Boiler steam-water side two-dimensional mechanism model establishing method and system

Abstract

The invention discloses a boiler steam-water side two-dimensional mechanism model establishing method and system. The method comprises the steps that all operation data of a boiler are comprehensively collected and preprocessed; establishing two-dimensional discretization modeling of each area of the boiler according to the operation data; predicting temperature distribution of each region of the boiler and formulating a temperature control strategy of each region; constructing a virtual boiler operation model and dynamically optimizing a control decision of the boiler; verifying the model through experiment and actual operation data; the system comprises a data acquisition preprocessing module, a two-dimensional discretization modeling module, a prediction module, a construction control decision module, a verification module and a real-time monitoring module. According to the method, the efficiency of the decision-making process is improved, meanwhile, the complex coupling relation between different areas can be flexibly processed, an optimization scheme is found in a multi-area and multi-dimensional control space, the energy efficiency, stability and economical efficiency of the boiler can be improved, the requirement for human intervention is lowered, and the overall operation efficiency and environmental protection performance of the boiler are remarkably improved.

Description

Technical Field

[0001] The invention belongs to the technical field of boiler control, and in particular relates to a method and system for establishing a two-dimensional mechanism model of a boiler steam-water side. Background Art

[0002] Boilers are core equipment in the power and industrial fields, and their efficient and safe operation directly affects energy utilization and production costs. In modern complex boiler systems, precise control of parameters such as temperature, pressure, and flow is crucial to achieving optimal operation of the boiler. In addition, in the boiler system, the thermodynamic process on the steam-water side has an important impact on the overall performance of the boiler. Traditional boiler control methods often rely on experience and simple control algorithms, which are difficult to cope with the complex thermodynamic processes and nonlinear characteristics inside the boiler. In addition, existing boiler modeling methods are usually based on one-dimensional models, which cannot fully capture the complex multidimensional characteristics inside the boiler, resulting in limited prediction accuracy for actual operating conditions. In order to improve the precise control capability of boiler operation, it is of great theoretical and practical significance to establish an accurate two-dimensional mechanism model of the boiler steam-water side and to provide a detailed description and dynamic prediction of the boiler operating status; therefore, it is particularly important to invent a method for establishing a two-dimensional mechanism model of the boiler steam-water side.

[0003] The existing method for establishing a two-dimensional mechanism model on the steam-water side of a boiler has an inefficient decision-making process and cannot flexibly handle the complex coupling relationship between different regions, which reduces the applicability of the model. In addition, the existing method for establishing a two-dimensional mechanism model on the steam-water side of a boiler has low energy efficiency, stability and economy, and requires more human intervention, which reduces the overall operating efficiency and environmental performance of the boiler. Summary of the invention

[0004] The purpose of the present invention is to solve the defects existing in the prior art and provide a method and system for establishing a two-dimensional mechanism model of the steam-water side of a boiler.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: A method for establishing a two-dimensional mechanism model of a boiler steam-water side, comprising: Q1. Comprehensively collect and pre-process various boiler operation data; Q2. Establish two-dimensional discrete modeling of each area of ​​the boiler based on operating data; Q3. Based on the established two-dimensional discretization modeling of each boiler area, predict the temperature distribution of each boiler area and formulate the temperature control strategy for each area; Q4. Based on the temperature control strategy of each area, a virtual boiler operation model is constructed and the boiler control decision is dynamically optimized; Q5. Verify the virtual boiler operation model by implementing optimized control decisions and collecting actual operation data; Q6. Based on the verification results, monitor the boiler operating status in real time and dynamically adjust the boiler control strategy.

[0006] A further improvement of the present invention is that various boiler operation data are comprehensively collected and preprocessed in Q1, including: S1.1: Obtain the temperature and flow data of each area of ​​the boiler collected by each group of sensors from the DCS system, check whether there are missing values ​​in the data set, and if so, fill the missing values ​​through interpolation or mean filling method, then detect the outliers in the data through box plots, and replace or remove the outliers; S1.2: Check and remove possible duplicate records in the data set, unify the units and formats of various data, and unify various data through minimum-maximum normalization to unify the data range to between 0 and 1; S1.3: Determine the boundaries of different physical areas of the boiler based on the physical structure and thermodynamic properties of the boiler. Map the collected sensor data to the corresponding boiler area based on the physical location and data characteristics of each sensor obtained by the DCS system, and mark the boiler area to which it belongs in each record of the data set.

