Boiler Operation Simulation Method and System Based on Multi-Model Coupling

Abstract

This invention relates to a boiler operation simulation method and system based on multi-model coupling. The method includes: establishing a high-precision mechanistic model of the boiler based on its specific type and operating conditions; establishing a CFD-AI model of the combustion process using CFD technology combined with AI algorithms; constructing a data-driven model based on operational data and jointly building a dynamic model of the boiler; and establishing a controller based on the dynamic model of the boiler for multi-variable coupled control. The system includes a model building module, a coupled control module, a key influencing factor identification module, and a fault diagnosis module. This invention can accurately describe the flow, combustion, and heat exchange processes within the boiler, while considering multiple influencing factors, providing realistic dynamic characteristics and a comprehensive analysis of the boiler system. It enables early warning and fault diagnosis, improves boiler operating efficiency and safety, reduces downtime, increases boiler thermal efficiency, reduces energy consumption and emissions, supports sustainable development, and reduces costs and time during testing and commissioning.

Description

Technical Field

[0001] This invention belongs to the field of simulation management technology, and in particular relates to a boiler operation simulation method and system based on multi-model coupling. Background Technology

[0002] Boilers, as crucial heat energy production equipment, are widely used in industry and the energy sector. Their operating efficiency and environmental impact directly affect the effective utilization of energy and sustainable development. In recent years, with the continuous increase in global energy demand and the gradual enhancement of environmental protection awareness, the efficient and stable operation of boiler systems has become particularly important. Boilers, as important heat energy production equipment, are widely used in various fields such as industry, power generation, and heating. Their performance not only affects production efficiency but also has a profound impact on energy consumption and environmental emissions. However, traditional boiler control methods often rely on experience and simple control algorithms, making it difficult to adapt to complex operating conditions and various disturbances, leading to low energy efficiency and excessive emissions. Therefore, inventing a boiler operation simulation method based on multi-model coupling has become particularly important.

[0003] A search revealed that Chinese patent number CN118313282A discloses a boiler operation simulation method based on complex calculations. Although this invention improves the accuracy of genetic algorithms and thus accurately obtains the best individuals, enabling accurate boiler operation simulation, it cannot accurately describe the flow, combustion, and heat exchange processes within the boiler. Furthermore, it does not comprehensively consider influencing factors, cannot provide realistic dynamic characteristics, and cannot provide a comprehensive analysis of the boiler system, thereby reducing the boiler's operating efficiency and safety. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a boiler operation simulation method and system based on multi-model coupling.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] Boiler operation simulation methods based on multi-model coupling include:

[0007] S1. Establish a high-precision mechanism model of the boiler based on the specific type and operating conditions of the boiler;

[0008] S2. Based on the established high-precision boiler mechanism model, a CFD-AI model of the combustion process is established using CFD technology combined with AI algorithms.

[0009] S3. Based on the high-precision mechanism model and CFD-AI model of the boiler, and based on the operation data, a data-driven model is constructed and a dynamic model of the boiler is jointly constructed.

[0010] S4. Establish a controller based on the boiler dynamic model to perform multivariable coupled control;

[0011] S5. Simulate system response under different operating conditions and identify key influencing factors through multivariable coupling control;

[0012] S6. Based on key influencing factors, identify potential faults to provide corresponding early warnings and optimize boiler operating parameters.

[0013] A further improvement of this invention lies in that, in S2, based on the established high-precision boiler mechanism model, a CFD-AI model of the combustion process is established using CFD technology combined with AI algorithms, including:

[0014] Q1.1: Based on the established high-precision mechanism model of the boiler, obtain the corresponding geometric model inside the boiler, use a mesh generation tool to divide the geometric model into multiple discrete mesh units, and refine the mesh in the combustion zone and the airflow turbulence zone;

[0015] Q1.2: Obtain the properties of the flow inside the boiler in each grid cell by using data from each set of sensors, including steady-state or unsteady-state, compressible or incompressible, and laminar or turbulent flow. Then, establish a continuity equation based on the mass conservation of the fluid, and establish a momentum conservation equation by combining the conservation of mass and momentum. Establish the combustion reaction kinetic equation based on the finite rate reaction model.

[0016] Q1.3: Based on the combustion reaction kinetic equation, the turbulent region generated during the combustion process is collected in real time, and the corresponding k-ε turbulence model is constructed. Then, the k-ε turbulence model is combined with each set of equations to construct the corresponding CFD model. Boundary conditions and initial conditions are set for the CFD model, and CFD simulation is performed. The governing equations of the CFD model are solved iteratively using ANSYS Fluent software.

[0017] Q1.4: Collect the data generated by CFD simulation, remove noise and outliers from each set of data, scale the processed data to the same range, use correlation analysis to evaluate feature importance, determine the key features in each set of data, and then define the number of neurons in the input layer, hidden layer and output layer of the AI ​​model.

