Prediction Method, Device and Terminal for Coal Ash Composition

Abstract

The present invention provides a method, device and terminal for predicting the ash composition of coal. The method includes: obtaining the coal parameters of the target coal, where the coal parameters include coal composition, coal shape and coal quantity; obtaining the combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume and secondary air damper opening; inputting the coal parameters and combustion parameters into a trained composition prediction model to obtain the ash composition of the ash produced after the target coal burns under the combustion parameters; the composition prediction model is a BP neural network, and the ash composition includes the mass fractions of various metal oxides in the ash. By predicting the ash composition after coal combustion through a BP neural network, the present invention can determine the ash composition before coal combustion, so that the ash composition can be changed by adjusting the combustion parameters of the coal, and the slagging and coking phenomena on the water-cooled wall of the tail flue of the power station boiler can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of coal ash composition detection, and particularly to a prediction method, device and terminal for coal ash composition. Background Art

[0002] During the operation of a power station boiler, slagging and fouling phenomena will occur on the water-cooled wall in the tail flue. This is due to the relatively low ash melting point of pulverized coal. Therefore, the influence of the ash melting point of the coal type used needs to be considered in the design and operation of the boiler. The chemical composition of coal ash is the main factor affecting the ash melting point. Therefore, to determine the ash melting point of coal ash, it is necessary to first determine the chemical composition of coal ash.

[0003] Currently, the chemical composition of coal ash can only be determined after the coal combustion is completed. Therefore, it is also necessary to determine the ash melting point after the coal combustion is completed, and it is difficult to avoid the slagging and fouling phenomena on the water-cooled wall in the tail flue of the power station boiler by adjusting the parameters during the coal combustion process. Summary of the Invention

[0004] The present invention provides a prediction method, device and terminal for coal ash composition to solve the problem of determining the coal ash composition before coal combustion.

[0005] In a first aspect, the present invention provides a prediction method for coal ash composition, including:

[0006] Obtaining the coal parameters of the target coal, where the coal parameters include coal composition, coal shape and coal quantity;

[0007] Obtaining the combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume and secondary air damper opening;

[0008] Inputting the coal parameters and combustion parameters into a trained composition prediction model to obtain the coal ash composition of the coal ash generated after the target coal burns under the combustion parameters; the composition prediction model is a BP neural network, and the coal ash composition includes the mass fractions of various metal oxides in the coal ash.

[0009] In a possible implementation manner, before inputting the coal parameters and combustion parameters into a trained composition prediction model to obtain the coal ash composition of the coal ash generated after the target coal burns under the combustion parameters, the method further includes:

[0010] Establishing an initial composition prediction model;

[0011] Training the initial composition prediction model based on an improved particle swarm optimization algorithm and a training sample set to obtain a trained composition prediction model.

[0012] In a possible implementation, training an initial component prediction model based on an improved particle swarm optimization algorithm and a training sample set to obtain a trained component prediction model includes:

[0013] Optimizing the weights and thresholds in the initial component prediction model through the improved particle swarm optimization algorithm to obtain target weights and target thresholds; wherein, the items to be optimized by the improved particle swarm optimization algorithm are the weights and thresholds in the initial component prediction model, the objective function is the reciprocal of the prediction error of the component prediction model, and the optimization objective is to maximize the objective function.

[0014] Training the component prediction model with the target weights and target thresholds based on the training sample set to obtain a trained component prediction model.

[0015] In a possible implementation, optimizing the weights and thresholds in the initial component prediction model through the improved particle swarm optimization algorithm to obtain target weights and target thresholds includes:

[0016] Step 1: Establish an initial particle swarm. The initial particle swarm includes a preset number of particles. Each particle has a random initial velocity and an initial position, and the position of each particle corresponds to a set of weights and thresholds.

[0017] Step 2: Calculate the fitness value of each particle.

[0018] Step 3: For each particle, compare the current fitness value of the particle with the historical best fitness value of the particle. If the current fitness value of the particle is higher than the historical best fitness value of the particle, then use the current fitness value of the particle as the historical best fitness value of the particle.

[0019] Step 4: For each particle, compare the current fitness value of the particle with the global best fitness value of the particle swarm. If the current fitness value of the particle is higher than the global best fitness value of the particle swarm, then use the current fitness value of the particle as the global best fitness value of the particle swarm.

[0020] Step 5: Update the velocity and position of each particle based on the historical best fitness value of each particle and the global best fitness value of the particle swarm.

