Coal blending combustion proportion analysis method, system and equipment for thermal power plant and medium

Abstract

The invention provides a thermal power plant coal blending combustion proportion analysis method, system equipment and a medium. According to the embodiment of the invention, physical and chemical parameters of to-be-blended-combusted fuel are collected; fuel characteristic data are determined according to the basic coal type, the standard combustion parameters corresponding to the coal types and the physical and chemical parameters of the fuel to be subjected to blending combustion; preliminarily setting a blending combustion proportion range by adopting a multi-objective optimization algorithm according to the fuel characteristic data; performing a small-scale blending combustion test based on the blending combustion proportion range, and collecting test data; and analyzing test data and adjusting a blending combustion ratio to optimize a preset combustion model so as to analyze the coal blending combustion ratio, so that through accurate blending combustion ratio control, the utilization of fuel energy is maximized, the power generation efficiency is improved, the emission of harmful gases such as NOx and SOx is reduced, the stricter environmental protection standard is met, the fuel cost and the combustion benefit are balanced, and the total operation cost is reduced. The market competitiveness of the thermal power plant is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method, system equipment and medium for analyzing the proportion of coal blending in a thermal power plant. Background Art

[0002] In the context of the current global energy structure transformation, coal, as the main component of traditional energy, has become the focus of attention in the energy industry for its efficient and clean use. With increasingly stringent environmental protection requirements and the advancement of sustainable development goals, how to reduce the impact of coal-fired power generation on the environment while ensuring stable power supply is an important challenge facing thermal power plants. Therefore, exploring and implementing coal blending technology, that is, adding a certain proportion of other fuels (such as biomass fuel, coal gangue, inferior coal, etc.) during the coal combustion process, has become an important direction of technological innovation.

[0003] With the advancement of combustion technology, material handling technology and pollutant control technology, the feasibility and benefits of coal blending have been significantly improved. Accurately controlling the blending ratio, optimizing the combustion process, and effectively managing the physical and chemical properties of the blended fuel have become the key to improving blending efficiency and reducing environmental risks.

[0004] The calorific value, volatile matter, ash and other physical and chemical properties of different fuels vary greatly, making it difficult to achieve accurate blending ratio control in actual operation. Too high a blending ratio may lead to unstable combustion, reduced efficiency, and even affect equipment safety; while too low a ratio may not fully utilize the environmental and economic advantages of blending. The current control system and algorithm still need to be optimized in dynamically adjusting the blending ratio to cope with fuel changes. Summary of the invention

[0005] The embodiments of the present invention provide a method, system equipment and medium for analyzing the proportion of coal blending in a thermal power plant, which can at least solve some problems in the prior art.

[0006] A first aspect of an embodiment of the present invention provides a method for analyzing the proportion of coal blending in a thermal power plant.

[0007] Collecting the physical and chemical parameters of the fuel to be blended, wherein the physical and chemical parameters include calorific value, volatile matter, ash content, and moisture content;

[0008] Determine fuel characteristic data based on basic coal types and standard combustion parameters corresponding to each coal type and physicochemical parameters of the fuel to be blended;

[0009] According to the fuel characteristic data, a multi-objective optimization algorithm is used to preliminarily set the blending ratio range;

[0010] Conduct a small-scale blending test based on the blending ratio range and collect test data;

[0011] Analyze test data and adjust the blending ratio to optimize the preset combustion model for analyzing coal blending ratio.

[0012] Further, the step of preliminarily setting the blending ratio range by using a multi-objective optimization algorithm according to the fuel characteristic data includes:

[0013] Calculate the theoretical heat of combustion of fuel mixtures;

[0014] Evaluating the emission potential of a fuel mixture, including predicting NOx and SOx generation;

[0015] The upper and lower limits of the blending ratio are determined based on preset environmental protection standards and economic benefits.

[0016] Further, a small-scale blending test is conducted based on the blending ratio range, and the step of collecting test data includes:

[0017] Design a pretreatment scheme for the blended fuel to improve the fuel mixing uniformity;

[0018] Collecting processing information after physical pre-treatment of the fuel;

[0019] The processing information is mixed according to a preliminarily set ratio to obtain test data.

[0020] Further, the test data is analyzed and the blending ratio is adjusted to optimize the preset combustion model. The steps for analyzing the coal blending ratio include:

[0021] Establishing a combustion model of the fuel mixture dynamics, the combustion model taking into account oxygen concentration, temperature and other factors that affect combustion efficiency;

[0022] Apply machine learning algorithms to iteratively optimize model parameters based on test data feedback;

[0023] Adjust the blending ratio to the optimal state to achieve the predetermined combustion efficiency and emission standards.