[0018] Q1.5: Divide the extracted key features into training set, validation set and test set, use training set data to train the model, adjust the model parameters through backpropagation algorithm, evaluate the model performance through k-fold cross-validation, adjust the original hyperparameters of the AI ​​model, use validation set to evaluate the generalization ability of the model, and calculate accuracy and F1 score.

[0019] Q1.6: Repeatedly train and validate the AI ​​model until the model's loss function value converges to a preset range or the accuracy and F1 score reach preset standards. Combine the prediction results of multiple models using a voting method, and then validate the performance of the final model using a test set. If the model performance meets the preset indicators, deploy the model in a practical application and couple the AI ​​model and the CFD model to form a CFD-AI combustion model.

[0020] A further improvement of this invention lies in that, in S3, a data-driven model is constructed based on a high-precision boiler mechanism model and a CFD-AI model, and a dynamic boiler model is jointly constructed based on operational data, including:

[0021] Q2.1: Collect the parameter information of the constructed high-precision boiler mechanism model, CFD-AI model and data-driven model, construct the corresponding parameter space based on the key parameters of each model and their value range, and then generate multiple parameter combinations through a random number generator according to the predefined parameter range.

[0022] Q2.2: Take the current state and parameter combination of the current boiler as the root node, calculate the boiler state corresponding to each randomly generated parameter combination, and take each parameter combination and its corresponding state as a new child node. Construct a search tree based on the root node and child nodes.

[0023] Q2.3: Starting from the root node, select child nodes along the path of the current search tree using a greedy strategy until a node that is not fully expanded is reached. When a node that is not fully expanded is reached, the parameters of the node are perturbed by a random generator, and one or more child nodes are generated from the node to represent new combinations of states and parameters.

[0024] Q2.4: Perform the simulation by randomly selecting actions or parameter combinations until the termination state is reached, and record the reward value and performance indicators obtained from the simulation. Feed back the simulation results to all nodes on the same path and update the reward value of each node.

[0025] Q2.5: Repeat the selection, expansion, simulation, and backtracking steps until the set computational resource limit is reached. Starting from the root node, select the child node with the highest reward value as the optimal parameter combination, and integrate the output results of different models under this parameter combination to form a boiler dynamic model.

[0026] A further improvement of this invention is that, in S4, a controller is established based on the boiler dynamic model to perform multivariable coupled control, including:

[0027] Q3.1: Based on the boiler dynamic model, obtain the set of variables of boiler temperature, pressure and flow rate that need to be controlled, as well as the relevant system state, and set the target value based on the needs of the staff. Then, establish a state transition model according to the dynamic characteristics of the boiler, and associate the system state with the observation data to construct the corresponding observation model.

[0028] Q3.2: Select a PID closed-loop control strategy, monitor the error between the system output and the set value in real time, then calculate the prior probability distribution of multiple variables and states based on the prior knowledge of the boiler system, collect the observation data generated by the observation model, and update the posterior probability through Bayes' theorem.

[0029] Q3.3: Based on posterior probability, calculate the cost of boiler state and control input, and adjust each group of control variables according to the calculated cost. Based on the optimized control variables, design a multivariate coupled control strategy, implement the control strategy in the boiler system, monitor the system response in real time, and continuously adjust the control input and state estimation according to the feedback data.

[0030] A further improvement of this invention is that, in S5, multivariable coupling control is used to simulate the system response under different operating conditions and identify key influencing factors, including:

[0031] Q4.1: Collect the range of simulable operating conditions, and set different combinations of operating conditions according to the boiler's operating conditions. Simulate the boiler dynamic model under different operating conditions and output the dynamic characteristics and response of the boiler under different combinations of operating conditions in real time.

[0032] Q4.2: Record the changes of each input parameter and its corresponding output results. Preprocess and organize the input and output data in the simulation experiment to form a dataset. Perform sensitivity analysis on the collected data through variance analysis to evaluate the degree of influence of different input parameters on the output response.

[0033] Q4.3: Based on the simulation data, establish a linear regression model for the relationship between each group of inputs and outputs, and collect the regression coefficients of the regression model. Arrange the collected coefficients from largest to smallest. The larger the absolute value of the coefficient, the more significant the influence of the factor on the output. Based on the arrangement results, identify the key factors that have the greatest impact on the system response, and then analyze the identified key influencing factors to verify their actual impact on the boiler system.

[0034] A further improvement of this invention lies in that, based on key influencing factors in S6, potential faults are identified to provide corresponding early warnings and optimize boiler operating parameters, including:

[0035] Q5.1: Based on the optimization objectives of improving boiler thermal efficiency, reducing emissions, and reducing fuel consumption, design an optimization objective function, randomly generate multiple sets of particles, each set of particles represents a parameter combination, and randomly set the initial position and velocity of each particle.

[0036] Q5.2: Calculate the fitness value of each group of particles using the set optimization objective function, initialize the global best position as the initial position of the first particle, update the particle position according to the current particle velocity, the current velocity, the historical best position and the global best position, record the fitness value of each group of particles in each iteration, and compare the current fitness with the historical best fitness. If the current fitness is better than its historical best fitness, then update the historical best position.

