Operation scheme determination method and device of power plant desulfurization system, computer equipment, readable storage medium and program product

By obtaining and training multiple sets of working conditions of the desulfurization system, we provide accurate and optimal operating solutions, and solve the problem of low accuracy of the desulfurization system of thermal power plants under variable operating conditions, achieving efficient, intelligent and accurate Desulfurization system operation.

CN119026953APending Publication Date: 2024-11-26GUANGDONG YUDEAN BOHE COAL POWER CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411128378.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The existing thermal power plants rely on manual experience and fixed rules in the operation of desulfurization systems, and have poor adaptability, resulting in low accuracy of the desulfurization system operation plan under variable working conditions.

Method used

By obtaining multiple sets of working condition information, inputting the preset operating plan prediction model, obtaining the optimal operating plan corresponding to each set of working condition information, and training the initial operating plan determination model to obtain the target operating plan determination model so that in different working conditions The optimal operating solution is provided below.

Benefits of technology

It improves the accuracy and optimization effect of the desulfurization system operation plan, reduces the errors of manual intervention and empirical judgment, reduces operating costs and energy consumption, ensures the achievement of environmental protection indicators, and improves the environmental protection and economic benefits of power plants.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026953A_ABST
    Figure CN119026953A_ABST
Patent Text Reader

Abstract

The invention relates to an operation scheme determination method and device of a power plant desulfurization system, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring multiple groups of working condition information of a desulfurization system to be determined; inputting the multiple groups of working condition information into a preset operation scheme prediction model to obtain an optimal operation scheme corresponding to each group of working condition information; training the initial operation scheme determination model by adopting the multiple groups of working condition information and the optimal operation scheme corresponding to each group of working condition information to obtain a trained target operation scheme determination model; and inputting the current working condition information of the desulfurization system to be determined into the target operation scheme determination model to obtain a target operation scheme corresponding to the current working condition information. By adopting the method, a more accurate operation scheme of the desulfurization system of the power plant can be obtained.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for determining an operation plan of a desulfurization system of a power plant. Background Art

[0002] Thermal power generation is one of the main ways of producing electricity in the world, but its combustion process produces a large amount of sulfur dioxide (SO 2 ) and other harmful gases, which seriously pollute the environment. Desulfurization technology removes SO in flue gas through chemical reactions. 2 , reducing the harm of acid rain and air pollution. Therefore, the implementation of effective desulfurization measures is essential to protecting the environment and public health, and is also a key link in ensuring that power plants comply with environmental regulations.

[0003] At present, thermal power plants usually rely on manual experience and fixed rules to determine the operation plan during the operation of the desulfurization system. However, this method has poor adaptability and the accuracy of the desulfurization system operation plan obtained under variable operating conditions is low. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for determining an operation plan of a power plant desulfurization system, which can obtain a more accurate operation plan of the power plant desulfurization system in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for determining an operation scheme of a desulfurization system of a power plant, comprising:

[0006] Acquire multiple groups of operating condition information of the desulfurization system to be determined;

[0007] Inputting the plurality of groups of operating condition information into a preset operation scheme prediction model respectively to obtain the optimal operation scheme corresponding to each group of operating condition information;

[0008] Using the multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information, training an initial operating scheme determination model to obtain a trained target operating scheme determination model;

[0009] The current operating condition information of the desulfurization system to be determined is input into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

[0010] In one embodiment, before the plurality of groups of operating condition information are respectively input into a preset operation scheme prediction model to obtain the optimal operation scheme corresponding to each group of operating condition information, the method further includes:

[0011] Obtain the initial operation plan prediction model, as well as the target constraint conditions and objective function;

[0012] According to the target constraint condition and the objective function, the initial constraint condition and the initial objective function of the initial operation scheme prediction model are updated to obtain the preset operation scheme prediction model.

[0013] In one embodiment, obtaining the target constraint condition and the target function includes:

[0014] Obtain historical operation plans and plan evaluation information corresponding to the historical operation plans;

[0015] The objective function is constructed according to the historical operation plan and the plan evaluation information, and the objective constraint condition is constructed according to the historical operation plan and the preset constraint information.

[0016] In one embodiment, the adopting of the multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information to train an initial operating scheme determination model to obtain a trained target operating scheme determination model includes:

[0017] Using the difference information between the optimal operation scheme corresponding to each set of operating condition information and the model output of the initial operation scheme determination model as the initial loss function of the initial operation scheme determination model;

[0018] Determining penalty information for a model output of the initial operation scheme determination model based on the constraint conditions in the preset operation scheme prediction model;

[0019] Determine the target loss function of the initial operation scheme determination model according to the penalty information and the initial loss function;

[0020] The multiple groups of operating condition information, the optimal operating scheme corresponding to each group of operating condition information and the target loss function are used to iteratively train the initial operating scheme determination model to obtain a trained target operating scheme determination model.

[0021] In one embodiment, the adopting of the multiple groups of operating condition information, the optimal operating scheme corresponding to each group of operating condition information and the target loss function, iteratively training the initial operating scheme determination model to obtain the trained target operating scheme determination model includes:

[0022] Determine the model of the initial operation scheme as the current training model;

[0023] Determine a loss value corresponding to the current training model according to the multiple groups of operating condition information, each group of operating condition information, the target loss function and the current training model;

[0024] According to the loss value, adjust the model parameters of the current training model to obtain a new current training model, and return to execute the step of determining the loss value corresponding to the current training model according to the multiple groups of working condition information, each group of working condition information, the target loss function and the current training model, until a current training model that meets the preset training conditions is obtained;

[0025] The current training model that meets the preset training conditions is used as the target operation plan determination model for completing the training.

