Method for adjusting resistance balance of dust removal pipe network based on analog simulation and artificial intelligence

Abstract

The invention provides a method for adjusting the resistance balance of a dust removal pipe network based on analog simulation and artificial intelligence, and the method comprises the steps: carrying out the equivalent replacement of a bent pipe in a dust removal pipe network system through a jumping boundary model, and then carrying out the equivalent replacement of the bent pipe through a mode of combining numerical simulation with an artificial neural network. The optimal design scheme for adjusting the resistance balance of the dust removal pipe network system is obtained efficiently at low cost, excessive dependence on traditional design manual and experience of designers in the resistance balance adjusting process is avoided, and the method is suitable for large-scale application and popularization.

Description

Technical Field

[0001] The invention relates to the technical field of numerical simulation, and in particular to a method for adjusting the resistance balance of a dust removal pipe network based on simulation and artificial intelligence. Background Art

[0002] The dust removal pipe network system is an effective method for controlling and reducing dust pollution generated in industrial production processes. Its main function is to discharge the dirty, harmful particles and dust emitted from local areas or equipment in industrial plants to the outside after being treated by purification equipment, and at the same time, to deliver fresh air that meets the standards to various required air supply points indoors. Among them, the design of the pipe network system involves factors such as the spatial position, type, size of pipes and pipe fittings, and the resistance loss generated when the fluid flows through the pipe fittings. At the same time, the pipe network system has problems such as complex structure, numerous branch pipes, and limited installation space, which leads to the operating air volume not being consistent with the design and it is difficult to achieve the expected design effect. In order to ensure that the pipe network system can effectively collect dust and meet the design standards, a higher exhaust fan power is usually required. This not only increases the cost, but also causes unnecessary energy waste. Therefore, optimizing the theoretical design scheme to ensure that the pipe network system effectively collects dust and achieves the design goals has become a crucial link in adjusting the resistance balance of the system.

[0003] In a complex pipe network system, the resistance loss of the pipeline is divided into two parts: along-the-line resistance loss and local resistance loss. The along-the-line resistance loss refers to the resistance generated when the fluid flows in the straight pipe section. However, in order to ensure the integrity of the entire pipe network system, it is usually impossible to adjust the resistance balance by directly adjusting the along-the-line resistance loss caused by the straight pipe section. The local resistance loss is mainly caused by the drastic change in flow velocity or direction when the fluid flows through the pipe fittings, which is also the key factor that causes the theoretical design air volume of the pipe network to fail. Therefore, the local resistance loss can be changed by manually controlling the opening of the pipeline valve, thereby achieving the resistance balance of the actual pipe network system.

[0004] The dust removal pipe network system itself has the characteristics of complex structure and numerous branch pipes, which makes the numerical modeling process complicated and the calculation efficiency of numerical simulation experiments low. Among them, elbows, as the most common pipe fittings in the pipe network system, are also one of the main factors leading to pipeline resistance loss and high energy consumption of fans.

[0005] For some common local resistance losses in elbows, predecessors have conducted a large number of experiments and summarized relatively complete theories. However, due to the limitations of experiments and incomplete data, it is usually necessary to rely on design manuals and engineers' experience when designing pipelines and adjusting resistance balance.

[0006] CN116144862A discloses a method and device for balancing and adjusting the air volume of a blast furnace dust removal system, the method comprising: determining the resistance of the branch pipe with the largest resistance in the pipeline network of the blast furnace dust removal system as the maximum resistance; determining the resistance of the branch pipe to be adjusted; calculating the difference between the maximum resistance and the resistance of the branch pipe to be adjusted as the control resistance; adjusting the resistance of the branch pipe to be adjusted based on the control resistance. The method uses the control resistance to adjust the resistance value of the branch pipe, quickly and effectively balances the air volume distribution of each branch pipe, and improves the dust collection effect.