[0037] Q5.3: Compare the fitness values ​​of other particles with the current global best fitness value. If the fitness value of the current particle is better than the global best fitness value, then update the global best position and the global best fitness value. Then, based on the current velocity, the historical best position and the global best position, update the velocity of each group of particles, and update the position of the particles based on the updated velocity.

[0038] Q5.4: Check whether the updated particle position is within the preset boundary range. If it exceeds the range, adjust its position to the boundary value. Repeatedly update the particle velocity, historical best position, and global best position until the fitness change value reaches the preset range, and then stop updating.

[0039] Q5.5: Extract the final global optimal position and the corresponding fitness value, apply the optimized control parameters to the boiler dynamic model, conduct simulation tests, calculate and evaluate the boiler's thermal efficiency, emission level and fuel consumption based on the simulation results, compare the optimized operating parameters with the unoptimized operating parameters, analyze the differences in boiler performance before and after optimization, and evaluate the degree of influence of each optimized parameter on boiler performance.

[0040] A boiler operation simulation system based on multi-model coupling includes:

[0041] The first model building module establishes a high-precision mechanism model of the boiler based on the specific type and operating conditions of the boiler.

[0042] The second model building module, based on the established high-precision boiler mechanism model, uses CFD technology combined with AI algorithms to build a CFD-AI model of the combustion process;

[0043] The third model building module is based on the high-precision mechanism model of the boiler and the CFD-AI model, and constructs a data-driven model based on the operating data and jointly constructs the dynamic model of the boiler.

[0044] The coupling control module establishes a controller based on the boiler dynamic model to perform multivariable coupled control.

[0045] The module for identifying key influencing factors uses multivariable coupling control to simulate system response under different operating conditions and identify key influencing factors.

[0046] The fault diagnosis module, based on key influencing factors, identifies potential faults to provide corresponding early warnings and optimize boiler operating parameters.

[0047] A further improvement of this invention lies in that, based on the established high-precision boiler mechanism model, the second model building module utilizes CFD technology combined with AI algorithms to establish a CFD-AI model of the combustion process, including:

[0048] Q1.1: Based on the established high-precision mechanism model of the boiler, obtain the corresponding geometric model inside the boiler, use a mesh generation tool to divide the geometric model into multiple discrete mesh units, and refine the mesh in the combustion zone and the airflow turbulence zone;

[0049] Q1.2: Obtain the properties of the flow inside the boiler in each grid cell by using data from each set of sensors, including steady-state or unsteady-state, compressible or incompressible, and laminar or turbulent flow. Then, establish a continuity equation based on the mass conservation of the fluid, and establish a momentum conservation equation by combining the conservation of mass and momentum. Establish the combustion reaction kinetic equation based on the finite rate reaction model.

[0050] Q1.3: Based on the combustion reaction kinetic equation, the turbulent region generated during the combustion process is collected in real time, and the corresponding k-ε turbulence model is constructed. Then, the k-ε turbulence model is combined with each set of equations to construct the corresponding CFD model. Boundary conditions and initial conditions are set for the CFD model, and CFD simulation is performed. The governing equations of the CFD model are solved iteratively using ANSYS Fluent software.

[0051] Q1.4: Collect the data generated by CFD simulation, remove noise and outliers from each set of data, scale the processed data to the same range, use correlation analysis to evaluate feature importance, determine the key features in each set of data, and then define the number of neurons in the input layer, hidden layer and output layer of the AI ​​model.

[0052] Q1.5: Divide the extracted key features into training set, validation set and test set, use training set data to train the model, adjust the model parameters through backpropagation algorithm, evaluate the model performance through k-fold cross-validation, adjust the original hyperparameters of the AI ​​model, use validation set to evaluate the generalization ability of the model, and calculate accuracy and F1 score.

[0053] Q1.6: Repeatedly train and validate the AI ​​model until the model's loss function value converges to a preset range or the accuracy and F1 score reach preset standards. Combine the prediction results of multiple models using a voting method, and then validate the performance of the final model using a test set. If the model performance meets the preset indicators, deploy the model in a practical application and couple the AI ​​model and the CFD model to form a CFD-AI combustion model.

[0054] A further improvement of this invention lies in that the third model building module is based on a high-precision boiler mechanism model and a CFD-AI model, and constructs a data-driven model based on operational data, and jointly constructs a boiler dynamic model, including:

[0055] Q2.1: Collect the parameter information of the constructed high-precision boiler mechanism model, CFD-AI model and data-driven model, construct the corresponding parameter space based on the key parameters of each model and their value range, and then generate multiple parameter combinations through a random number generator according to the predefined parameter range.

[0056] Q2.2: Take the current state and parameter combination of the current boiler as the root node, calculate the boiler state corresponding to each randomly generated parameter combination, and take each parameter combination and its corresponding state as a new child node. Construct a search tree based on the root node and child nodes.