[0026] In one embodiment, after obtaining the target operation plan corresponding to the current operating condition information, the method further includes:

[0027] Input the current operating condition information into a preset operation plan prediction model to obtain the optimal operation plan corresponding to the current operating condition information;

[0028] Based on the current operating condition information and the optimal operating solution corresponding to the current operating condition information, the target operating solution determination model is optimized to obtain an optimized target operating solution determination model.

[0029] In a second aspect, the present application also provides a device for determining an operation scheme of a desulfurization system of a power plant, comprising:

[0030] An operating condition information acquisition module, used to acquire multiple groups of operating condition information of the desulfurization system to be determined;

[0031] An operation scheme prediction module is used to input the multiple groups of operating condition information into a preset operation scheme prediction model to obtain the optimal operation scheme corresponding to each group of operating condition information;

[0032] A scheme model training module, used to train an initial operation scheme determination model by using the multiple groups of operating condition information and the optimal operation scheme corresponding to each group of operating condition information, and obtain a trained target operation scheme determination model;

[0033] The target scheme determination module is used to input the current operating condition information of the desulfurization system to be determined into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

[0034] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0035] Acquire multiple groups of operating condition information of the desulfurization system to be determined;

[0036] Inputting the plurality of groups of operating condition information into a preset operation scheme prediction model respectively to obtain the optimal operation scheme corresponding to each group of operating condition information;

[0037] Using the multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information, training an initial operating scheme determination model to obtain a trained target operating scheme determination model;

[0038] The current operating condition information of the desulfurization system to be determined is input into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

[0039] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0040] Acquire multiple groups of operating condition information of the desulfurization system to be determined;

[0041] Inputting the plurality of groups of operating condition information into a preset operation scheme prediction model respectively to obtain the optimal operation scheme corresponding to each group of operating condition information;

[0042] Using the multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information, training an initial operating scheme determination model to obtain a trained target operating scheme determination model;

[0043] The current operating condition information of the desulfurization system to be determined is input into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

[0044] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0045] Acquire multiple groups of operating condition information of the desulfurization system to be determined;

[0046] Inputting the plurality of groups of operating condition information into a preset operation scheme prediction model respectively to obtain the optimal operation scheme corresponding to each group of operating condition information;

[0047] Using the multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information, training an initial operating scheme determination model to obtain a trained target operating scheme determination model;

[0048] The current operating condition information of the desulfurization system to be determined is input into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

[0049] The above-mentioned method, device, computer equipment, computer-readable storage medium and computer program product for determining the operation plan of the desulfurization system of a power plant first obtain multiple groups of operating condition information of the desulfurization system to be determined, and comprehensively grasp the performance data of the desulfurization system under different operating conditions, so as to provide rich basic data for subsequent optimization and prediction, and ensure that the model can cover a wide range of actual operating conditions; then, the multiple groups of operating condition information are respectively input into the preset operation plan prediction model to obtain the optimal operation plan corresponding to each group of operating condition information. The preset operation plan prediction model can be used to calculate the optimal operation plan for each group of operating condition information, which can fully consider the specific conditions of different operating conditions and provide optimal operating parameters based on precise calculation, thereby improving the accuracy and optimization effect of the operation plan of the desulfurization system; then, multiple groups of operating condition information and the optimal operation plan corresponding to each group of operating condition information are used to train the initial An operation scheme determination model is obtained to obtain a trained target operation scheme determination model. The initial operation scheme determination model is trained by using multiple sets of operating condition information and corresponding optimal operation schemes to enable it to learn the optimal operation rules under different operating conditions. The obtained target operation scheme determination model can quickly predict the optimal operation scheme under the current operating conditions without the need to re-apply the operation scheme prediction model, thereby improving the calculation efficiency and response speed. Finally, the current operating condition information of the desulfurization system to be determined is input into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information. By inputting the current operating condition information into the trained machine learning model, the optimal operation scheme under the current operating condition can be obtained in real time, which not only improves the response speed and decision-making accuracy of the system, but also can quickly adjust the operating parameters when the operating conditions change, thereby ensuring the best balance between desulfurization effect and operation economy. In the above method, by utilizing the combination of the operation plan prediction model and the target operation plan determination model, it is ensured that an accurate optimal operation plan can be provided under variable operating conditions, reducing the errors of manual intervention and experience judgment, which not only improves the operation efficiency and desulfurization effect, but also significantly reduces the operation cost and energy consumption, ensures the achievement of environmental protection indicators, improves the environmental protection and economic benefits of the power plant, and ultimately efficiently determines a high-accuracy operation plan to achieve efficient, intelligent and precise operation of the desulfurization system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A schematic flow chart of a method for determining an operation scheme of a desulfurization system of a power plant in one embodiment;

[0052] Figure 2 A schematic flow chart of a method for determining an operation scheme of a desulfurization system of a power plant in another embodiment;

[0053] Figure 3 A structural block diagram of a system for determining an operation scheme of a desulfurization system of a power plant in one embodiment;

[0054] Figure 4 A structural block diagram of a device for determining an operation scheme of a desulfurization system of a power plant in one embodiment;

[0055] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] In one embodiment, Figure 1 As shown, a method for determining an operation plan of a desulfurization system of a power plant is provided. This embodiment uses the method applied to a terminal as an example for illustration. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, smart phones, and tablet computers. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. In this embodiment, the method includes the following steps:

[0058] Step S101, obtaining multiple groups of operating condition information of the desulfurization system to be determined.

