Cyclone separator optimization system and method based on machine learning

Abstract

The invention relates to the technical field of separation devices, and discloses a cyclone separator optimization system and method based on machine learning, and the system comprises a user operation module, a numerical simulation module and a deep learning module. The method comprises the steps that when a user terminal inputs cyclone separator parameters in a cloud platform, a system calls a cyclone separator parameter database; if no corresponding cyclone separator parameter exists in the cyclone separator parameter database, performing analogue simulation calculation according to the cyclone separator parameter input by a user, and establishing a flow prediction model according to a simulation result; the flow prediction model performs prediction according to the simulation result; after the prediction is finished, performing comparison according to a prediction result; and finally obtaining optimized parameters required by the parameters of the cyclone separator, and transmitting the optimized parameters to a user terminal through the cloud platform. According to the method, the separation efficiency of cyclone separators with different designs is predicted, and manpower, material resources, time, cost and the like are saved in actual production.

Description

Technical Field

[0001] The present invention relates to the technical field of separation devices, and in particular to a cyclone separator optimization system and method based on machine learning. Background Art

[0002] The cyclone separator uses the centrifugal force generated by the high-speed rotation of the dust-laden airflow inside it to separate the dust particles from the airflow and capture them on the wall. Under the action of gravity, the dust particles finally fall into the ash hopper, thereby achieving the purpose of dust removal. This type of separation and dust removal equipment is widely used in steel metallurgy, chemical industry, thermal power generation and other industrial fields due to its high dust removal efficiency, simple structure, low cost, easy operation and high reliability. In recent years, environmental problems have become increasingly prominent, especially smoke and dust pollution can cause a variety of respiratory diseases, so the development of efficient cyclone separation equipment is of great significance. Separation efficiency is one of the main performances of the cyclone separator, so most inventions focus on solving the separation efficiency of the cyclone separator.

[0003] In addition to being closely related to the particle size, the separation efficiency of the cyclone separator is also affected by the separator barrel diameter, separator barrel length, separator inlet height and other structures. For operations in different scenarios, the cyclone separator needs to be designed separately. The existing design method is to determine the optimal design parameters by repeating a series of processes, which include: collecting, analyzing and processing the initial process parameters; calculating the actual process parameters under the working conditions; designing the cyclone single tube distribution structure; designing and calculating the cyclone separator structure size; estimating the separator pressure drop; estimating the separator separation particle efficiency; calculating the separator wall thickness and strength verification; all these calculations are completed by the designer manually. This method has problems such as large workload, long adjustment time, long design cycle, low calculation accuracy, easy deviation, high cost, etc. At the same time, this method generally requires designers to have a deep professional knowledge background and rich design experience to conduct empirical speculation design. Not everyone can use this method.

[0004] Prior art 1, Chinese patent, application number: CN202322225931.9 relates to a cyclone separator dust removal circulation system, including a cyclone separator, a fan and a raw material mechanism, one end of the raw material mechanism is provided with a feed port, the bottom of the other end of the raw material mechanism is provided with a hard impurity outlet and a clean cotton seed outlet respectively, the top of the other end of the raw material mechanism is connected to the inlet of the cyclone separator through an air outlet duct; the fan is connected to the part of the raw material mechanism close to the hard impurity outlet through an air supply duct, and the top of the cyclone separator is connected to the air inlet of the fan through a return air duct. The beneficial effect is that the structure is simple and the design is reasonable. Although the cyclone separator tail gas is recycled and treated, the discharged gas has a higher cleanliness, is energy-saving and environmentally friendly, and has a lower cost, but the dust removal efficiency of the cyclone separator is low, especially when dealing with fine particles, it is difficult to achieve an ideal separation effect.

[0005] Prior art 2, Chinese patent, application number: CN202410012033.1 discloses a cyclone separator, cyclone separation system and operation method, the cyclone separator includes: a separator body, provided with an inlet end and an outlet end; an air riser, provided at the outlet end and extending into the separator body; an adjusting member, provided inside the air riser and movably connected to the air riser; a moving assembly, provided on the outside of the air riser and capable of driving the adjusting member to move, the moving assembly can drive the adjusting member to move up and down in the separator body, so as to adjust the separation efficiency of the separator body by changing the insertion depth of the adjusting member. The above-mentioned cyclone separator can realize online adjustment of the particle escape rate. Although the adjustment of the separation efficiency of the separator is realized, the structure is simple and effective, and it can be applied to different working conditions, and has good application value, but in the case of significant amplification effect, the performance of the cyclone separator may be negatively affected, resulting in a decrease in separation efficiency.

[0006] Prior art three, Chinese patent, application number CN202310923160.2 discloses a cyclone separator control method, system and equipment that adapts to flow changes, and relates to the field of cyclone separator technology. The method includes: constructing a cyclone separator performance prediction model based on a machine learning method; inputting the gas flow rate at the inlet of the cyclone separator into the cyclone separator performance prediction model to predict the separation efficiency of the cyclone separator; determining the flow control signal according to the difference between the predicted separation efficiency of the cyclone separator and the optimal separation efficiency of the cyclone separator; based on the flow control signal, the fan controller controls the motor frequency, thereby controlling the jet flow entering the cyclone separator, so that the cyclone separator is in a flow mode with optimal separation efficiency. Although the separator can always be in a flow mode with optimal separation efficiency, avoiding the problem of low separation efficiency of the cyclone separator caused by changes in flow parameters during the process, there is no better and precise control of gas flow conditions, which affects the separation effect.

[0007] At present, the dust removal efficiency of the cyclone separators in the prior art one, the prior art two and the prior art three is low, especially when dealing with fine particles, it is difficult to achieve an ideal separation effect; when the amplification effect is significant, the performance of the cyclone separator may be negatively affected, resulting in a decrease in the separation efficiency; there is no better and more precise control of the gas flow conditions, which affects the separation effect. In order to solve the above problems, the present invention provides an optimization system and method for a cyclone separator based on machine learning. Summary of the invention

[0008] The main purpose of the present invention is to provide a cyclone separator optimization system and method based on machine learning to solve the problems in the prior art that the dust removal efficiency of the cyclone separator is low, especially when dealing with fine particles, it is difficult to achieve an ideal separation effect; when the amplification effect is significant, the performance of the cyclone separator may be negatively affected, resulting in a decrease in separation efficiency; there is no better and more precise control of gas flow conditions, which affects the separation effect.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A cyclone separator optimization system based on machine learning, comprising:

[0011] The user operation module is responsible for registering an account when the user logs in for the first time. After successful login, the user inputs the relevant parameters of the cyclone separator through the user terminal operation interface; retrieves the data stored in the database according to the relevant parameters of the cyclone separator input by the user; determines whether the corresponding cyclone separator related parameter data is stored; if it is corresponding, retrieves the database data and outputs it to the user terminal; if not, imports the new data into the numerical simulation module for calculation;

[0012] Among them, the cyclone separator related parameters include: cyclone separator design parameters, working environment, operating parameters, etc.; the cyclone separator parameters include cyclone separator parameters, working environment, operating parameters; the working environment includes working pressure, working temperature, humidity and air density; the operating parameters include inlet velocity, particle size, particle density, particle viscosity, outlet pressure and inlet pressure;

[0013] The numerical simulation module is responsible for receiving the cyclone separator related parameters imported by the user terminal; using SolidWorks to draw the cyclone separator geometric model according to the design parameters of the cyclone separator; simulating the cyclone separator model to calculate the separation efficiency; and sending the separator simulation results and separation efficiency to the deep learning module;

[0014] The deep learning module is responsible for data preprocessing of the simulation results and separation efficiency to establish a data set; sorting the data set according to timestamps, and training the flow prediction model based on the data after timestamp sorting; combining the simulation results with the prediction results to identify the areas that need to be optimized in the flow prediction model; iteratively adjusting the optimized areas to obtain the final cyclone separator parameters; and feeding back the adjustments to the user terminal.