[0021] Step 6: Repeat Step 2 to Step 5 until a preset number of iterations is reached, and output the current global best position as the target weights and target thresholds.

[0022] In a possible implementation, before training the component prediction model with the target weights and target thresholds based on the training sample set to obtain a trained component prediction model, the method further includes:

[0023] Obtain the coal ash data of multiple coals; the coal ash data includes coal parameters, combustion parameters, and coal ash components;

[0024] For each coal ash data, use the coal parameters and combustion parameters of the coal ash data as samples, and the coal ash components as the labels of the samples;

[0025] Form a training sample set with each sample and its corresponding label.

[0026] In a second aspect, the present invention provides a device for predicting coal ash components, including:

[0027] A first acquisition module for obtaining the coal parameters of the target coal, where the coal parameters include coal components, coal shape, and coal quantity;

[0028] A second acquisition module for obtaining the combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume, and secondary air damper opening;

[0029] A prediction module for inputting the coal parameters and combustion parameters into a trained component prediction model to obtain the coal ash components of the coal ash generated after the target coal burns under the combustion parameters; the component prediction model is a BP neural network, and the coal ash components include the mass fractions of various metal oxides in the coal ash.

[0030] In a third aspect, the present invention provides a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method for predicting coal ash components as shown in the first aspect or any possible implementation manner of the first aspect above.

[0031] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, it implements the steps of the method for predicting coal ash components as shown in the first aspect or any possible implementation manner of the first aspect above.

[0032] The present invention provides a method, apparatus, and terminal for predicting the composition of coal ash. The method includes: obtaining the coal parameters of the target coal, where the coal parameters include coal composition, coal shape, and coal quantity; obtaining the combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume, and secondary air damper opening; inputting the coal parameters and combustion parameters into a trained composition prediction model to obtain the coal ash composition of the coal ash generated after the target coal burns under the combustion parameters; the composition prediction model is a BP neural network, and the coal ash composition includes the mass fractions of various metal oxides in the coal ash. By predicting the coal ash composition after coal combustion through a BP neural network, the present invention can determine the coal ash composition before coal combustion, so that the coal ash composition can be changed by adjusting the combustion parameters of the coal, and the slagging and coking phenomena on the water-cooled wall of the tail flue of the power station boiler can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts.

[0034] Figure 1 is a flowchart of the implementation of the method for predicting the composition of coal ash provided by the embodiment of the present invention;

[0035] Figure 2 is a schematic structural diagram of the apparatus for predicting the composition of coal ash provided by the embodiment of the present invention;

[0036] Figure 3 is a schematic diagram of the terminal provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0038] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will be described through specific embodiments with reference to the drawings.

[0039] See Figure 1 , which shows a flowchart of the implementation of the method for predicting the composition of coal ash provided by the embodiment of the present invention, and is described in detail as follows:

[0040] Step 101: Obtain the coal parameters of the target coal. The coal parameters include coal composition, coal shape, and coal quantity.

[0041] In this embodiment, the coal composition is used to reflect the types of metal elements contained in the coal and the content of each metal element. The coal composition can affect the types of metal oxides in the coal ash and the upper limit of the mass fraction of metal oxides in the coal ash. The coal shape is the form when the coal is burned, which can specifically be pulverized coal, medium-sized particles, etc. The burning degrees of coals with different shapes may be different under the same combustion conditions, thus affecting the content of metal oxides in the coal ash. The coal quantity is the amount of coal input for combustion at one time. If the combustion method is staged combustion, the coal quantity is the coal feeding quantity.

[0042] Step 102: Obtain the combustion parameters of the target coal. The combustion parameters include combustion method, combustion duration, primary air volume, and secondary air damper opening.

[0043] In this embodiment, different combustion parameters will affect the furnace temperature and combustion speed during coal combustion, thereby affecting the coal combustion degree and the formation of metal oxides. The combustion methods include stoker firing, suspension combustion, and fluidized bed combustion. The primary air volume and secondary air damper opening will affect the air content in the coal-fired boiler, and thus affect the contact degree between the coal and the air.

[0044] Step 103: Input the coal parameters and combustion parameters into the trained composition prediction model to obtain the coal ash composition of the coal ash generated after the target coal is burned under the combustion parameters. The composition prediction model is a BP neural network, and the coal ash composition includes the mass fractions of various metal oxides in the coal ash.