[0059] Among them, the operating condition information is to describe various status parameters of the desulfurization system under different operating conditions, including but not limited to unit load, original flue gas sulfur dioxide concentration, current pH value of the absorption tower, current and voltage of each major equipment and other data.

[0060] For example, the user can install high-precision sensors and monitoring equipment at various key locations of the desulfurization system in advance to collect system operating parameters in real time. The terminal obtains multiple sets of operating information of the desulfurization system under different operating conditions from the sensors and monitoring equipment as needed.

[0061] Step S102 , inputting multiple groups of operating condition information into a preset operation plan prediction model respectively to obtain the optimal operation plan corresponding to each group of operating condition information.

[0062] Among them, the operation plan prediction model adopts optimization algorithms, such as genetic algorithms, simulated annealing algorithms or other heuristic algorithms.

[0063] Exemplarily, first, the terminal inputs the acquired multiple groups of operating condition information into the operation scheme prediction model. The model uses an optimization algorithm for calculation. The processing process of each group of operating condition information in the model includes parsing the state parameters of the current operating condition, performing optimization calculations based on constraints, so that the operation scheme continues to move closer to the optimization target, so as to determine the operation scheme that can achieve the highest desulfurization efficiency and the lowest energy consumption under the current conditions. Finally, the model outputs the optimal operation scheme corresponding to each group of operating condition information, and the scheme may include specific operation parameter settings, such as the optimal combination of slurry circulation pumps and the optimal pH value setting of the absorption tower.

[0064] Step S103, using multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information, train the initial operating scheme determination model to obtain a trained target operating scheme determination model. Commonly used machine learning algorithms include but are not limited to linear regression, support vector machine, decision tree and neural network.

[0065] Among them, the operation plan determination model adopts a machine learning model. By learning multiple sets of operating condition information and the corresponding optimal operation plans, it can predict the optimal operation plan when new operating condition information is given.

[0066] Exemplarily, first, the terminal uses the acquired multiple groups of operating condition information and the optimal operating plan corresponding to each group of operating condition information as a training data set. The initial operating plan determination model is trained using the data set, and the training process may include steps such as data preprocessing, feature extraction, and model training. During the training process, the model parameters are continuously adjusted so that the model can accurately predict the optimal operating plan under different operating condition information. After multiple iterations of training and verification, the trained target operating plan determination model is finally obtained, which can efficiently and accurately provide the optimal operating plan in practical applications.

[0067] Step S104, inputting the current operating condition information of the desulfurization system to be determined into a target operation scheme determination model to obtain a target operation scheme corresponding to the current operating condition information.

[0068] Exemplarily, first, the terminal obtains the current operating condition information of the desulfurization system to be determined, which includes parameters such as unit load, raw flue gas sulfur dioxide concentration, current pH value of the absorption tower, and current and voltage of each major equipment. Input these current operating condition information into the trained target operation plan determination model. The model uses the knowledge and rules learned in the previous training process to analyze and calculate the current operating conditions, and predicts the operating plan that can achieve the best desulfurization effect and economic benefits under the current conditions. The target operation plan output by the model may include specific operating parameter settings, such as the optimal combination of slurry circulation pumps and the optimal pH value (acidity and alkalinity) setting of the absorption tower. Finally, the terminal will display the obtained target operation plan to the user, or automatically apply it to actual operation to ensure that the desulfurization system operates efficiently and stably under the current operating conditions.

[0069] In the above-mentioned method for determining the operation plan of the desulfurization system of a power plant, first, multiple groups of operating condition information of the desulfurization system to be determined are obtained, and the performance data of the desulfurization system under different operating conditions are fully understood, which provides rich basic data for subsequent optimization and prediction, and ensures that the model can cover a wide range of actual operating conditions; then, the multiple groups of operating condition information are respectively input into the preset operation plan prediction model to obtain the optimal operation plan corresponding to each group of operating condition information. By using the preset operation plan prediction model, the optimal operation plan can be calculated for each group of operating condition information, which can fully consider the specific conditions of different operating conditions and provide the optimal operating parameters based on precise calculation, thereby improving the accuracy and optimization effect of the operation plan of the desulfurization system; then, the multiple groups of operating condition information and the optimal operation plan corresponding to each group of operating condition information are used to train the initial operation plan determination model to obtain the training The completed target operation plan determination model uses multiple sets of operating condition information and corresponding optimal operation plans to train the initial operation plan determination model, so that it can learn the optimal operation rules under different operating conditions. The obtained target operation plan determination model can quickly predict the optimal operation plan under the current operating conditions without the need to re-apply the operation plan prediction model, thereby improving the calculation efficiency and response speed. Finally, the current operating condition information of the desulfurization system to be determined is input into the target operation plan determination model to obtain the target operation plan corresponding to the current operating condition information. By inputting the current operating condition information into the trained machine learning model, the optimal operation plan under the current operating condition can be obtained in real time, which not only improves the response speed and decision-making accuracy of the system, but also can quickly adjust the operating parameters when the operating conditions change, to ensure the best balance between desulfurization effect and operation economy. In the above method, by utilizing the combination of the operation plan prediction model and the target operation plan determination model, it is ensured that an accurate optimal operation plan can be provided under variable operating conditions, reducing the errors of manual intervention and experience judgment, which not only improves the operation efficiency and desulfurization effect, but also significantly reduces the operation cost and energy consumption, ensures the achievement of environmental protection indicators, improves the environmental protection and economic benefits of the power plant, and ultimately efficiently determines a high-accuracy operation plan to achieve efficient, intelligent and precise operation of the desulfurization system.