[0015] As a further improvement of the present invention, the user operation module includes:

[0016] The registration submodule is responsible for collecting the face image of the user to be registered, extracting the face features of the face image, generating a face feature vector, reducing the dimension of the face vector, generating a face feature vector; calibrating the feature vector; if it meets the set conditions, inputting the feature vector into the feature library and completing the registration;

[0017] If not, vice versa;

[0018] Among them, the associated feature vector of the user to be registered is compared with all the key features of the feature database where the registered users are located. If the similarity meets the set conditions, a temporary identity is created. If the similarity does not meet the set conditions, identity authentication is performed. If identity authentication fails, the user's face information is collected again;

[0019] The operation submodule is responsible for responding to the data parameters initiated by the user terminal to retrieve the cyclone separator target database, determining whether the data parameters initiated by the user terminal correspond to the data parameters of the cyclone separator target database, and judging whether there are data parameters corresponding to the cyclone separator target database; and outputting the judgment result to the data transmission submodule;

[0020] The data transmission submodule is responsible for outputting the data parameters to the user terminal if there are data parameters corresponding to the cyclone separator target database, and transmitting the cyclone separator parameters transmitted by the user terminal to the numerical simulation module if there are no data parameters corresponding to the cyclone separator target database;

[0021] The generation unit is responsible for generating control instructions from data parameters that do not correspond to the cyclone separator target database, processing the control instructions, and returning if the instructions are wrong, and prompting the user terminal that the input of the control instruction is wrong;

[0022] The command transmission unit is responsible for obtaining the user terminal control instructions and storing the control instructions; processing the transmission data according to the control instructions; and running the compilation device to compile the transmission data;

[0023] The parsing unit is responsible for transmitting the compilation results to the numerical simulation module; during the transmission process, the transmission data is sent to the numerical simulation module through the pin unit; the numerical simulation module performs control simulation according to the compilation results.

[0024] As a further improvement of the present invention, the numerical simulation module includes:

[0025] The first preprocessing submodule is responsible for preprocessing the relevant parameters of the cyclone separator imported by the user terminal, and drawing the geometric model of the cyclone separator using SolidWorks according to the preprocessing results; preprocessing the relevant parameters of the cyclone separator and meshing the cyclone separator model;

[0026] The simulation submodule is responsible for setting the relevant parameters of the simulation equipment through the relevant parameters of the cyclone separator of the preprocessing results; simulating the gas-solid two-phase flow process inside the cyclone separator to determine the flow mode of the cyclone separator; and calculating the separation efficiency according to the flow mode;

[0027] The simulation result transmission submodule is responsible for comparing the separation efficiency with the historical experimental separation efficiency, evaluating the separation efficiency, and sending the simulation results, separation efficiency and evaluation results to the deep learning module.

[0028] As a further improvement of the present invention, the first preprocessing submodule includes:

[0029] The drawing unit is responsible for verifying the relevant parameters of the cyclone separator imported by the user terminal, comparing the relevant parameters with the standard data, verifying the accuracy and completeness of the relevant parameters, and drawing the geometric model of the cyclone separator according to the relevant data after the verification;

[0030] The meshing unit is responsible for dividing the geometric model of the cyclone separator into multiple non-overlapping sub-regions, the node position of each sub-region, and the control volume represented by the node; setting mesh parameters and generating meshes;

[0031] The mesh verification unit is responsible for checking the mesh after it is generated; if the mesh is qualified, the mesh is output to the simulation submodule; if the mesh is unqualified, the unqualified area is checked and the mesh is re-divided in the area.

[0032] As a further improvement of the present invention, the simulation submodule includes:

[0033] The flow judging unit is responsible for judging the flow type according to the target parameters of the cyclone separator after grid division;

[0034] Formula for determining flow type:

[0035]

[0036] Wherein, ρ is the density of the fluid, U is the flow velocity of the fluid, L is the characteristic size, and μ is the dynamic viscosity of the fluid; for the cyclone separator of the present invention, when Re>4000, it is a turbulent flow; otherwise, it is a cyclone;

[0037] The flow calculation unit is responsible for predicting the motion characteristics of the swirl flow in the three-dimensional spray gun using the discretized Navier-Stokes equations, and assumes that the gas phase flow field is steady, compressible and turbulent. When there is no other external force acting on the flow field, the continuity and momentum conservation equations of the swirl flow are as follows:

[0038]

[0039]

[0040] Where u, ρ, P, and g represent the velocity vector, density, pressure, and gravitational acceleration of the gas phase in the computational domain, respectively. The symbol v represents the kinematic viscosity of the gas phase, v t Indicates the gas flow viscosity;

[0041] SSTk-ω is used as the unknown term in the Navier-Stokes equations for turbulence closure;

[0042] Calculate turbulence function (k g )The formula is:

[0043]

[0044] Calculate the eddy dissipation rate (ω g )The formula is:

[0045]

[0046] The separation efficiency unit is responsible for calculating the separation efficiency according to the above calculation results;

[0047] The separation efficiency calculation formula is:

[0048]

[0049] Among them, M in is the mass of particles flowing into the cyclone separator inlet, M out It is the mass of particles flowing out from the upper outlet of the cyclone separator.

[0050] As a further improvement of the present invention, the deep learning module includes:

[0051] The second preprocessing submodule is responsible for obtaining the simulation results and separation efficiency, preprocessing the simulation results and separation efficiency to form a data set; sorting the data set according to the timestamp to obtain a time series data set, and dividing the time series data set into a training set, a test set, and a validation set;

[0052] The flow prediction model submodule is responsible for inputting the flow prediction model according to the time series data, taking the environmental data information of the simulation results as input and the cyclone separator design parameter data as output to train the flow prediction model, extract the relationship between the component deformation and the surrounding environment change, and finally obtain the trained flow prediction model;

[0053] The first optimization submodule is responsible for comparing the simulation results with the prediction results, identifying the areas that need to be optimized in the flow prediction model, and continuously adjusting the optimization area parameters until the maximum number of iterations is reached or the qualified area is reached; the adjusted optimization area parameters are fed back to the user terminal.

[0054] As a further improvement of the present invention, the second preprocessing submodule includes:

[0055] The third preprocessing unit is responsible for extracting relevant features and reducing the dimension and complexity of the data structure through the dimension reduction device; the extracted data is normalized so that the data has similar scale and distribution;

[0056] The feature extraction unit is responsible for extracting the timestamps of each data, arranging the normalized data in sequence according to the timestamps to form time series data; and setting the time series data as a time series data set;

[0057] The data set division unit is responsible for dividing the time series data set into a training set, a test set and a validation set; and sending the training set, the test set and the validation set divided into the time series data set to the flow prediction model submodule 32.

[0058] As a further improvement of the present invention, the flow prediction model submodule includes:

[0059] The structural unit is responsible for inputting the flow prediction model into the flow prediction model according to the time series data, setting the flow prediction model architecture into the input layer, hidden layer and output layer; taking the pipe diameter data of the simulation results as input, the hidden layer performs feature extraction and processing, and the cyclone separator design parameter data as output to train the flow prediction model, and using the minimum batch data for iterative training;

[0060] The training unit is responsible for verifying the process of training the cyclone separator design parameter model using the verification set; stopping the training if it is not improved for multiple consecutive cycles; analyzing the relationship between the building deformation and the environmental parameters; extracting the influence of the key pipe diameter parameters on the building deformation, and generating a correlation data table;

[0061] The test model unit is responsible for adjusting the relevant hyperparameters of the flow prediction model according to the relevant relationship data table, and finally obtaining the final flow prediction model.

[0062] As a further improvement of the present invention, the first optimization submodule includes:

[0063] The identification and optimization unit is responsible for comparing the simulation results with the prediction results, identifying the area that needs to be optimized in the flow prediction model, and adjusting the area through continuous iterations. After each iteration, the simulation and prediction are repeated until the maximum number of iterations is reached or the qualified area is reached;

[0064] The second optimization unit is responsible for transmitting the adjusted optimization area parameters to the user terminal through the cloud platform. After receiving the optimization parameters, the user terminal can further adjust the parameters or confirm the final cyclone separator according to the actual situation. The adjustment parameters include: manual adjustment or re-request optimization;

[0065] The feedback unit is responsible for receiving the re-request for optimization operation initiated by the user terminal, retrieving the cyclone separator information from the database, and re-simulating and predicting the cyclone separator according to the target parameters re-given by the user until it is qualified.