[0007] CN117019810A discloses an environmental dust removal system and a resistance balance adjustment method thereof. The environmental dust removal system includes a chimney, a fan and a dust collector connected in sequence. The dust collector is connected to a dust suction pipe. A plurality of dust suction points are arranged on the dust suction pipe. Each dust suction point is equipped with an electric resistance balance valve and a wind speed detector. The wind speed detector at each dust suction point detects the wind speed at the dust suction point. According to the wind speed at each dust suction point, the opening of the corresponding electric resistance balance valve is controlled so that the deviation value between the wind speed at each dust suction point and the wind speed setting value is within the set range. The scheme adopts automatic control, by setting a wind speed detector at each dust suction point, each dust suction point adopts an electric resistance balance valve, the wind speed detector collects the wind speed at each dust suction point and compares the wind speed with the set value, and controls and adjusts the opening of the resistance balance valve according to the comparison result to achieve resistance balance of the pipe network system.

[0008] CN103185610A discloses a method for evaluating and debugging the resistance balance of a dust removal system network. Based on the test of the operating parameters of the dust removal system network, the operating parameters of the dust removal system network are compared with the original design parameters of the network, thereby evaluating the resistance balance of the dust removal system network, and rebalancing the resistance of the dust removal system network through on-site debugging of the resistance of the dust removal system network, ensuring that the dust removal system efficiently captures the dust generated by each dust suction point and purifies the workplace environment. It not only provides reliable technical parameters and theoretical basis for the management of the factory's dust removal system, making production operation, management, maintenance and repair more targeted, but also has theoretical and practical guiding significance for on-site debugging and major and medium-sized modifications of the dust removal system network.

[0009] However, the above method for balancing the resistance of the dust removal pipe network is relatively complicated to operate, and the optimal design solution for resistance balance cannot be obtained quickly. Summary of the invention

[0010] In view of the problems existing in the prior art, the present invention provides a method for adjusting the resistance balance of the dust removal pipe network based on simulation and artificial intelligence. First, the elbows in the dust removal pipe network system are equivalently replaced to obtain a simplified model, and then numerical simulation is used and an artificial neural network is constructed to obtain the optimal design scheme for adjusting the resistance balance of the dust removal pipe network system. The method of the present invention reduces the modeling difficulty of the dust removal pipe network system and provides the computational efficiency of the numerical simulation experiment of the entire dust removal pipe network system.

[0011] To achieve this object, the present invention adopts the following technical solutions:

[0012] The present invention provides a method for adjusting the resistance balance of a dust removal pipe network based on simulation and artificial intelligence, and the method comprises the following steps:

[0013] (1) Based on the jumping boundary model of numerical simulation, the elbows in the dust removal pipe network system are replaced equivalently to obtain an equivalent simplified model of the elbows;

[0014] (2) Conduct numerical simulation experiments on the dust removal pipe network system under different working conditions to obtain the resistance distribution information of each pipeline in the dust removal pipe network system and form a numerical simulation data set;

[0015] (3) An artificial neural network is constructed based on the numerical simulation data set to obtain the optimal design solution for adjusting the resistance balance of the dust removal pipe network system.

[0016] The method for adjusting the resistance balance of the dust removal pipe network based on simulation and artificial intelligence described in the present invention performs equivalent replacement of the curved pipes in the dust removal pipe network system through the jumping boundary model, thereby achieving model simplification, which can greatly improve the calculation efficiency of the numerical simulation, and then conducts a numerical simulation experiment to obtain a numerical simulation data set; finally, an artificial neural network is constructed to predict whether the pipeline valve needs to be adjusted, and form an optimal design scheme for adjusting the resistance balance of the dust removal pipe network system. The present invention obtains the optimal design scheme for adjusting the resistance balance of the dust removal pipe network system efficiently and at low cost by combining numerical simulation with artificial neural networks. The method described in the present invention efficiently and accurately adjusts the resistance balance of the pipe network system, and avoids excessive reliance on traditional design manuals and the experience of designers in the process of adjusting the resistance balance, and is suitable for large-scale promotion and application.