[0045] In this embodiment, since the parameters input into the composition prediction model include parameters other than numerical values, normalization calculation needs to be performed before each parameter is input into the composition prediction model to map the data to between [0, 1]. Correspondingly, when obtaining the coal ash composition output by the composition prediction model, inverse normalization also needs to be performed to obtain the actual mass fractions of each metal oxide.

[0046] The component prediction model in this embodiment is an error backpropagation (BP) neural network with a structure of 15-5-3-9. Among them, the 15 input nodes respectively correspond to the contents of Si, Al, Fe, S, Ca, K, Na, Ti, and Mg in coal, the shape of coal, the quantity of coal, the combustion mode, the combustion time, the primary air volume, and the opening degree of the secondary air damper. The hidden layer has two layers, and the number of nodes is 5 and 3 respectively. Increasing the number of hidden layers instead of the number of nodes can improve the prediction accuracy and avoid the situation that all nodes cannot be trained due to too many hidden layer nodes. The output nodes are 9, which respectively correspond to the mass fractions of SiO2, Al2O3, Fe2O3, SO3, CaO, K2O, Na2O, TiO2, and MgO in coal ash.

[0047] In some embodiments, before inputting the coal parameters and combustion parameters into the trained component prediction model to obtain the coal ash components of the target coal after combustion under the combustion parameters, the method further includes:

[0048] Establish an initial component prediction model;

[0049] Train the initial component prediction model based on the improved particle swarm optimization algorithm and the training sample set to obtain the trained component prediction model.

[0050] In this embodiment,

[0051] In some embodiments, training the initial component prediction model based on the improved particle swarm optimization algorithm and the training sample set to obtain the trained component prediction model includes:

[0052] Optimize the weights and thresholds in the initial component prediction model through the improved particle swarm optimization algorithm to obtain the target weights and target thresholds; among them, the items to be optimized by the improved particle swarm optimization algorithm are the weights and thresholds in the initial component prediction model, the objective function is the reciprocal of the prediction error of the component prediction model, and the optimization objective is to maximize the objective function;

[0053] Train the component prediction model with the target weights and target thresholds based on the training sample set to obtain the trained component prediction model.

[0054] In this embodiment, the neural network is a better method to solve the "black box" problem in the modeling process. The neural network is trained by the steepest descent method. By continuously training to establish a non-linear mapping mathematical model between samples and labels, the connection weights and thresholds of the network are adjusted to reduce the network error value and reach the expected error. The neural network needs to be trained before use to optimize the connection coefficients between each node. In order to improve the training speed and effect of the neural network, in this embodiment, an improved particle swarm optimization algorithm is used to optimize the BP neural network. The improved particle swarm optimization algorithm adopts a bionic method of the group behavior of birds searching for food, searches for the weights and thresholds of the component prediction model, can quickly find the global optimal solution, and avoid falling into the local optimum.

[0055] In some embodiments, optimizing the weights and thresholds in the initial component prediction model by the improved particle swarm optimization algorithm to obtain the target weights and target thresholds includes:

[0056] Step 1: Establish an initial particle swarm. The initial particle swarm includes a preset number of particles. Each particle has a random initial velocity and initial position, and the position of each particle corresponds to a set of weights and thresholds;

[0057] Step 2: Calculate the fitness value of each particle;

[0058] Step 3: For each particle, compare the current fitness value of the particle with the historical best fitness value of the particle. If the current fitness value of the particle is higher than the historical best fitness value of the particle, then use the current fitness value of the particle as the historical best fitness value of the particle;

[0059] Step 4: For each particle, compare the current fitness value of the particle with the global best fitness value of the particle swarm. If the current fitness value of the particle is higher than the global best fitness value of the particle swarm, then use the current fitness value of the particle as the global best fitness value of the particle swarm;

[0060] Step 5: Update the velocity and position of each particle based on the historical best fitness value of each particle and the global best fitness value of the particle swarm;

[0061] Step 6: Repeat Step 2 to Step 5 until the preset number of iterations is reached, and output the current global best position as the target weights and target thresholds.

[0062] In this embodiment, specifically, the number of particles can be set to 100, the initial positions of each particle are random numbers between [0, 1], the initial velocity is 1, and the number of iterations is 500. At the same time, the learning rate of the component prediction model is set to 0.03.

[0063] In some embodiments, before training a component prediction model with target weights and a target threshold based on a training sample set to obtain a trained component prediction model, the method further includes:

[0064] Obtain the coal ash data of various coals; the coal ash data includes coal parameters, combustion parameters, and coal ash components;

[0065] For each coal ash data, use the coal parameters and combustion parameters of the coal ash data as samples, and the coal ash components as the labels of the samples;

[0066] Form a training sample set from the samples and their corresponding labels.