[0024] Further, the steps of applying a machine learning algorithm to iteratively optimize model parameters according to test data feedback include:

[0025] Define the state space, including the current blending ratio, combustion efficiency index, and pollutant emission level;

[0026] Set the action space to determine the adjustment range of the blending ratio;

[0027] A reward function is designed to adjust parameters of the optimization model based on the test data.

[0028] Furthermore, the step of designing the reward function specifically includes:

[0029] Determine weight factors, where each factor is dynamically adjusted based on the operational strategy;

[0030] Calculate economic benefit indicators;

[0031] Evaluate the effectiveness of pollutant emission reduction;

[0032] Calculating a reward value based on the weight factor, the economic benefit index, and the pollutant emission reduction effect, wherein the reward value reflects the comprehensive performance of the current blending strategy in terms of economic benefit and pollutant emission reduction effect;

[0033] The calculated reward value is fed back to the machine learning algorithm as a basis for optimizing the model parameters and iterating the blending ratio parameters.

[0034] Furthermore, the step of setting the action space includes:

[0035] Define the minimum step length for adjusting the blending ratio to ensure the accuracy and feasibility of the adjustment;

[0036] Determine the exploration scope of the blending ratio adjustment, and the exploration scope is dynamically adjusted according to the historical optimal blending ratio, the current blending test feedback and the changes in environmental protection and economic policies;

[0037] Combined with the current blending ratio in the state space, a specific blending ratio adjustment sequence is generated as a set of optional actions in the iterative process of the machine learning algorithm.

[0038] In a second aspect, an embodiment of the present invention provides a system for analyzing the proportion of coal blending in a thermal power plant, comprising:

[0039] A collection module, used to collect the physical and chemical parameters of the fuel to be blended, wherein the physical and chemical parameters include calorific value, volatile matter, ash content, and moisture content;

[0040] A blending module, used to determine fuel characteristic data based on the basic coal type and the standard combustion parameters corresponding to each coal type and the physical and chemical parameters of the fuel to be blended;

[0041] A determination module is used to preliminarily set the blending ratio range based on fuel characteristic data using a multi-objective optimization algorithm;

[0042] A test module, used to conduct a small-scale blending test based on the blending ratio range and collect test data;

[0043] The analysis module is used to analyze the test data and adjust the blending ratio to optimize the preset combustion model and analyze the coal blending ratio.

[0044] According to a third aspect of the embodiments of the present invention,

[0045] An electronic device is provided, comprising:

[0046] processor;

[0047] a memory for storing processor-executable instructions;

[0048] The processor is configured to call the instructions stored in the memory to execute the method described in the first aspect.

[0049] A fourth aspect of the embodiments of the present invention is:

[0050] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described in the first aspect is implemented.

[0051] The embodiment of the present invention collects the physical and chemical parameters of the fuel to be blended; determines the fuel characteristic data according to the basic coal type and the standard combustion parameters corresponding to each coal type and the physical and chemical parameters of the fuel to be blended; preliminarily sets the blending ratio range according to the fuel characteristic data using a multi-objective optimization algorithm; conducts a small-scale blending test based on the blending ratio range and collects test data; analyzes the test data and adjusts the blending ratio to optimize a preset combustion model for analyzing the coal blending ratio, so as to maximize the use of fuel energy through precise blending ratio control, improve power generation efficiency, reduce emissions of harmful gases such as NOx and SOx, meet stricter environmental protection standards, balance fuel costs and combustion benefits, reduce overall operating costs, and enhance the market competitiveness of thermal power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of a flow chart of a method for analyzing the proportion of coal blending in a thermal power plant according to an embodiment of the present invention;

[0053] Figure 2 The present invention is a schematic structural diagram of a method system for analyzing the proportion of coal blending in a thermal power plant according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] With the advancement of combustion technology, material handling technology and pollutant control technology, the feasibility and benefits of coal blending have been significantly improved. Accurately controlling the blending ratio, optimizing the combustion process, and effectively managing the physical and chemical properties of the blended fuel have become the key to improving blending efficiency and reducing environmental risks.

