Boiler operation load dynamic distribution method and system

Abstract

The invention discloses a boiler operation load dynamic distribution method and system, and the method comprises the steps: collecting and preprocessing the boiler operation data of each boiler in a mother system boiler system according to a preset period through employing a preset standard communication protocol, and obtaining the to-be-analyzed boiler operation data; according to the boiler operation data to be analyzed, dynamic characteristics of each boiler are analyzed to serve as a dynamic characteristic model of each boiler; the dynamic characteristics are used for representing change information of performance parameters along with time in the operation process of each boiler; based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm, dynamically planning the load of each boiler to obtain the optimal load distribution proportion of each boiler; and distributing the operation load of each boiler through the optimal load distribution proportion of each boiler. Therefore, by adopting the embodiment of the invention, the operation cost of the header system boiler system can be reduced, and the reliability and economical efficiency of the system are improved.

Description

Technical Field

[0001] This application relates to the technical field of equipment control, and particularly to a method and system for dynamically allocating the operating load of boilers. Background Art

[0002] In modern industrial production, the header-type boiler systems of thermal power plants and heat and power plants are widely used. In such a system, the steam output ends of multiple boilers are connected to a common steam header to supply steam to steam-consuming equipment such as downstream steam turbines and heaters.

[0003] In the related art, the load distribution in the header-type boiler system mainly adopts the average distribution or empirical distribution mode. The average distribution mode evenly distributes the total load to each boiler, while the empirical distribution mode manually adjusts the load of each boiler according to the experience of the operating personnel.

[0004] However, the operating efficiencies of different boilers are different, and the health states of the equipment are also different. In addition, the pollutant emission characteristics of different boilers may also vary. Due to these factors not being fully considered, the existing distribution methods result in a relatively high overall coal consumption and increased equipment wear. These problems not only increase the operating cost of the header-type boiler system but also reduce the reliability and economy of the system. Summary of the Invention

[0005] Embodiments of this application provide a method and system for dynamically allocating the operating load of boilers. To provide a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary part is not a general review, nor is it intended to identify key / important constituent elements or delineate the protection scope of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the subsequent detailed description.

[0006] In a first aspect, embodiments of this application provide a method for dynamically allocating the operating load of boilers, the method including: Collecting and preprocessing the boiler operation data of each boiler in the header-type boiler system according to a preset standard communication protocol at a preset period to obtain the boiler operation data to be analyzed; Analyzing the dynamic characteristics of each boiler based on the boiler operation data to be analyzed as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler during operation over time; Based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm, dynamically programming the load of each boiler to obtain the optimal load distribution ratio of each boiler; Allocating the operating load of each boiler through the optimal load distribution ratio of each boiler.

[0007] Second aspect, an embodiment of the present application provides a dynamic load distribution system for a boiler operation, and the system includes: A data processing module, configured to collect and preprocess the boiler operation data of each boiler in a header pipe boiler system according to a preset standard communication protocol at a preset period, so as to obtain the boiler operation data to be analyzed; An analysis module, configured to analyze the dynamic characteristics of each boiler according to the boiler operation data to be analyzed, as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler changing with time during the operation process; A dynamic programming module, configured to perform dynamic programming on the load of each boiler based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm, so as to obtain the optimal load distribution ratio of each boiler; An adjustment module, configured to distribute the operation load of each boiler through the optimal load distribution ratio of each boiler.

[0008] The technical solution provided by the embodiment of the present application may include the following beneficial effects: In the embodiment of the present application, on the one hand, by accurately collecting and analyzing the operation data of each boiler, a dynamic characteristic model reflecting its actual operation state can be constructed. This model enables the system to deeply understand the performance change law of each boiler under different working conditions, fully consider the individual differences of each boiler, avoid the problems of low operation efficiency and increased equipment loss caused by blind distribution, and can reduce the operation cost of the header pipe boiler system. On the other hand, using the dynamic characteristic model and the multi-objective optimization algorithm for load distribution can comprehensively consider multi-dimensional factors such as coal consumption cost, equipment life loss, and pollutant emissions, realize the optimal load distribution among boilers, and the optimized load distribution can improve the operation stability and reliability of the system and reduce the operation risk caused by unreasonable load distribution.

[0009] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings here are incorporated into the description and form a part of this description, showing embodiments consistent with the present application, and are used together with the description to explain the principles of the present application.

[0011] Figure 1 is a schematic flowchart of a method for dynamically distributing the operation load of a boiler provided by an embodiment of the present application; Figure 2 is a schematic diagram of the distribution of the furnace temperature field provided by an embodiment of the present application; Figure 3 is a schematic diagram of the vibration spectrum of a steam drum provided by an embodiment of the present application; Figure 4 It is a schematic diagram of an efficiency curve provided by an embodiment of the present application; Figure 5 It is a schematic diagram of a scenario of dynamic load distribution for a boiler operation load provided by an embodiment of the present application; Figure 6 It is a schematic diagram of a UI interface of an administrator client provided by an embodiment of the present application; Figure 7 It is a schematic flowchart of a method for training a load demand prediction model provided by an embodiment of the present application; Figure 8 It is a schematic structural diagram of a dynamic load distribution system for a boiler operation provided by the present application; Figure 9 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0012] The following description and drawings fully illustrate specific implementation manners of the present application, enabling those skilled in the art to practice them.

[0013] It should be clear that the described embodiments are only a part of the embodiments of the present application, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts belong to the scope of protection of the present application.

[0014] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are only examples of systems and methods consistent with some aspects of the present application as detailed in the appended claims.

[0015] In the description of the present application, it should be understood that terms such as "first", "second", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances. In addition, in the description of the present application, unless otherwise stated, "a plurality of" means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0016] Currently, in the aspect of load distribution, the header pipe boiler system mainly adopts an average distribution or empirical distribution mode. The average distribution mode evenly distributes the total load to each boiler, while the empirical distribution mode manually adjusts the load of each boiler according to the experience of the operating personnel.