[0070] In an exemplary embodiment, before the above step S102 inputs multiple groups of operating condition information into a preset operating plan prediction model to obtain the optimal operating plan corresponding to each group of operating condition information, it also includes: obtaining an initial operating plan prediction model, and obtaining target constraints and objective functions; according to the target constraints and objective functions, updating the initial constraints and initial objective functions of the initial operating plan prediction model to obtain a preset operating plan prediction model.

[0071] Among them, the target constraints are various restrictive conditions for the operation of the desulfurization system, including environmental protection indicators, power consumption limits, pH value range of the absorption tower, etc.; the objective function is an evaluation criterion for measuring the pros and cons of the operation plan, including but not limited to desulfurization efficiency, power consumption, operating costs, etc.

[0072] Exemplarily, first, the terminal can obtain an initial operation plan prediction model, which is a general optimization model with basic optimization capabilities, but has not been customized for specific working conditions. Then, determine the target constraints and objective functions. For example, the target constraints may include requirements such as the maximum allowable sulfur dioxide emission concentration, power consumption limit, and pH range of the absorption tower; the objective function may include minimizing power consumption, maximizing desulfurization efficiency, etc. Next, according to the target constraints and objective functions, the initial operation plan prediction model is updated, that is, by replacing or adding the target constraints and objective function parameters to the general optimization model to ensure that the model can obtain the operation plan that best meets the objective function while meeting all constraints. After the update is completed, a preset operation plan prediction model is obtained, which can calculate the optimal operation plan corresponding to each set of operating condition information when multiple sets of operating condition information are input.

[0073] In addition, the use of different optimization models also requires adaptive encoding of the parameters of the application scenario. For example, in the particle swarm optimization algorithm model, the key parameters in the operating condition information are encoded as the dimensions of the particles, the speed and position update rules of the particles are defined, and the fitness value is calculated according to the objective function. Through multiple iterations, the particle swarm gradually converges to the optimal solution; in the simulated annealing algorithm model, the key parameters in the operating condition information are represented as the dimensions of the solution, and a new solution is generated through neighborhood search. The acceptance probability determines whether to accept the new solution, and the temperature is gradually lowered to find the global optimal solution; in the genetic algorithm model, the key parameters in the operating condition information are encoded as genetic genes, and the optimization objectives (such as minimum power consumption, maximum desulfurization efficiency) are defined as traits. The gene encoding can be represented by binary encoding, real number encoding, etc. The genetic algorithm synthesizes these genes into chromosomes, and iteratively optimizes the operation plan corresponding to each set of operating condition information through operations such as selection, crossover and mutation, etc.

[0074] In this embodiment, by obtaining and updating the target constraints and objective functions, it is ensured that the model fully considers all the constraints and objectives in actual operation during the optimization process, thereby improving the accuracy of the calculation results. The general optimization model is transformed into an operation plan prediction model that adapts to the current working conditions by updating the target constraints and objective functions, thereby enhancing the applicability of the model under different working conditions.

[0075] In an exemplary embodiment, the above-mentioned acquisition of target constraints and target functions also includes: obtaining historical operation plans, and obtaining plan evaluation information corresponding to the historical operation plans; constructing a target function based on the historical operation plans and the plan evaluation information, and constructing target constraints based on the historical operation plans and preset constraint information.

[0076] Among them, the historical operation plan is the actual operation data and operation records of the desulfurization system under different working conditions, and the plan evaluation information is the effect evaluation indicators of these operation plans, such as desulfurization efficiency, power consumption, operating cost, emission concentration, etc.

[0077] Exemplarily, the terminal collects and organizes the historical operation plans of the desulfurization system under different working conditions. Then, the program evaluation information corresponding to these historical operation plans is obtained, for example, the effect of each plan is evaluated by calculating indicators such as desulfurization efficiency, power consumption and operating costs. Next, the terminal constructs a mathematical model of the objective function based on the collected historical operation plans and program evaluation information. The process of constructing the objective function includes analyzing the relationship between the evaluation information of each plan, using a multi-objective optimization method, combining desulfurization efficiency and economic benefits to define the objective function, for example, weighted calculation of multiple indicators. At the same time, the terminal constructs a mathematical model of the target constraint conditions based on the historical operation plans and preset constraint information. The preset constraint information includes environmental protection regulations, equipment operation restrictions, and specific operating parameter ranges. By analyzing the constraints in the historical operation plans and the corresponding operating effects, the impact of these constraints on the system operation is determined, thereby defining the target constraints of the model.

[0078] In this embodiment, by analyzing historical operation plans and plan evaluation information, the objective function mathematical model constructed can more accurately reflect the actual operation effect of the desulfurization system under different operating conditions, thereby improving the reliability of the optimization results. According to the historical operation plan and preset constraint information, the target constraint condition mathematical model constructed can more comprehensively cover the constraints in actual operation, ensuring that the model meets all actual restriction requirements during the optimization process. Through accurate objective functions and reasonable constraints, the optimization model can provide better operation plans under different operating conditions and improve the desulfurization efficiency and economic benefits of the system. By constructing an objective function and constraint condition mathematical model based on historical data, the model can continuously learn and update, improve the intelligence level of the system, and ensure that the desulfurization system always maintains the optimal operating state under changing operating conditions.