[0056] Although the feasibility and benefits of coal blending have been significantly improved with the advancement of relevant technologies, there are still some technical challenges and defects in practical applications. These challenges limit the widespread application of blending technology and the maximization of its efficiency. The calorific value, volatile matter, ash content and other physical and chemical properties of different fuels vary greatly, making it difficult to achieve accurate blending ratio control in actual operations. Too high a blending ratio may lead to unstable combustion, reduced efficiency, and even affect equipment safety; while too low a ratio may not fully exert the environmental and economic advantages of blending. The current control systems and algorithms still need to be optimized in terms of dynamically adjusting the blending ratio to cope with fuel changes. More specifically, as an important global power supply method, thermal power has always been the focus of industry attention in terms of fuel costs and environmental issues. Coal blending technology aims to balance fuel costs and combustion efficiency while reducing harmful gas emissions by mixing different types of coal or other fuels. However, how to accurately control the blending ratio to achieve optimal combustion efficiency and environmental benefits is a complex technical challenge. Based on this, an embodiment of the present invention first provides a method for analyzing the coal blending ratio in a thermal power plant, such as Figure 1 As shown, including:

[0057] Step S1, collecting the physical and chemical parameters of the fuel to be blended, wherein the physical and chemical parameters include calorific value, volatile matter, ash content, and moisture content;

[0058] In this step, it is first required to collect and measure the physical and chemical properties of all planned fuels, including but not limited to calorific value (unit energy release), volatile matter (key factor affecting combustion rate), ash (proportion of useless substances remaining after combustion) and moisture content (affecting combustion efficiency). For example, the calorific value of a low-sulfur coal collected is 5000 kcal / kg, the volatile matter is 25%, the ash content is 15%, and the moisture content is 8%.

[0059] Example: Comparing the parameters of base coal (such as bituminous coal) and biomass fuel, it is found that biomass fuel has a lower calorific value (about 4000 kcal / kg) but a high volatile matter, which suggests that adjustments may be needed during blending to maintain a stable total calorific value.

[0060] Step S2, determining fuel characteristic data according to the basic coal type and the standard combustion parameters corresponding to each coal type and the physical and chemical parameters of the fuel to be blended;

[0061] In this step, a comprehensive fuel property database is formed based on the basic coal type and its standard combustion parameters (such as theoretical combustion temperature, combustion product composition, etc.), combined with the physical and chemical parameters of the fuel to be blended. For example, the basic data of bituminous coal and the characteristics of biomass fuel are combined to evaluate the expected combustion characteristics of the mixed fuel through a mathematical model.

[0062] Step S3, preliminarily setting the blending ratio range using a multi-objective optimization algorithm based on the fuel characteristic data;

[0063] In this step, a series of possible blending ratios are initially set using methods such as multi-objective genetic algorithms or particle swarm optimization, under multiple objectives such as ensuring combustion efficiency, reducing pollutant emissions, and controlling costs. For example, the algorithm calculates that the optimal blending ratio of bituminous coal and biomass fuel is between 60% and 80% bituminous coal, in order to maintain combustion efficiency while reducing SOx emissions.

[0064] Step S4, performing a small-scale blending test based on the blending ratio range and collecting test data;

[0065] In this step, under laboratory conditions, several typical ratios recommended by the algorithm are selected for small-scale blending tests, and key indicators such as combustion efficiency and emission components (such as NOx and SOx concentrations) are recorded. For example, a test is conducted with a ratio of 70% bituminous coal and 30% biomass, and the combustion temperature and emission data are observed.

[0066] Step S5, analyzing the test data and adjusting the blending ratio to optimize the preset combustion model for analyzing the coal blending ratio.

[0067] In this step, the test data is analyzed, the theoretical prediction is compared with the actual effect, the blending ratio is adjusted, and the combustion model is further optimized. The experimental results are input into the model through the feedback mechanism, and iterative optimization is performed until the most suitable blending scheme is found. For example, it is found that SOx emissions are better than expected at a ratio of 70%, but the combustion efficiency is slightly reduced, so it can be fine-tuned to 72% for further verification.

[0068] The embodiments of the present invention maximize the use of fuel energy, improve power generation efficiency, reduce emissions of harmful gases such as NOx and SOx, meet stricter environmental protection standards, balance fuel costs and combustion benefits, reduce overall operating costs, and enhance the market competitiveness of thermal power plants through precise blending and combustion ratio control.