[0057] Q2.3: Starting from the root node, select child nodes along the path of the current search tree using a greedy strategy until a node that is not fully expanded is reached. When a node that is not fully expanded is reached, the parameters of the node are perturbed by a random generator, and one or more child nodes are generated from the node to represent new combinations of states and parameters.

[0058] Q2.4: Perform the simulation by randomly selecting actions or parameter combinations until the termination state is reached, and record the reward value and performance indicators obtained from the simulation. Feed back the simulation results to all nodes on the same path and update the reward value of each node.

[0059] Q2.5: Repeat the selection, expansion, simulation, and backtracking steps until the set computational resource limit is reached. Starting from the root node, select the child node with the highest reward value as the optimal parameter combination, and integrate the output results of different models under this parameter combination to form a boiler dynamic model.

[0060] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the boiler operation simulation method based on multi-model coupling.

[0061] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0062] This invention provides a boiler operation simulation method and system based on multi-model coupling. It constructs a high-precision mechanistic model based on physical laws, combines CFD and AI algorithms to build a CFD-AI combustion model to simulate the combustion process, and learns the boiler's dynamic characteristics from historical data using machine learning methods to build a data-driven model. Then, based on the key parameters and their value ranges of each model, a corresponding parameter space is constructed. Next, according to the predefined parameter range, multiple parameter combinations are generated using a random number generator, and a search tree is constructed to select the optimal parameter combination. The output results of different models under this parameter combination are integrated to form a comprehensive boiler dynamic model. Based on the boiler dynamic model, the required controllable boiler temperature, pressure, and flow rate variables, as well as related system states, are obtained. Target values ​​are set based on operator needs. Then, a state transition model is established based on the boiler's dynamic characteristics, and the system state is correlated with observation data to construct a corresponding observation model. The PID closed-loop control strategy monitors the error between the system output and the setpoint in real time. It randomly generates multiple groups of particles, each representing a parameter combination, and randomly sets the initial position and velocity of each particle. The fitness value of each group of particles is calculated using a defined optimization objective function, and the position, velocity, historical best position, and global best position of each group of particles are continuously updated and iterated. The optimized operating parameters are compared with those before optimization to analyze the differences in boiler performance and assess the impact of each optimized parameter on boiler performance. This accurately describes the flow, combustion, and heat exchange processes within the boiler, considering multiple influencing factors, providing realistic dynamic characteristics and a comprehensive analysis of the boiler system. It enables early warning and fault diagnosis, improves boiler operating efficiency and safety, reduces downtime, increases boiler thermal efficiency, reduces energy consumption and emissions, supports sustainable development, and reduces costs and time during testing and commissioning. Attached Figure Description

[0063] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0064] Figure 1 This is a flowchart of the boiler operation simulation method based on multi-model coupling according to the present invention.

[0065] Figure 2 This is a structural block diagram of the boiler operation simulation system based on multi-model coupling of the present invention. Detailed Implementation

[0066] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0067] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0068] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0069] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0070] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0071] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0072] Example 1

[0073] The boiler operation simulation method based on multi-model coupling provided by this invention includes:

[0074] S1. Establish a high-precision mechanism model of the boiler based on the specific type and operating conditions of the boiler;

[0075] S2. Based on the established high-precision boiler mechanism model, a CFD-AI model of the combustion process is established using CFD technology combined with AI algorithms.

[0076] S3. Based on the high-precision mechanism model and CFD-AI model of the boiler, and based on the operation data, a data-driven model is constructed and a dynamic model of the boiler is jointly constructed.

[0077] S4. Establish a controller based on the boiler dynamic model to perform multivariable coupled control;

[0078] S5. Simulate system response under different operating conditions and identify key influencing factors through multivariable coupling control;

[0079] S6. Based on key influencing factors, identify potential faults to provide corresponding early warnings and optimize boiler operating parameters.

[0080] Example 2

[0081] Reference Figure 1 The present invention provides a boiler operation simulation method based on multi-model coupling, the specific steps of which are as follows:

[0082] A high-precision mechanism model of the boiler is established based on the specific type and operating conditions of the boiler.

[0083] Based on the established high-precision boiler mechanism model, a CFD-AI model of the combustion process is established using CFD technology combined with AI algorithms.