[0066] To achieve the above object, the present invention also provides the following technical solutions:

[0067] A cyclone separator optimization method based on machine learning:

[0068] The first login requires registration of an account. After successful login, input the relevant parameters of the cyclone separator through the user terminal operation interface; retrieve the data stored in the database according to the relevant parameters of the cyclone separator input by the user terminal; determine whether the corresponding cyclone separator related parameter data is stored; if corresponding, retrieve the database data and output it to the user terminal; if not, import the new data into the simulation model;

[0069] The parameters of the cyclone separator include cyclone separator parameters, working environment, and operating parameters; the working environment includes working pressure, working temperature, humidity, and air density; the operating parameters include inlet velocity, particle size, particle density, particle viscosity, outlet pressure, and inlet pressure;

[0070] Receive the relevant parameters of the cyclone separator imported by the user terminal; use SolidWorks to draw the geometric model of the cyclone separator according to the design parameters of the cyclone separator; simulate the cyclone separator model and calculate the separation efficiency; and pre-process the separator simulation results and the separation efficiency;

[0071] Establish a data set based on the preprocessing results; sort the data set according to timestamps, and train the flow prediction model based on the data after timestamp sorting; combine the simulation results with the prediction results to identify the area that needs to be optimized in the flow prediction model; continuously iterate and adjust the optimized area to obtain the final cyclone separator parameters; and feed back the adjustments to the user terminal.

[0072] The present invention can predict the separation efficiency of cyclone separators of different designs according to the production working environment and different operating parameters, provide certain suggestions for actual production, and greatly save manpower, material resources, time, cost, etc. It is not necessary for operators to have rich professional knowledge in fluid mechanics, separator design, etc. Even if they are not familiar with cyclone separators, they can use the system to adjust the design input to control the relevant parameters of the design output, and can significantly reduce the design time and workload of cyclone separators. By adding online systems and deep learning, real-time data updates are achieved, ensuring the richness, pertinence, and real-time nature of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a functional module schematic diagram of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0074] Figure 2 A structural diagram of a cyclone separator of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0075] Figure 3 This is a functional module diagram of a user operation module of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0076] Figure 4 A schematic diagram of a user login process of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0077] Figure 5 A schematic diagram of the functional modules of a data transmission submodule of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0078] Figure 6 A schematic diagram of functional modules of a numerical simulation module of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0079] Figure 7An embodiment of the cyclone separator optimization system based on machine learning of the present invention Figure 3 and Figure 5 Schematic diagram;

[0080] Figure 8 This is a functional module schematic diagram of a first preprocessing submodule of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0081] Fig. 9 A block diagram of a simulation of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0082] Fig.10 A schematic diagram of the functional modules of a deep learning module of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0083] Fig.11 A schematic diagram of a deep learning module of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0084] Fig.12 This is a functional module schematic diagram of a second preprocessing submodule of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0085] Fig.13 This is a functional module schematic diagram of a flow prediction model submodule of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0086] Fig.14 This is a functional module schematic diagram of a first optimization submodule of an embodiment of a cyclone separator optimization system based on machine learning of the present invention;

[0087] Fig.15 A flowchart of the steps of an embodiment of a cyclone separator optimization method based on machine learning of the present invention;

[0088] Fig.16 It is a structural schematic diagram of an embodiment of an electronic device of the present invention;

[0089] Fig.17 It is a schematic structural diagram of an embodiment of the storage medium of the present invention. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] The terms "first", "second" and "third" in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0092] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0093] like Figure 1 As shown, this embodiment provides an embodiment of a cyclone separator optimization system based on machine learning, and the cyclone separator optimization system based on machine learning specifically includes:

[0094] User operation module 1 is responsible for the user's first login and registration of an account. After successful login, the user enters the relevant parameters of the cyclone separator through the user terminal operation interface; according to the user input of the relevant parameters of the cyclone separator, the data stored in the database is retrieved; it is determined whether the corresponding cyclone separator related parameter data is stored; if it is corresponding, the database data is retrieved and output to the user terminal; if not, the new data is imported into the numerical simulation module 2 for calculation;

[0095] The parameters of the cyclone separator include cyclone separator parameters, working environment, and operating parameters; the working environment includes working pressure, working temperature, humidity, and air density; the operating parameters include inlet velocity, particle size, particle density, particle viscosity, outlet pressure, and inlet pressure; the design parameters of the cyclone separator are as shown in the attached figure. Figure 2 ;

[0096] Numerical simulation module 2 is responsible for receiving the relevant parameters of the cyclone separator imported by the user terminal; using SolidWorks to draw the geometric model of the cyclone separator according to the design parameters of the cyclone separator; simulating the cyclone separator model to calculate the separation efficiency; and sending the separator simulation results and separation efficiency to deep learning module 3;

[0097] Deep learning module 3 is responsible for data preprocessing of simulation results and separation efficiency to establish a data set; sorting the data set according to timestamps, and training the flow prediction model based on the data after timestamp sorting; combining simulation results with prediction results to identify the areas that need to be optimized in the flow prediction model; iteratively adjusting the optimized areas to obtain the final cyclone separator parameters; and feeding back the adjustments to the user terminal.

[0098] Preferably, the user operation module 1 of this embodiment provides an intuitive operation interface to facilitate user input and modification of parameters, quick retrieval and display of historical data, and improved efficiency; data accuracy and consistency; direct display of historical data to reduce repeated calculations; processing and analysis of new data; numerical simulation module 2, accurate transmission of data; using the powerful modeling function of SolidWorks, the accuracy of the geometric model; through numerical simulation, predict the performance of the cyclone separator; provide basic data for subsequent data analysis and optimization. Deep learning module 3, data quality and consistency. Improve the accuracy and reliability of the model; accurately locate the optimization point and improve the design efficiency; achieve the best design parameters through iterative optimization; users can understand the optimization results in time to facilitate subsequent operations. The cyclone separator is mainly composed of a dusty gas inlet pipe, a purified gas outlet pipe (exhaust pipe), a simplified body, an ash discharge pipe and an ash hopper. Figure 2 In the figure, a-inlet height, b-inlet width, D0-cylinder diameter, D2-ash outlet diameter, α-semi-cone angle, h-cylinder height, Hh-cone height, de-exhaust pipe diameter. The structural dimensions of the cyclone separator have a great influence on its performance.

[0099] In summary, the automation and intelligence of cyclone separator design and optimization have been jointly promoted, and the design efficiency and performance have been improved. The present invention can predict the separation efficiency of cyclone separators of different designs according to the production working environment and different operating parameters, provide certain suggestions for actual production, and can greatly save manpower, material resources, time, cost, etc. There is no need for operators to have rich professional knowledge in fluid mechanics, separator design, etc. Even if they are not familiar with cyclone separators, they can use the system to adjust the design input to control the relevant parameters of the design output, and can significantly reduce the design time and workload of cyclone separators. Through the addition of online systems and deep learning, real-time updating of data is achieved, ensuring the richness, pertinence and real-time nature of the data.

[0100] Further, if Figure 3 As shown, the user operation module 1 specifically includes:

[0101] The registration submodule 11 is responsible for collecting the face image of the user to be registered, extracting the face features of the face image, generating a face feature vector, reducing the dimension of the face vector, generating a face feature vector; calibrating the feature vector; if it meets the set conditions, inputting the feature vector into the target database and completing the registration;

[0102] If not, on the contrary:

[0103] Among them, the associated feature vector of the user to be registered is compared with all the key features of the feature database where the registered users are located. If the similarity meets the set conditions, a temporary identity is created. If the similarity does not meet the set conditions, identity authentication is performed. If identity authentication fails, the user's face information is collected again;

[0104] F(x)=W·x+b

[0105] Where: F(x) represents the output face feature vector, which is a multidimensional vector containing multiple features extracted from the face image. W is a weight matrix that contains the conversion rules from the face image to the face feature vector. X is the input face image data, which can be a matrix of pixel values ​​or data reduced in dimension by some method (such as PCA). b is a bias vector used to adjust the overall position of the output feature vector.