[0017] Preferably, the jumping boundary model based on numerical simulation in step (1) performs equivalent replacement on the elbow in the dust removal pipe network system to obtain an equivalent simplified model of the elbow, which specifically includes:

[0018] Firstly, CFD software is used to simulate the resistance loss of the fluid in the elbow. By setting the jump boundary model parameters, the fluid will produce the same resistance loss as the elbow after passing through the jump boundary. Then, a boundary is used to replace the elbow position in the dust removal pipe network system to obtain an equivalent simplified model of the elbow.

[0019] The elbow equivalent simplified model described in the present invention realizes accurate simulation of the pressure loss caused by the elbow in the dust removal pipe network system by reasonably setting the jump boundary model parameters, thereby effectively replacing the actual position of the elbow in the dust removal pipe network system. It not only significantly reduces the modeling difficulty of the dust removal pipe network system, but also greatly improves the computational efficiency of the numerical simulation experiment of the entire dust removal pipe network system.

[0020] Preferably, the jump boundary model parameters include surface permeability, porous medium thickness and pressure jump coefficient;

[0021] Preferably, the method for calculating the surface permeability and the pressure jump coefficient specifically includes:

[0022] By fitting the resistance loss obtained from the elbow simulation and its corresponding Reynolds number, a quadratic function without a constant term is constructed:

[0023] ΔP=n1Re 2 +n2Re (1) Then, the function is compared with Formula 2 to determine the surface permeability and pressure jump coefficient required for the jump boundary model:

[0024]

[0025] Among them, n1 and n2 are parameters obtained after fitting the function, ΔP represents the resistance loss, Re represents the Reynolds number, μ represents the dynamic viscosity of the fluid, ρ represents the fluid density, α represents the surface permeability, C2 represents the pressure jump coefficient, and Δm represents the thickness of the porous medium, and its default value is set to 0.01.

[0026] Preferably, the different operating conditions in step (2) are operating conditions formed by shutting down and stopping different pipelines in the dust removal pipe network system.

[0027] Preferably, the step (2) of conducting a numerical simulation experiment on the dust removal pipe network system under different working conditions to obtain the resistance distribution information of each pipeline of the dust removal pipe network system specifically includes:

[0028] The dust removal pipe network system is three-dimensionally modeled using modeling software, and then the model is meshed and iteratively solved using CFD software to simulate the flow process of the fluid in the complex pipe network system and obtain the resistance distribution information of each pipeline in the dust removal pipe network system.

[0029] Preferably, the method of forming a numerical simulation data set in step (2) comprises:

[0030] Based on simulation examples under different working conditions, the resistance distribution information of each pipeline in the dust removal pipe network system is obtained, and the resistance is balanced by adjusting the valves between different pipelines to obtain the numerical simulation data set under the current working conditions.

[0031] Preferably, the artificial neural network in step (3) is composed of an input layer, a hidden layer and an output layer, which includes input nodes and output nodes with the same number as the import points, and a BPNN model of a hidden layer; the input nodes are the switch status of each import point, and the output nodes are whether the currently predicted pipeline valve needs to be adjusted.

[0032] Preferably, the artificial neural network in step (3) is trained using the numerical simulation data set as a training sample.

[0033] Preferably, the basis for determining the best design scheme for adjusting the resistance balance of the dust removal pipe network system in step (3) is that the error between the operating air volume of the simulation result and the design air volume meets the operating requirements of the dust removal pipe network system.

[0034] Preferably, the step (3) constructs an artificial neural network based on the numerical simulation data set to obtain the optimal design scheme for adjusting the resistance balance of the dust removal pipe network system, specifically including:

[0035] Artificial neural networks are used to predict the adjustment requirements of valves in each pipeline of the dust removal pipe network system, and the optimal design scheme is formed based on the prediction results.

[0036] As a preferred technical solution of the present invention, the method comprises the following steps:

[0037] (1) First, CFD software was used to simulate the resistance loss of the fluid in the elbow. By setting the jump boundary model parameters, the fluid produced the same resistance loss as the elbow after passing through the jump boundary. Then, a boundary was used to replace the elbow position in the dust removal pipe network system to obtain an equivalent simplified model of the elbow.