[0007] A further improvement of the present invention is that in Q2, a two-dimensional discretization model of each area of ​​the boiler is established according to the operation data, including: S2.1: Analyze the overall structure of the boiler, identify the physical location and function of each area, determine the three-dimensional boundaries of each area in terms of height, width and depth according to the boiler design drawings and operating characteristics, then obtain the boundary information between the areas, and map the three-dimensional space inside the boiler into various functional areas to form a preliminary regional division model; S2.2: Calculate the total volume of each region according to the boundary size of each region in the region partition model, calculate the number of cells to be divided in each direction according to the total size of the region and the preset cell size, and evenly distribute the cell boundaries in each direction of the region to obtain a grid system composed of multiple groups of cubic cells; S2.3: Number each grid node and cell according to the spatial coordinate order in the grid system, and after completing the grid division, obtain the discretized model of each area of ​​the boiler, check the consistency and quality of the grid, and then output the node coordinates, cell number and boundary information of the grid as a file, and store and back up the model grid data at the same time.

[0008] A further improvement of the present invention is that, based on the two-dimensional discretization modeling of each area of ​​the boiler established in Q3, the temperature distribution of each area of ​​the boiler is predicted and the temperature control strategy of each area is formulated, including: S3.1: The operating state of the boiler is taken as an element of the state space, where each state represents the overall operating condition of the boiler at each moment, and the action space of the control strategy is constructed according to the adjustable control variables of the boiler, and different actions in the action space are constructed according to different combinations of control variables; S3.2: Initialize a set of root nodes according to the current operating state of the boiler, and based on the root nodes, select each set of action strategies that can be used from the action space, and obtain the operating state of the state space corresponding to each action strategy, and construct a corresponding number of child nodes based on the selected action strategy and the corresponding boiler operating state; S3.3: Starting from the root node, calculate the UCB value of each child node, and select the strategy based on the upper confidence limit, repeatedly select the child node with the highest UCB value, and select layer by layer until you reach an incompletely expanded node, and expand the incompletely expanded node reached during the selection process. In the current state, select an action that has not been explored, apply the action to get a new state, and add a new node to the tree; S3.4: Starting from the expanded node, randomly select actions to simulate until the set time point or temperature target is reached, and evaluate the temperature change of the boiler under this action sequence, calculate the benefit of the path, and backpropagate the simulated benefit information back to the entire path, and update the number of visits and average benefit values ​​of each node on the path; S3.5: Repeat the selection, expansion, simulation and backtracking steps. After each simulation, evaluate the current optimal path and compare it with the previous optimal strategy to determine the optimization effect. According to the stability of the optimal strategy and the trend of the change in benefits, determine whether the search has converged. If the strategy change value is within the preset range in multiple iterations, it is considered that the optimal control strategy has been achieved. S3.6: Extract the path with the highest benefit and use it as the optimal control strategy. Then convert the extracted optimal control strategy into actual control instructions and apply it to the distributed control system of the boiler to optimize the temperature of each area. At the same time, record the key indicators of the boiler in actual operation in real time and readjust the control strategy according to the actual operation effect.

[0009] A further improvement of the present invention is that the operating state of the boiler described in S3.1 specifically includes the real-time temperature, pressure, fuel flow, air flow, flue gas composition, steam flow and water level of the boiler combustion zone, burnout zone, water wall zone, screen pass zone, high pass / high re-zone and low pass / low re-zone; The adjustable control variables of the boiler described in S3.1 specifically include fuel supply, air volume control, blower speed, exhaust fan speed, feed water flow, recirculated flue gas volume and burner position.

[0010] A further improvement of the present invention is that the key indicators described in S3.6 specifically include outlet steam temperature, boiler efficiency, emission indicators, fuel consumption, steam pressure and heat load distribution.