[0084] Specifically, based on the established high-precision boiler mechanism model, the corresponding internal geometric model of the boiler is obtained. A mesh generation tool is used to divide the geometric model into multiple discrete mesh elements, with mesh refinement in the combustion and turbulent flow zones. Data from various sensors is used to acquire the flow properties within each mesh element, including steady-state or unsteady-state, compressible or incompressible, and laminar or turbulent flow. Then, a continuity equation is established based on fluid mass conservation, and a momentum conservation equation is established by combining mass and momentum conservation. Combustion reaction kinetics equations are established based on the finite-rate reaction model. Turbulent flow generated during combustion is collected in real-time based on these equations, and a corresponding k-ε turbulence model is constructed. The k-ε turbulence model is then integrated with the various equations to construct the corresponding CFD model. Boundary and initial conditions are then set for the CFD model, and CFD simulation is performed using ANSYS. Fluent software iteratively solves the governing equations of the CFD model, collects data from CFD simulations, removes noise and outliers, scales the processed data to the same range, uses correlation analysis to assess feature importance and identify key features, defines the number of neurons in the input, hidden, and output layers of the AI ​​model, divides the extracted key features into training, validation, and test sets, trains the model using the training set, adjusts model parameters using backpropagation, evaluates model performance using k-fold cross-validation, adjusts the original hyperparameters of the AI ​​model, evaluates the model's generalization ability using the validation set, and calculates accuracy and F1 score. The training and validation of the AI ​​model are repeated until the model's loss function converges to a preset range or the accuracy and F1 score meet preset standards. The prediction results of multiple models are combined using a voting method, and the performance of the final model is verified using the test set. If the model performance meets preset indicators, it is deployed in practical applications, and the AI ​​model and CFD model are coupled to form a CFD-AI combustion model.

[0085] Based on the high-precision boiler mechanism model and CFD-AI model, and based on the operational data, a data-driven model is constructed and a boiler dynamic model is jointly constructed.

[0086] Specifically, the process involves collecting parameter information from the constructed high-precision boiler mechanism model, CFD-AI model, and data-driven model. Based on the key parameters and their value ranges for each model, a corresponding parameter space is constructed. Then, according to the predefined parameter ranges, multiple parameter combinations are generated using a random number generator. The current boiler state and parameter combinations are used as the root node. The boiler state corresponding to each randomly generated parameter combination is calculated, and each parameter combination and its corresponding state are designated as new child nodes. A search tree is constructed based on the root node and child nodes. Starting from the root node, a greedy strategy is used to select child nodes along the path of the current search tree until an incompletely expanded node is reached. When a node is reached that is not fully expanded, the parameters of that node are perturbed by a random generator, and one or more child nodes are generated from that node, representing new states and parameter combinations. Actions or parameter combinations are randomly selected and performed until a termination state is reached. The reward value and performance index obtained from the simulation are recorded, and the simulation results are fed back to all nodes on the same path, and the reward value of each node is updated. The steps of selection, expansion, simulation and backtracking are repeated until the set computing resource limit is reached. Starting from the root node, the child node with the highest reward value is selected as the optimal parameter combination, and the output results of different models under this parameter combination are integrated to form the boiler dynamic model.

[0087] Example 3

[0088] Reference Figure 1 A boiler operation simulation method based on multi-model coupling is described, and the specific steps of this simulation method are as follows:

[0089] A controller is established based on the boiler dynamic model to achieve multivariable coupled control.

[0090] Specifically, based on the boiler dynamic model, the system acquires the sets of variables for the boiler temperature, pressure, and flow rate that need to be controlled, as well as the relevant system states. Target values ​​are set based on the needs of the operators. A state transition model is then established according to the boiler's dynamic characteristics, and the system states are correlated with the observed data to construct a corresponding observation model. A PID closed-loop control strategy is selected, and the error between the system output and the setpoint is monitored in real time. Then, based on prior knowledge of the boiler system, the prior probability distributions for multiple variables and states are calculated. Next, the observation data generated by the observation model are collected, and the posterior probabilities are updated using Bayes' theorem. Based on the posterior probabilities, the costs of the boiler state and control inputs are calculated. Based on the calculated costs, the control variables are adjusted. Based on the optimized control variables, a multivariate coupled control strategy is designed and implemented in the boiler system. Simultaneously, the system response is monitored in real time, and the control inputs and state estimates are continuously adjusted based on feedback data.

[0091] The system response under different operating conditions is simulated and key influencing factors are identified through multivariable coupling control.

[0092] Specifically, a range of simulable operating conditions is collected, and different combinations of operating conditions are set according to the boiler's operating conditions. Simulations are then performed under different operating conditions using a constructed boiler dynamic model, and the dynamic characteristics and responses of the boiler under different operating condition combinations are output in real time. Changes in each input parameter and their corresponding output results are recorded. The input and output data from the simulation experiment are preprocessed and organized to form a dataset. Sensitivity analysis of the collected data is performed using variance analysis to assess the influence of different input parameters on the output response. Based on the simulation data, a linear regression model of the relationship between each group of inputs and outputs is established, and the regression coefficients of the regression model are collected. The collected coefficients are arranged from largest to smallest; the larger the absolute value of the coefficient, the more significant the influence of that factor on the output. Based on the arrangement results, the key factors with the greatest impact on the system response are identified, and then the identified key influencing factors are analyzed to verify their actual impact on the boiler system.

[0093] Based on key influencing factors, potential faults are identified to provide corresponding early warnings and optimize boiler operating parameters.