[0017] The inventors realized that the operating efficiencies of different boilers are different, and the equipment health states are also different. In addition, there may be differences in the pollutant emission characteristics of different boilers. Due to these factors not being fully considered, the existing allocation method results in a relatively high overall coal consumption and increased equipment wear. These problems not only increase the operating costs of the header-type boiler system but also reduce the reliability and economy of the system.

[0018] To solve the above problems, the present application provides a method and system for dynamically allocating the operating load of boilers to solve the problems existing in the above-related technical problems. In the embodiments of the present application, on the one hand, by accurately collecting and analyzing the operating data of each boiler, a dynamic characteristic model reflecting its actual operating state can be constructed. This model enables the system to deeply understand the performance change laws of each boiler under different working conditions, fully consider the individual differences of each boiler, avoid problems such as low operating efficiency and increased equipment wear caused by blind allocation, and can reduce the operating costs of the header-type boiler system. On the other hand, using the dynamic characteristic model and multi-objective optimization algorithm for load allocation can comprehensively consider multi-dimensional factors such as coal consumption cost, equipment life loss, and pollutant emission, achieve the optimal load allocation among boilers, and the optimized load allocation can improve the operating stability and reliability of the system and reduce the operating risks caused by unreasonable load allocation. The following uses exemplary embodiments for detailed description.

[0019] The following will combine with the attached Figure 1 - attached Figure 7 drawings to introduce in detail the method for dynamically allocating the operating load of boilers provided by the embodiments of the present application. This method can be implemented depending on a computer program and can run on a boiler operating load dynamic allocation system based on the von Neumann architecture. This computer program can be integrated in an application or run as an independent tool-type application.

[0020] Please refer to Figure 1 FIG. Figure 1 shown, the method of the embodiments of the present application may include the following steps: S101, collect and preprocess the boiler operating data of each boiler in the header-type boiler system according to a preset standard communication protocol at a preset period to obtain the boiler operating data to be analyzed; Among them, the preset standard communication protocol refers to a pre-set communication protocol that conforms to industrial standards. This communication protocol can include the OPC (OLE for Process Control) protocol, the Modbus protocol, and the Profibus protocol. By adopting the standard communication protocol, it can be ensured that the data acquisition module in the boiler system can stably and reliably obtain the operation data from the control system of each boiler. The preset period refers to the pre-set time interval for data acquisition. This time interval can be adjusted according to actual needs, such as collecting data every minute, every 5 minutes, or every 10 minutes. Regularly collecting data can ensure that the system can monitor the operation status of the boiler in real time, detect abnormal situations in a timely manner, and make adjustments. The header pipe boiler system refers to a steam output end of multiple boilers connected to a common steam header pipe to form a unified steam supply system. The boiler operation data of each boiler includes but is not limited to the following key parameters: steam pressure, steam flow, flue gas temperature, metal wall temperature, coal consumption, and pollutant emissions. Pretreatment refers to the preliminary processing of the collected raw data to ensure the quality and usability of the data. The operations of pretreatment can include data cleaning, data standardization, and data noise reduction. Part of the boiler operation data of each boiler is shown in Table 1.

[0021] Table 1

[0022] In some embodiments, a header pipe boiler system of a thermal power plant includes 3 boilers and adopts OPC server software (such as OPC UA) for communicating with the boiler control system. The server software collects the operation data of each boiler every 5 minutes through the OPC protocol, checks for outliers in the data (such as data with a flue gas temperature exceeding 170°C) and marks them as invalid, removes duplicate data, normalizes all data to the range of 0 - 1, uses the moving average method to smooth the coal consumption and pollutant emissions, and stores the preprocessed data in a CSV file to obtain the boiler operation data to be analyzed.

[0023] Specifically, during normalization, assuming the data range is , the normalized value is:

[0024] For example, the range of steam pressure is 1.1 - 1.35 MPa, and the normalization expression is:

[0025] S102. Analyze the dynamic characteristics of each boiler based on the operating data of the boiler to be analyzed, and use them as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler over time during the operation process. Among them, the dynamic characteristic analysis is to extract the characteristics from the operating data to be analyzed that can characterize the change of the boiler operating performance over time, and to reflect the change law of the performance parameters of the boiler over time during the operation process.

[0026] In some embodiments of the present application, the specific process of analyzing the dynamic characteristics of each boiler based on the operating data of the boiler to be analyzed includes: classifying the operating data of the boiler to be analyzed to extract the furnace temperature field distribution, drum vibration spectrum, and the time-series change of flue gas composition of each boiler; aligning the time scales of the furnace temperature field distribution, drum vibration spectrum, and the time-series change of flue gas composition to obtain a target data matrix with synchronized time scales; analyzing the dynamic derivative of the combustion efficiency and the characteristics of the equipment health status at each moment within a preset period for each boiler according to the target data matrix; determining the efficiency curve, pollutant emission characteristics, and equipment health evaluation index of each boiler based on the dynamic derivative of the combustion efficiency and the characteristics of the equipment health status; using the efficiency curve, pollutant emission characteristics, and equipment health evaluation index of each boiler as the dynamic characteristics of each boiler. Among them, time-scale alignment is to align the time of different types of data to ensure their synchronization in time.

[0027] Among them, the furnace temperature field distribution is used to reflect the spatial distribution of the temperature in the furnace. The drum vibration spectrum is used to reflect the frequency and amplitude of the drum vibration. The time-series change of flue gas composition is used to record the change of flue gas composition over time. Time-scale alignment is to align the timestamps of different data through technologies such as interpolation and timestamp matching. The target data matrix is the data after time-scale alignment, forming a unified matrix, where each row represents a time point and each column represents a characteristic. The dynamic derivative of the combustion efficiency is the rate of change of the combustion efficiency over time, which is used to reflect the dynamic change trend of the combustion efficiency. A positive value indicates an increase in efficiency, and a negative value indicates a decrease in efficiency. The characteristics of the equipment health status are used to reflect the characteristics of the equipment health status, such as vibration frequency, amplitude, etc. The efficiency curve is used to describe the relationship between the boiler efficiency and the load change. The pollutant emission characteristics are used to describe the pollutant emission situation of the boiler under different loads. The equipment health evaluation index is an index used to evaluate the equipment health status, such as vibration amplitude weight, temperature gradient, wear index, etc. Among them, the furnace temperature field distribution, for example Figure 2 is shown, where different furnace temperatures have different colors in space. The drum vibration spectrum, for example Figure 3 is shown.