[0011] A further improvement of the present invention is that in Q4, a virtual boiler operation model is constructed according to the temperature control strategy of each area and the control decision of the boiler is dynamically optimized, including: S4.1: Collect the historical operation data, control parameters and environmental variables of the boiler from the DCS system, and after preprocessing each set of data, extract the physical structure and working principle of the boiler from the data, and write the heat transfer and mass transfer equations of each part of the boiler, as well as the fluid flow equations, to build the corresponding boiler physical model, set the boundary conditions and initial conditions of the model according to the actual operating conditions of the boiler, use computer-aided design software to draw the three-dimensional structural model of the boiler, mesh the boiler model in three-dimensional space, decompose it into many small units, adjust the mesh density according to the characteristics of each area of ​​the boiler, and set the input parameters and solution accuracy of the simulation model according to the actual parameters of the boiler to complete the construction of the virtual boiler operation model; S4.2: Collect the key variables of the boiler operation conditions simulated by the virtual boiler operation model, and form the state space of the boiler with different combinations of each group of variables. Then list the adjustable control variables in the boiler operation, and each action corresponds to adjusting one or more control variables, and construct the corresponding action space; S4.3: Based on the historical operation data, analyze the probability distribution of the boiler transferring to other states after taking different actions in different states, and determine the transfer probability of each state through frequency statistics. Then, the current optimal control strategy is used as the starting point, and the strategy is applied to all possible states, and a complete initial strategy table is constructed; S4.4: Calculate the total return expected from following the current strategy in different states through the Bellman equation, and then recursively calculate the value function of each state until the change in the value function is less than the preset threshold. At the same time, in each iteration, update the value function of each state according to the current strategy and state transition probability; S4.5: Compare the value functions of different actions under each state, and select the action with the highest value function result, and use the optimal action to replace the action in the original strategy, and record the optimal action selected in each state to form an updated strategy table, and then repeat the iteration; S4.6: Compare the current strategy with the strategy in the previous iteration. If the actions in all states no longer change, it means that the strategy has converged and the iteration process is terminated. Otherwise, continue the iteration, convert the optimal strategy into an operational control instruction, and perform simulation testing.

[0012] A two-dimensional mechanism model building system for boiler steam-water side, comprising: Data collection and preprocessing module, comprehensively collects and preprocesses various boiler operation data; Two-dimensional discrete modeling module, which establishes two-dimensional discrete modeling of each area of ​​the boiler based on operating data; The prediction module predicts the temperature distribution of each area of ​​the boiler and formulates the temperature control strategy for each area based on the established two-dimensional discretization model of each area of ​​the boiler; Build a control decision module, build a virtual boiler operation model and dynamically optimize the boiler's control decision based on the temperature control strategy of each area; A verification module verifies the virtual boiler operation model by implementing optimized control decisions and collecting actual operation data; The real-time monitoring module monitors the boiler operation status in real time and dynamically adjusts the boiler control strategy based on the verification results.

[0013] A further improvement of the present invention is that in the data acquisition and preprocessing module, various boiler operation data are comprehensively acquired and preprocessed, including: S1.1: Obtain the temperature and flow data of each area of ​​the boiler collected by each group of sensors from the DCS system, check whether there are missing values ​​in the data set, and if so, fill the missing values ​​through interpolation or mean filling method, then detect the outliers in the data through box plots, and replace or remove the outliers; S1.2: Check and remove possible duplicate records in the data set, unify the units and formats of various data, and unify various data through minimum-maximum normalization to unify the data range to between 0 and 1; S1.3: Determine the boundaries of different physical areas of the boiler based on the physical structure and thermodynamic properties of the boiler. Map the collected sensor data to the corresponding boiler area based on the physical location and data characteristics of each sensor obtained by the DCS system, and mark the boiler area to which it belongs in each record of the data set.

[0014] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the steps of a method for establishing a two-dimensional mechanism model of a boiler steam-water side.

