of the risk of excessive dust in advance, and takes targeted measures to improve the reliability of the safe and environmental protection operation of the unit. Conducting research on the prediction modeling of the dust concentration at the outlet of the dry electrostatic precipitator based on data-driven methods can realize the intelligent automatic control optimization and closed-loop control of the dry dust removal system, with the lowest total energy consumption of the dry dust removal system under the condition of meeting the dust standard, thus achieving the purpose of energy conservation and carbon reduction. Regulating the operation optimization of the dry dust removal system, fully exploring the energy-saving potential of the dry dust removal, and ensuring that the dust collector always operates in the state of the lowest energy consumption and the highest dust removal efficiency. At the same time, for different operating conditions of the unit, the intelligent dust removal system will realize the automatic adjustment and control of all high-voltage power supply operating states according to the collected unit load and dust concentration, and realize the load linkage automatic control function.

[0151] In a second aspect, an embodiment of the present application provides an electrostatic precipitator energy-saving optimization system for coal-fired units, which is used to execute any one of the above-mentioned electrostatic precipitator energy-saving optimization methods for coal-fired units. The system includes:

[0152] A data acquisition module, which is used to obtain the system operation data of the electrostatic precipitator system of the coal-fired unit collected historically;

[0153] A prediction model construction module, which is used to construct an outlet NOx concentration prediction model and an outlet dust concentration dynamic model according to the system operation data;

[0154] An optimization model construction module, which is used to obtain the power optimization control strategy of the high-voltage power supply of each stage of the electric field in the dust removal system, and construct a model based on the outlet NOx concentration prediction model, the outlet dust concentration dynamic model, and the optimization control strategy to obtain an electrostatic precipitator energy-saving optimization model;

[0155] An optimization analysis module, which is used to obtain the current outlet dust concentration, analyze the current outlet dust concentration and the preset environmental protection compliance concentration through the electrostatic precipitator energy-saving optimization model to obtain the optimal dust removal parameters; and adjust the current or voltage of each stage of the electric field in the dust removal system according to the optimal dust removal parameters.

[0156] In a third aspect, an embodiment of the present application provides an electronic device, Figure 3 which is a block diagram of an electronic device shown according to an exemplary embodiment. As Figure 3 shown, the electronic device may include a processor 11 and a memory 12 storing computer program instructions.

[0157] Specifically, the above-mentioned processor 11 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0158] Among them, the memory 12 may include a mass memory for data or instructions. By way of example and not limitation, the memory 12 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In suitable cases, the memory 12 may include removable or non-removable (or fixed) media. In suitable cases, the memory 12 may be internal or external to the data processing device. In a particular embodiment, the memory 12 is a non-volatile memory. In a particular embodiment, the memory 12 includes a read-only memory (ROM) and a random access memory (RAM). In suitable cases, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or a flash memory, or a combination of two or more of these. In suitable cases, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM may be a fast page mode dynamic random access memory (FPMDRAM), an extended date out dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0159] The memory 12 can be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 11.

[0160] The processor 11 reads and executes the computer program instructions stored in the memory 12 to implement any one of the electrostatic precipitation energy-saving optimization methods for coal-fired units in the above embodiments.

[0161] In one embodiment, the electronic device may further include a communication interface 13 and a bus 10. Among them, as Figure 3 shown, the processor 11, the memory 12, and the communication interface 13 are connected through the bus 10 and complete communication with each other.

[0162] The communication interface 13 is used to implement communication between various modules, devices, units, and / or devices in the embodiments of the present application. The communication port 13 can also implement data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0163] Bus 10 includes hardware, software, or both, and couples components of an electronic device to each other. Bus 10 includes, but is not limited to, at least one of the following: Data Bus, Address Bus, Control Bus, Expansion Bus, Local Bus. By way of example and not limitation, Bus 10 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable bus or a combination of two or more of these. In suitable cases, Bus 10 may include one or more buses. Although embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0164] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, the energy-saving optimization method for electrostatic precipitation of a coal-fired unit provided in the first aspect is implemented.

[0165] Among them, more specifically, the readable storage medium may include, but is not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination of the above.

[0166] In a possible implementation manner, the present invention can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps of implementing the electrostatic precipitator energy-saving optimization method for a coal-fired unit provided in the first aspect.

[0167] Among them, the program code for executing the present invention can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, executed as an independent software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0168] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0169] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several deformations and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for optimizing energy saving of electrostatic dust removal for coal-fired units, characterized in that: The method is applied in an electrostatic dust removal system of a coal-fired unit, and the method comprises: Acquire historically collected system operation data of the electrostatic precipitator system of the coal-fired unit, and construct an outlet NOx concentration prediction model and an outlet dust concentration dynamic model based on the system operation data; Obtaining the power optimization control strategy of the high-voltage power supply of each level of the electric field in the dust removal system, constructing a model based on the outlet NOx concentration prediction model, the outlet dust concentration dynamic model and the optimization control strategy, and obtaining an electrostatic dust removal energy-saving optimization model; Obtaining a preset environmental protection standard concentration, analyzing the environmental protection standard concentration through the electrostatic dust removal energy-saving optimization model, and obtaining optimal dust removal parameters; The current or voltage of each level of electric field in the dust removal system is adjusted according to the optimal dust removal parameters.