[0079] In an exemplary embodiment, the above-mentioned step S103 uses multiple groups of operating condition information and the optimal operating plan corresponding to each group of operating condition information to train the initial operating plan determination model to obtain a trained target operating plan determination model, and also includes: using the difference information between the optimal operating plan corresponding to each group of operating condition information and the model output of the initial operating plan determination model as the initial loss function of the initial operating plan determination model; based on the constraints in the preset operating plan prediction model, determining the penalty information for the model output of the initial operating plan determination model; determining the target loss function of the initial operating plan determination model based on the penalty information and the initial loss function; using multiple groups of operating condition information, the optimal operating plan corresponding to each group of operating condition information and the target loss function, iteratively training the initial operating plan determination model to obtain a trained target operating plan determination model.

[0080] Among them, the initial loss function is to measure the difference between the model output and the optimal operating plan. Commonly used loss functions include mean square error, etc.

[0081] The penalty information is the penalty value when the model output violates the preset constraint conditions. Common methods include adding penalty items to the loss function. For example, if the pH value output by the model exceeds the preset range, the corresponding penalty value is increased.

[0082] Exemplarily, the terminal uses multiple sets of operating condition information, the optimal operating scheme corresponding to each set of operating condition information, and the target loss function to train the initial operating scheme determination model through an iterative optimization method. For example, gradient descent or other optimization algorithms can be used to continuously adjust model parameters to minimize the target loss function. After multiple iterative trainings, a trained target operating scheme determination model is finally obtained, which can efficiently and accurately provide the optimal operating scheme in practical applications.

[0083] In this embodiment, the difference between the model prediction result and the optimal operation plan is measured by the initial loss function, and the model parameters are continuously adjusted to improve the prediction accuracy. By converting the preset constraints into penalty information, it is ensured that the model output always meets the restriction requirements in the actual operation during the optimization process. The target loss function comprehensively considers the prediction accuracy and constraint satisfaction, and through iterative training, the model can provide the optimal operation plan under different working conditions.

[0084] In an exemplary embodiment, the above-mentioned use of multiple groups of operating condition information, the optimal operating plan corresponding to each group of operating condition information, and the target loss function, iteratively trains the initial operating plan determination model to obtain the trained target operating plan determination model, and also includes: using the initial operating plan determination model as the current training model; determining the loss value corresponding to the current training model according to the multiple groups of operating condition information, each group of operating condition information, the target loss function and the current training model; adjusting the model parameters of the current training model according to the loss value to obtain a new current training model, and returning to execute the step of determining the loss value corresponding to the current training model according to the multiple groups of operating condition information, each group of operating condition information, the target loss function and the current training model, until the current training model that meets the preset training conditions is obtained; using the current training model that meets the preset training conditions as the trained target operating plan determination model.

[0085] Among them, the preset training condition can be that the loss value converges to a certain threshold or reaches a maximum number of iterations.

[0086] Exemplarily, the terminal inputs multiple groups of operating condition information and the optimal operating plan corresponding to each group of operating condition information into the current training model, calculates the difference between the model output and the actual optimal plan according to the target loss function, and the penalty information of the constraint conditions, and comprehensively obtains the loss value of the current training model. The terminal can use gradient descent or other optimization algorithms to adjust the parameters of the current training model according to the calculated loss value and its gradient to obtain a new current training model. Repeat the above steps, iteratively calculate the loss value of the current training model and adjust the model parameters until the loss value or number of iterations of the current training model meets the preset conditions. Finally, the current training model that meets the preset training conditions is the target operating plan determination model that has completed the training, which can be used to provide the optimal operating plan in practical applications.

[0087] In this embodiment, by iteratively calculating the loss value and adjusting the model parameters, the prediction error is gradually reduced and the prediction accuracy of the model is improved. Through the penalty information in the target loss function, it is ensured that the model output always meets the preset constraints during the optimization process. The target loss function comprehensively considers the prediction accuracy and constraint satisfaction. Through iterative training, the model can provide the best operation plan under different working conditions. By combining multiple sets of working condition information and the optimal operation plan for training, the model can adapt to different working conditions and improve the flexibility and reliability in practical applications. Through iterative training and optimization process, the model can continuously learn and improve, improve the intelligence level of the desulfurization system, and ensure that it always maintains the optimal operating state under changing working conditions.

[0088] In an exemplary embodiment, after obtaining the target operating plan corresponding to the current operating condition information in the above step S104, it also includes: inputting the current operating condition information into a preset operating plan prediction model to obtain the optimal operating plan corresponding to the current operating condition information; based on the current operating condition information and the optimal operating plan corresponding to the current operating condition information, optimizing the target operating plan determination model to obtain an optimized target operating plan determination model.

[0089] Exemplarily, when the target operation plan corresponding to the current operating condition information is obtained, the terminal first applies the target operation plan to achieve real-time optimized operation of the desulfurization system. Afterwards, the terminal uses the optimal operation plan obtained through the operation plan prediction model together with the current operating condition information as a new training sample for dynamically optimizing the target operation plan determination model, which can be carried out by incremental learning or retraining the model to gradually improve the prediction accuracy of the model while maintaining its rapid response and real-time performance. It should be noted that the target operation plan determination model is a trained machine learning model that can quickly predict the optimal operation plan and has good real-time performance; the preset operation plan prediction model is the optimal operation plan calculated by a complex optimization algorithm. Although the result is more accurate, the calculation time is longer and the real-time performance is poor. Therefore, the accurate results of the preset operation plan prediction model can be used to optimize the target operation plan determination model, and the accuracy of the plan can be improved while maintaining the efficiency of plan determination.