[0015] Compared with the prior art, the present invention has at least the following beneficial technical effects: The present invention provides a method and system for establishing a two-dimensional mechanism model of a boiler steam-water side, which initializes a group of root nodes, and based on the root nodes, selects each group of action strategies that can be used from the action space, and constructs a corresponding number of child nodes based on the selected action strategies and the corresponding boiler operation status. Starting from the root node, according to the upper confidence limit selection strategy, the child node with the highest UCB value is repeatedly selected, and the selection is performed downward layer by layer until an incompletely expanded node is reached, and the incompletely expanded node reached during the selection process is expanded. Starting from the expanded node, actions are randomly selected for simulation until a set time point or temperature target is reached, and the temperature change of the boiler under this action sequence is evaluated, the benefit of the path is calculated, and the simulated benefit information is back-propagated back to the entire path, and the nodes on the path are updated. The number of visits and average benefit values ​​of the point are calculated, and the selection, expansion, simulation and backtracking steps are repeated. After each simulation, the current optimal path is evaluated and compared with the previous optimal strategy to determine the optimization effect. After the iteration, the path with the highest benefit is extracted and used as the optimal control strategy. The extracted optimal control strategy is then converted into actual control instructions and applied to the distributed control system of the boiler to optimize the temperature of each area. At the same time, various key indicators of the boiler in actual operation are recorded in real time. According to the actual operation effect, the control strategy is readjusted to improve the efficiency of the decision-making process. At the same time, it can flexibly handle the complex coupling relationship between different areas, find the optimization plan in the multi-region and multi-dimensional control space, improve the applicability of the model, and effectively reduce the demand for computing resources.

[0016] The present invention provides a method and system for establishing a two-dimensional mechanism model of the steam-water side of a boiler, which constructs the state space and action space of the boiler, analyzes the probability distribution of the boiler possibly transferring to other states after taking different actions under different states based on historical operation data, and determines the transfer probability of each state through frequency statistics. After that, the current optimal control strategy is used as the starting point, and the strategy is applied to all possible states, and a complete initial strategy table is constructed. The total return expected to be obtained by following the current strategy under different states is calculated through the Bellman equation, and then the value function of each state is calculated in a recursive manner until the change in the value function is less than a preset threshold. At the same time, in each iteration, according to the current strategy and state transfer, the value function of each state is calculated. Probability, update the value function of each state, compare the value functions of different actions under each state, and select the action with the highest value function result, and use the optimal action to replace the action in the original strategy, and record the optimal action selected in each state to form an updated strategy table, and then repeat the iteration, compare the current strategy with the strategy in the previous iteration. If the actions in all states no longer change, it means that the strategy converges and the iteration process is terminated. Otherwise, continue to iterate, convert the optimal strategy into an operational control instruction, and conduct simulation tests, which can improve the energy efficiency, stability and economy of the boiler, reduce the need for human intervention, and significantly improve the overall operating efficiency and environmental performance of the boiler. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0018] Figure 1 The present invention is a flow chart of a method for establishing a two-dimensional mechanism model of a boiler steam-water side.

[0019] Figure 2 The present invention is a flow chart of a system for establishing a two-dimensional mechanism model of the steam-water side of a boiler. DETAILED DESCRIPTION

[0020] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0021] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0022] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0023] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0024] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0025] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] Example 1 The present invention provides a method for establishing a two-dimensional mechanism model of a boiler steam-water side, comprising: Q1. Comprehensively collect and pre-process various boiler operation data; Q2. Establish two-dimensional discrete modeling of each area of ​​the boiler based on operating data; Q3. Based on the established two-dimensional discretization modeling of each boiler area, predict the temperature distribution of each boiler area and formulate the temperature control strategy for each area; Q4. Based on the temperature control strategy of each area, a virtual boiler operation model is constructed and the boiler control decision is dynamically optimized; Q5. Verify the virtual boiler operation model by implementing optimized control decisions and collecting actual operation data; Q6. Based on the verification results, monitor the boiler operating status in real time and dynamically adjust the boiler control strategy.

[0027] Example 2 Reference Figure 1 The present invention provides a method for establishing a two-dimensional mechanism model of a boiler steam-water side, and the specific steps of the model establishment method are as follows: Comprehensively collect and pre-process various boiler operation data.

[0028] Specifically, the temperature and flow data of each area of ​​the boiler collected by each group of sensors are obtained from the DCS system, and it is checked whether there are missing values ​​in the data set. If so, the missing values ​​are filled by interpolation or mean filling method. Then, the outliers in the data are detected by box plot, and the outliers are replaced or eliminated. The duplicate records that may exist in the data set are checked and removed, and then the units and formats of various data are unified. Various types of data are unified through minimum-maximum normalization processing to unify the data range to between 0 and 1. According to the physical structure and thermodynamic characteristics of the boiler, the boundaries of different physical areas of the boiler are determined. According to the physical position and data characteristics of each sensor obtained by the DCS system, the collected sensor data are mapped to the corresponding boiler area, and the boiler area to which it belongs is marked in each record of the data set.