[0038] The jump boundary model parameters include surface permeability, porous medium thickness and pressure jump coefficient;

[0039] The method for calculating the surface permeability and the pressure jump coefficient specifically includes:

[0040] By fitting the resistance loss obtained from the elbow simulation and its corresponding Reynolds number, a quadratic function without a constant term is constructed:

[0041] ΔP=n1Re 2 +n2Re (1)

[0042] Then, the function is compared with Formula 2 to determine the surface permeability and pressure jump coefficient required for the jump boundary model:

[0043]

[0044] Among them, n1 and n2 are parameters obtained after fitting the function, ΔP represents the resistance loss, Re represents the Reynolds number, μ represents the dynamic viscosity of the fluid, ρ represents the fluid density, α represents the surface permeability, C2 represents the pressure jump coefficient, and Δm represents the thickness of the porous medium, and its default value is set to 0.01;

[0045] (2) The dust removal pipe network system is three-dimensionally modeled using modeling software, and then the model is meshed and iteratively solved using CFD software to simulate the flow process of the fluid in the complex pipe network system and obtain the resistance distribution information of each pipeline in the dust removal pipe network system; based on simulation examples under different working conditions, the resistance distribution information of each pipeline in the dust removal pipe network system is obtained, and the resistance is balanced by adjusting the valves between different pipelines to obtain the numerical simulation data set under the current working conditions;

[0046] The different working conditions are the working conditions formed by the shutdown operation of different pipelines in the dust removal pipe network system;

[0047] (3) Use artificial neural networks to predict the adjustment requirements of valves in each pipeline of the dust removal pipe network system, and form the best design plan based on the prediction results;

[0048] The artificial neural network is composed of an input layer, a hidden layer and an output layer, which includes input nodes and output nodes with the same number as the inlet points, and a BPNN model of a hidden layer; the input nodes are the switch states of each inlet point, and the output nodes are whether the currently predicted pipeline valve needs to be adjusted; the artificial neural network is trained by using the numerical simulation data set as a training sample;

[0049] The basis for determining the best design scheme for adjusting the resistance balance of the dust removal pipe network system is that the error between the operating air volume of the simulation result and the design air volume meets the operating requirements of the dust removal pipe network system.

[0050] Compared with the prior art, the present invention has at least the following beneficial effects:

[0051] (1) The method for adjusting the resistance balance of the dust removal pipe network based on simulation and artificial intelligence provided by the present invention forms an equivalent alternative model for the bent pipe by setting the jump boundary parameters, which not only reduces the modeling difficulty of the pipe network system, but also greatly improves the computational efficiency of the numerical simulation experiment of the entire dust removal pipe network system.

[0052] (2) The method for adjusting the resistance balance of the dust removal pipe network based on simulation and artificial intelligence provided by the present invention can simulate the entire dust removal pipe network system and can deeply participate in the design and optimization process of such pipe network system.

[0053] (3) The method for adjusting the resistance balance of the dust removal pipe network based on simulation and artificial intelligence provided by the present invention can predict whether the pipeline valve needs to be adjusted through an artificial neural network, and form an optimal design scheme based on the prediction results. This method can not only efficiently and accurately adjust the resistance balance of the pipe network system, but also avoid over-reliance on traditional design manuals and designers' experience in the process of adjusting the resistance balance. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of a method for adjusting the resistance balance of a dust removal pipe network based on simulation and artificial intelligence provided by the present invention.

[0055] Figure 2 It is a pipe bend model with corresponding upstream and downstream pipelines established by using modeling software in a specific embodiment of the present invention.

[0056] Figure 3 It is a simplified model of a bent pipe in a specific embodiment of the present invention.

[0057] Figure 4 It is a schematic diagram of the principle of the jumping boundary model in a specific implementation manner of the present invention.

[0058] Figure 5 It is a comparison diagram of the error of the resistance loss of different types of elbows and their corresponding simplified models in the specific implementation mode of the present invention.

[0059] Figure 6 It is a schematic diagram of a complex pipe network established based on an actual engineering project in a specific implementation manner of the present invention.

[0060] Figure 7 It is a simplified model of a complex pipe network in a specific implementation manner of the present invention.