[0067] In this embodiment, the number of samples can specifically be 400, and it includes 150 coal types, including single coals and blended coals. The number of samples can also be modified according to the actual situation, but it should depend on the number of nodes of the component prediction model, and should not be too high or too low to avoid overfitting of the trained component prediction model.

[0068] The prediction method for coal ash components provided by the embodiments of the present invention includes: obtaining the coal parameters of the target coal, where the coal parameters include coal components, coal shape, and coal quantity; obtaining the combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume, and secondary air baffle opening; inputting the coal parameters and combustion parameters into the trained component prediction model to obtain the coal ash components of the coal ash generated after the target coal burns under the combustion parameters; the component prediction model is a BP neural network, and the coal ash components include the mass fractions of various metal oxides in the coal ash. The present invention predicts the coal ash components after coal combustion through a BP neural network, can determine the coal ash components before coal combustion, and thus can change the coal ash components by adjusting the combustion parameters of the coal, improving the slagging and fouling phenomena on the water-cooled wall of the tail flue of the power station boiler.

[0069] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution is prior or subsequent. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0070] The following is an apparatus embodiment of the present invention. For the details not described in detail therein, reference can be made to the corresponding method embodiments above.

[0071] Figure 2 The structural schematic diagram of the prediction apparatus for coal ash components provided by the embodiments of the present invention is shown. For the sake of convenience of description, only the parts related to the embodiments of the present invention are shown and are described in detail as follows:

[0072] As Figure 2 shown, the prediction apparatus 2 for coal ash components includes:

[0073] The first acquisition module 21 is configured to acquire the coal parameters of the target coal, where the coal parameters include coal composition, coal shape, and coal quantity;

[0074] The second acquisition module 22 is configured to acquire the combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume, and secondary air damper opening;

[0075] The prediction module 23 is configured to input the coal parameters and the combustion parameters into a trained composition prediction model to obtain the ash composition of the coal ash generated after the target coal burns under the combustion parameters; the composition prediction model is a BP neural network, and the ash composition includes the mass fractions of various metal oxides in the coal ash.

[0076] In some embodiments, the ash composition prediction device 2 further includes:

[0077] The model establishment module is configured to establish an initial composition prediction model;

[0078] The model training module is configured to train the initial composition prediction model based on an improved particle swarm optimization algorithm and a training sample set to obtain a trained composition prediction model.

[0079] In some embodiments, the model training module includes:

[0080] The optimization unit is configured to optimize the weights and thresholds in the initial composition prediction model through an improved particle swarm optimization algorithm to obtain target weights and target thresholds; wherein, the items to be optimized by the improved particle swarm optimization algorithm are the weights and thresholds in the initial composition prediction model, the objective function is the reciprocal of the prediction error of the composition prediction model, and the optimization goal is to maximize the objective function;

[0081] The training unit is configured to train the composition prediction model using the target weights and target thresholds based on the training sample set to obtain a trained composition prediction model.

[0082] In some embodiments, the optimization unit is specifically configured to:

[0083] Step 1: Establish an initial particle swarm, where the initial particle swarm includes a preset number of particles, each particle has a random initial velocity and an initial position, and the position of each particle corresponds to a set of weights and thresholds;

[0084] Step 2: Calculate the fitness value of each particle;

[0085] Step 3: For each particle, compare the current fitness value of the particle with the historical best fitness value of the particle. If the current fitness value of the particle is higher than the historical best fitness value of the particle, then use the current fitness value of the particle as the historical best fitness value of the particle;

[0086] Step 4: For each particle, compare the current fitness value of the particle with the global best fitness value of the particle swarm. If the current fitness value of the particle is higher than the global best fitness value of the particle swarm, then use the current fitness value of the particle as the global best fitness value of the particle swarm;

[0087] Step 5: Update the velocity and position of each particle based on the historical best fitness value of each particle and the global best fitness value of the particle swarm;

[0088] Step 6: Repeat Step 2 to Step 5 until a preset number of iterations is reached, and output the current global best position as the target weight and target threshold.