[0029] A two-dimensional discrete model of each area of ​​the boiler is established based on the operating data.

[0030] Specifically, the overall structure of the boiler is analyzed to identify the physical location and function of each area. According to the boiler design drawings and operating characteristics, three sets of three-dimensional boundaries of height, width and depth of each area are determined, and then the boundary information between the areas is obtained. At the same time, the three-dimensional space inside the boiler is mapped into various functional areas to form a preliminary regional division model. The total volume of each area is calculated according to the boundary size of each area in the regional division model. According to the total size of the area and the preset cell size, the number of cells to be divided in each direction is calculated, and the cell boundaries are evenly distributed in each direction of the area to obtain a grid system composed of multiple groups of cubic cells. Each grid node and cell are numbered according to the spatial coordinate order in the grid system. After the grid division is completed, a discretized model of each area of ​​the boiler is obtained, and the consistency and quality of the grid are checked. Then, the node coordinates, cell number and boundary information of the grid are output as a file, and the model grid data is stored and backed up.

[0031] Based on the established two-dimensional discretization modeling of each boiler area, the temperature distribution of each boiler area is predicted and the temperature control strategy of each area is formulated.

[0032] Specifically, the operating state of the boiler is taken as an element of the state space, where each state represents the overall operating status of the boiler at each moment, and the action space of the control strategy is constructed according to the adjustable control variables of the boiler. At the same time, different actions in the action space are constructed according to different combinations of the control variables. According to the current operating state of the boiler, a set of root nodes are initialized, and based on the root node, each set of action strategies that can be used is selected from the action space, and the operating state of the state space corresponding to each action strategy is obtained. Based on the selected action strategy and the corresponding boiler operating state, a corresponding number of child nodes are constructed. Starting from the root node, the UCB value of each child node is calculated, and a strategy is selected according to the upper confidence limit. The child node with the highest UCB value is repeatedly selected, and the selection is performed downward layer by layer until an incompletely expanded node is reached, and the incompletely expanded nodes reached during the selection process are expanded. In the current state, an action that has not been explored is selected, and a new state is obtained after applying the action, and a new node is added to the tree. From the expansion Starting from the node, randomly select actions for simulation until the set time point or temperature target is reached, and evaluate the temperature change of the boiler under this action sequence, calculate the benefit of the path, and backpropagate the simulated benefit information back to the entire path, update the number of visits and average benefit values ​​of each node on the path, repeat the selection, expansion, simulation and backtracking steps, evaluate the current optimal path after each simulation, and compare it with the previous optimal strategy to judge the optimization effect, judge whether the search converges according to the stability of the optimal strategy and the change trend of the benefit, if the strategy change value is within the preset range in multiple iterations, it is considered that the optimal control strategy is reached, extract the path with the highest benefit, and use it as the optimal control strategy, then convert the extracted optimal control strategy into actual control instructions, and apply it to the distributed control system of the boiler to optimize the temperature of each area, and at the same time record the key indicators of the boiler in actual operation in real time, and readjust the control strategy according to the actual operation effect.

[0033] In this embodiment, the operating status of the boiler specifically includes the real-time temperature, pressure, fuel flow, air flow, flue gas composition, steam flow and water level of the boiler combustion zone, burnout zone, water-cooled wall zone, screen pass zone, high pass / high re-zone and low pass / low re-zone; the adjustable control variables of the boiler specifically include fuel supply, air volume control, blower speed, smoke exhaust fan speed, feed water flow, recycled flue gas volume and burner position; the key indicators specifically include outlet steam temperature, boiler efficiency, emission indicators, fuel consumption, steam pressure and heat load distribution.

[0034] Example 3 Reference Figure 1 The present invention provides a method for establishing a two-dimensional mechanism model of a boiler steam-water side, and the specific steps of the model establishment method are as follows: According to the temperature control strategy of each area, a virtual boiler operation model is constructed and the boiler control decision is dynamically optimized.