[0094] Specifically, based on the optimization objectives of improving boiler thermal efficiency, reducing emissions, and reducing fuel consumption, an optimization objective function is designed. Multiple groups of particles are randomly generated, each representing a parameter combination. The initial position and velocity of each particle are randomly set. The fitness value of each group of particles is calculated using the defined optimization objective function. The global optimal position is initialized as the initial position of the first particle. The particle position is updated based on the current particle velocity, the historical optimal position, and the global optimal position. The fitness value of each group of particles is recorded in each iteration, and the current fitness is compared with the historical optimal fitness. If the current fitness is better than its historical optimal fitness, the historical optimal position is updated. The relationship between the fitness values ​​of other particles and the current global optimal fitness value is compared. If the current particle's fitness value is better than the global optimal fitness value, the global optimal position is updated, and the global optimal fitness value is also updated. The fitness value is calculated, and then the velocity of each group of particles is updated based on the current velocity, historical best position, and global best position. The position of the particles is also updated based on the updated velocity. It is checked whether the updated particle position is within the preset boundary range. If it is outside the range, its position is adjusted to the boundary value. The particle velocity, historical best position, and global best position are updated repeatedly until the fitness change value reaches the preset range. The update is then stopped, and the final global best position and corresponding fitness value are extracted. The optimized control parameters are applied to the boiler dynamic model for simulation testing. Based on the simulation results, the boiler's thermal efficiency, emission level, and fuel consumption are calculated and evaluated. The optimized operating parameters are compared with the unoptimized operating parameters to analyze the differences in boiler performance before and after optimization and to evaluate the degree of influence of each optimized parameter on boiler performance.

[0095] Example 4

[0096] Reference Figure 2 The boiler operation simulation system based on multi-model coupling provided by this invention includes:

[0097] The first model building module establishes a high-precision mechanism model of the boiler based on the specific type and operating conditions of the boiler.

[0098] The second model building module, based on the established high-precision boiler mechanism model, uses CFD technology combined with AI algorithms to build a CFD-AI model of the combustion process;

[0099] The third model building module is based on the high-precision mechanism model of the boiler and the CFD-AI model, and constructs a data-driven model based on the operating data and jointly constructs the dynamic model of the boiler.

[0100] The coupling control module establishes a controller based on the boiler dynamic model to perform multivariable coupled control.

[0101] The module for identifying key influencing factors uses multivariable coupling control to simulate system response under different operating conditions and identify key influencing factors.

[0102] The fault diagnosis module, based on key influencing factors, identifies potential faults to provide corresponding early warnings and optimize boiler operating parameters.

[0103] Example 5

[0104] The present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the boiler operation simulation method based on multi-model coupling.

[0105] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0106] This application is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0107] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0108] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0109] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0110] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for boiler operation simulation based on multi-model coupling, characterized in that, Comprise: S1, based on the specific type and working condition of the boiler to establish a high-precision mechanism model of the boiler; S2, based on the established high-precision mechanism model of the boiler, the CFD-AI model of the combustion process is established by using CFD technology combined with AI algorithm; S3, based on the high-precision mechanism model and CFD-AI model of the boiler, and according to the operation data, a data-driven model is constructed and a boiler dynamic model is jointly constructed, including: Q2.1: Collect the parameter information of the established high-precision mechanism model, CFD-AI model and data-driven model of the boiler, build the corresponding parameter space based on the key parameters of each group of models and their value range, and then generate multiple parameter combinations through a random number generator according to the predefined parameter range; Q2.2: Take the current state and parameter combination of the current boiler as the root node, calculate the corresponding state of each group of randomly generated parameter combinations, and respectively take each group of parameter combinations and the corresponding state as a new child node, and build a search tree according to the root node and the child node; Q2.3: From the root node, select the child node along the path of the current search tree through the greedy strategy until a node that is not fully expanded is reached, and when a node that is not fully expanded is reached, disturb the parameters of the node through a random generator to generate one or more child nodes from the node, representing new state and parameter combination; Q2.4: Select and combine parameters randomly to proceed until the termination state is reached, and record the reward value and performance index obtained by simulation, and feed back the simulation results to all nodes on the same path and update the reward value of each node; Q2.5: Repeat the selection, expansion, simulation and backtracking steps until the set calculation resource limit is reached, select the child node with the highest reward value from the root node as the best parameter combination, and integrate the output results of different models under the parameter combination to form a boiler dynamic model; S4, based on the boiler dynamic model, a controller is established for multivariate coupling control; S5, simulate the system response under different working conditions and identify the key influencing factors through multivariate coupling control; S6, based on the key influencing factors, judge the potential fault to provide corresponding early warning and optimize the boiler operation parameters.