[0106] Introduce some nonlinear transformations and activation functions as follows:

[0107] F(x)=σ(W·X+b)

[0108] σ represents the activation function, which performs a nonlinear transformation on the result of multiplying the weight matrix and the input image.

[0109] Furthermore, this process can be extended to a multi-layer network:

[0110] F(x)=σ(W2·σ(W1·X+b1)+b2)

[0111] In the equation, W1 and W2 are weight matrices of different layers, and b1 and b2 are bias vectors of different layers;

[0112] The operation submodule 12 is responsible for responding to the data parameters initiated by the user terminal to retrieve the cyclone separator target database, determining whether the data parameters initiated by the user terminal correspond to the data parameters of the cyclone separator target database, and judging whether there are data parameters corresponding to the cyclone separator target database; and outputting the judgment result to the data transmission submodule 13;

[0113] The data transmission submodule 13 is responsible for outputting the data parameters to the user terminal if there are data parameters corresponding to the cyclone separator target database, and transmitting the cyclone separator parameters transmitted by the user terminal to the numerical simulation module 2 if there are no data parameters corresponding to the cyclone separator target database.

[0114] Preferably, the registration submodule 11 of this embodiment is responsible for collecting the facial image of the user to be registered, extracting the face features of the facial image, generating a facial feature vector, and reducing the dimension of the facial vector to generate a facial feature vector. Then, the feature vector is calibrated. If it meets the set conditions, the feature vector is input into the feature library and the registration is completed. If it does not meet the conditions, a similarity comparison and identity authentication are performed. The uniqueness and security of the user identity improve the accuracy and reliability of the system through a multi-step verification process. The operation submodule 12 is responsible for responding to the data parameters initiated by the user terminal to retrieve the cyclone separator target database, determine whether the data parameters initiated by the user terminal correspond to the data parameters of the cyclone separator target database, and determine whether there are data parameters corresponding to the cyclone separator target database. The judgment result is output to the data transmission submodule 13. The accuracy and timeliness of the data improve the response speed and data processing capability of the system through an efficient database query and data transmission mechanism. The data transmission submodule 13 is responsible for outputting the data parameters to the user terminal if there are data parameters corresponding to the cyclone separator target database, and for transmitting the cyclone separator parameters transmitted by the user terminal to the numerical simulation module 2 if there are no data parameters corresponding to the cyclone separator target database. Through the flexible data transmission mechanism, the accurate transmission and processing of data improves the overall efficiency of the system and the user experience (for the specific principle, refer to the attached Figure 4 ).

[0115] In summary, the present embodiments jointly construct an efficient, accurate and secure face recognition and data processing system.

[0116] Further, if Figure 5 As shown, the data transmission submodule 13 specifically includes:

[0117] The generating unit 131 is responsible for generating a control instruction from the data parameter that does not correspond to the cyclone separator target database, processing the control instruction, and returning if the instruction is wrong, and prompting the user terminal that the control instruction input is wrong;

[0118] The command transmission unit 132 is responsible for obtaining the user terminal control instructions and storing the control instructions; processing the transmission data according to the control instructions; and running the compilation device to compile the transmission data;

[0119] The parsing unit 133 is responsible for transmitting the compilation result to the numerical simulation module 2; during the transmission process, the transmission data is sent to the numerical simulation module 2 through the pin unit; the numerical simulation module 2 performs control simulation according to the compilation result.

[0120] Preferably, the generation unit 131 of this embodiment ensures the accuracy and validity of the instruction by generating and processing the control instruction. If the instruction is wrong, the module will return and prompt the user terminal that the control instruction input is wrong, thereby improving the reliability of the system and the accuracy of user operations. The command transmission unit 132 obtains and stores the control instructions of the user terminal, and then processes the transmission data according to the instructions. The transmission data is compiled by running the compilation device to ensure the integrity and consistency of the data. Parsing unit 133: The compilation result is transmitted to the numerical simulation module 2, and the transmission data is sent to the numerical simulation module 2 through the pin unit; the numerical simulation module 2 performs control simulation according to the compilation result, thereby realizing precise control and simulation of the system.

[0121] In summary, the generation unit ensures the accuracy and reliability of instructions; the command transmission end element ensures the integrity and consistency of data; and the parsing unit realizes precise control and simulation of the system.

[0122] Further, if Figure 6 As shown, the numerical simulation module 2 specifically includes:

[0123] The first preprocessing submodule 21 is responsible for preprocessing the relevant parameters of the cyclone separator imported by the user terminal, and drawing the geometric model of the cyclone separator using SolidWorks according to the preprocessing results; preprocessing the relevant parameters of the cyclone separator and meshing the cyclone separator model;

[0124] The simulation submodule 22 is responsible for setting the relevant parameters of the simulation device through the relevant parameters of the cyclone separator of the preprocessing results; simulating the gas-solid two-phase flow process inside the cyclone separator to determine the flow mode of the cyclone separator; and calculating the separation efficiency according to the flow mode;

[0125] The simulation result transmission submodule 23 is responsible for comparing the separation efficiency with the historical experimental separation efficiency, evaluating the separation efficiency, and sending the simulation results, separation efficiency and evaluation results to the deep learning module 3.

[0126] Preferably, the first preprocessing submodule 21 of this embodiment preprocesses the relevant parameters of the cyclone separator imported by the user terminal to ensure the accuracy and consistency of the parameters. The preprocessed parameters are used for subsequent geometric model drawing and meshing, providing reliable basic data for subsequent simulation. In addition, by drawing the geometric model of the cyclone separator in SolidWorks, the structure of the separator can be intuitively displayed, which is convenient for subsequent analysis and optimization. The simulation submodule 22 sets the relevant parameters of the simulation device through the relevant parameters of the cyclone separator of the preprocessing results, and simulates the gas-solid two-phase flow inside the cyclone separator using computational fluid dynamics software. During the simulation process, the flow mode of the cyclone separator is determined, and the separation efficiency is calculated based on the flow mode. It helps to understand the working principle and performance of the cyclone separator and provide a basis for optimizing the design. The simulation result transmission submodule 23 sends the simulation results, separation efficiency and evaluation results to the deep learning module 3. By comparing with the separation efficiency of historical experiments, the separation efficiency is evaluated to optimize the design and performance of the cyclone separator. Artificial intelligence technology, especially deep learning, is used to predict and optimize the final quality of the product, improving design efficiency and accuracy. (For specific principles, please refer to the attached Figure 7 )

[0127] In summary, each module is mainly reflected in data preprocessing, simulation, result evaluation and optimization design. Through technical means, the design efficiency and performance of cyclone separators can be effectively improved.

[0128] Further, if Figure 8 As shown, the first preprocessing submodule 21 specifically includes

[0129] The drawing unit 211 is responsible for verifying the relevant parameters of the cyclone separator imported by the user terminal, comparing the relevant parameters with the standard data, verifying the accuracy and completeness of the relevant parameters, and drawing the geometric model of the cyclone separator according to the relevant data after the verification;

[0130] The mesh division unit 212 is responsible for dividing the geometric model of the cyclone separator into a plurality of non-overlapping sub-regions, the node position of each sub-region, and the control volume represented by the node; setting mesh parameters and generating meshes;

[0131] The mesh verification unit 213 is responsible for checking the mesh after it is generated; if the mesh is qualified, the mesh is output to the simulation submodule 22; if the mesh is unqualified, the unqualified area is checked and the mesh is re-divided in the area.