[0061] Figure 8 It is a comparison diagram of the errors between the complex pipe network model and its corresponding simplified model in a specific implementation manner of the present invention.

[0062] Fig. 9 It is a schematic diagram of the artificial neural network BPNN model in a specific implementation manner of the present invention.

[0063] Fig.10 It is a schematic diagram of the prediction results of the artificial neural network model in a specific implementation manner of the present invention.

[0064] In the figure: 1 to 61 represent the numbers of the corresponding dust collection point entrances. DETAILED DESCRIPTION

[0065] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0066] The present invention provides a method for adjusting the resistance balance of a dust removal pipe network based on simulation and artificial intelligence, and its flow chart is as follows: Figure 1 The method comprises the following steps:

[0067] (1) Based on the jumping boundary model of numerical simulation, the elbows in the dust removal pipe network system are replaced equivalently to obtain an equivalent simplified model of the elbows;

[0068] (2) Conduct numerical simulation experiments on the dust removal pipe network system under different working conditions to obtain the resistance distribution information of each pipeline in the dust removal pipe network system and form a numerical simulation data set;

[0069] (3) An artificial neural network is constructed based on the numerical simulation data set to obtain the optimal design solution for adjusting the resistance balance of the dust removal pipe network system.

[0070] The present invention is further described in detail below. However, the following examples are only simplified examples of the present invention and do not represent or limit the scope of protection of the present invention. The scope of protection of the present invention shall be subject to the claims.

[0071] As a specific embodiment of the present invention, a method for adjusting the resistance balance of a dust removal pipe network based on simulation and artificial intelligence is provided, and the method comprises the following steps:

[0072] (1) According to the actual shape of the elbow, a three-dimensional model is created using modeling software, and corresponding upstream and downstream pipelines are established to ensure that turbulence can be fully developed. The lengths of the upstream and downstream pipelines are set to 20 and 10 times the pipe diameter, respectively. The elbow model is shown in the figure below. Figure 2 Then, CFD software is used to mesh the model and solve it iteratively to simulate the resistance loss generated by the fluid in the entire elbow flow process.

[0073] (2) Establish a horizontal pipeline with the same length as the upstream and downstream pipelines of the elbow, ignore the geometric dimensions of the elbow, and use a special "boundary", that is, the jump boundary in the numerical simulation, so as to replace the position of the elbow in the pipeline with a plane.

[0074] (3) By setting the jump boundary model parameters in the CFD software, the fluid in the horizontal pipeline can generate the same resistance loss as the elbow after passing through the jump boundary.

[0075] The jump boundary model parameters include surface permeability, porous medium thickness and pressure jump coefficient;

[0076] The method for calculating the surface permeability and the pressure jump coefficient specifically includes:

[0077] By fitting the resistance loss obtained from the elbow simulation and its corresponding Reynolds number, a quadratic function without a constant term is constructed:

[0078] ΔP=n1Re 2 +n2Re (1)

[0079] Then, the function is compared with Formula 2 to determine the surface permeability and pressure jump coefficient required for the jump boundary model:

[0080]

[0081] Among them, n1 and n2 are parameters obtained after fitting the function, ΔP represents the resistance loss, Re represents the Reynolds number, μ represents the dynamic viscosity of the fluid, ρ represents the fluid density, α represents the surface permeability, C2 represents the pressure jump coefficient, and Δm represents the thickness of the porous medium, and its default value is set to 0.01.

[0082] Therefore, a set of planes with set parameters can be used to replace the elbow position in the pipe network system, thereby forming an equivalent simplified model of the elbow, such as Figure 3 As shown in the figure, the schematic diagram of the jumping boundary model is as follows Figure 4 shown.

[0083] The error comparison diagram of the resistance loss of different types of elbows and their corresponding simplified models in this specific implementation is as follows: Figure 5 As shown in the figure, it can be seen that the error between the resistance loss generated by the simplified model and the resistance loss actually generated by the elbow is within ±10%, indicating that the simplified model has a higher accuracy and meets the engineering application requirement of an error of ±30%.