2. The energy-saving optimization method according to claim 1, characterized in that: The system operation data includes boiler operation data, outlet flue gas data and electrostatic dust removal system parameters, and constructing an outlet NOx concentration prediction model and an outlet dust concentration dynamic model based on the system operation data includes: Preprocessing the boiler operation data, outlet flue gas data and electrostatic precipitator system parameters, wherein the preprocessing process includes data cleaning, standardization / normalization, missing value filling and outlier removal processing; Extract characteristic data related to the outlet dust concentration or NOx concentration based on the pre-processed data, and select features through correlation analysis or principal component analysis to obtain high-precision characteristic data; A deep learning model is selected according to the sequence characteristics of historical dust concentration data or historical NOx concentration data, and the deep learning model is trained, verified and optimized using the high-precision feature data as training data to obtain the outlet NOx concentration prediction model or the outlet dust concentration dynamic model.

3. The energy-saving optimization method according to claim 1, characterized in that: The power optimization control strategy of the high-voltage power supply of each level of electric field in the dust removal system is: When the unit load of the dust removal system increases, the voltage and / or current of the high-voltage power supply is increased according to a preset step size; when the unit load of the dust removal system decreases, the voltage and / or current of the high-voltage power supply is reduced according to a preset step size; When the dust concentration at the outlet of the dust removal system increases, the voltage and / or current of the high-voltage power supply is increased according to a preset step size; when the dust concentration at the outlet of the dust removal system decreases, the voltage and / or current of the high-voltage power supply is reduced according to a preset step size; When the dust specific resistance of the dust removal system becomes higher, the voltage of the high-voltage power supply is reduced according to a preset step size; when the dust specific resistance of the dust removal system becomes lower, the voltage of the electric field high-voltage power supply is increased according to a preset step size.

4. The energy-saving optimization method according to claim 3, characterized in that: The dust removal system includes a front-stage electric field and a rear-stage electric field; wherein, based on the outlet NOx concentration prediction model and the optimization control strategy, a model is constructed to obtain an electrostatic dust removal energy-saving optimization model, including: According to the relationship between the unit load, inlet dust concentration and dust specific resistance and the voltage and current of the high-voltage power supply, a front-stage power optimization function of the front-stage electric field high-voltage power supply is constructed, and the output values ​​of the outlet NOx concentration prediction model and the outlet dust concentration dynamic model are used as feedforward control of the front-stage power optimization function to obtain a front-stage electric field energy-saving optimization model; The rear-stage electric field adopts PID negative feedback closed-loop regulation, and introduces a disturbance observer to establish a rear-stage power optimization function of the rear-stage electric field high-voltage power supply. The rear-stage power optimization function is combined with a fuzzy weighted method to obtain a rear-stage electric field energy-saving optimization model; The front-stage electric field energy-saving optimization model and the rear-stage electric field energy-saving optimization model are combined to obtain the electrostatic precipitator energy-saving optimization model.

5. The energy-saving optimization method according to claim 4, characterized in that: The step of obtaining a preset environmental protection standard concentration and analyzing the environmental protection standard concentration through the electrostatic dust removal energy-saving optimization model to obtain an optimal dust removal parameter includes: Obtaining the current unit load of the dust removal system, and obtaining the current dust concentration and current dust resistivity of the dust removal system through a pre-built dust concentration online measurement system; The current unit load, current dust concentration and current dust specific resistance are input into the front-stage electric field energy-saving optimization model to obtain the front-stage electric field adjustment voltage and the front-stage electric field adjustment current.

6. The energy-saving optimization method according to claim 4, characterized in that: The step of obtaining a preset environmental protection standard concentration and analyzing the environmental protection standard concentration through the electrostatic dust removal energy-saving optimization model to obtain an optimal dust removal parameter includes: Predicting the outlet NOx concentration prediction value and the outlet dust concentration prediction value of the dust removal system according to the outlet NOx concentration prediction model and the outlet dust concentration dynamic model; The environmental protection standard concentration, the outlet NOx concentration prediction value and the dust concentration prediction value are input into the subsequent electric field energy-saving optimization model to obtain the subsequent electric field adjustment voltage and the subsequent electric field adjustment current.

7. The energy-saving optimization method according to claim 1, characterized in that: The method further comprises: The optimization model is deployed to the artificial intelligence server, and data is bidirectionally read and written with the process controller through the OPC station, so that the operating parameters of the electrostatic precipitator can be adjusted in real time to ensure that the dust emission concentration meets environmental protection standards and minimizes energy consumption.

8. An electrostatic dust removal energy-saving optimization system for coal-fired units, characterized in that: Used to implement the energy-saving optimization method for electrostatic precipitator of any one of claims 1 to 7 for coal-fired units, the system comprising: A data acquisition module is used to obtain historically collected system operation data of the electrostatic precipitator system of the coal-fired unit; A prediction model building module, used to build an outlet NOx concentration prediction model and an outlet dust concentration dynamic model according to the system operation data; An optimization model building module is used to obtain a power optimization control strategy for high-voltage power supplies at various levels of electric fields in the dust removal system, and to build a model based on the outlet NOx concentration prediction model, the outlet dust concentration dynamic model and the optimization control strategy to obtain an electrostatic precipitator energy-saving optimization model; The optimization analysis module is used to obtain the current outlet dust concentration, analyze the current outlet dust concentration and the preset environmental protection standard concentration through the electrostatic dust removal energy-saving optimization model, and obtain the optimal dust removal parameters; according to the optimal dust removal parameters, the current or voltage of each level of the electric field in the dust removal system is adjusted.

9. An electronic device, characterized in that: It comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the energy-saving optimization method for electrostatic precipitator of a coal-fired unit as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by the processor, the electrostatic precipitator energy-saving optimization method for a coal-fired unit as described in any one of claims 1 to 7 is implemented.

Images