2. The multi-model coupling-based boiler operation simulation method according to claim 1, characterized in that, In S2, based on the established high-precision mechanism model of the boiler, the CFD-AI model of the combustion process is established by using CFD technology combined with AI algorithm, including: Q1.1: Based on the established high-precision mechanism model of the boiler, the corresponding geometric model of the boiler is obtained, and the geometric model is divided into a plurality of discrete grid units using a grid division tool, and the grid is encrypted in the combustion zone and the airflow turbulence zone; Q1.2: Obtain the properties of the flow in the boiler of each grid unit through the data of each group of sensors, including steady or unsteady, compressible or incompressible, and laminar or turbulent flow, then establish the continuity equation for the mass conservation of the fluid, and establish the momentum conservation equation combined with the conservation of mass and momentum, and establish the combustion reaction kinetics equation according to the finite rate reaction model; Q1.3: Based on the combustion reaction kinetics equation, the turbulent flow region generated during the combustion process is collected in real time, and the corresponding k-ε turbulent flow model is constructed. Then, the k-ε turbulent flow model is integrated with each group of equations to construct the corresponding CFD model. After that, the boundary conditions and initial conditions of the CFD model are set, and the CFD simulation is performed. The CFD model control equations are solved by iteration through ANSYS Fluent software; Q1.4: Collect the data generated by CFD simulation, remove the noise and outliers of each group of data, and then scale the processed data to the same range. Use correlation analysis to evaluate the importance of features to determine the key features in each group of data. Then define the number of neurons in the input layer, hidden layer, and output layer of the AI model; Q1.5: Divide the extracted key features into training set, validation set and test set. Use the training set data to train the model, and adjust the model parameters through the back propagation algorithm. Evaluate the model performance through k-fold cross-validation and adjust the original hyperparameters of the AI model. Use the validation set to evaluate the generalization ability of the model and calculate the accuracy and F1 score indicators; Q1.6: Repeat the training and validation of the AI model until the model loss function value converges to the preset range or the accuracy and F1 score indicators reach the preset standard. Combine the prediction results of multiple models through voting method, and then verify the performance of the final model through the test set. If the model performance meets the preset indicators, deploy the model to actual application, and couple the AI model and CFD model to form a CFD-AI combustion model.

3. The multi-model coupling-based boiler operation simulation method according to claim 2, characterized in that, In S4, a controller is established based on the boiler dynamic model for multivariate coupling control, including: Q3.1: Based on the boiler dynamic model, obtain the boiler temperature, pressure and flow variables that need to be controlled, as well as the related system states. Set the target value based on the staff's demand, and establish a state transition model based on the dynamic characteristics of the boiler. Relate the system state to the observation data to construct the corresponding observation model; Q3.2: Select the PID closed-loop control strategy to monitor the error between the system output and the set value in real time. Then, based on the prior knowledge of the boiler system, calculate the prior probability distribution of multiple variables and states. Then collect the observation data generated by the observation model, and update the posterior probability through Bayes' theorem; Q3.3: Based on the posterior probability, calculate the cost of the boiler state and control input, and adjust the control variables according to the calculated cost. Based on the optimized control variables, design a multivariate coupling control strategy, implement the control strategy in the boiler system, and monitor the system response in real time. Continuously adjust the control input and state estimation based on the feedback data.

4. The multi-model coupling-based boiler operation simulation method according to claim 3, characterized in that, In S5, the system response under different working conditions is simulated and the key influencing factors are identified through multivariate coupling control, including: Q4.1: Collect the simulation range of working conditions, and set different working condition combinations according to the operating conditions of the boiler. Perform simulation under different working condition combinations through the constructed boiler dynamic model, and output the dynamic characteristics and responses of the boiler under different working condition combinations in real time; Q4.2: Record the changes of each input parameter and its corresponding output result, preprocess and organize the input and output data in the simulation experiment to form a data set, and perform sensitivity analysis on the collected data through variance analysis to evaluate the influence degree of different input parameters on the output response; Q4.3: According to the simulation data, a linear regression model is established between each group of inputs and outputs, and the regression coefficients of the regression model are collected. Arrange the collected coefficients in descending order. The larger the absolute value of the coefficient, the more significant the factor's impact on the output. According to the arrangement result, identify the key factors that have the greatest impact on the system response, and then analyze the identified key factors to verify their actual impact on the boiler system.

5. The multi-model coupling-based boiler operation simulation method according to claim 4, characterized in that, Based on the key influencing factors in S6, potential faults are judged to provide corresponding early warnings and optimize the boiler operating parameters, including: Q5.1: Based on the optimization objectives of improving the thermal efficiency of the boiler, reducing emissions, and reducing fuel consumption, design an optimization objective function, randomly generate multiple groups of particles, each particle representing a parameter combination, and randomly set the initial position and velocity of each particle; Q5.2: Calculate the fitness value of each particle using the set optimization objective function, initialize the global optimal position as the initial position of the first particle, update the particle position according to the current particle velocity, current velocity, historical optimal position, and global optimal position, record the fitness value of each particle at each iteration, and compare the current fitness and historical optimal fitness. If the current fitness is better than its historical optimal fitness, update the historical optimal position; Q5.3: Compare the relationship between the fitness value of the remaining particles and the current global optimal fitness value. If the fitness value of the current particle is better than the global optimal fitness value, update the global optimal position and the global optimal fitness value. Then update the velocity of each particle according to the current velocity, historical optimal position, and global optimal position, and update the position of the particle according to the updated velocity; Q5.4: Check if the updated particle position is within the pre-set boundary range. If it exceeds the range, adjust its position to the boundary value. Repeat the update of particle velocity, historical optimal position, and global optimal position until the fitness change value reaches the pre-set range, and stop updating; Q5.5: Extract the final global optimal position and corresponding fitness value, apply the optimized control parameters to the boiler dynamic model for simulation testing, calculate and evaluate the key performance indicators of the boiler's thermal efficiency, emission level, and fuel consumption based on the simulation results, compare the optimized operating parameters with the unoptimized operating parameters, analyze the differences in boiler performance before and after optimization, and evaluate the influence degree of each optimization parameter on the boiler performance.