[0132] Preferably, the drawing unit 211 of this embodiment compares the relevant parameters of the cyclone separator imported by the user terminal with the standard data to ensure the accuracy and completeness of the parameters. Provide a reliable data basis for the subsequent geometric model drawing. Draw the geometric model of the cyclone separator according to the verified relevant data. The accuracy and consistency of the model provide an accurate geometric basis for the subsequent simulation analysis. The grid division unit 212 divides the geometric model of the cyclone separator into multiple non-overlapping sub-regions, and the node position and control volume of each sub-region are clear. Improve the accuracy and calculation efficiency of the grid; generate high-quality grids by setting appropriate grid parameters. The uniformity and consistency of the grid, thereby improving the accuracy of the simulation results. The role of grid division is to divide the calculation area into small discrete units, namely grid units, so as to facilitate the numerical calculation of fluid flow. After the calculation area is discretized, the physical quantities in each grid unit, such as velocity, pressure, temperature, etc., can be calculated by numerical methods. At the same time, gridding can also describe the complexity of the flow field. Complex flow field phenomena such as curvature, suspension, boundary layer, etc. can be described by reasonable grid division. After the mesh is generated, the mesh verification unit 213 checks the quality of the mesh and whether it is qualified. This includes checking the mesh's distortion, orthogonality and other parameters. If the mesh is unqualified, the unqualified area is checked and the mesh is re-divided in the area. The quality of the entire mesh is improved to avoid simulation result deviations caused by local mesh quality problems.

[0133] In summary, this embodiment improves the accuracy and reliability of cyclone separator simulation analysis through the steps of parameter verification, geometric model drawing, mesh division and mesh verification, which not only improves the accuracy of simulation results, but also optimizes the efficiency of computing resources.

[0134] Further, if Fig. 9 As shown, the simulation 22 specifically includes:

[0135] The flow determination unit 221 is responsible for determining the flow type according to the target parameters of the cyclone separator after grid division;

[0136] Formula for determining flow type:

[0137]

[0138] Wherein, ρ is the density of the fluid, U is the flow velocity of the fluid, L is the characteristic size, and μ is the dynamic viscosity of the fluid; for the cyclone separator of the present invention, when Re>4000, it is a turbulent flow; otherwise, it is a cyclone;

[0139] The flow calculation unit 222 is responsible for predicting the motion characteristics of the swirl flow in the three-dimensional spray gun using the discretized Navier-Stokes equations, and assumes that the gas phase flow field is steady, compressible and turbulent. When there is no other external force acting on the flow field, the continuity and momentum conservation equations of the swirl flow are as follows:

[0140]

[0141]

[0142] Where u, ρ, P and g represent the velocity vector, density, pressure and gravitational acceleration of the gas phase in the computational domain, respectively. The symbol v represents the kinematic viscosity of the gas phase, and vt represents the gas flow viscosity;

[0143] SSTk-ω is used as the unknown term in the Navier-Stokes equations for turbulence closure;

[0144] The formula for calculating turbulence function (kg) is:

[0145]

[0146] The formula for calculating the eddy dissipation rate (ωg) is:

[0147]

[0148] In the formula, ρ g Indicates the density of the fluid (in kg / m 3 ), k g represents the turbulent kinetic energy (unit: J / kg), u j represents the jth component of velocity (in m / s), u j represents the turbulence generation term (unit: W / kg), β represents the ratio of eddy dissipation rate to turbulent kinetic energy, ω g represents the eddy dissipation rate (unit: s -1 ), μ g represents turbulent viscosity (unit: Pa·s), μ t represents the turbulent stress (unit: Pa), σ k Prandtl number, which represents the turbulent kinetic energy, is used to control the eddy viscosity coefficient, α g represents the ratio of eddy dissipation rate to turbulent kinetic energy, P a represents the term of eddy dissipation (unit: W / kg), β b represents the eddy dissipation rate and the self-dissipation coefficient, σ a represents the Prandtl number of eddy dissipation rate, and F1 represents an empirical coefficient in the eddy dissipation rate model;

[0149] The separation efficiency unit 223 is responsible for calculating the separation efficiency according to the above calculation results;

[0150] The calculation formula of separation efficiency η is:

[0151]

[0152] Among them, M in is the mass of particles flowing into the cyclone separator inlet, and Mout is the mass of particles flowing out of the cyclone separator upper outlet.

[0153] Preferably, the turbulence of the present embodiment is a very random, instantaneous flow state, and the flow velocity and flow direction will change at every moment. The energy dissipation process of turbulence is to separate a large vortex into many small vortices. For small vortices with very small scales, energy dissipation will occur, converting kinetic energy into internal energy. If you want to accurately simulate turbulent flow, the amount of calculation required is very, very large, and the time-consuming will be very, very long. In engineering applications, we usually focus on the average flow state of the fluid over a period of time. In summary, in engineering calculations, time-averaged methods are used to obtain time-averaged solutions. In the numerical simulation process, the discretized Navier-Stokes equations are used to predict the motion characteristics of the vortex in the three-dimensional spray gun, and the gas phase flow field is assumed to be steady, compressible and turbulent. When there is no other external force acting on the flow field, the turbulence model plays a vital role in the prediction accuracy of the motion characteristics of the vortex in the spray gun. Large eddy simulation (LES) and direct numerical simulation (DNS) are two numerical models that can accurately simulate fluid flow. However, due to the high computational cost, the calculation of these two models usually requires a lot of computing resources, and it is difficult to set a sufficient grid resolution to fully evaluate the turbulent motion characteristics within industrial equipment. The Reynolds averaged Navier-Stokes (RANS) equations substitute the average and random components of velocity and pressure into the Navier-Stokes equations to achieve the prediction of gas phase turbulent flow. As two representative flow models in RANS, the Reynolds stress model (RSM) and the shear stress transport (SST) k-ω model are often used to predict the dynamic characteristics of swirl flow. The RSM model shows accurate prediction capabilities in solving axial and swirl motion, Reynolds stress anisotropy, streamline curvature and swirl flow effects. However, since RSM involves solving six stress transport equations and one turbulence dissipation equation, its computational cost is significantly greater than the SST k-ω model, which only needs to solve two equations. The SST model combines the advantages of the k-ε and k-ω methods in predicting free flow and viscous sublayers. In summary, this model has the ability to simulate fast strain flow and track the flow characteristics of complex swirl systems while ensuring low computational costs. This system uses SSTk-ω as the unknown term in the Navier-Stokes equation for turbulent closure. Turbulent kinetic energy (kg) and eddy dissipation rate (ωg) are two important variables in the SSTk-ω model. Simulation calculation program operation: Run the software with the above settings in the server.

[0154] In summary, this embodiment calculates the separation efficiency based on the operation results. This technology provides a large number of working conditions with different particle sizes, different design parameters, and different feed conditions, and imports each working condition into the numerical simulation module for calculation, and evaluates the effect of the calculation results of each working condition.

[0155] Furthermore, if Fig.10As shown, the deep learning module 3 specifically includes:

[0156] The second preprocessing submodule 31 is responsible for obtaining the simulation results and separation efficiency, preprocessing the simulation results and separation efficiency to form a data set; sorting the data set according to the timestamp to obtain a time series data set, and dividing the time series data set into a training set, a test set and a validation set;

[0157] The flow prediction model submodule 32 is responsible for inputting the flow prediction model according to the time series data, taking the environmental data information of the simulation results as input and the cyclone separator design parameter data as output to train the flow prediction model, extract the relationship between the component deformation and the surrounding environment change, and finally obtain the trained flow prediction model;

[0158] The first optimization submodule 33 is responsible for comparing the simulation results with the prediction results, identifying the area that needs to be optimized in the flow prediction model, and continuously adjusting the optimization area parameters until the maximum number of iterations is reached or the qualified area is reached; and feeding back the adjusted optimization area parameters to the user terminal.