[0069] Furthermore, in another embodiment provided by the present invention, the step of preliminarily setting the range of the blending ratio by using a multi-objective optimization algorithm is further refined. The specific operation process is as follows, to ensure that both environmental protection regulations and economic benefits are met, and the step of preliminarily setting the range of the blending ratio by using a multi-objective optimization algorithm according to the fuel characteristic data includes:

[0070] Calculate the theoretical heat of combustion of fuel mixtures;

[0071] This step first considers the impact of different fuel ratios on the calorific value of the overall mixed fuel through a calculation model based on the calorific value data collected for various types of fuels. When calculating, factors such as the calorific value and volatile matter of the fuel need to be considered to estimate the total energy that can be released by the combustion of the fuel mixture at a specific blending ratio. For example, if the calorific value of the base coal is 25MJ / kg and the calorific value of the blended fuel (assuming it is biomass) is 18MJ / kg, the theoretical calorific value range after mixing in different ratios is calculated through linear interpolation or more complex thermodynamic models.

[0072] Evaluating the emission potential of a fuel mixture, including predicting NOx and SOx generation;

[0073] In this step, the environmental chemistry model is used to predict the NOx (nitrogen oxides) and SOx (sulfur oxides) emissions generated when the fuel is mixed and burned based on the physicochemical properties of the fuel (such as sulfur content, nitrogen content, etc.). This step is crucial for controlling air pollution and requires considering the effects of factors such as temperature, oxygen concentration, and combustion time on emissions. For example, through a chemical reaction kinetic model, the fuel components at different proportions are input to simulate the expected generation concentrations of NOx and SOx.

[0074] The upper and lower limits of the blending ratio are determined based on preset environmental protection standards and economic benefits.

[0075] In this step, the optimization range of the blending ratio can be set in combination with the national or regional environmental protection regulations on NOx and SOx emission limits and the economic efficiency goals of the thermal power plant itself (such as cost minimization and power generation efficiency maximization). For example, if the environmental protection standard requires SOx emissions not to exceed 100mg / m 3 The economic benefit analysis shows that when the blending ratio is between 70% bituminous coal and 30% biomass, it can meet the emission requirements and maintain good economy. Then this ratio range will become the upper and lower limits of the preferred blending range.

[0076] The embodiment of the present invention not only ensures the combustion efficiency of the fuel mixture, but also strictly controls pollutant emissions, achieving dual optimization of environmental protection and economic benefits. This methodology provides scientific decision-making support for thermal power plants, helping them to achieve the best operating state by carefully managing the fuel blending ratio while complying with environmental protection regulations.

[0077] In another embodiment provided by the present invention, detailed steps of theoretical basis are provided, and steps of conducting a small-scale blending test based on the blending ratio range and collecting test data include:

[0078] Design a pretreatment scheme for the blended fuel to improve the fuel mixing uniformity;

[0079] In this step, in order to improve the uniformity and combustion efficiency of the fuel mixture, a fuel pretreatment solution needs to be designed in advance. This may include physical treatment processes such as crushing, screening, drying, and possible use of additives to improve the compatibility and reactivity of the fuel. For example, coal particles may need to be crushed to a certain particle size, and biomass fuels need to be dried to reduce moisture. Before the two are mixed, combustion aids can be added to promote uniform mixing and complete combustion.

[0080] Collecting processing information after physical pre-treatment of the fuel;

[0081] In this step, after the pretreatment of the fuel is completed, it is necessary to systematically record and analyze the characteristics of the treated fuel, including but not limited to particle size distribution, moisture content, ash ratio, volatile matter change, etc. This information is crucial to verify the effectiveness of the pretreatment scheme and is the basic data for subsequent experimental design. It can be measured by laboratory testing equipment such as particle size analyzers, moisture meters, etc.

[0082] The processing information is mixed according to a preliminarily set ratio to obtain test data.

[0083] In this step, the collected information is used to accurately proportion the fuel mixture according to the blending ratio range previously set by the multi-objective optimization algorithm. Small-scale combustion tests are carried out, during which key parameters such as combustion temperature, combustion efficiency, pollutant emission levels (NOx, SOx, etc.), heat output, etc. need to be monitored and recorded. These test data have direct guiding significance for verifying the accuracy of theoretical calculations and adjusting and optimizing blending strategies.

[0084] The embodiment of the present invention verifies the feasibility of the theoretically calculated blending ratio through actual small-scale tests, and provides valuable practical experience, which helps to further fine-tune the blending ratio and optimize the fuel processing process to ensure that the blending scheme is both environmentally friendly and efficient and economically feasible before large-scale application. Through feedback iteration, a stable and efficient coal blending operation guidance scheme can eventually be formed.