[0090] In this embodiment, the preset operation scheme prediction model calculates a more accurate optimal operation scheme through a complex optimization algorithm. These accurate results can be used to optimize the target operation scheme determination model and improve its prediction accuracy. The target operation scheme determination model has a rapid response capability and can quickly provide an operation scheme when the real-time working conditions change. By combining the working condition information obtained in real time and the accurate optimal operation scheme, the target operation scheme determination model is continuously and dynamically optimized, so that the model gradually improves the prediction accuracy while maintaining real-time performance.

[0091] In another exemplary embodiment, Figure 2 As shown, the present application provides a method for determining an operation scheme of a desulfurization system of a power plant, the method comprising the following steps:

[0092] Step S201, obtaining historical operation plans, and obtaining plan evaluation information corresponding to the historical operation plans.

[0093] Step S202: constructing an objective function based on historical operation plans and plan evaluation information, and constructing an objective constraint condition based on historical operation plans and preset constraint information.

[0094] Step S203, obtaining an initial operation plan prediction model, and updating the initial constraints and the initial objective function of the initial operation plan prediction model according to the target constraints and the objective function, to obtain a preset operation plan prediction model.

[0095] Step S204, obtaining multiple groups of operating condition information of the desulfurization system to be determined, and inputting the multiple groups of operating condition information into a preset operation plan prediction model respectively to obtain the optimal operation plan corresponding to each group of operating condition information.

[0096] Step S205 , using the difference information between the optimal operation scheme corresponding to each set of operating condition information and the model output of the initial operation scheme determination model as the initial loss function of the initial operation scheme determination model.

[0097] Step S206, based on the constraint conditions in the preset operation plan prediction model, determine the penalty information for the model output of the initial operation plan determination model.

[0098] Step S207, determining the target loss function of the model for the initial operation scheme according to the penalty information and the initial loss function.

[0099] Step S208: determine the model of the initial operation scheme as the current training model.

[0100] Step S209, determining the loss value corresponding to the current training model according to the multiple groups of operating condition information, each group of operating condition information, the target loss function and the current training model.

[0101] Step S210, adjust the model parameters of the current training model according to the loss value to obtain a new current training model, and return to execute step S209 until a current training model that meets the preset training conditions is obtained.

[0102] Step S211, taking the current training model that meets the preset training conditions as the target operation plan determination model after training is completed.

[0103] Step S212, inputting the current operating condition information of the desulfurization system to be determined into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

[0104] Step S213, input the current operating condition information into a preset operation plan prediction model to obtain the optimal operation plan corresponding to the current operating condition information.

[0105] Step S214, based on the current operating condition information and the optimal operating solution corresponding to the current operating condition information, the target operating solution determination model is optimized to obtain an optimized target operating solution determination model.

[0106] In this embodiment, by utilizing the combination of the operation plan prediction model and the target operation plan determination model, it is ensured that an accurate optimal operation plan can be provided under variable operating conditions, reducing the errors of manual intervention and experience judgment, not only improving the operating efficiency and desulfurization effect, but also significantly reducing the operating cost and energy consumption, ensuring the achievement of environmental protection indicators, improving the environmental protection and economic benefits of the power plant, and ultimately efficiently determining a high-accuracy operation plan to achieve efficient, intelligent and accurate operation of the desulfurization system. Through the penalty information in the target loss function, it is ensured that the model output always meets the preset constraints during the optimization process. By combining the real-time acquired operating condition information and the accurate optimal operation plan, the target operation plan determination model is continuously and dynamically optimized, so that the model gradually improves the prediction accuracy while maintaining real-time performance.

[0107] In an exemplary embodiment, Figure 3 As shown, the present application provides a system for determining an operation plan of a desulfurization system of a power plant, the system comprising an intelligent control platform 301 , a communication unit 302 , a storage server 303 and an intelligent data platform 304 .

[0108] Among them, the intelligent data platform 304 includes a working condition data analysis and performance calculation module and a unit model optimization operation module. The intelligent data platform 304 continuously collects relevant technical parameters of the desulfurization system from the desulfurization DCS control system, namely the desulfurization distributed control system (DCS), including unit load, slurry circulation pump operation combination, absorption tower pH value, inlet and outlet sulfur dioxide concentration, current and voltage of each major equipment and other technical parameters;

[0109] Through the working condition data analysis and performance calculation module, the desulfurization system is analyzed for energy efficiency and consumption difference, environmental protection index analysis, economic index analysis, etc., and stored in the form of economic reports, so that the relevant departments or leaders of the power plant can grasp the production information of the entire desulfurization island and its ancillary systems at any time;

[0110] Through the unit model optimization operation module, the operation plans of the system under various load conditions are continuously recorded. By calculating the corresponding desulfurization efficiency, power consumption, powder consumption and other indicators under each operation plan, intelligent analysis is performed in combination with energy consumption considerations. The preset operation plan prediction model is used to optimize the operation plan, find the best operation mode, and establish and train the operation plan determination model. It also has the ability to dynamically update the operation plan determination model, and regularly provide the latest operation plan determination model to the intelligent control platform 301, making the intelligent control more reasonable.

[0111] The intelligent control platform 301 determines the model according to the operation plan sent by the intelligent data platform 304, and determines the current optimal operation mode according to the real-time load conditions. For example, it determines the best slurry circulation pump combination and absorption tower slurry pH value under comprehensive consideration of economic indicators, and recommends the optimal operation mode to the operator.

