Intelligent raw coal bunker separation system

Abstract

The invention provides an intelligent raw coal bunker separation system which comprises a plurality of independent storage bins, a plurality of lower insertion plate doors, a dynamic mixing device, a data acquisition device and an intelligent control device, and the intelligent control device controls the distribution of a temperature field according to the distribution of the temperature field, the smoke concentration, the output load of a boiler and current environmental protection requirement information. Generating a current demand matrix of the boiler; matching the coal quality data of each storage bin with the current demand matrix to obtain a plurality of target coal types meeting boiler demands; and based on the current demand matrix, solving a multivariable optimization equation to obtain an optimal feeding ratio of the plurality of target coal types so as to control the lower plugboard gates corresponding to the plurality of target coal types based on the optimal feeding ratio. Therefore, fuel storage flexibility can be improved, boiler combustion efficiency can be improved, pollutant emission can be reduced, and coal consumption cost can be optimized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal-fired boiler fuel supply, and particularly relates to an intelligent raw coal bunker sub-bunker system. BACKGROUND

[0002] In the field of thermal power generation, industrial boilers, etc., the storage and distribution of raw coal are key links affecting the combustion efficiency, pollutant emissions and operating costs. Traditional raw coal bunker systems mostly use single storage bunkers or simple sub-bunker designs, which have problems such as extensive coal management, insufficient coal blending accuracy, etc., and are difficult to adapt to the dynamic operation requirements of boilers and strict environmental protection requirements.

[0003] Specifically, the traditional intelligent raw coal bunker sub-bunker system usually relies on manual experience for coal blending and cannot respond in real time to changes in boiler load, furnace temperature field fluctuations and flue gas composition. When the boiler output load is adjusted, if the parameters such as calorific value and volatile matter of the coal do not match the demand, it is easy to cause insufficient combustion, which not only reduces energy utilization efficiency, but also increases the emission of pollutants such as nitrogen oxides and sulfur dioxide, making it difficult to meet the strict upper limit of pollutant emissions required by current environmental protection regulations. At the same time, single coal or fixed proportion coal blending mode is difficult to balance the stability of combustion and cost control. For example, high-volatile coal is beneficial to ignition but has high cost, and high-ash coal may cause problems such as furnace slagging and heating surface wear, which seriously affect the safe and stable operation of the boiler.

[0004] In addition, with the diversification of coal resources and the increasing environmental protection requirements, there are higher requirements for accurate proportioning and dynamic adjustment of different coal types. The traditional system lacks real-time collection and intelligent analysis capabilities for coal quality data and boiler operating parameters, and cannot optimize the proportioning of multiple coal types according to the current dynamic requirements of the boiler (such as target calorific value, pollutant emission limit, etc.), resulting in high coal consumption costs and difficulty in achieving coordinated optimization of combustion efficiency and environmental performance.

[0005] Therefore, it is of great significance to develop an intelligent raw coal bunker sub-bunker system capable of realizing accurate storage, real-time monitoring and intelligent proportioning of multiple coal types, to improve the combustion efficiency of the boiler, reduce pollutant emissions and optimize the consumption cost of coal types. SUMMARY

[0006] The present application provides an intelligent raw coal bunker sub-bunker system to improve the combustion efficiency of the boiler, reduce pollutant emissions and optimize the consumption cost of coal types.

[0007] In one aspect, the present application provides an intelligent raw coal bunker sub-bunker system, which comprises: a plurality of independent storage bunkers, each storage bunker being used to store a specific coal type; a plurality of lower plug gate doors, each lower plug gate door being arranged at the bottom of each storage bunker and being used to control the discharge amount of each coal type; A dynamic mixing device is used for mixing different coal types to obtain mixed coal types. A data acquisition device is used for acquiring coal quality data of each storage bin, temperature field distribution in a boiler furnace, flue gas concentration, and output load of the boiler. An intelligent control device comprises: A generation module is configured to generate a current demand matrix of the boiler according to the temperature field distribution, the flue gas concentration, the output load of the boiler, and current environmental protection requirement information. A screening module is configured to match the coal quality data of each storage bin with the current demand matrix to obtain a plurality of target coal types meeting the demand of the boiler. An optimization control module is configured to solve a multivariable optimization equation based on the current demand matrix to obtain an optimal supply ratio of the plurality of target coal types, so as to control the corresponding under-inserting plate door of the plurality of target coal types based on the optimal supply ratio, wherein the multivariable optimization equation is an equation constructed with the highest combustion efficiency, the lowest pollutant emission, and the optimal coal consumption cost as the target.