[0085] In another embodiment of the present invention, the test data is analyzed and the blending ratio is adjusted to optimize the preset combustion model. The step of analyzing the coal blending ratio includes:

[0086] Establishing a combustion model of the fuel mixture dynamics, the combustion model taking into account oxygen concentration, temperature and other factors that affect combustion efficiency;

[0087] In this step, advanced thermodynamics and chemical kinetics principles are used to build a dynamic combustion model that can reflect the combustion behavior of the fuel mixture under different oxygen concentrations, temperatures, and other key operating parameters (such as pressure and turbulence intensity). The model should take into account changes in the physical and chemical properties of the fuel, such as pyrolysis, volatile release, coke combustion, and other stages to ensure the accuracy and practicality of the model.

[0088] Apply machine learning algorithms to iteratively optimize model parameters based on test data feedback;

[0089] In this step, machine learning techniques such as neural networks, support vector machines or reinforcement learning are introduced to iteratively optimize the parameters of the combustion model based on the collected test data (including combustion efficiency, pollutant emissions, temperature distribution, etc.). The algorithm needs to be able to identify and learn the complex nonlinear relationship between different blending ratios, combustion conditions and performance indicators, and gradually adjust the model parameters to approach the ideal combustion state.

[0090] Adjust the blending ratio to the optimal state to achieve the predetermined combustion efficiency and emission standards.

[0091] In this step, based on the prediction of the optimized combustion model, the fuel blending ratio is continuously adjusted to find the optimal ratio configuration that can maximize the combustion efficiency while ensuring that the pollutant emissions are lower than the specified standards. This process may involve multiple iterations, and the combustion effect needs to be re-evaluated after each adjustment until the predetermined combustion efficiency and emission standards are achieved. For example, through algorithm guidance, it may be found that the highest combustion efficiency and the lowest SOx emissions can be achieved when the blending ratio of bituminous coal and biomass fuel is adjusted to 65:35.

[0092] Through the implementation of the embodiments of the present invention, thermal power plants can continuously optimize the coal blending process in the close interaction between theory and practice, which not only improves energy utilization efficiency and reduces operating costs, but more importantly, significantly reduces environmental pollution and meets the requirements of sustainable development. The application of machine learning algorithms makes the method highly adaptable and optimized, and can quickly adjust strategies as external conditions change (such as changes in fuel quality and upgrades in environmental protection standards) to maintain the optimal blending effect.

[0093] Specifically, the steps of applying a machine learning algorithm to iteratively optimize model parameters based on test data feedback include:

[0094] Define the state space, including the current blending ratio, combustion efficiency index, and pollutant emission level;

[0095] In this step, the state space contains all the key variables that affect the efficiency and environmental performance of coal blending. Specifically, it includes but is not limited to the current blending ratio (for example, the ratio of coal to biomass), combustion efficiency indicators (such as thermal efficiency, energy utilization), and pollutant emission levels (NOx, SOx concentration, dust emissions, etc.). These state variables reflect the real-time status of the combustion process and are the basic information for the machine learning algorithm to make decisions.

[0096] Set the action space to determine the adjustment range of the blending ratio;

[0097] In this step, the action space defines all possible actions that the algorithm can take, that is, specific adjustments to the blending ratio. This may include, but is not limited to, increasing or decreasing the proportion of a certain fuel, changing the pretreatment method of the fuel, etc. When setting the action space, it is necessary to consider the feasibility and economy of the actual operation to ensure that the proposed adjustment plan is both reasonable and easy to implement. For example, the action space can be set so that the blending ratio fluctuates within the range of ±5% to explore a better blending ratio.

[0098] A reward function is designed to adjust parameters of the optimization model based on the test data.

[0099] In this step, the reward function is an evaluation criterion used to measure the degree of improvement in system performance caused by the actions proposed by the algorithm. It needs to comprehensively consider factors such as economic benefits (such as fuel cost savings), environmental benefits (pollutant reduction effects) and combustion stability. For example, the reward function can be defined as:

[0100] [R = α\times (economic benefits) - β\times (NOx emissions) - γ\times (SOx emissions) + δ\times (combustion stability)];

[0101] Among them, α, β, γ, and δ are weight coefficients adjusted according to actual conditions, reflecting the importance of different objectives respectively.