[0112] The communication unit 302 acts as an information transmission medium and is responsible for transmitting the data collected from the DCS to the storage server 303 and feeding back the information of the intelligent data platform 304 database to the DCS system.

[0113] The storage server 303 serves as a node of the entire system network, stores and processes a large amount of data transmitted by the communication unit 302, and provides safe and reliable data for the intelligent data platform 304.

[0114] The process of the above-mentioned power plant desulfurization system operation plan determination system in executing the power plant desulfurization system operation plan determination method is as follows:

[0115] ① According to the measurement points set up by the user in advance on the DCS system, the intelligent data platform 304 automatically collects real-time operating parameters (such as unit load, raw / clean flue gas sulfur dioxide concentration, absorption tower pH, current and voltage of each major equipment, etc.), and transmits the required parameters to the storage server 303 for storage and processing through the communication unit 302.

[0116] ② The storage server 303 performs preliminary processing on the data transmitted by the communication unit 302, and transmits the processed data to the intelligent data platform 304. The intelligent data platform 304 includes an operating condition data analysis and performance calculation module and a unit model optimization operation module:

[0117] Among them, through the operating data analysis and performance calculation module, the desulfurization system is analyzed for energy efficiency and consumption difference, environmental protection index analysis, economic index analysis, etc., and stored in the form of economic reports, so that the relevant departments or leaders of the power plant can grasp the production information of the entire desulfurization island and its ancillary systems at any time.

[0118] Through the unit model optimization operation module, the operation plans of the system under various load conditions are continuously recorded. By calculating the corresponding desulfurization efficiency, power consumption, powder consumption and other indicators under each operation plan, intelligent analysis is performed in combination with energy consumption considerations. The preset operation plan prediction model is used to optimize the operation plan, find the best operation mode, and establish and train the operation plan determination model. It also has the ability to dynamically update the operation plan determination model, and regularly provide the latest operation plan determination model to the intelligent control platform 301, making the intelligent control more reasonable.

[0119] ③ The intelligent data platform 304 feeds back the latest operation plan determination model to the intelligent control platform 301, and the intelligent control platform 301 receives and stores the latest operation plan determination model.

[0120] ④ The intelligent control platform 301 uses the latest operation plan determination model to determine the current optimal operation mode according to the real-time load conditions, such as the specific slurry circulation pump combination and the corresponding pH setting value.

[0121] In this embodiment, under the load of different flue gas volumes of the unit, the system can simultaneously consider the outlet SO2 concentration and the factors of optimizing the operating cost of the desulfurization equipment, and provide suggestions such as the optimal slurry circulation pump combination and the pH value of the absorber slurry under any load, so as to reduce power consumption and powder consumption as much as possible. The optimized operation of the desulfurization system is comprehensively considered with the energy consumption, material consumption and operating cost of the desulfurization system, and a complete desulfurization system evaluation system is established, including but not limited to the unit energy consumption of the desulfurization system, the unit material consumption of the desulfurization system, and the cost of the desulfurization system, so that users can fully grasp the production information of the entire desulfurization island and its ancillary systems. In summary, the above system can help users reduce the operating burden, and more importantly, it can optimize the operating plan of the desulfurization system and avoid the emergence of unreasonable operating plans. Compared with traditional control methods, the control method is more intelligent and will achieve better results in energy saving and consumption reduction of the desulfurization system.

[0122] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0123] Based on the same inventive concept, the embodiment of the present application also provides an operation scheme determination device for a power plant desulfurization system for implementing the above-mentioned operation scheme determination method for a power plant desulfurization system. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in the embodiments of one or more operation scheme determination devices for power plant desulfurization systems provided below can refer to the limitations of the operation scheme determination method for a power plant desulfurization system above, and will not be repeated here.

[0124] In an exemplary embodiment, Figure 4As shown, a device for determining an operation scheme of a desulfurization system of a power plant is provided, comprising: an operating condition information acquisition module 401, an operation scheme prediction module 402, a scheme model training module 403 and a target scheme determination module 404, wherein:

[0125] The operating condition information acquisition module 401 is used to acquire multiple groups of operating condition information of the desulfurization system to be determined;

[0126] The operation scheme prediction module 402 is used to input multiple groups of operating condition information into a preset operation scheme prediction model to obtain the optimal operation scheme corresponding to each group of operating condition information;

[0127] A solution model training module 403 is used to train an initial operation solution determination model using multiple groups of operating condition information and the optimal operation solution corresponding to each group of operating condition information, and obtain a trained target operation solution determination model;

[0128] The target scheme determination module 404 is used to input the current operating condition information of the desulfurization system to be determined into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

[0129] In one of the embodiments, the operation plan determination device of the above-mentioned power plant desulfurization system also includes a prediction model determination module, which is used to obtain an initial operation plan prediction model, as well as to obtain target constraints and objective functions; according to the target constraints and objective functions, the initial constraints and initial objective functions of the initial operation plan prediction model are updated to obtain a preset operation plan prediction model.

[0130] In one of the embodiments, the above-mentioned prediction model determination module is used to obtain historical operation plans and scheme evaluation information corresponding to the historical operation plans; construct an objective function based on the historical operation plans and scheme evaluation information, and construct target constraint conditions based on the historical operation plans and preset constraint information.