[0028] For example, after obtaining the operating data of the boiler to be analyzed, extract the parameters related to the furnace temperature from the data, such as the metal wall temperature. Extract the flue gas temperature, pollutant emission amount, and drum vibration spectrum from the data.

[0029] It should be noted that the drum vibration spectrum is obtained by measuring the vibration of the drum using an acceleration sensor (accelerometer). The sensor is installed at key positions of the drum, such as the support points, connection points of the drum, or other parts prone to vibration. The sensor is fixed on the surface of the drum using a special installation bracket or magnetic base to ensure close contact between the sensor and the drum.

[0030] In the embodiments of the present application, by analyzing the boiler operation data, key dynamic features are extracted and a dynamic characteristic model is constructed. The efficiency curve in the dynamic characteristic model can accurately describe the efficiency performance of the boiler under different loads. The equipment health evaluation index in the dynamic characteristic model can quantify the health state of the equipment. The pollutant emission characteristics in the dynamic characteristic model can accurately describe the emission performance of the boiler under different loads. The above information provides a basis for the analysis of the optimal load distribution ratio, thereby enabling the optimization of combustion efficiency, the enhancement of equipment reliability, the reduction of pollutant emissions, and the refinement of system management, thus significantly improving the overall performance and adaptability of the boiler system and reducing the operation cost.

[0031] In some embodiments of the present application, according to the target data matrix, the specific process of analyzing the combustion efficiency dynamic derivative and the equipment health state characteristics of each boiler at each moment within a preset period includes: using the furnace temperature field distribution and the sequential change of flue gas composition in the target data matrix to calculate the combustion efficiency of each boiler at each moment; performing numerical differentiation processing on the combustion efficiency of each boiler at each moment to determine the combustion efficiency dynamic derivative of each boiler at each moment within the preset period; performing fast Fourier transform processing on the drum vibration spectrum in the target data matrix to identify the characteristic frequency and amplitude of each boiler as the equipment health state characteristics of each boiler at each moment within the preset period.

[0032] Among them, the target data matrix is a matrix integrating various data, including the furnace temperature field distribution, the drum vibration spectrum, the sequential change of flue gas composition, etc. These data have been time - scale aligned to ensure time synchronization. Numerical differentiation processing is a mathematical method used to calculate the rate of change of a function. The rate of change of combustion efficiency over time (i.e., the combustion efficiency dynamic derivative) can be obtained by performing numerical differentiation on the combustion efficiency data. Fast Fourier transform (FFT) is a mathematical algorithm used to convert a time - domain signal into a frequency - domain signal. By performing FFT processing on the drum vibration spectrum, the characteristic frequency and amplitude in the signal can be identified.

[0033] In a possible implementation, the furnace temperature and flue gas composition at each time point are extracted from the target data matrix. Based on the pre-established mapping relationship between the flue gas composition and the flue gas temperature under different types of flue gas compositions, the actual flue gas temperature corresponding to the flue gas composition at different times is queried. The calculation formula for the combustion efficiency of each boiler at each moment is: ; When calculating the dynamic derivative of the combustion efficiency through numerical differentiation, first calculate the difference in combustion efficiency between adjacent time points, and then divide it by the time interval to obtain the rate of change of the combustion efficiency. The expression is: Where, is the combustion efficiency with respect to time is the rate of change, that is, the dynamic derivative, represents the combustion efficiency at time , is the current time point, is the time interval, is the current time is the combustion efficiency at time

[0034] For example, for the first time point (2025-04-15 00:00) and the second time point (2025-04-15 00:05), the rate of change of the combustion efficiency is:

[0035] For example Figure 3 as shown, assuming that Figure 3 the drum vibration spectrum data in is: ; Perform FFT processing: ; Identify the characteristic frequencies and amplitudes in the FFT results: ; .

[0036] Among them, the dynamic derivative of the combustion efficiency is used to characterize the rate of change of the combustion efficiency of each boiler with respect to time.

[0037] In some embodiments of the present application, the specific process of determining the efficiency curve, pollutant emission characteristics, and equipment health evaluation index of each boiler based on the dynamic derivative of combustion efficiency and equipment health status characteristics includes: within a preset time window, numerically integrating the change rate of the combustion efficiency of each boiler over time to obtain the efficiency time series of each boiler; aligning the efficiency time series of each boiler with the historical operating load series within a preset period to obtain the first aligned data; constructing the efficiency curve of each boiler based on the first aligned data; aligning the flue gas composition data within a preset period with the historical operating load series to obtain the second aligned data; establishing an emission-load relationship model based on the second aligned data as the pollutant emission characteristics of each boiler; determining the vibration amplitude weight, temperature gradient, and wear index of each boiler based on the equipment health status characteristics and the historical maintenance records of each boiler as the equipment health evaluation index of each boiler. Among them, the expression for numerically integrating the change rate of the combustion efficiency of each boiler over time is:

[0038] Among them, is the real-time combustion efficiency of the boiler at time , is the initial efficiency reference value of the boiler , is the dynamic derivative of combustion efficiency, a positive value indicates an efficiency increase, and a negative value indicates an efficiency decrease, is the length of the integration time window.

[0039] Among them, the pre-established emission-load relationship model is: ; Among them, is the pollutant emission concentration of the th boiler at load , and are the emission-load linear relationship parameters in the low load section , is the low load emission coefficient, is the low load base emission, is the emission-load linear relationship parameter in the high load section , is the inflection point load of the emission characteristics, and the inflection point load of the emission characteristics is the load at the position point where the emission slope changes due to the sudden change of the combustion state under different loads.