[0035] Specifically, the historical operation data, control parameters and environmental variables of the boiler are collected from the DCS system, and after preprocessing each group of data, the physical structure and working principle of the boiler in the data are extracted, and the heat transfer, mass transfer equations and fluid flow equations of each part of the boiler are compiled to construct the corresponding boiler physical model, and the boundary conditions and initial conditions of the model are set according to the actual operating conditions of the boiler. The three-dimensional structural model of the boiler is drawn using computer-aided design software, and the boiler model is meshed in three-dimensional space and decomposed into many small units. The mesh density is adjusted according to the characteristics of each area of ​​the boiler, and the input parameters and solution accuracy of the simulation model are set according to the actual parameters of the boiler to complete the construction of the virtual boiler operation model, collect the key variables of the boiler operation conditions simulated by the virtual boiler operation model, and form the state space of the boiler with different combinations of each group of variables, and then list the adjustable control variables in the boiler operation, and each action corresponds to adjusting one or more control variables, and construct the corresponding action space, and analyze the effects of taking different actions under different states according to the historical operation data. , the probability distribution of the boiler's possible transfer to other states, and determine the transfer probability of each state through frequency statistics. Then, the current optimal control strategy is used as the starting point, and the strategy is applied to all possible states, and a complete initial strategy table is constructed. The Bellman equation is used to calculate the total return expected to be obtained by following the current strategy in different states, and then the value function of each state is calculated recursively until the change of the value function is less than the preset threshold. At the same time, in each iteration, according to the current strategy and state transfer probability, the value function of each state is updated, the value function of different actions under each state is compared, and the action with the highest value function result is selected, and the action in the original strategy is replaced by the optimal action, and the optimal action selected in each state is recorded to form an updated strategy table, and then the iteration is repeated, and the current strategy is compared with the strategy in the previous iteration. If the actions in all states no longer change, it means that the strategy converges and the iteration process is terminated. Otherwise, the iteration continues, the optimal strategy is converted into an operational control instruction, and a simulation test is performed.

[0036] The virtual boiler operation model is validated by implementing optimized control decisions and collecting actual operation data.

[0037] Based on the verification results, the boiler operating status is monitored in real time and the boiler control strategy is adjusted dynamically.

[0038] Example 4 Reference Figure 2 The present invention provides a boiler steam-water side two-dimensional mechanism model establishment system, comprising: Data collection and preprocessing module, comprehensively collects and preprocesses various boiler operation data; Two-dimensional discrete modeling module, which establishes two-dimensional discrete modeling of each area of ​​the boiler based on operating data; The prediction module predicts the temperature distribution of each area of ​​the boiler and formulates the temperature control strategy for each area based on the established two-dimensional discretization model of each area of ​​the boiler; Build a control decision module, build a virtual boiler operation model and dynamically optimize the boiler's control decision based on the temperature control strategy of each area; A verification module verifies the virtual boiler operation model by implementing optimized control decisions and collecting actual operation data; The real-time monitoring module monitors the boiler operation status in real time and dynamically adjusts the boiler control strategy based on the verification results.

[0039] Example 5 The present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method for establishing a two-dimensional mechanism model of the steam-water side of a boiler are implemented.

[0040] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0041] The present application is described with reference to the flowcharts and / or block diagrams of the methods, systems and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or boxes.

[0042] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0043] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0044] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0045] In addition, it should be understood that although this specification is described in accordance with the implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that can be understood by those skilled in the art. The above content is only to illustrate the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution according to the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for establishing a two-dimensional mechanism model of a boiler steam-water side, characterized in that: include: Q1. Comprehensively collect and pre-process various boiler operation data; Q2. Establish two-dimensional discrete modeling of each area of ​​the boiler based on operating data; Q3. Based on the established two-dimensional discretization modeling of each area of ​​the boiler, predict the temperature distribution of each area of ​​the boiler and formulate the temperature control strategy for each area; Q4. Based on the temperature control strategy of each area, a virtual boiler operation model is constructed and the boiler control decision is dynamically optimized; Q5. Verify the virtual boiler operation model by implementing optimized control decisions and collecting actual operation data; Q6. Based on the verification results, monitor the boiler operating status in real time and dynamically adjust the boiler control strategy.