[0089] In some embodiments, the prediction device 2 for coal ash composition further includes:

[0090] A third acquisition module, configured to acquire coal ash data of multiple coals before training a composition prediction model using the target weight and target threshold based on a training sample set to obtain a trained composition prediction model; the coal ash data includes coal parameters, combustion parameters, and coal ash composition;

[0091] A sample generation module, configured to use the coal parameters and combustion parameters of the coal ash data as samples for each coal ash data, and the coal ash composition as the label of the sample;

[0092] A sample set establishment module, configured to form a training sample set from each sample and the corresponding label.

[0093] The prediction device for coal ash composition provided by the embodiments of the present invention includes: a first acquisition module, configured to acquire coal parameters of target coal, where the coal parameters include coal composition, coal shape, and coal quantity; a second acquisition module, configured to acquire combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume, and secondary air damper opening; a prediction module, configured to input the coal parameters and combustion parameters into a trained composition prediction model to obtain the coal ash composition of the coal ash generated after the target coal burns under the combustion parameters; the composition prediction model is a BP neural network, and the coal ash composition includes the mass fractions of various metal oxides in the coal ash. The present invention predicts the coal ash composition after coal combustion through a BP neural network, can determine the coal ash composition before coal combustion, and thus can change the coal ash composition by adjusting the combustion parameters of the coal, improving the slagging and fouling phenomena on the water-cooled wall in the tail flue of a power station boiler.

[0094] Figure 3 is a schematic diagram of the terminal provided by the embodiments of the present invention. As Figure 3As shown, the terminal 3 of this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30. When the processor 30 executes the computer program 32, it implements the steps in the embodiments of the prediction method for each coal ash component described above. For example Figure 1 the steps 101 to 103 shown. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module in the above device embodiments. For example Figure 2 the functions of the modules 21 to 23 shown.

[0095] Exemplarily, the computer program 32 can be divided into one or more modules. The one or more modules are stored in the memory 31 and executed by the processor 30 to complete the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program 32 in the terminal 3. For example, the computer program 32 can be divided into Figure 2 the modules 21 to 23 shown.

[0096] The terminal 3 can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art can understand that Figure 3 merely examples of the terminal 3, which do not constitute a limitation on the terminal 3. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the terminal may further include input / output devices, network access devices, a bus, etc.

[0097] The so-called processor 30 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0098] The memory 31 may be an internal storage unit of the terminal 3, such as a hard disk or a memory of the terminal 3. The memory 31 may also be an external storage device of the terminal 3, such as a plug-in hard disk equipped on the terminal 3, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 31 may also include both the internal storage unit of the terminal 3 and the external storage device. The memory 31 is used to store the computer program and other programs and data required by the terminal. The memory 31 may also be used to temporarily store the data that has been output or will be output.

[0099] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be described in detail here.

[0100] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0101] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0102] In the embodiments provided by the present invention, it should be understood that the disclosed device / terminal and method can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in electrical, mechanical or other forms.

[0103] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0104] In addition, each functional unit in various embodiments of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0105] If the integrated module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments for predicting various coal ash components can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0106] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for predicting the composition of coal ash, characterized in that, Including: Obtain the coal parameters of the target coal, where the coal parameters include coal composition, coal shape, and coal quantity; among them, the coal shape includes pulverized coal and medium particles; Obtain the combustion parameters of the target coal, where the combustion parameters include combustion mode, combustion duration, primary air volume, and secondary air damper opening; among them, the combustion mode includes grate firing, suspension combustion, and fluidized bed combustion; Input the coal parameters and the combustion parameters into a trained composition prediction model to obtain the ash composition of the ash generated after the target coal burns under the combustion parameters; the composition prediction model is a BP neural network, and the ash composition includes the mass fractions of various metal oxides in the ash. The composition prediction model is a BP neural network with a structure of 15-5-3-9. Among them, 15 input nodes respectively correspond to the contents of Si, Al, Fe, S, Ca, K, Na, Ti, and Mg in the coal, coal shape, coal quantity, combustion mode, combustion time, primary air volume, and secondary air damper opening, and the output nodes are 9, respectively corresponding to the mass fractions of SiO2, Al2O3, Fe2O3, SO3, CaO, K2O, Na2O, TiO2, and MgO in the ash.

2. The method for predicting the composition of coal ash according to claim 1, characterized in that, Before inputting the coal parameters and the combustion parameters into a trained composition prediction model to obtain the ash composition of the ash generated after the target coal burns under the combustion parameters, the method further includes: Establish an initial composition prediction model; Train the initial composition prediction model based on an improved particle swarm optimization algorithm and a training sample set to obtain a trained composition prediction model.