[0159] Preferably, the second preprocessing submodule 31 of this embodiment obtains the simulation results and separation efficiency, and preprocesses the data to form a data set. Then the data set is sorted according to the timestamp to obtain a time series data set. It helps the time consistency of the data and provides a reliable data basis for subsequent model training. The time series data set is divided into a training set, a test set and a verification set. It helps to evaluate the performance of the model on different data sets and the generalization ability of the model. Through steps such as data cleaning, data integration, data reduction and data transformation, the quality and availability of the data are improved. For example, using a sliding window algorithm to intercept time series data and construct a data set required for model training can effectively extract key information and reduce redundant data. The flow prediction model submodule 32 uses the environmental data information of the simulation results as input and the cyclone separator design parameter data as output to train the flow prediction model. The combination method helps the model better understand the relationship between environmental changes and design parameters, thereby improving the accuracy of the prediction. The flow prediction model is trained by extracting the relationship between component deformation and changes in the surrounding environment. The training method of this embodiment helps the model learn complex nonlinear relationships and improve the accuracy of prediction. The first optimization submodule 33 compares the simulation results with the prediction results to identify the areas of the flow prediction model that need to be optimized. The comparison method of this embodiment helps to optimize the performance of the model in practical applications. The parameters of the optimization area are continuously adjusted until the maximum number of iterations or the qualified area is reached. The adjustment method of this embodiment helps to improve the adaptability of the model under different working conditions and improve the robustness of the model. The adjusted optimization area parameters are fed back to the user terminal so that the user can understand the optimization progress of the model in real time and make further adjustments and optimizations based on the feedback. (For specific principles, please refer to the attached Fig.11 )

[0160] In summary, this embodiment mainly reflects in aspects such as data preprocessing, model training and optimization, and parameter adjustment and feedback, which helps to improve the accuracy and robustness of the model, thereby better serving practical applications.

[0161] Further, if Fig.12 As shown, the second preprocessing submodule 31 specifically includes:

[0162] The third preprocessing unit 311 is responsible for extracting relevant features, reducing the dimension and complexity of the data structure through a dimension reduction device; normalizing the extracted data so that the data has a similar scale and distribution;

[0163] The feature extraction unit 312 is responsible for extracting the timestamps of each data, arranging the normalized data in sequence according to the timestamps to form time series data; and setting the time series data as a time series data set;

[0164] The data set division unit 313 is responsible for dividing the time series data set into a training set, a test set and a validation set; and sending the training set, the test set and the validation set divided into the time series data set to the flow prediction model submodule 32.

[0165] Preferably, the third preprocessing unit 311 of this embodiment can effectively reduce the complexity of data by reducing the dimensionality of the device, making data processing more efficient and convenient. Improve the speed and efficiency of subsequent model training. Normalized processing data has similar scales and distributions, eliminates the dimensional effects between different features, and improves the comparability of data and training efficiency. The feature extraction unit 312 extracts the timestamps of each data and arranges them sequentially according to the timestamps to form time series data, which helps to capture the time dependency of the data and provides a basis for subsequent time series analysis. The time series data is set as a time series data set to facilitate the training and prediction of subsequent time series models. The data set partitioning unit 313 divides the time series data set into a training set, a test set, and a validation set, and the time period of the test set is after the time period of the training set to simulate the prediction situation in the real scene. Data partitioning should ensure that the data distribution of each subset is consistent as much as possible to ensure that the model can perform well in various situations. By partitioning the data set, the generalization ability of the model can be better evaluated, and the model can also perform well on unseen data.

[0166] In summary, this embodiment improves the efficiency of data processing and model training by reducing data dimension and normalizing data. Feature extraction of timestamps and setting of time series data sets provide a basis for time series analysis. By reasonably dividing the data set, the generalization ability and prediction accuracy of the model are improved.

[0167] Further, if Fig.13 As shown, the flow prediction model submodule 32 specifically includes:

[0168] The structural unit 321 is responsible for inputting the flow prediction model into the flow prediction model according to the time series data, setting the flow prediction model architecture into an input layer, a hidden layer and an output layer; taking the pipe diameter data of the simulation result as input, the hidden layer performs feature extraction and processing, and the cyclone separator design parameter data as output to train the flow prediction model, and using the minimum batch data for iterative training;

[0169] The training unit 322 is responsible for verifying the process of training the cyclone separator design parameter model using the verification set; stopping the training if it is not improved in multiple consecutive cycles; analyzing the relationship between the building deformation and the environmental parameters; extracting the influence of the key pipe diameter parameters on the building deformation, and generating a correlation data table;

[0170] The test model unit 323 is responsible for adjusting the relevant hyper parameters of the flow prediction model according to the relevant relationship data table, and finally obtaining the final flow prediction model.

[0171] Preferably, the structural unit 321 of this embodiment can effectively extract and process features from the pipe diameter data of the simulation results by setting the flow prediction model architecture of the input layer, the hidden layer and the output layer. Improve the model's understanding and prediction ability of complex flow phenomena. Iterative training using minimum batch data can accelerate the convergence speed of the model and improve the generalization ability of the model. By using the cyclone separator design parameter data as output, the model can predict the parameters more accurately, thereby optimizing the design process. The training unit 322 verifies the model using a validation set, and can dynamically adjust the hyperparameters during the training process, so that the model performs consistently on different data sets. If the model performance is no longer improved for multiple consecutive cycles, the training is automatically stopped to avoid overfitting and improve the stability and reliability of the model. By analyzing the relationship between the deformation and the environmental parameters, the influence of environmental factors on the deformation can be better understood, providing a basis for subsequent design and optimization. The test model unit adjusts the relevant hyperparameters of the flow prediction model according to the relevant relationship data table, which can further optimize the performance of the model and improve the prediction accuracy. Through the above steps, the flow prediction model finally obtained has high accuracy and generalization ability, and can be effectively applied to actual engineering.

[0172] In summary, this embodiment mainly improves the feature extraction capability of the model, accelerates the training process, optimizes the design parameter prediction, dynamically adjusts the hyperparameters, analyzes the impact of environmental parameters, and ultimately improves the prediction accuracy and reliability of the model.

[0173] Further, if Fig.14 As shown, the first optimization submodule 33 specifically includes:

[0174] The identification and optimization unit 331 is responsible for comparing the simulation results with the prediction results, identifying the area that needs to be optimized in the flow prediction model, and adjusting the area through continuous iteration. After each iteration, the simulation and prediction are re-performed until the maximum number of iterations is reached or the qualified area is reached;

[0175] The second optimization unit 332 is responsible for transmitting the adjusted optimization area parameters to the user terminal through the cloud platform. After receiving the optimization parameters, the user terminal can further adjust the parameters or confirm the final cyclone separator according to the actual situation. The adjustment parameters include: manual adjustment or re-request optimization;

[0176] The feedback unit 333 is responsible for receiving the re-request for optimization operation initiated by the user terminal, retrieving the cyclone separator information from the database, and re-simulating and predicting the cyclone separator according to the target parameters re-given by the user until it is qualified.

[0177] Preferably, the optimization unit 331 of this embodiment is to adjust the structural parameters of the cyclone separator in a continuous iterative manner to improve its separation efficiency and reduce the operating pressure drop. Specifically, the unit compares the simulation results with the prediction results, identifies the area that needs to be optimized, and iterates the area for multiple times until the qualified area is reached or the maximum number of iterations is reached. The second optimization unit 332 is to efficiently transmit the optimized parameters to the user terminal through the cloud platform, so that the user can further adjust the parameters or confirm the final cyclone separator design according to the actual situation. The data transmission management of the cloud platform can quickly and reliably handle large-scale data transmission requirements, reduce data transmission delays and network bandwidth occupancy. The feedback unit 333 is that when the user terminal initiates a re-request for optimization operation, the system can quickly retrieve the cyclone separator information from the database, and re-simulate and predict according to the target parameters re-given by the user until the qualified standard is reached. The flexibility and adaptability of the cyclone separator design can be quickly adjusted and optimized according to user needs.

[0178] In summary, this embodiment is mainly reflected in significantly improving the performance and design efficiency of the cyclone separator through iterative optimization, efficient data transmission and fast feedback mechanism.