[0084] (4) Use CFD software to mesh the model and iteratively solve it to simulate the flow process of the fluid in the complex pipe network system and obtain the resistance distribution information of each pipe in the dust removal pipe network system.

[0085] (5) The simulation of complex pipe network is based on the dust removal pipe network system established in the actual engineering project, and is constructed by using the proportional accurate modeling method. The complex pipe network consists of two main roads and 61 branches. The schematic diagram is shown in Figure 6 shown.

[0086] (6) Using the established equivalent simplified model, the six elbows in the two main roads of the complex pipe network are equivalently replaced with the set plane using parameters, and the simplified model of the complex pipe network is obtained as follows: Figure 7 As shown, the purple part is the location of the jump boundary. CFD software was then used to perform numerical simulation on the simplified pipe network system and compared with the iterative solution rate of the original model.

[0087] (7) Based on the actual project operation (closing the dust collection points numbered 19, 23, 28, 29, 32, and 33), the error comparison between the simplified model and the actual model is shown in the figure below: Figure 8 As shown in the figure, the error of each branch of the entire pipe network system is basically within ±20%, which meets the engineering application requirement of ±30%. Then, by shutting down one dust collection point in sequence from 1 to 61, 55 different working conditions are generated. Subsequently, the numerical simulation method is used to simulate these different working conditions to obtain the air volume information of each branch dust collection point.

[0088] (8) By comparing the simulated air volume at each branch suction point with the actual design air volume, when the error between the simulated air volume and the design air volume exceeds 30%, it indicates that there is a resistance imbalance problem, causing the air volume of some branches to exceed the design value. Therefore, it is necessary to adjust the valves of these pipelines that exceed the design air volume, and reduce the simulated air volume by reducing the valve opening. Subsequently, the adjusted pipeline network is simulated again, and the air volume is compared. After multiple adjustments to the valves that exceed the design air volume, an optimization plan for the current working conditions is finally formed. Based on the final optimization plan results under these 55 different working conditions, a numerical simulation data set for adjusting the resistance balance of the pipeline network under different working conditions is formed.

[0089] (9) Construct a BP neural network (BPNN) model consisting of an input layer, a hidden layer, and an output layer. The schematic diagram is shown in the figure. Fig. 9 As shown. The number of input nodes of the model is the same as the number of import points. The input nodes correspond to the switch status of each import point, and the output nodes indicate whether the currently predicted pipeline valve needs to be adjusted. The artificial neural network is trained through the constructed numerical simulation data set. After the training is completed, the model can be used to directly predict the adjustment requirements of pipeline valves without additional numerical simulation experiments, and form the best design scheme for adjusting the resistance balance of the pipeline network according to the adjustment requirements of each pipeline valve, so that the pipeline can achieve resistance balance and meet the air volume design requirements of each branch.

[0090] The schematic diagram of the prediction result of the artificial neural network model in this specific implementation method is as follows Fig.10 As shown. The training model based on the artificial neural network model for closing a single outlet, and then the prediction results of randomly closing two and three outlets, because the model output is an array, it is necessary to round the prediction results. When outputting, 0 means that the valve of the current inlet pipeline does not need to be adjusted, and 1 means that it needs to be adjusted. It can be seen from the figure that this model can accurately predict the results.

[0091] In summary, the method for adjusting the resistance balance of the dust removal pipe network based on simulation and artificial intelligence provided by the present invention forms an equivalent substitution model for the bent pipe by setting the jump boundary parameters, which greatly reduces the difficulty of modeling the pipe network system and improves the computational efficiency of the numerical simulation experiment of the entire dust removal pipe network system; then, through the combination of numerical simulation and artificial neural network, the optimal design scheme for adjusting the resistance balance of the dust removal pipe network system is obtained efficiently and at low cost, avoiding excessive reliance on traditional design manuals and the experience of designers in the process of adjusting the resistance balance.

[0092] The applicant declares that the above is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention are within the protection scope and disclosure scope of the present invention.