[0102] The machine learning algorithm provided by the above steps can continuously pursue higher combustion efficiency and lower pollutant emissions by intelligently adjusting the blending ratio based on clearly defined state space and action space, while ensuring the economy and stability of the combustion process. The design of the reward function ensures that the optimization direction is consistent with the comprehensive goals of the thermal power plant, and promotes the intelligence and optimization of the coal blending process. This dynamic adjustment strategy enables thermal power plants to respond quickly to complex changes in operating conditions, continuously optimize operating strategies, and ultimately achieve a win-win situation of energy conservation, emission reduction and economic benefits.

[0103] The step of designing the reward function specifically includes:

[0104] Determine weight factors, where each factor is dynamically adjusted based on the operational strategy;

[0105] In this step, the weight factors (such as α, β, γ, etc.) are set according to the operation strategy of the thermal power plant and the changes in the external environment. For example, when environmental protection policies are tightened, the weight of the pollutant reduction factor may be increased, and when fuel costs rise, the proportion of the economic benefit factor may be increased to ensure that the reward function can flexibly reflect the company's current top priority.

[0106] Calculate economic benefit indicators;

[0107] In this step, economic benefit indicators include but are not limited to fuel cost savings, benefits from improved power generation efficiency, etc. By comparing the total fuel cost and power generation benefits under different blending ratios, the incremental economic benefits are calculated as an important part of the reward function.

[0108] Evaluate the effectiveness of pollutant emission reduction;

[0109] In this step, based on test data and model predictions, the reduction in emissions of pollutants such as NOx, SOx, and dust caused by the blending strategy is quantified, and its contribution to environmental improvement is considered. This step may involve statistical analysis of emission data and comparison with environmental standards to ensure accurate assessment of emission reduction effects.

[0110] Calculating a reward value based on the weight factor, the economic benefit index, and the pollutant emission reduction effect, wherein the reward value reflects the comprehensive performance of the current blending strategy in terms of economic benefit and pollutant emission reduction effect;

[0111] In this step, according to the determined weight factors (α, β, γ, etc.), economic benefit indicators and pollutant emission reduction effects, the formula is applied:

[0112] [R = α\times economic benefit index - β\times NOx emission reduction - γ\times SOx emission reduction + ...];

[0113] A comprehensive reward value is calculated, which directly reflects the comprehensive benefits of the current blending strategy. The larger the value, the better the strategy.

[0114] The calculated reward value is fed back to the machine learning algorithm as a basis for optimizing the model parameters and iterating the blending ratio parameters.

[0115] In this step, the calculated reward value is used as a feedback signal and input into the machine learning algorithm. The algorithm automatically adjusts the blending ratio parameters according to the size of the reward value, hoping to obtain a higher reward value in the next iteration. Through continuous iterations, the algorithm gradually optimizes the blending strategy to maximize economic and environmental benefits.

[0116] The embodiment of the present invention can more scientifically and accurately balance economic benefits and environmental protection responsibilities in the process of optimizing the coal blending ratio in thermal power plants. The design of the reward function and the feedback mechanism ensure that the machine learning algorithm continuously searches for and executes the optimal blending strategy in a dynamically changing operating environment, which not only improves fuel utilization efficiency, but also significantly reduces environmental pollution, achieving a win-win situation for economic and environmental benefits.

[0117] In addition, the step of setting the action space includes:

[0118] Define the minimum step length for adjusting the blending ratio to ensure the accuracy and feasibility of the adjustment;

[0119] In this step, ensure that each adjustment is a refined operation based on technical feasibility and economic rationality. The setting of the minimum step size needs to take into account the response characteristics of the combustion equipment, the fuel mixing uniformity requirements and the accuracy limitations of the monitoring system. For example, if the minimum coal type ratio change for stable operation of the equipment is 1%, the step size should not be less than this value to prevent the adjustment from being too fine, which will increase the control difficulty or reduce the economy.

[0120] Determine the exploration scope of the blending ratio adjustment, and the exploration scope is dynamically adjusted according to the historical optimal blending ratio, the current blending test feedback and the changes in environmental protection and economic policies;

[0121] In this step, the exploration scope should not be fixed, but should be flexibly adjusted based on three factors: "historical optimal blending ratio", "current blending test feedback" and "environmental and economic policy changes". The historical optimal ratio provides a verified high-efficiency range reference; the current test feedback helps to capture the changing trend of the blending effect in a timely manner; and policy changes ensure that the strategy adapts to the latest environmental requirements and economic incentives.

[0122] Combined with the current blending ratio in the state space, a specific blending ratio adjustment sequence is generated as a set of optional actions in the iterative process of the machine learning algorithm.