[0040] In some embodiments of the present application, the specific process of constructing the efficiency curve of each boiler according to the first alignment data includes: using a sliding window with a preset window length, statistically calculating the average efficiency in different load intervals of the historical operating load from the first alignment data; using the average efficiency in different load intervals of the historical operating load to fit a second-order polynomial fitting function; the second-order polynomial fitting function is used to characterize the quantitative relationship between the efficiency of each boiler and the load change; solving the polynomial coefficients of the second-order polynomial fitting function by the least squares method to obtain the objective function; simulating the change curve of the objective function over time to obtain the efficiency-load curve of each boiler as the efficiency curve. The efficiency curve is, for example Figure 4 as shown

[0041] Among them, the second-order polynomial fitting function is:

[0042] Among them, is the efficiency of the th boiler changing with the load quantitative relationship, , , are polynomial coefficients, and the polynomial coefficients are obtained by least squares fitting, characterizes the curvature of the efficiency change with the load, reflects the degree of linear influence of the load on the efficiency, represents the theoretical efficiency intercept at low load.

[0043] Among them, the preset window length = 10 minutes.

[0044] S103, based on the dynamic characteristic model of each boiler and the preset multi-objective optimization algorithm, perform dynamic programming on the load of each boiler to obtain the optimal load distribution ratio of each boiler; Among them, the multi-objective optimization algorithm is an optimization method used to optimize the objective function simultaneously. The multi-objective optimization algorithm includes non-dominated sorting genetic algorithm (NSGA-II), multi-objective particle swarm optimization (MOPSO), etc. Dynamic programming is an optimization method used to find the optimal solution in a multi-stage decision-making process. The optimal load distribution ratio refers to the load distribution ratio of each boiler that makes the objective function optimal under all constraint conditions.

[0045] Among them, the dynamic characteristic model of each boiler includes an efficiency curve, pollutant emission characteristics, and equipment health evaluation indicators.

[0046] In some embodiments of the present application, based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm, the specific process of dynamically programming the load of each boiler to obtain the optimal load distribution ratio of each boiler includes: using the load prediction data of the downstream steam-using equipment within a future preset time period as the current total load demand; creating an objective function, which is used to maximize efficiency, equipment health, and minimize pollutant emissions; obtaining the constraint conditions of the objective function, which are used to constrain the total load balance, the load range of a single boiler, and pollutant emissions; randomly generating a preset number of groups of load distribution ratios according to the number of boilers in the header-tank boiler system and the current total load demand; querying the target efficiency and target emission information corresponding to each group of load distribution ratios from the efficiency curve and pollutant emission characteristics of each boiler; using the target efficiency, target emission information, equipment health evaluation index, and constraint conditions as the parameters of the preset multi-objective optimization algorithm, and executing the preset multi-objective optimization algorithm to obtain the function value of each group of load distribution ratios; the preset multi-objective optimization algorithm is a non-dominated sorting genetic algorithm; using the load distribution ratio with the largest function value as the optimal load combination.

[0047] Among them, the objective function designed in the present application is: ; Among them, is the load distribution ratio allocation function, is the efficiency weight coefficient, is the emission weight coefficient, is the health degree weight coefficient, is the th boiler under the group of loads is the th boiler under the group of loads is the th boiler's equipment health evaluation index.

[0048] S104, allocate the operating load of each boiler according to the optimal load distribution ratio of each boiler.

[0049] In some embodiments of the present application, the specific process of allocating the operating load of each boiler through the optimal load allocation ratio of each boiler includes: obtaining the historical load demand of downstream steam-using equipment; based on the historical load demand, predicting the load prediction data of the downstream steam-using equipment within a preset future period through a pre-trained load demand prediction model; according to the load prediction data, fine-tuning the optimal load allocation ratio of each boiler at a preset step length to obtain the final load allocation ratio of each boiler; and allocating the operating load of each boiler using the final load allocation ratio of each boiler.

[0050] Among them, the dynamic characteristics of each boiler include an efficiency curve, pollutant emission characteristics, and equipment health evaluation indicators.

[0051] In some embodiments of the present application, it also includes fine-tuning the parameters of the model. The specific process is as follows: extracting the efficiency change rate, pollutant emission trend information, and equipment health status indicators related to load demand prediction from the efficiency curve, pollutant emission characteristics, and equipment health evaluation indicators to obtain fine-tuning feature information; extracting historical model parameters from the pre-trained load demand prediction model; using the fine-tuning feature information and historical model parameters to perform model fine-tuning on the pre-trained load demand prediction model to obtain fine-tuned model parameters; and deploying the fine-tuned model parameters to the pre-trained load demand prediction model.

[0052] In some embodiments of the present application, the specific process of generating a pre-trained load demand prediction model includes: collecting the historical operation data of the header-type boiler system and the historical load demand data of downstream steam-using equipment; obtaining data features related to load demand from the historical operation data and historical load demand data; creating a load demand prediction model using a long short-term memory network; inputting the data features related to load demand into the prediction model and outputting the loss value of the model; and obtaining the pre-trained load demand prediction model when the loss value reaches the minimum.

[0053] Among them, the optimal load allocation ratio and the original load allocation ratio of each boiler are shown in Table 2, for example.

[0054] Table 2

[0055] Among them, after obtaining the optimal load allocation ratio of each boiler, the boiler control system can be controlled to replace the original load allocation ratio with the optimal load allocation ratio of each boiler to operate according to the optimal load allocation ratio.