[0008] According to the intelligent raw coal bin sub-bank system provided by the application, the screening module is specifically configured to: The coal quality data of each storage bin is normalized to obtain a plurality of coal quality parameters of each storage bin. A multi-feature similarity algorithm is used to determine a first similarity of each coal quality parameter and a corresponding demand index in the current demand matrix, and a comprehensive similarity score is obtained by weighted summation, wherein the weight of each coal quality parameter is determined according to the working condition demand. The coal type corresponding to the comprehensive similarity score greater than the first preset similarity is selected as the target coal type.

[0009] According to the intelligent raw coal bin sub-bank system provided by the application, the screening module is further configured to: If there is only one historical demand matrix corresponding to the plurality of target coal types, a second similarity of the only one historical demand matrix and the current demand matrix is calculated. If the second similarity is greater than a second preset similarity, a historical ratio of the plurality of target coal types corresponding to the only one historical demand matrix is taken as an initial ratio, wherein the initial ratio is taken as a starting point for solving the multivariable optimization equation. If the second similarity is less than or equal to the second preset similarity, a difference value of each demand index between the only one historical demand matrix and the current demand matrix is determined, a proportion adjustment coefficient of each target coal type is determined according to the difference value, and an initial ratio is obtained according to the proportion adjustment coefficient and the historical ratio of the plurality of target coal types.

[0010] According to the intelligent raw coal bunker warehouse system provided by the application, each demand index corresponds to different coal quality parameters, and the screening module is further used for: According to the difference value, demand adjustment information for each coal quality parameter is determined; the demand adjustment information includes an adjustment direction and an adjustment amplitude of each coal quality parameter; The contribution degree of each target coal type to each coal quality parameter under the historical proportion of the plurality of target coal types is determined; According to the demand adjustment information and the contribution degree, a proportion adjustment coefficient of each target coal type is determined.

[0011] According to the intelligent raw coal bunker warehouse system provided by the application, the screening module is further used for: The product of the contribution degree of the jth coal quality parameter in the ith target coal type and the adjustment amplitude of the jth coal quality parameter is calculated, and the sum of the product after being given a positive or negative sign according to the adjustment direction of the jth coal quality parameter is added, and the product of the weight of the jth coal quality parameter and the sum value is obtained to obtain an influence factor of the ith target coal type; The ratio of the influence factor of the ith target coal type to the influence factors of all target coal types is taken as the proportion adjustment coefficient of the ith target coal type; Wherein, i=1, 2, …, N, N is the number of the plurality of target coal types, j=1, 2, …, M, M is the number of the coal quality parameters.

[0012] According to the intelligent raw coal bunker warehouse system provided by the application, the screening module is further used for: The product of the content of the jth coal quality parameter in the ith target coal type and the historical proportion of the ith target coal type is calculated to obtain an actual contribution amount of the jth coal quality parameter in the ith target coal type; The ratio of the actual contribution amount of the jth coal quality parameter to all actual contribution amounts of the jth coal quality parameter is taken as the contribution degree of the jth coal quality parameter.

[0013] According to the intelligent raw coal bunker warehouse system provided by the application, the screening module is further used for: If there are a plurality of historical demand matrices corresponding to the plurality of target coal types, a third similarity between the plurality of historical demand matrices and the current demand matrix is calculated; If the maximum third similarity is greater than a second preset similarity, the historical proportion of the plurality of target coal types corresponding to the historical demand matrix of the maximum third similarity is taken as the initial proportion; If the maximum third similarity is less than or equal to the second preset similarity, the historical proportions of the plurality of target coal types of all historical demand matrices are weightedly averaged to obtain an average proportion value as the initial proportion.

[0014] The data acquisition device is further used for collecting granularity distribution of the mixed coal and density uniformity of the mixed coal. The dynamic mixing device is further used for: According to the granularity distribution data and the density uniformity data, and in combination with a preset mixing uniformity model, the stirring rotating speed, the stirring blade angle and the mixing time of the dynamic mixing device are automatically adjusted, so that the uniformity of the mixed coal reaches a preset uniformity.

[0015] The intelligent control device is further used for: After the mixing is stopped, according to the granularity distribution data and the density uniformity data, and in combination with a preset mixing uniformity model, if it is detected that the uniformity of the mixed coal does not reach the preset uniformity, an alarm information is generated.

[0016] The intelligent control device further comprises: A man-machine interaction module is used for displaying operation information of the intelligent raw coal bunker division system, and receiving control information of an operator.

[0017] The intelligent raw coal bunker division system provided by the present application stores different coal in multiple independent storage bins, and combines a dynamic mixing device and an intelligent control device, so that the highest combustion efficiency, the lowest pollutant emission and the optimal coal consumption cost are taken as the target, the boiler demand is matched in real time and the feeding proportion is optimized, so that the flexibility of fuel storage is improved, the boiler combustion efficiency is improved, the pollutant emission is reduced and the coal consumption cost is optimized. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0019] Figure 1 FIG. 1 is a structural schematic diagram of an intelligent raw coal bunker division system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below in conjunction with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0021] In the prior art, the conventional coal-fired system usually uses a single coal bunker or simple sub-bunker to store coal, resulting in insufficient fuel flexibility. When the boiler operating conditions change, the coal ratio cannot be dynamically adjusted, the combustion efficiency is limited, and the pollutant emission is difficult to control. For example, when the boiler load fluctuates or the environmental protection standard upgrades, a single coal type cannot meet the real-time demand of parameters such as calorific value and sulfur content, causing unstable combustion or emission exceeding the standard.