2. A method for establishing a two-dimensional mechanism model of a boiler steam-water side according to claim 1, characterized in that: In Q1, all boiler operation data are collected and pre-processed, including: S1.1: Obtain the temperature and flow data of each area of ​​the boiler collected by each group of sensors from the DCS system, check whether there are missing values ​​in the data set, and if so, fill the missing values ​​through interpolation or mean filling method, then detect the outliers in the data through box plots, and replace or remove the outliers; S1.2: Check and remove possible duplicate records in the data set, unify the units and formats of various data, and unify various data through minimum-maximum normalization to unify the data range to between 0 and 1; S1.3: Determine the boundaries of different physical areas of the boiler based on the physical structure and thermodynamic properties of the boiler. Map the collected sensor data to the corresponding boiler area based on the physical location and data characteristics of each sensor obtained by the DCS system, and mark the boiler area to which it belongs in each record of the data set.

3. A method for establishing a two-dimensional mechanism model of a boiler steam-water side according to claim 2, characterized in that: In Q2, two-dimensional discretization modeling of each area of ​​the boiler is established based on the operating data, including: S2.1: Analyze the overall structure of the boiler, identify the physical location and function of each area, determine the three-dimensional boundaries of each area in terms of height, width and depth according to the boiler design drawings and operating characteristics, then obtain the boundary information between the areas, and map the three-dimensional space inside the boiler into various functional areas to form a preliminary regional division model; S2.2: Calculate the total volume of each region according to the boundary size of each region in the region partition model, calculate the number of cells to be divided in each direction according to the total size of the region and the preset cell size, and evenly distribute the cell boundaries in each direction of the region to obtain a grid system composed of multiple groups of cubic cells; S2.3: Number each grid node and cell according to the spatial coordinate order in the grid system, and after completing the grid division, obtain the discretized model of each area of ​​the boiler, check the consistency and quality of the grid, and then output the node coordinates, cell number and boundary information of the grid as a file, and store and back up the model grid data at the same time.

4. A method for establishing a two-dimensional mechanism model of a boiler steam-water side according to claim 3, characterized in that: In Q3, based on the two-dimensional discretization modeling of each boiler area, the temperature distribution of each boiler area is predicted and the temperature control strategy of each area is formulated, including: S3.1: The operating state of the boiler is taken as an element of the state space, where each state represents the overall operating condition of the boiler at each moment, and the action space of the control strategy is constructed according to the adjustable control variables of the boiler, and different actions in the action space are constructed according to different combinations of control variables; S3.2: Initialize a set of root nodes according to the current operating state of the boiler, and based on the root nodes, select each set of action strategies that can be used from the action space, and obtain the operating state of the state space corresponding to each action strategy, and construct a corresponding number of child nodes based on the selected action strategy and the corresponding boiler operating state; S3.3: Starting from the root node, calculate the UCB value of each child node, and select the strategy based on the upper confidence limit, repeatedly select the child node with the highest UCB value, and select layer by layer until you reach an incompletely expanded node, and expand the incompletely expanded node reached during the selection process. In the current state, select an action that has not been explored, apply the action to get a new state, and add a new node to the tree; S3.4: Starting from the expanded node, randomly select actions to simulate until the set time point or temperature target is reached, and evaluate the temperature change of the boiler under this action sequence, calculate the benefit of the path, and backpropagate the simulated benefit information back to the entire path, and update the number of visits and average benefit values ​​of each node on the path; S3.5: Repeat the selection, expansion, simulation and backtracking steps. After each simulation, evaluate the current optimal path and compare it with the previous optimal strategy to determine the optimization effect. According to the stability of the optimal strategy and the trend of the change in benefits, determine whether the search has converged. If the strategy change value is within the preset range in multiple iterations, it is considered that the optimal control strategy has been achieved. S3.6: Extract the path with the highest benefit and use it as the optimal control strategy. Then convert the extracted optimal control strategy into actual control instructions and apply it to the distributed control system of the boiler to optimize the temperature of each area. At the same time, record the key indicators of the boiler in actual operation in real time and readjust the control strategy according to the actual operation effect.

5. A method for establishing a two-dimensional mechanism model of a boiler steam-water side according to claim 4, characterized in that: The operating status of the boiler described in S3.1 specifically includes the real-time temperature, pressure, fuel flow, air flow, flue gas composition, steam flow and water level of the boiler combustion zone, burnout zone, water wall zone, screen pass zone, high pass / high re-zone and low pass / low re-zone; The adjustable control variables of the boiler described in S3.1 specifically include fuel supply, air volume control, blower speed, exhaust fan speed, feed water flow, recirculated flue gas volume and burner position.