[0056] For example Figure 5 as shown Figure 5It is a schematic diagram of a scenario for the dynamic distribution of the boiler operation load provided by this application, including an OPC server software, a boiler control system, and a boiler controller. The OPC server software collects and preprocesses the boiler operation data of each boiler in the header pipe boiler system from the boiler control system according to a preset standard communication protocol at a preset cycle to obtain the boiler operation data to be analyzed. The OPC server software analyzes the dynamic characteristics of each boiler based on the boiler operation data to be analyzed as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler over time during the operation process; the OPC server software performs dynamic programming on the load of each boiler based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm to obtain the optimal load distribution ratio of each boiler; the OPC server software distributes the operation load of each boiler through the optimal load distribution ratio of each boiler, controls the boiler control system to replace the original load distribution ratio with the optimal load distribution ratio of each boiler, so that the boiler controller executes according to the optimal load distribution ratio. After executing according to the optimal load distribution ratio, the interface-related data of the administrator client has been updated with parameters, such as Figure 6 as shown

[0057] In the embodiment of this application, on the one hand, by accurately collecting and analyzing the operation data of each boiler, a dynamic characteristic model reflecting its actual operation state can be constructed. This model enables the system to deeply understand the performance change rules of each boiler under different working conditions, fully consider the individual differences of each boiler, avoid problems such as low operation efficiency and increased equipment loss caused by blind distribution, and can reduce the operation cost of the header pipe boiler system. On the other hand, using the dynamic characteristic model and multi-objective optimization algorithm for load distribution can comprehensively consider multi-dimensional factors such as coal consumption cost, equipment life loss, and pollutant emissions, achieve the optimal load distribution among boilers, and the optimized load distribution can improve the operation stability and reliability of the system and reduce the operation risk caused by unreasonable load distribution.

[0058] Please refer to Figure 7 , which is a schematic flow diagram of a method for training a load demand prediction model provided by an embodiment of this application. As Figure 7 shown, the method of the embodiment of this application may include the following steps: S201, collect the historical operation data of the header pipe boiler system and the historical load demand data of the downstream steam-using equipment; In some embodiments, collect the historical operation data of the header pipe boiler system and the historical load demand data of the downstream steam-using equipment. These data may include the operation parameters of the boiler (such as temperature, pressure, fuel consumption, etc.) and the load demand of the downstream equipment (such as steam flow, temperature, pressure, etc.).

[0059] S202, Obtain data features related to load demand from historical operation data and historical load demand data; In some embodiments, data features related to load demand are obtained from historical operation data and historical load demand data. These features include time series data (such as hours, days, months) and external factors affecting the load (such as weather conditions, weekdays and non - weekdays, etc.).

[0060] S203, Create a load demand prediction model using a long short - term memory network; In some embodiments, a long short - term memory network (LSTM) is used to create a load demand prediction model. LSTM is a special type of recurrent neural network (RNN) that can learn long - term dependency information.

[0061] S204, Input the data features related to load demand into the prediction model and output the loss value of the model; In some embodiments, the data features related to load demand are input into the prediction model. The model predicts future load demand by learning the patterns and relationships in the data. During the training process, the loss value of the model (representing the difference between the predicted value and the actual value) is calculated.

[0062] S205, When the loss value reaches the minimum, obtain the pre - trained load demand prediction model.

[0063] In some embodiments, during the training process, the loss value is minimized by adjusting the model parameters and optimization algorithms. When the loss value reaches the minimum, it means the highest prediction accuracy of the model, and at this time, the pre - trained load demand prediction model can be obtained.

[0064] In the embodiments of the present application, on the one hand, by accurately collecting and analyzing the operation data of each boiler, a dynamic characteristic model reflecting its actual operation state can be constructed. This model enables the system to deeply understand the performance change rules of each boiler under different working conditions, fully consider the individual differences of each boiler, avoid problems such as low operation efficiency and increased equipment wear caused by blind allocation, and can reduce the operation cost of the header - type boiler system. On the other hand, using the dynamic characteristic model and multi - objective optimization algorithm for load distribution can comprehensively consider multi - dimensional factors such as coal consumption cost, equipment life loss, and pollutant emissions, achieve the optimal load distribution among boilers, and the optimized load distribution can improve the operation stability and reliability of the system and reduce the operation risks caused by unreasonable load distribution.

[0065] The following is the system embodiment of the present application, which can be used to execute the method embodiment of the present application. For the details not disclosed in the system embodiment of the present application, please refer to the method embodiment of the present application.

[0066] Please refer to Figure 8 , which shows a schematic structural diagram of a boiler operation load dynamic distribution system provided by an exemplary embodiment of the present application. The boiler operation load dynamic distribution system can be implemented as all or part of an electronic device through software, hardware, or a combination of both. The system 1 includes a data processing module 10, an analysis module 20, a dynamic programming module 30, and an adjustment module 40.

[0067] The data processing module 10 is configured to collect and preprocess the boiler operation data of each boiler in the header pipe boiler system according to a preset standard communication protocol at a preset period, so as to obtain the boiler operation data to be analyzed; The analysis module 20 is configured to analyze the dynamic characteristics of each boiler according to the boiler operation data to be analyzed, as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler during operation over time; The dynamic programming module 30 is configured to perform dynamic programming on the load of each boiler based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm, so as to obtain the optimal load distribution ratio of each boiler; The adjustment module 40 is configured to distribute the operation load of each boiler through the optimal load distribution ratio of each boiler.

[0068] It should be noted that when the boiler operation load dynamic distribution system provided in the above embodiment executes the boiler operation load dynamic distribution method, only the above-mentioned division of each functional module is used for illustration. In practical applications, the above functions can be allocated by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the boiler operation load dynamic distribution system provided in the above embodiment and the embodiment of the boiler operation load dynamic distribution method belong to the same concept, and the implementation process is shown in detail in the method embodiment, which will not be elaborated here.

[0069] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0070] In the embodiments of the present application, on the one hand, by accurately collecting and analyzing the operation data of each boiler, a dynamic characteristic model reflecting its actual operation state can be constructed. This model enables the system to deeply understand the performance change rules of each boiler under different working conditions, fully consider the individual differences of each boiler, avoid problems such as low operation efficiency and increased equipment wear caused by blind allocation, and reduce the operation cost of the header-type boiler system. On the other hand, using the dynamic characteristic model and multi-objective optimization algorithm for load distribution can comprehensively consider multi-dimensional factors such as coal consumption cost, equipment life loss, and pollutant emissions, realize the optimal load distribution among boilers, and the optimized load distribution can improve the operation stability and reliability of the system and reduce the operation risks caused by unreasonable load distribution.