3. The method for predicting the composition of coal ash according to claim 2, characterized in that, The training of the initial composition prediction model based on the improved particle swarm optimization algorithm and the training sample set to obtain a trained composition prediction model includes: Optimize the weights and thresholds in the initial composition prediction model through the improved particle swarm optimization algorithm to obtain target weights and target thresholds; among them, the items to be optimized by the improved particle swarm optimization algorithm are the weights and thresholds in the initial composition prediction model, the objective function is the reciprocal of the prediction error of the composition prediction model, and the optimization objective is to maximize the objective function; Train the composition prediction model with the target weights and the target thresholds based on the training sample set to obtain a trained composition prediction model.

4. The method for predicting the composition of coal ash according to claim 3, characterized in that, The optimizing the weights and thresholds in the initial composition prediction model through the improved particle swarm optimization algorithm to obtain target weights and target thresholds includes: Step 1: Establish an initial particle swarm, where the initial particle swarm includes a preset number of particles, each particle has a random initial velocity and initial position, and the position of each particle corresponds to a set of weights and thresholds; Step 2: Calculate the fitness value of each particle; Step 3: For each particle, compare the current fitness value of the particle with the historical best fitness value of the particle. If the current fitness value of the particle is higher than the historical best fitness value of the particle, then use the current fitness value of the particle as the historical best fitness value of the particle; Step 4: For each particle, compare the current fitness value of the particle with the global best fitness value of the particle swarm. If the current fitness value of the particle is higher than the global best fitness value of the particle swarm, then use the current fitness value of the particle as the global best fitness value of the particle swarm; Step 5: Update the velocity and position of each particle based on the historical best fitness value of each particle and the global best fitness value of the particle swarm; Step 6: Repeat Step 2 to Step 5 until a preset number of iterations is reached, and output the current global best position as the target weight and the target threshold.

5. The method for predicting the composition of coal ash according to claim 3, characterized in that, Before training the component prediction model using the target weight and the target threshold based on the training sample set to obtain a trained component prediction model, the method further includes: Obtain the coal ash data of multiple coals; the coal ash data includes coal parameters, combustion parameters, and coal ash components; For each coal ash data, use the coal parameters and combustion parameters of the coal ash data as samples, and the coal ash components as the labels of the samples; Form a training sample set with each sample and its corresponding label.

6. A device for predicting the composition of coal ash, characterized in that, Includes: The first acquisition module is used to acquire the coal parameters of the target coal, and the coal parameters include coal components, coal shape, and coal quantity; wherein, the coal shape includes pulverized coal and medium particles; The second acquisition module is used to acquire the combustion parameters of the target coal, and the combustion parameters include combustion mode, combustion duration, primary air volume, and secondary air damper opening; wherein, the combustion mode includes stoker firing, suspension combustion, and fluidized bed combustion; The prediction module is used to input the coal parameters and the combustion parameters into the trained component prediction model to obtain the coal ash components of the coal ash generated after the target coal burns under the combustion parameters; the component prediction model is a BP neural network, and the coal ash components include the mass fractions of various metal oxides in the coal ash. The component prediction model is a BP neural network with a structure of 15-5-3-9. Among them, 15 input nodes respectively correspond to the contents of Si, Al, Fe, S, Ca, K, Na, Ti, Mg in the coal, coal shape, coal quantity, combustion mode, combustion time, primary air volume, and secondary air damper opening, and the output nodes are 9, respectively corresponding to the mass fractions of SiO2, Al2O3, Fe2O3, SO3, CaO, K2O, Na2O, TiO2, MgO in the coal ash.

7. The device for predicting the composition of coal ash according to claim 6, characterized in that, The device further includes: The model establishment module is used to establish an initial component prediction model; The model training module is used to train the initial component prediction model based on the improved particle swarm algorithm and the training sample set to obtain a trained component prediction model.

8. The device for predicting the composition of coal ash according to claim 7, characterized in that, The model training module includes: The optimization unit is used to optimize the weights and thresholds in the initial component prediction model through the improved particle swarm algorithm to obtain the target weights and the target threshold; wherein, the items to be optimized by the improved particle swarm algorithm are the weights and thresholds in the initial component prediction model, the objective function is the reciprocal of the prediction error of the component prediction model, and the optimization objective is to maximize the objective function; A training unit trains a component prediction model using the target weight and the target threshold based on the training sample set to obtain a trained component prediction model.

9. A terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the prediction method for coal ash components described in any one of claims 1 to 5 above.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the prediction method for coal ash components described in any one of claims 1 to 5 above.

Images