[0179] Further, if Fig.15 As shown, this embodiment further provides a cyclone separator optimization method based on machine learning. In this embodiment, a cyclone separator design and optimization method based on machine learning is applied to a cyclone separator design and optimization system based on machine learning in the above-mentioned embodiment. The cyclone separator optimization method based on machine learning specifically includes the following steps:

[0180] Step S1: The first login requires an account registration. After successful login, input the cyclone separator related parameters through the user terminal operation interface; retrieve the data stored in the database according to the cyclone separator related parameters input by the user terminal; determine whether the corresponding cyclone separator related parameter data is stored; if corresponding, retrieve the database data and output it to the user terminal; if not, import the new data into the simulation model;

[0181] The parameters of the cyclone separator include cyclone separator parameters, working environment, and operating parameters; the working environment includes working pressure, working temperature, humidity, and air density; the operating parameters include inlet velocity, particle size, particle density, particle viscosity, outlet pressure, and inlet pressure;

[0182] Step S2: receiving the relevant parameters of the cyclone separator imported by the user terminal; drawing the geometric model of the cyclone separator using SolidWorks according to the design parameters of the cyclone separator; simulating the cyclone separator model to calculate the separation efficiency; and preprocessing the separator simulation results and the separation efficiency;

[0183] Step S3: Establish a data set according to the preprocessing results; sort the data set according to the timestamp, and train the flow prediction model according to the data after the timestamp sorting; combine the simulation results with the prediction results to identify the area that needs to be optimized in the flow prediction model; continuously iterate and adjust the optimization area to obtain the final cyclone separator parameters; and feed back the adjustment to the user terminal.

[0184] Preferably, in step S1, an account needs to be registered for the first login. After successful login, the relevant parameters of the cyclone separator are input through the user terminal operation interface. The verification of user identity and the accurate input of parameters provide a basis for subsequent data processing and simulation. According to the relevant parameters of the cyclone separator input by the user terminal, the data stored in the database is retrieved to determine whether the corresponding cyclone separator related parameter data is stored. If it is corresponding, the database data is retrieved and output to the user terminal; if not, the new data is imported into the simulation model. Through the use of the database, rapid data retrieval and storage of new data are achieved, and the response speed and data management efficiency of the system are improved. Step S2 receives the relevant parameters of the cyclone separator imported by the user terminal, and uses SolidWorks to draw the geometric model of the cyclone separator according to the design parameters of the cyclone separator. Through the three-dimensional modeling technology, the precise design and visual display of the cyclone separator are realized. The cyclone separator model is simulated, the separation efficiency is calculated, and the separator simulation results and separation efficiency are preprocessed. Through numerical simulation and simulation technology, the performance of the cyclone separator is evaluated, and data support is provided for subsequent optimization. Step S3 establishes a data set according to the preprocessing results, sorts the data set according to the timestamp, trains the flow prediction model according to the data after the timestamp sorting, combines the simulation results with the prediction results, identifies the area that needs to be optimized in the flow prediction model, continuously iterates and adjusts the optimization area, obtains the final cyclone separator parameters, and feeds back the adjustment to the user terminal. Through the establishment of the data set and the training of the flow prediction model, the continuous optimization and feedback of the cyclone separator performance are achieved, and the adaptive ability and performance of the system are improved.

[0185] In summary, this embodiment achieves comprehensive improvement and optimization of the cyclone separator performance through the synergy of multiple technical links such as data storage and retrieval, geometric model drawing and simulation, data preprocessing and model training, structure optimization and iterative adjustment, and user feedback and terminal interaction.

[0186] like Fig.16As shown, this embodiment provides an embodiment of an electronic device. In this embodiment, the electronic device 8 includes a processor 81 and a memory 82 coupled to the processor 81.

[0187] The memory 82 stores program instructions for implementing the layout method of the chemical pump body processing equipment of any of the above-mentioned embodiments.

[0188] The processor 81 is used to execute the program instructions stored in the memory 82 to perform the layout of the chemical pump body processing equipment.

[0189] The processor 81 may also be referred to as a CPU (Central Processing Unit). The processor 81 may be an integrated circuit chip having signal processing capabilities. The processor 81 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0190] Further, Fig.17 This is a schematic diagram of the structure of a storage medium of an embodiment of the present application. The storage medium 9 of the embodiment of the present application stores program instructions 91 that can implement all the above methods, wherein the program instructions 91 can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0191] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0192] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units. The above is only an implementation mode of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present invention.

[0193] The specific implementation methods of the invention are described in detail above, but they are only examples, and the invention is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the invention. In summary, the equalization, modification, improvement, etc. made without departing from the spirit and principle of the invention should be included in the scope of the invention.

Claims

1. A cyclone separator optimization system based on machine learning, characterized in that: The cyclone separator optimization system based on machine learning includes: The user operation module is responsible for registering an account when the user logs in for the first time. After successful login, the user inputs the relevant parameters of the cyclone separator through the user terminal operation interface; retrieves the data stored in the database according to the relevant parameters of the cyclone separator input by the user; determines whether the corresponding cyclone separator related parameter data is stored; if it is corresponding, retrieves the database data and outputs it to the user terminal; if not, imports the new data into the numerical simulation module for calculation; Among them, the cyclone separator related parameters include: cyclone separator design parameters, working environment, and operating parameters; the cyclone separator parameters include cyclone separator parameters, working environment, and operating parameters; the working environment includes working pressure, working temperature, humidity, and air density; the operating parameters include inlet velocity, particle size, particle density, particle viscosity, outlet pressure, and inlet pressure; The numerical simulation module is responsible for receiving the cyclone separator related parameters imported by the user terminal; using SolidWorks to draw the cyclone separator geometric model according to the design parameters of the cyclone separator; simulating the cyclone separator model to calculate the separation efficiency; and sending the separator simulation results and separation efficiency to the deep learning module; The deep learning module is responsible for data preprocessing of the simulation results and separation efficiency to establish a data set; sorting the data set according to timestamps, and training the flow prediction model based on the data after timestamp sorting; combining the simulation results with the prediction results to identify the areas that need to be optimized in the flow prediction model; iteratively adjusting the optimized areas to obtain the final cyclone separator parameters; and feeding back the adjustments to the user terminal.

2. The cyclone separator optimization system based on machine learning according to claim 1, characterized in that: User operation module, including: The registration submodule is responsible for collecting the face image of the user to be registered, extracting the face features of the face image, generating a face feature vector, reducing the dimension of the face vector, generating a face feature vector; calibrating the feature vector; if it meets the set conditions, inputting the feature vector into the feature library and completing the registration; If not, vice versa; Among them, the associated feature vector of the user to be registered is compared with all the key features of the feature database where the registered users are located. If the similarity meets the set conditions, a temporary identity is created. If the similarity does not meet the set conditions, identity authentication is performed. If identity authentication fails, the user's face information is collected again; The operation submodule is responsible for responding to the data parameters initiated by the user terminal to retrieve the cyclone separator target database, determining whether the data parameters initiated by the user terminal correspond to the data parameters of the cyclone separator target database, and judging whether there are data parameters corresponding to the cyclone separator target database; and outputting the judgment result to the data transmission submodule; The data transmission submodule is responsible for outputting the data parameters to the user terminal if there are data parameters corresponding to the cyclone separator target database, and transmitting the cyclone separator parameters transmitted by the user terminal to the numerical simulation module if there are no data parameters corresponding to the cyclone separator target database; The generation unit is responsible for generating control instructions from data parameters that do not correspond to the cyclone separator target database, processing the control instructions, and returning if the instructions are wrong, and prompting the user terminal that the input of the control instruction is wrong; The command transmission unit is responsible for obtaining the user terminal control instructions and storing the control instructions; processing the transmission data according to the control instructions; and running the compilation device to compile the transmission data; The parsing unit is responsible for transmitting the compilation results to the numerical simulation module; during the transmission process, the transmission data is sent to the numerical simulation module through the pin unit; the numerical simulation module performs control simulation according to the compilation results.

3. The cyclone separator optimization system based on machine learning according to claim 1, characterized in that: Numerical simulation module, including: The first preprocessing submodule is responsible for preprocessing the relevant parameters of the cyclone separator imported by the user terminal, and drawing the geometric model of the cyclone separator using SolidWorks according to the preprocessing results; preprocessing the relevant parameters of the cyclone separator and meshing the cyclone separator model; The simulation submodule is responsible for setting the relevant parameters of the simulation equipment through the relevant parameters of the cyclone separator of the preprocessing results; simulating the gas-solid two-phase flow process inside the cyclone separator to determine the flow mode of the cyclone separator; and calculating the separation efficiency according to the flow mode; The simulation result transmission submodule is responsible for comparing the separation efficiency with the historical experimental separation efficiency, evaluating the separation efficiency, and sending the simulation results, separation efficiency and evaluation results to the deep learning module.