[0131] In one of the embodiments, the above-mentioned scheme model training module 403 is also used to use the difference information between the optimal operating scheme corresponding to each set of operating condition information and the model output of the initial operating scheme determination model as the initial loss function of the initial operating scheme determination model; based on the constraints in the preset operating scheme prediction model, determine the penalty information for the model output of the initial operating scheme determination model; determine the target loss function of the initial operating scheme determination model based on the penalty information and the initial loss function; use multiple sets of operating condition information, the optimal operating scheme corresponding to each set of operating condition information, and the target loss function to iteratively train the initial operating scheme determination model to obtain a trained target operating scheme determination model.

[0132] In one of the embodiments, the above-mentioned scheme model training module 403 is also used to determine the model of the initial operation scheme as the current training model; determine the loss value corresponding to the current training model according to multiple groups of operating condition information, each group of operating condition information, the target loss function and the current training model; adjust the model parameters of the current training model according to the loss value to obtain a new current training model, and return to execute the step of determining the loss value corresponding to the current training model according to multiple groups of operating condition information, each group of operating condition information, the target loss function and the current training model until the current training model that meets the preset training conditions is obtained; and use the current training model that meets the preset training conditions as the target operation scheme determination model after the training is completed.

[0133] In one of the embodiments, the operation plan determination device of the above-mentioned power plant desulfurization system also includes a model dynamic optimization module, which is used to input the current operating condition information into a preset operation plan prediction model to obtain the optimal operation plan corresponding to the current operating condition information; based on the current operating condition information and the optimal operation plan corresponding to the current operating condition information, the target operation plan determination model is optimized to obtain an optimized target operation plan determination model.

[0134] Each module in the above-mentioned operation scheme determination device of the power plant desulfurization system can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0135] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 5As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for determining an operation plan of a power plant desulfurization system is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0136] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0137] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0138] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0139] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0140] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0141] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0142] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0143] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for determining an operation plan of a desulfurization system of a power plant, characterized in that: The method comprises: Acquire multiple groups of operating condition information of the desulfurization system to be determined; Inputting the plurality of groups of operating condition information into a preset operation scheme prediction model respectively to obtain the optimal operation scheme corresponding to each group of operating condition information; Using the multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information, training an initial operating scheme determination model to obtain a trained target operating scheme determination model; The current operating condition information of the desulfurization system to be determined is input into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

2. The method according to claim 1, characterized in that: Before respectively inputting the plurality of groups of operating condition information into a preset operation scheme prediction model to obtain the optimal operation scheme corresponding to each group of operating condition information, the method further includes: Obtain the initial operation plan prediction model, as well as the target constraint conditions and objective function; According to the target constraint condition and the objective function, the initial constraint condition and the initial objective function of the initial operation scheme prediction model are updated to obtain the preset operation scheme prediction model.

3. The method according to claim 2, characterized in that The obtaining of target constraints and target functions includes: Obtain historical operation plans and plan evaluation information corresponding to the historical operation plans; The objective function is constructed according to the historical operation plan and the plan evaluation information, and the objective constraint condition is constructed according to the historical operation plan and the preset constraint information.

4. The method according to claim 1, characterized in that: The adopting of the multiple groups of operating condition information and the optimal operating scheme corresponding to each group of operating condition information to train the initial operating scheme determination model to obtain the trained target operating scheme determination model includes: Using the difference information between the optimal operation scheme corresponding to each set of operating condition information and the model output of the initial operation scheme determination model as the initial loss function of the initial operation scheme determination model; Determining penalty information for a model output of the initial operation scheme determination model based on the constraint conditions in the preset operation scheme prediction model; Determine the target loss function of the initial operation scheme determination model according to the penalty information and the initial loss function; The multiple groups of operating condition information, the optimal operating scheme corresponding to each group of operating condition information and the target loss function are used to iteratively train the initial operating scheme determination model to obtain a trained target operating scheme determination model.

5. The method according to claim 4, characterized in that The adopting of the multiple groups of operating condition information, the optimal operating scheme corresponding to each group of operating condition information and the target loss function, iteratively training the initial operating scheme determination model to obtain the trained target operating scheme determination model, includes: Determine the model of the initial operation scheme as the current training model; Determine a loss value corresponding to the current training model according to the multiple groups of operating condition information, each group of operating condition information, the target loss function and the current training model; According to the loss value, adjust the model parameters of the current training model to obtain a new current training model, and return to execute the step of determining the loss value corresponding to the current training model according to the multiple groups of working condition information, each group of working condition information, the target loss function and the current training model, until a current training model that meets the preset training conditions is obtained; The current training model that meets the preset training conditions is used as the target operation plan determination model for completing the training.

6. The method according to claim 1, characterized in that After obtaining the target operation plan corresponding to the current operating condition information, the method further includes: Input the current operating condition information into a preset operation plan prediction model to obtain the optimal operation plan corresponding to the current operating condition information; Based on the current operating condition information and the optimal operating solution corresponding to the current operating condition information, the target operating solution determination model is optimized to obtain an optimized target operating solution determination model.

7. A device for determining an operation plan of a desulfurization system in a power plant, characterized in that: The device comprises: An operating condition information acquisition module, used to acquire multiple groups of operating condition information of the desulfurization system to be determined; An operation scheme prediction module is used to input the multiple groups of operating condition information into a preset operation scheme prediction model to obtain the optimal operation scheme corresponding to each group of operating condition information; A scheme model training module, used to train an initial operation scheme determination model by using the multiple groups of operating condition information and the optimal operation scheme corresponding to each group of operating condition information, and obtain a trained target operation scheme determination model; The target scheme determination module is used to input the current operating condition information of the desulfurization system to be determined into the target operation scheme determination model to obtain the target operation scheme corresponding to the current operating condition information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.