[0123] In this step, combined with the blending ratio recorded in the current state space (i.e., the ratio currently actually executed), a series of possible blending ratio adjustment schemes are systematically generated according to the minimum step size and exploration range. These schemes constitute a set of actions in the iterative process of the machine learning algorithm, and each action represents an attempt to adjust the current blending ratio. For example, if the current blending ratio is A%, the minimum step size is 1%, and the upper and lower limits of the exploration range are [A-5%, A+5%], then the possible action set contains all legal ratios from [A-5%, A-4%,..., A, A+1%,..., A+5%].

[0124] Through the above steps, the setting of the action space not only ensures the comprehensiveness of the exploration, but also maintains the rationality and efficiency of the adjustment, provides a wealth of decision-making options for subsequent machine learning algorithms, and promotes the continuous optimization and innovation of the blending strategy. This methodology of dynamic adjustment and strategy exploration is the key to achieving efficient and environmentally friendly operation of thermal power plants.

[0125] In practical applications, the steps of setting the action space are explained in detail through a specific example:

[0126] Assume that a thermal power plant is using coal blending technology to optimize costs and reduce pollutant emissions. Currently, the main fuels are ordinary coal (labeled as Coal A) and a low-sulfur, low-cost alternative fuel (labeled as Fuel B). The goal is to achieve the best balance between economic benefits and environmental protection requirements by adjusting the blending ratio of these two fuels.

[0127] Step 1: Define the minimum step length for adjusting the blending ratio;

[0128] Based on the characteristics of the plant equipment and previous operating experience, the minimum step size is determined to be 2%, which means that each adjustment of the blending ratio of FuelB increases or decreases by at least 2%.

[0129] Step 2: Determine the exploration scope of adjusting the blending ratio;

[0130] Historically optimal blending ratio: Historical data shows that when the blending ratio of Fuel B is 30%, the economic and environmental indicators are optimal.

[0131] Feedback from current blending tests: The latest test shows that increasing the proportion of Fuel B to 33% will further reduce costs, but SO 2 Emissions rose slightly.

[0132] Changes in environmental economic policies: New environmental policies require SO 2 Emissions need to be reduced by 5% from current levels.

[0133] In summary, the scope of exploration should be adjusted to ensure SO 2 Explore cost savings potential while meeting emission standards. Taking into account the historical best and latest environmental requirements, the exploration range is set to [Fuel B 28%, Fuel B 35%] to ensure that we can try to reduce costs while meeting stricter environmental standards.

[0134] Step 3: Generate a specific blending ratio adjustment sequence;

[0135] The current blending ratio is set to Fuel B 30%.

[0136] Based on the minimum step size of 2% and the exploration range [Fuel B 28%, Fuel B 35%], the generated blending ratio adjustment sequence is as follows:

[0137] Fuel B 28% (adjust the lowest point downward)

[0138] Fuel B 30% (current ratio, can also be considered as one of the adjustment options)

[0139] Fuel B 32%

[0140] Fuel B 34% (adjusted upwards towards the upper limit, excluding 35% to keep the step length consistent)

[0141] The above sequence constitutes a series of specific action options, which can be used to guide the subsequent blending ratio adjustment experiment. These actions are evaluated and selected through machine learning algorithms (such as reinforcement learning). After each round of experiments, the state space is updated according to the feedback data (such as cost, emissions, etc.), and the exploration scope is dynamically adjusted and a new action sequence is generated accordingly, forming a closed-loop optimization process.

[0142] This example shows how to translate theoretical steps into practical operations to ensure that the blending strategy can be effectively and efficiently implemented under complex and changing conditions.

[0143] The embodiment of the present invention provides a system for analyzing the proportion of coal blending in a thermal power plant. Figure 2 As shown, including:

[0144] A collection module, used to collect the physical and chemical parameters of the fuel to be blended, wherein the physical and chemical parameters include calorific value, volatile matter, ash content, and moisture content;

[0145] A blending module, used to determine fuel characteristic data based on the basic coal type and the standard combustion parameters corresponding to each coal type and the physical and chemical parameters of the fuel to be blended;

[0146] A determination module is used to preliminarily set the blending ratio range based on fuel characteristic data using a multi-objective optimization algorithm;

[0147] A test module, used to conduct a small-scale blending test based on the blending ratio range and collect test data;

[0148] The analysis module is used to analyze the test data and adjust the blending ratio to optimize the preset combustion model and analyze the coal blending ratio.