4. The cyclone separator optimization system based on machine learning according to claim 3, characterized in that: The first preprocessing submodule includes: The drawing unit is responsible for verifying the relevant parameters of the cyclone separator imported by the user terminal, comparing the relevant parameters with the standard data, verifying the accuracy and completeness of the relevant parameters, and drawing the geometric model of the cyclone separator according to the relevant data after the verification; The meshing unit is responsible for dividing the geometric model of the cyclone separator into multiple non-overlapping sub-regions, the node position of each sub-region, and the control volume represented by the node; setting mesh parameters and generating meshes; The mesh verification unit is responsible for checking the mesh after it is generated; if the mesh is qualified, the mesh is output to the simulation submodule; if the mesh is unqualified, the unqualified area is checked and the mesh is re-divided in the area.

5. The cyclone separator optimization system based on machine learning according to claim 3, characterized in that: Simulation submodule, including: The flow judging unit is responsible for judging the flow type according to the target parameters of the cyclone separator after grid division; Formula for determining flow type: Among them, ρ is the density of the fluid, U is the flow velocity of the fluid, L is the characteristic size, and v is the dynamic viscosity of the fluid. For the cyclone separator, when Re>4000, it is a turbulent flow; otherwise, it is a steady flow. The flow calculation unit is responsible for predicting the motion characteristics of the swirl flow in the three-dimensional spray gun using the discretized Navier-Stokes equations, and assumes that the gas phase flow field is steady, compressible and turbulent; when there is no other external force acting on the flow field, the continuity and momentum conservation equations of the swirl flow are as follows: Where u, ρ, P, and g represent the velocity vector, density, pressure, and gravitational acceleration of the gas phase in the computational domain, respectively; the symbol v represents the kinematic viscosity of the gas phase, v t Indicates the gas flow viscosity; SST k-ω is used as the unknown term in the Navier-Stokes equations for turbulence closure; Calculate the turbulence function k g The formula is: Calculate the eddy dissipation rate (ω g )The formula is: In the formula, ρ g represents the density of the fluid, k g represents the turbulent kinetic energy, u j represents the jth component of velocity, u j represents the turbulence generation term, β represents the eddy dissipation rate and the turbulent kinetic energy ratio, ω g represents the eddy dissipation rate, μ g represents the turbulent viscosity, μ t represents the turbulent stress, σ k Prandtl number, which represents the turbulent kinetic energy, is used to control the eddy viscosity coefficient, α g represents the ratio of eddy dissipation rate to turbulent kinetic energy, P a represents the term of eddy dissipation, β b represents the eddy dissipation rate and the self-dissipation coefficient, σ a represents the Prandtl number of eddy dissipation rate, and F1 represents an empirical coefficient in the eddy dissipation rate model; The separation efficiency unit is responsible for calculating the separation efficiency based on the calculation results; The separation efficiency calculation formula is: Among them, M in is the mass of particles flowing into the cyclone separator inlet, M out It is the mass of particles flowing out from the upper outlet of the cyclone separator.

6. The cyclone separator optimization system based on machine learning according to claim 1, characterized in that: Deep learning modules, including: The second preprocessing submodule is responsible for obtaining the simulation results and separation efficiency, preprocessing the simulation results and separation efficiency to form a data set; sorting the data set according to the timestamp to obtain a time series data set, and dividing the time series data set into a training set, a test set, and a validation set; The flow prediction model submodule is responsible for inputting the flow prediction model according to the time series data, taking the environmental data information of the simulation results as input and the cyclone separator design parameter data as output to train the flow prediction model, extract the relationship between the component deformation and the surrounding environment change, and finally obtain the trained flow prediction model; The first optimization submodule is responsible for comparing the simulation results with the prediction results, identifying the areas that need to be optimized in the flow prediction model, and continuously adjusting the optimization area parameters until the maximum number of iterations is reached or the qualified area is reached; the adjusted optimization area parameters are fed back to the user terminal.

7. The cyclone separator optimization system based on machine learning according to claim 6, characterized in that: The second preprocessing submodule includes: The third preprocessing unit is responsible for extracting relevant features and reducing the dimension and complexity of the data structure through the dimension reduction device; the extracted data is normalized so that the data has similar scale and distribution; The feature extraction unit is responsible for extracting the timestamps of each data, arranging the normalized data in sequence according to the timestamps to form time series data; and setting the time series data as a time series data set; The data set division unit is responsible for dividing the time series data set into a training set, a test set and a validation set; and sending the training set, the test set and the validation set divided into the time series data set to the flow prediction model submodule 32.

8. The cyclone separator optimization system based on machine learning according to claim 6, characterized in that: Flow prediction model submodules include: The structural unit is responsible for inputting the flow prediction model into the flow prediction model according to the time series data, setting the flow prediction model architecture into the input layer, hidden layer and output layer; taking the pipe diameter data of the simulation results as input, the hidden layer performs feature extraction and processing, and the cyclone separator design parameter data as output to train the flow prediction model, and using the minimum batch data for iterative training; The training unit is responsible for verifying the process of training the cyclone separator design parameter model using the verification set; stopping the training if it is not improved for multiple consecutive cycles; analyzing the relationship between the building deformation and the environmental parameters; extracting the influence of the key pipe diameter parameters on the building deformation, and generating a correlation data table; The test model unit is responsible for adjusting the relevant hyperparameters of the flow prediction model according to the relevant relationship data table, and finally obtaining the final flow prediction model.

9. The cyclone separator optimization system based on machine learning according to claim 6, characterized in that: The first optimization submodule includes: The identification and optimization unit is responsible for comparing the simulation results with the prediction results, identifying the area that needs to be optimized in the flow prediction model, and adjusting the area through continuous iterations. After each iteration, the simulation and prediction are repeated until the maximum number of iterations is reached or the qualified area is reached; The second optimization unit is responsible for transmitting the adjusted optimization area parameters to the user terminal through the cloud platform. After receiving the optimization parameters, the user terminal can further adjust the parameters or confirm the final cyclone separator according to the actual situation. The adjustment parameters include: manual adjustment or re-request optimization; The feedback unit is responsible for receiving the re-request for optimization operation initiated by the user terminal, retrieving the cyclone separator information from the database, and re-simulating and predicting the cyclone separator according to the target parameters re-given by the user until it is qualified.

10. A cyclone separator optimization method based on machine learning, applied to a cyclone separator optimization system based on machine learning as claimed in any one of claims 1 to 9, characterized in that: The cyclone separator optimization method based on machine learning comprises: The first login requires registration of an account. After successful login, input the relevant parameters of the cyclone separator through the user terminal operation interface; retrieve the data stored in the database according to the relevant parameters of the cyclone separator input by the user terminal; determine whether the corresponding cyclone separator related parameter data is stored; if corresponding, retrieve the database data and output it to the user terminal; if not, import the new data into the simulation model; The parameters of the cyclone separator include cyclone separator parameters, working environment, and operating parameters; the working environment includes working pressure, working temperature, humidity, and air density; the operating parameters include inlet velocity, particle size, particle density, particle viscosity, outlet pressure, and inlet pressure; Receive the relevant parameters of the cyclone separator imported by the user terminal; use SolidWorks to draw the geometric model of the cyclone separator according to the design parameters of the cyclone separator; simulate the cyclone separator model and calculate the separation efficiency; and pre-process the separator simulation results and the separation efficiency; Establish a data set based on the preprocessing results; sort the data set according to timestamps, and train the flow prediction model based on the data after timestamp sorting; combine the simulation results with the prediction results to identify the area that needs to be optimized in the flow prediction model; continuously iterate and adjust the optimized area to obtain the final cyclone separator parameters; and feed back the adjustments to the user terminal.

Images