[0071] The present application also provides a computer-readable medium, on which program instructions are stored. When the program instructions are executed by a processor, the dynamic load distribution method for boiler operation provided in each of the above method embodiments is implemented.

[0072] The present application also provides a computer program product containing instructions. When it runs on a computer, it causes the computer to execute the dynamic load distribution method for boiler operation in each of the above method embodiments.

[0073] Please refer to Figure 9 , which is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 9 shown, the electronic device 1000 may include: at least one processor 1001, at least one network interface 1004, a user interface 1003, a memory 1005, and at least one communication bus 1002.

[0074] Among them, the communication bus 1002 is used to realize the connection and communication between these components.

[0075] Among them, the user interface 1003 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 1003 may further include a standard wired interface and a wireless interface.

[0076] Among them, the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0077] Among them, the processor 1001 may include one or more processing cores. The processor 1001 connects various parts within the entire electronic device 1000 through various interfaces and circuits. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, and by invoking the data stored in the memory 1005, it executes various functions of the electronic device 1000 and processes data. Optionally, the processor 1001 may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), or programmable logic array (PLA). The processor 1001 may integrate one or a combination of several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor 1001 and may be implemented separately by a single chip.

[0078] Among them, the memory 1005 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 1005 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store the data involved in the above-mentioned various method embodiments. Optionally, the memory 1005 may also be at least one storage system located far from the aforementioned processor 1001. As Figure 9 shown, the memory 1005, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a boiler operation load dynamic allocation application program.

[0079] In Figure 9In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an interface for the user to input and obtain the data input by the user; and the processor 1001 can be used to call the boiler operation load dynamic allocation application program stored in the memory 1005 and specifically perform the following operations: Collect and preprocess the boiler operation data of each boiler in the header pipe boiler system according to a preset standard communication protocol at a preset period to obtain the boiler operation data to be analyzed; According to the boiler operation data to be analyzed, analyze the dynamic characteristics of each boiler as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler over time during the operation process; Based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm, perform dynamic programming on the load of each boiler to obtain the optimal load allocation ratio of each boiler; Allocate the operation load of each boiler through the optimal load allocation ratio of each boiler.

[0080] In one embodiment, when the processor 1001 executes the operation of allocating the operation load of each boiler through the optimal load allocation ratio of each boiler, it specifically performs the following operations: Obtain the historical load demand of the downstream steam-using equipment; Based on the historical load demand, predict the load prediction data of the downstream steam-using equipment within a preset future period through a pre-trained load demand prediction model; According to the load prediction data, fine-tune the optimal load allocation ratio of each boiler at a preset step size to obtain the final load allocation ratio of each boiler; Allocate the operation load of each boiler using the final load allocation ratio of each boiler.

[0081] In one embodiment, when the processor 1001 executes the operation of analyzing the dynamic characteristics of each boiler according to the boiler operation data to be analyzed, it specifically performs the following operations: Perform data classification on the boiler operation data to be analyzed to extract the furnace temperature field distribution, steam drum vibration spectrum, and flue gas composition time series change of each boiler; Align the time scales of the furnace temperature field distribution, steam drum vibration spectrum, and flue gas composition time series change to obtain a time-scale synchronized target data matrix; According to the target data matrix, analyze the combustion efficiency dynamic derivative and equipment health status characteristics of each boiler at each moment within a preset period; Based on the combustion efficiency dynamic derivative and equipment health status characteristics, determine the efficiency curve, pollutant emission characteristics, and equipment health evaluation index of each boiler; Take the efficiency curve, pollutant emission characteristics, and equipment health evaluation indicators of each boiler as the dynamic characteristics of each boiler.

[0082] In one embodiment, when the processor 1001 executes the analysis of the dynamic derivative of the combustion efficiency and the characteristics of the equipment health state of each boiler at each moment within a preset period according to the target data matrix, the following operations are specifically performed: Utilize the furnace temperature field distribution and the time-series change of flue gas composition in the target data matrix to calculate the combustion efficiency of each boiler at each moment; Perform numerical differentiation on the combustion efficiency of each boiler at each moment to determine the dynamic derivative of the combustion efficiency of each boiler at each moment within the preset period; Perform fast Fourier transform processing on the drum vibration spectrum in the target data matrix to identify the characteristic frequencies and amplitudes of each boiler, which are used as the characteristics of the equipment health state of each boiler at each moment within the preset period.

[0083] In one embodiment, when the processor 1001 executes the determination of the efficiency curve, pollutant emission characteristics, and equipment health evaluation indicators of each boiler based on the dynamic derivative of the combustion efficiency and the characteristics of the equipment health state, the following operations are specifically performed: Within a preset time window, perform numerical integration on the rate of change of the combustion efficiency of each boiler over time to obtain the efficiency time series of each boiler; Align the efficiency time series of each boiler with the historical operation load series within the preset period to obtain the first aligned data; Construct the efficiency curve of each boiler according to the first aligned data; Align the flue gas composition data within the preset period with the historical operation load series to obtain the second aligned data; Based on the second aligned data, establish an emission-load relationship model as the pollutant emission characteristics of each boiler; Determine the vibration amplitude weight, temperature gradient, and wear index of each boiler according to the characteristics of the equipment health state and the historical maintenance records of each boiler, which are used as the equipment health evaluation indicators of each boiler.

[0084] In one embodiment, when the processor 1001 executes the construction of the efficiency curve of each boiler according to the first aligned data, the following operations are specifically performed: Adopt a sliding window with a preset window length, and from the first aligned data, statistically calculate the average efficiency in different load intervals of the historical operation load; Use the average efficiency in different load intervals of the historical operation load to fit a second-order polynomial fitting function; the second-order polynomial fitting function is used to characterize the quantitative relationship between the efficiency of each boiler and the load change; Solve for the polynomial coefficients of the second-order polynomial fitting function by the least squares method to obtain the objective function; Simulate the change curve graph of the objective function over time to obtain the efficiency-load curve of each boiler as the efficiency curve.