Claims

1. A method for adjusting the resistance balance of a dust removal pipe network based on simulation and artificial intelligence, characterized in that: The method comprises the following steps: (1) Based on the jumping boundary model of numerical simulation, the elbows in the dust removal pipe network system are replaced equivalently to obtain an equivalent simplified model of the elbows; (2) Conduct numerical simulation experiments on the dust removal pipe network system under different working conditions to obtain the resistance distribution information of each pipeline in the dust removal pipe network system and form a numerical simulation data set; (3) An artificial neural network is constructed based on the numerical simulation data set to obtain the optimal design solution for adjusting the resistance balance of the dust removal pipe network system.

2. The method according to claim 1, characterized in that The jumping boundary model based on numerical simulation in step (1) performs equivalent replacement on the elbow in the dust removal pipe network system to obtain an equivalent simplified model of the elbow, which specifically includes: Firstly, CFD software is used to simulate the resistance loss of the fluid in the elbow. By setting the jump boundary model parameters, the fluid will produce the same resistance loss as the elbow after passing through the jump boundary. Then, a boundary is used to replace the elbow position in the dust removal pipe network system to obtain an equivalent simplified model of the elbow.

3. The method according to claim 2, characterized in that The jump boundary model parameters include surface permeability, porous medium thickness and pressure jump coefficient; Preferably, the method for calculating the surface permeability and the pressure jump coefficient specifically includes: By fitting the resistance loss obtained from the elbow simulation and its corresponding Reynolds number, a quadratic function without a constant term is constructed: ΔP=n1Re 2 +n2Re (1) Then, the function is compared with Formula 2 to determine the surface permeability and pressure jump coefficient required for the jump boundary model: Among them, n1 and n2 are parameters obtained after fitting the function, ΔP represents the resistance loss, Re represents the Reynolds number, μ represents the dynamic viscosity of the fluid, ρ represents the fluid density, α represents the surface permeability, C2 represents the pressure jump coefficient, and Δm represents the thickness of the porous medium, and its default value is set to 0.

01.

4. The method according to any one of claims 1 to 3, characterized in that: The different operating conditions described in step (2) are the operating conditions formed by shutting down and stopping different pipelines in the dust removal pipe network system.

5. The method according to any one of claims 1 to 4, characterized in that: The step (2) of conducting numerical simulation experiments on the dust removal pipe network system under different working conditions to obtain the resistance distribution information of each pipe of the dust removal pipe network system specifically includes: The dust removal pipe network system is three-dimensionally modeled using modeling software, and then the model is meshed and iteratively solved using CFD software to simulate the flow process of the fluid in the complex pipe network system and obtain the resistance distribution information of each pipeline in the dust removal pipe network system.

6. The method according to any one of claims 1 to 5, characterized in that: The method for forming a numerical simulation data set in step (2) includes: Based on simulation examples under different working conditions, the resistance distribution information of each pipeline in the dust removal pipe network system is obtained, and the resistance is balanced by adjusting the valves between different pipelines to obtain the numerical simulation data set under the current working conditions.

7. The method according to any one of claims 1 to 6, characterized in that: The artificial neural network in step (3) is composed of an input layer, a hidden layer and an output layer, which includes input nodes and output nodes with the same number as the import points, and a BPNN model of a hidden layer; the input nodes are the switch status of each import point, and the output nodes are whether the currently predicted pipeline valve needs to be adjusted.

8. The method according to any one of claims 1 to 7, characterized in that: In step (3), the artificial neural network is trained using the numerical simulation data set as a training sample.

9. The method according to any one of claims 1 to 8, characterized in that: The basis for determining the best design scheme for adjusting the resistance balance of the dust removal pipe network system in step (3) is that the error between the operating air volume of the simulation result and the design air volume meets the operating requirements of the dust removal pipe network system.

10. The method according to any one of claims 1 to 9, characterized in that: The step (3) constructs an artificial neural network based on the numerical simulation data set to obtain the optimal design scheme for adjusting the resistance balance of the dust removal pipe network system, specifically including: Artificial neural networks are used to predict the adjustment requirements of valves in each pipeline of the dust removal pipe network system, and the optimal design scheme is formed based on the prediction results.

Images