6. A boiler operation simulation system based on multi-model coupling, characterized by, Including: A first model establishment module establishes a high-precision mechanism model of the boiler based on the specific type and working condition of the boiler; A second model establishment module establishes a CFD-AI model of the combustion process using CFD technology combined with AI algorithms based on the established high-precision mechanism model of the boiler; A third model establishing module, based on the high-precision mechanism model and the CFD-AI model of the boiler, and according to the operation data, a data-driven model is constructed and a boiler dynamic model is jointly constructed, including: Q2.1: Collect the parameter information of the high-precision mechanism model, the CFD-AI model and the data-driven model of the boiler constructed, construct the corresponding parameter space based on the key parameters of each group of models and their value range, and then generate a plurality of parameter combinations through a random number generator according to the predefined parameter range; Q2.2: Take the current state and parameter combination of the current boiler as the root node, calculate the corresponding state of the boiler of each group of randomly generated parameter combinations, and respectively take each group of parameter combinations and the corresponding state as a new child node, and construct a search tree according to the root node and the child node; Q2.3: Starting from the root node, select the child node along the path of the current search tree through the greedy strategy until reaching a node that is not fully expanded, when reaching a node that is not fully expanded, disturb the parameters of the node through the random generator, and generate one or more child nodes from the node to represent new state and parameter combination; Q2.4: Select and combine parameters randomly to proceed until the termination state is reached, and record the reward value and performance index obtained by simulation, feed back the simulation results to all nodes on the same path, and update the reward value of each node; Q2.5: Repeat the selection, expansion, simulation and backtracking steps until the set calculation resource limit is reached, select the child node with the highest reward value from the root node as the best parameter combination, and integrate the output results of different models under the parameter combination to form a boiler dynamic model; A coupling control module, based on the boiler dynamic model, establishes a controller for multivariate coupling control; A key influence factor identification module, which simulates the system response under different working conditions and identifies the key influence factors through multivariate coupling control; A fault judgment module, based on the key influence factors, judges the potential faults to provide corresponding early warning and optimize the boiler operation parameters.

7. The multi-model coupling based boiler operation simulation system according to claim 6, characterized in that, In the second model establishing module, based on the established high-precision mechanism model of the boiler, a CFD-AI model of the combustion process is established by using CFD technology combined with AI algorithm, including: Q1.1: Based on the established high-precision mechanism model of the boiler, the corresponding geometric model of the boiler inside is obtained, the geometric model is divided into a plurality of discrete grid units using a grid division tool, and the grid is encrypted in the combustion zone and the airflow turbulence zone; Q1.2: Obtain the properties of the boiler flow in each grid unit through the data of each group of sensors, including steady state or non-steady state, compressible or incompressible, and laminar or turbulent flow, then establish the continuity equation for the mass conservation of the fluid, and establish the momentum conservation equation combined with the conservation of mass and momentum, and establish the combustion reaction kinetics equation according to the finite rate reaction model; Q1.3: Based on the combustion reaction kinetics equation, the turbulent flow area generated in the combustion process is collected in real time, and the corresponding k-ε turbulent flow model is constructed, then the k-ε turbulent flow model is integrated with each group of equations to construct the corresponding CFD model, then the boundary conditions and initial conditions of the CFD model are set, and the CFD simulation is carried out, and the CFD model control equation is solved by iteration through ANSYS Fluent software; Q1.4: Collecting each group of data generated by CFD simulation, removing noise and outliers from each group of data, scaling each group of processed data to the same range, using correlation analysis to evaluate feature importance to determine key features in each group of data, and defining the number of neurons in the input layer, hidden layer and output layer of the AI model; Q1.5: Divide the extracted key features into training set, validation set and test set, use training set data for model training, adjust model parameters through back propagation algorithm, evaluate model performance through k-fold cross validation, adjust original hyperparameters of AI model, use validation set to evaluate generalization ability of model, and calculate accuracy and F1 score indicators; Q1.6: Repeat the training and validation of the AI model until the model loss function value converges to the preset range or the accuracy and F1 score indicators reach the preset standard, combine the prediction results of multiple models through voting method, then verify the performance of the final model through the test set, if the model performance meets the preset indicators, deploy the model to actual application, and couple the AI model and the CFD model to form a CFD-AI combustion model.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, which when executed by the processor, implements the steps of the boiler operation simulation method based on multi-model coupling according to any one of claims 1-5.

Images