[0085] In one embodiment, when the processor 1001 executes dynamic programming on the load of each boiler based on the dynamic characteristic model of each boiler and the preset multi-objective optimization algorithm to obtain the optimal load distribution ratio of each boiler, the following operations are specifically executed: Use the load prediction data of the downstream steam-using equipment within a future preset time period as the current total load demand; Create an objective function, which is used to maximize efficiency and equipment health and minimize pollutant emissions; Obtain the constraint conditions of the objective function, which are used to constrain the total load balance, the load range of a single boiler, and pollutant emissions; According to the number of boilers in the header boiler system and the current total load demand, randomly generate multiple groups of load distribution ratios with a preset quantity; Query the target efficiency and target emission information corresponding to each group of load distribution ratios from the efficiency curve and pollutant emission characteristics of each boiler; Use the target efficiency, target emission information, equipment health evaluation index, and constraint conditions as the parameters of the preset multi-objective optimization algorithm, and execute the preset multi-objective optimization algorithm to obtain the function value of each group of load distribution ratios; the preset multi-objective optimization algorithm is the non-dominated sorting genetic algorithm; Use the load distribution ratio with the largest function value as the optimal load combination.

[0086] In one embodiment, the processor 1001 also executes the following operations: Extract the efficiency change rate, pollutant emission trend information, and equipment health status index related to the load demand prediction from the efficiency curve, pollutant emission characteristics, and equipment health evaluation index to obtain the fine-tuning feature information; Extract the historical model parameters from the pre-trained load demand prediction model; Use the fine-tuning feature information and historical model parameters to fine-tune the pre-trained load demand prediction model to obtain the fine-tuned model parameters; Deploy the fine-tuned model parameters to the pre-trained load demand prediction model.

[0087] In one embodiment, the processor 1001 also executes the following operations: Collect the historical operation data of the header boiler system and the historical load demand data of the downstream steam-using equipment; Obtain data features related to load demand from historical operation data and historical load demand data; Create a load demand prediction model using a long short-term memory network; Input the data features related to load demand into the prediction model and output the loss value of the model; Obtain a pre-trained load demand prediction model when the loss value reaches the minimum.

[0088] In the embodiment of the present application, on the one hand, by accurately collecting and analyzing the operation data of each boiler, a dynamic characteristic model reflecting its actual operation state can be constructed. This model enables the system to deeply understand the performance change rules of each boiler under different working conditions, fully consider the individual differences of each boiler, avoid problems such as low operation efficiency and increased equipment wear caused by blind allocation, and reduce the operation cost of the header-type boiler system. On the other hand, using the dynamic characteristic model and multi-objective optimization algorithm for load distribution can comprehensively consider multi-dimensional factors such as coal consumption cost, equipment life loss, and pollutant emissions, achieve the optimal load distribution among boilers, and the optimized load distribution can improve the operation stability and reliability of the system and reduce the operation risk caused by unreasonable load distribution.

[0089] Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing relevant hardware through a computer program. The program for dynamic load distribution of boiler operation can be stored in a computer-readable storage medium. When this program is executed, it can include the processes of the above method embodiments. Among them, the storage medium of the program for dynamic load distribution of boiler operation can be a magnetic disk, an optical disk, a read-only memory, or a random access memory, etc.

[0090] The above-disclosed are only the preferred embodiments of the present application. Of course, the scope of the rights of the present application cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A dynamic load distribution method for boiler operation, characterized in that, The method includes: Collecting and preprocessing the boiler operation data of each boiler in the header - type boiler system according to a preset standard communication protocol at a preset period to obtain the boiler operation data to be analyzed; Analyzing the dynamic characteristics of each boiler based on the boiler operation data to be analyzed as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler over time during operation; Based on the dynamic characteristic model of each boiler and a preset multi - objective optimization algorithm, dynamically programming the load of each boiler to obtain the optimal load distribution ratio of each boiler; Allocating the operation load of each boiler through the optimal load distribution ratio of each boiler.

2. The method according to claim 1, wherein The allocating the operation load of each boiler through the optimal load distribution ratio of each boiler includes: Obtaining the historical load demand of the downstream steam - using equipment; Based on the historical load demand, predicting the load prediction data of the downstream steam - using equipment within a preset future period through a pre - trained load demand prediction model; According to the load prediction data, finely adjusting the optimal load distribution ratio of each boiler at a preset step size to obtain the final load distribution ratio of each boiler; Using the final load distribution ratio of each boiler to allocate the operation load of each boiler.

3. The method according to claim 1, wherein The analyzing the dynamic characteristics of each boiler based on the boiler operation data to be analyzed includes: Classifying the boiler operation data to be analyzed to extract the furnace temperature field distribution, steam drum vibration spectrum, and the time - series change of flue gas composition of each boiler; Aligning the time scales of the furnace temperature field distribution, steam drum vibration spectrum, and the time - series change of flue gas composition to obtain a time - scale synchronized target data matrix; Based on the target data matrix, analyzing the dynamic derivative of the combustion efficiency and the equipment health status characteristics of each boiler at each moment within a preset period; Based on the dynamic derivative of the combustion efficiency and the equipment health status characteristics, determining the efficiency curve, pollutant emission characteristics, and equipment health evaluation index of each boiler; Taking the efficiency curve, pollutant emission characteristics, and equipment health evaluation index of each boiler as the dynamic characteristics of each boiler.

4. The method according to claim 3, wherein The analyzing the dynamic derivative of the combustion efficiency and the equipment health status characteristics of each boiler at each moment within a preset period based on the target data matrix includes: Using the furnace temperature field distribution and the time - series change of flue gas composition in the target data matrix to calculate the combustion efficiency of each boiler at each moment; Performing numerical differential processing on the combustion efficiency of each boiler at each moment to determine the dynamic derivative of the combustion efficiency of each boiler at each moment within a preset period; Performing fast Fourier transform processing on the steam drum vibration spectrum in the target data matrix to identify the characteristic frequency and amplitude of each boiler as the equipment health status characteristics of each boiler at each moment within a preset period.