[0149] An embodiment of the present invention further provides an electronic device, including:

[0150] processor;

[0151] a memory for storing processor-executable instructions;

[0152] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0153] An embodiment of the present invention further provides a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0154] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing the proportion of coal blending in a thermal power plant, characterized in that: include: Collecting the physical and chemical parameters of the fuel to be blended, wherein the physical and chemical parameters include calorific value, volatile matter, ash content, and moisture content; Determine fuel characteristic data based on basic coal types and standard combustion parameters corresponding to each coal type and physicochemical parameters of the fuel to be blended; According to the fuel characteristic data, a multi-objective optimization algorithm is used to preliminarily set the blending ratio range; Conduct a small-scale blending test based on the blending ratio range and collect test data; Analyze test data and adjust the blending ratio to optimize the preset combustion model for analyzing coal blending ratio.

2. The method for analyzing the proportion of coal blending in a thermal power plant according to claim 1, characterized in that: According to the fuel characteristic data, the steps of preliminarily setting the blending ratio range using a multi-objective optimization algorithm include: Calculate the theoretical heat of combustion of fuel mixtures; Evaluating the emission potential of a fuel mixture, including predicting NOx and SOx generation; The upper and lower limits of the blending ratio are determined based on preset environmental protection standards and economic benefits.

3. The method for analyzing the proportion of coal blending in a thermal power plant according to claim 1, characterized in that: The steps of conducting a small-scale blending test based on the blending ratio range and collecting test data include: Design a pretreatment scheme for the blended fuel to improve the fuel mixing uniformity; Collecting processing information after physical pre-treatment of the fuel; The processing information is mixed according to a preliminarily set ratio to obtain test data.

4. The method for analyzing the proportion of coal blending in a thermal power plant according to claim 1, characterized in that: Analyze test data and adjust the blending ratio to optimize the preset combustion model. The steps for analyzing the coal blending ratio include: Establishing a combustion model of the fuel mixture dynamics, the combustion model taking into account oxygen concentration, temperature and other factors that affect combustion efficiency; Apply machine learning algorithms to iteratively optimize model parameters based on test data feedback; Adjust the blending ratio to the optimal state to achieve the predetermined combustion efficiency and emission standards.

5. A method for analyzing the proportion of coal blending in a thermal power plant according to claim 4, characterized in that: The steps of applying machine learning algorithms to iteratively optimize model parameters based on test data feedback include: Define the state space, including the current blending ratio, combustion efficiency index, and pollutant emission level; Set the action space to determine the adjustment range of the blending ratio; A reward function is designed to adjust parameters of the optimization model based on the test data.

6. A method for analyzing the proportion of coal blending in a thermal power plant according to claim 5, characterized in that: The steps of designing the reward function specifically include: Determine weight factors, where each factor is dynamically adjusted based on the operational strategy; Calculate economic benefit indicators; Evaluate the effectiveness of pollutant emission reduction; Calculating a reward value based on the weight factor, the economic benefit index, and the pollutant emission reduction effect, wherein the reward value reflects the comprehensive performance of the current blending strategy in terms of economic benefit and pollutant emission reduction effect; The calculated reward value is fed back to the machine learning algorithm as a basis for optimizing the model parameters and iterating the blending ratio parameters.

7. The method for analyzing the proportion of coal blending in a thermal power plant according to claim 5, characterized in that: The step of setting the action space includes: Define the minimum step length for adjusting the blending ratio to ensure the accuracy and feasibility of the adjustment; Determine the exploration scope of the blending ratio adjustment, and the exploration scope is dynamically adjusted according to the historical optimal blending ratio, the current blending test feedback and the changes in environmental protection and economic policies; Combined with the current blending ratio in the state space, a specific blending ratio adjustment sequence is generated as a set of optional actions in the iterative process of the machine learning algorithm.

8. A coal blending ratio analysis system for a thermal power plant, characterized in that: include: A collection module, used to collect the physical and chemical parameters of the fuel to be blended, wherein the physical and chemical parameters include calorific value, volatile matter, ash content, and moisture content; A blending module, used to determine fuel characteristic data based on the basic coal type and the standard combustion parameters corresponding to each coal type and the physical and chemical parameters of the fuel to be blended; A determination module is used to preliminarily set the blending ratio range based on fuel characteristic data using a multi-objective optimization algorithm; A test module, used to conduct a small-scale blending test based on the blending ratio range and collect test data; The analysis module is used to analyze the test data and adjust the blending ratio to optimize the preset combustion model and analyze the coal blending ratio.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images