6. A method for establishing a two-dimensional mechanism model of a boiler steam-water side according to claim 4, characterized in that: The key indicators described in S3.6 specifically include outlet steam temperature, boiler efficiency, emission indicators, fuel consumption, steam pressure and heat load distribution.

7. A method for establishing a two-dimensional mechanism model of a boiler steam-water side according to claim 4, characterized in that: In Q4, a virtual boiler operation model is constructed based on the temperature control strategy of each area and the boiler control decision is dynamically optimized, including: S4.1: Collect the historical operation data, control parameters and environmental variables of the boiler from the DCS system, and after preprocessing each set of data, extract the physical structure and working principle of the boiler from the data, and write the heat transfer and mass transfer equations of each part of the boiler, as well as the fluid flow equations, to build the corresponding boiler physical model, set the boundary conditions and initial conditions of the model according to the actual operating conditions of the boiler, use computer-aided design software to draw the three-dimensional structural model of the boiler, mesh the boiler model in three-dimensional space, decompose it into many small units, adjust the mesh density according to the characteristics of each area of ​​the boiler, and set the input parameters and solution accuracy of the simulation model according to the actual parameters of the boiler to complete the construction of the virtual boiler operation model; S4.2: Collect the key variables of the boiler operation conditions simulated by the virtual boiler operation model, and form the state space of the boiler with different combinations of each group of variables. Then list the adjustable control variables in the boiler operation, and each action corresponds to adjusting one or more control variables, and construct the corresponding action space; S4.3: Based on the historical operation data, analyze the probability distribution of the boiler transferring to other states after taking different actions in different states, and determine the transfer probability of each state through frequency statistics. Then, the current optimal control strategy is used as the starting point, and the strategy is applied to all possible states, and a complete initial strategy table is constructed; S4.4: Calculate the total return expected from following the current strategy in different states through the Bellman equation, and then recursively calculate the value function of each state until the change in the value function is less than the preset threshold. At the same time, in each iteration, update the value function of each state according to the current strategy and state transition probability; S4.5: Compare the value functions of different actions under each state, and select the action with the highest value function result, and use the optimal action to replace the action in the original strategy, and record the optimal action selected in each state to form an updated strategy table, and then repeat the iteration; S4.6: Compare the current strategy with the strategy in the previous iteration. If the actions in all states no longer change, it means that the strategy has converged and the iteration process is terminated. Otherwise, continue the iteration, convert the optimal strategy into an operational control instruction, and perform simulation testing.

8. A two-dimensional mechanism model building system for boiler steam-water side, characterized in that: include: Data collection and preprocessing module, which comprehensively collects and preprocesses various boiler operation data; Two-dimensional discrete modeling module, which establishes two-dimensional discrete modeling of each area of ​​the boiler based on operating data; The prediction module predicts the temperature distribution of each area of ​​the boiler and formulates the temperature control strategy for each area based on the established two-dimensional discretization model of each area of ​​the boiler; Build a control decision module, build a virtual boiler operation model and dynamically optimize the boiler's control decision based on the temperature control strategy of each area; A verification module verifies the virtual boiler operation model by implementing optimized control decisions and collecting actual operation data; The real-time monitoring module monitors the boiler operation status in real time and dynamically adjusts the boiler control strategy based on the verification results.

9. A boiler steam-water side two-dimensional mechanism model establishment system according to claim 8, characterized in that: In the data acquisition and preprocessing module, all boiler operation data are collected and preprocessed, including: S1.1: Obtain the temperature and flow data of each area of ​​the boiler collected by each group of sensors from the DCS system, check whether there are missing values ​​in the data set, and if so, fill the missing values ​​through interpolation or mean filling method, then detect the outliers in the data through box plots, and replace or remove the outliers; S1.2: Check and remove possible duplicate records in the data set, unify the units and formats of various data, and unify various data through minimum-maximum normalization to unify the data range to between 0 and 1; S1.3: Determine the boundaries of different physical areas of the boiler based on the physical structure and thermodynamic properties of the boiler. Map the collected sensor data to the corresponding boiler area based on the physical location and data characteristics of each sensor obtained by the DCS system, and mark the boiler area to which it belongs in each record of the data set.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a method for establishing a two-dimensional mechanism model of a boiler steam-water side according to any one of claims 1 to 7 are implemented.

Images