5. The method according to claim 3, characterized in that, The dynamic derivative of the combustion efficiency is used to characterize the change rate of the combustion efficiency of each boiler over time; Based on the dynamic derivative of the combustion efficiency and the characteristics of the equipment health status, determine the efficiency curve, pollutant emission characteristics, and equipment health evaluation index of each boiler, including: Within a preset time window, perform numerical integration on the rate of change of the combustion efficiency of each boiler over time to obtain the efficiency time series of each boiler; Align the efficiency time series of each boiler with the historical operation load series within a preset period to obtain the first aligned data; Construct the efficiency curve of each boiler according to the first aligned data; Align the flue gas composition data within a preset period with the historical operation load series to obtain the second aligned data; Based on the second aligned data, establish an emission-load relationship model as the pollutant emission characteristics of each boiler; According to the characteristics of the equipment health status and the historical maintenance records of each boiler, determine the vibration amplitude weight, temperature gradient, and wear index of each boiler as the equipment health evaluation index of each boiler; where The expression for performing numerical integration on the rate of change of the combustion efficiency of each boiler over time is: Among them, is the real-time combustion efficiency of the boiler at time , is the initial efficiency reference value of the boiler , is the dynamic derivative of the combustion efficiency, a positive value indicates an efficiency increase, and a negative value indicates an efficiency decrease, is the integral time window length; among them, The pre-established emission-load relationship model is: ； Among them, is the pollutant emission concentration of the th boiler at the load . and are the emission-load linear relationship parameters in the low-load section . is the low-load emission coefficient, is the low-load base emission, are the emission-load linear relationship parameters in the high-load section . is the load at the inflection point of the emission characteristics. The load at the inflection point of the emission characteristics is the load at the position point where the emission slope changes due to the sudden change of the combustion state under different loads.

6. The method according to claim 5, wherein The constructing the efficiency curve of each boiler according to the first aligned data includes: Using a sliding window with a preset window length, statistically calculate the average efficiency in different load intervals of the historical operation load from the first aligned data; Using the average efficiency in different load intervals of the historical operation load, fit a second-order polynomial fitting function; the second-order polynomial fitting function is used to characterize the quantitative relationship between the efficiency of each boiler and the load change; Solve the polynomial coefficients of the second-order polynomial fitting function by the least squares method to obtain the objective function; Simulate the change curve graph of the objective function over time to obtain the efficiency-load curve of each boiler as the efficiency curve; where the second-order polynomial fitting function is: Among them, is the efficiency of the nth boiler varying with the load , , are polynomial coefficients obtained by least squares fitting, characterizes the curvature of the efficiency varying with the load, reflects the degree of linear influence of the load on the efficiency, represents the theoretical efficiency intercept at low load.

7. The method according to claim 2, characterized in that The dynamic characteristic model of each boiler includes an efficiency curve, pollutant emission characteristics, and an equipment health evaluation index; Based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm, perform dynamic programming on the load of each boiler to obtain the optimal load distribution ratio of each boiler, including: Use the load prediction data of the downstream steam-consuming equipment within a future preset time period as the current total load demand; Create an objective function, which is used to maximize efficiency and equipment health and minimize pollutant emissions; Obtain the constraint conditions of the objective function, which are used to constrain the total load balance, the load range of a single boiler, and pollutant emissions; According to the number of boilers in the header boiler system and the current total load demand, randomly generate a preset number of groups of load distribution ratios; Query the target efficiency and target emission information corresponding to each group of load distribution ratios from the efficiency curve and pollutant emission characteristics of each boiler; Taking the target efficiency, target emission information, the equipment health evaluation index, and the constraint conditions as the parameters of a preset multi-objective optimization algorithm, execute the preset multi-objective optimization algorithm to obtain the function values of each set of load distribution ratios; the preset multi-objective optimization algorithm is a non-dominated sorting genetic algorithm; Taking the load distribution ratio with the largest function value as the optimal load combination; wherein, The objective function is: ； Among them, is the load distribution ratio allocation function, is the efficiency weight coefficient, is the emission weight coefficient, is the health degree weight coefficient, is the th boiler at the th group of load under the efficiency, is the th boiler at the th group of load under the pollutant emission amount, is the th boiler's equipment health degree evaluation index.

8. The method according to claim 2, wherein The dynamic characteristics of each boiler include an efficiency curve, pollutant emission characteristics, and an equipment health evaluation index; The method further includes: Extracting the efficiency change rate, pollutant emission trend information, and equipment health status index related to the load demand prediction from the efficiency curve, pollutant emission characteristics, and equipment health evaluation index to obtain fine-tuning feature information; Extracting historical model parameters from a pre-trained load demand prediction model; Using the fine-tuning feature information and the historical model parameters to perform model fine-tuning on the pre-trained load demand prediction model to obtain fine-tuned model parameters; Deploying the fine-tuned model parameters to the pre-trained load demand prediction model.

9. The method according to claim 2, characterized in that Generating a pre-trained load demand prediction model according to the following steps, including: Collecting the historical operation data of the header-type boiler system and the historical load demand data of the downstream steam-using equipment; Obtaining data features related to the load demand from the historical operation data and the historical load demand data; Creating a load demand prediction model using a long short-term memory network; Inputting the data features related to the load demand into the prediction model and outputting the loss value of the model; When the loss value reaches the minimum, obtaining a pre-trained load demand prediction model.

10. A dynamic load distribution system for boiler operation, characterized in that, The system includes: A data processing module for collecting and preprocessing the boiler operation data of each boiler in the header-type boiler system according to a preset standard communication protocol at a preset period to obtain the boiler operation data to be analyzed; An analysis module for analyzing the dynamic characteristics of each boiler based on the boiler operation data to be analyzed as the dynamic characteristic model of each boiler; the dynamic characteristics are used to characterize the change information of the performance parameters of each boiler during operation over time; A dynamic programming module for performing dynamic programming on the load of each boiler based on the dynamic characteristic model of each boiler and a preset multi-objective optimization algorithm to obtain the optimal load distribution ratio of each boiler; An adjustment module for distributing the operating load of each boiler through the optimal load distribution ratio of each boiler.

Images