[0022] In order to solve the above problems, by analyzing the relationship between the boiler operating data and the coal quality parameters, it is found that the complementary characteristics of different coal types can be utilized to achieve efficient combustion. Therefore, it is proposed to store multiple coal types independently, and dynamically control the feeding ratio through real-time data acquisition and optimization algorithm, so as to balance the combustion efficiency and environmental protection requirements.

[0023] Therefore, the present application provides an intelligent raw coal bunker sub-bunker system, Figure 1 is a structural schematic diagram of the intelligent raw coal bunker sub-bunker system provided by the embodiments of the present application. As Figure 1 shown, the intelligent raw coal bunker sub-bunker system of the present embodiment can include a system of multiple independent storage bins 11, each storage bin 11 stores a specific coal type, and a lower plug door 12 is provided at the bottom to control the discharge amount. The dynamic mixing device 13 mixes different coal types, and the data acquisition device 14 acquires coal quality data and boiler operating parameters. The intelligent control device 15 generates a demand matrix according to the boiler demand, selects matching coal types, and determines the optimal feeding ratio through an optimization equation to control the lower plug door 12 to achieve precise feeding.

[0024] Specifically, the intelligent control device 15 includes a generation module 151, a screening module 152, and an optimization control module 153.

[0025] The generation module 151 is configured to generate a current demand matrix of the boiler according to the temperature field distribution, the flue gas concentration, the output load of the boiler, and current environmental protection requirement information; The screening module 152 is configured to match the coal quality data of each storage bin 11 with the current demand matrix to obtain a plurality of target coal types that meet the demand of the boiler; An optimization control module 153 is configured to solve a multivariable optimization equation based on the current demand matrix to obtain optimal feeding proportions of the target coal types, so as to control the corresponding lower plug doors 12 of the target coal types based on the optimal feeding proportions.

[0026] In a specific implementation process, the independent storage bin 11 refers to a physically isolated storage unit that can prevent different coal types (such as anthracite, lignite, coke, etc.) from being mixed, for example, by using a partition wall plate or an independent bin body structure. The lower plug door 12 uses a gate mechanism with adjustable opening degree, for example, a plug-in plate structure driven by an electric push rod, to adjust the coal flow speed by controlling the opening degree. The dynamic mixing device 13 includes a stirring mechanism and a conveying channel, for example, a combination of a screw conveyor and an air flow mixer, to ensure uniform mixing of coal types. The data acquisition device 14 includes a coal quality analyzer and a boiler sensor, for example, a near-infrared coal quality detector and a thermocouple array, to obtain coal type characteristics and combustion state in real time. The intelligent control device 15 has an optimization algorithm module built-in, for example, a linear programming model, to determine the lowest cost ratio that meets the calorific value, sulfur content and other constraints by solving a multivariable equation.

[0027] Specifically, when the system is running, the coal types in each storage bin 11 are discharged in proportion through the lower plug door 12 and enter the dynamic mixing device 13 for physical mixing. The data acquisition device 14 continuously monitors the boiler furnace temperature, flue gas composition and load demand to generate a demand matrix containing parameters such as target calorific value and maximum sulfur content. The control device matches the coal quality data of the storage bin 11 with the demand matrix, selects the coal type combination that meets the conditions, and then solves an equation constructed with the highest combustion efficiency, the lowest pollutant emission and the optimal coal type consumption cost as the target to optimize the calculation and determine the optimal feeding proportion of each coal type. The lower plug door 12 adjusts the opening degree according to the proportion, so that the mixed coal type parameters can match the boiler demand in real time.

[0028] The highest combustion efficiency target refers to minimizing the unburned carbon, slag heat loss and exhaust heat loss, and improving the heat absorption efficiency of the boiler to the maximum extent while meeting the boiler load demand. The lowest pollutant emission target refers to minimizing the emission of harmful substances such as sulfur dioxide, nitrogen oxides, particulate matter and heavy metals in the flue gas while meeting the environmental protection requirements. The optimal coal type consumption cost target refers to minimizing the total fuel cost by optimizing the purchase price, transportation cost, storage cost and comprehensive utilization benefit of each coal type while meeting the aforementioned combustion efficiency and pollutant emission requirements. The specific construction function can be constructed according to actual needs, which will not be described here.

[0029] In a specific implementation, this invention can quickly respond to changes in boiler operating conditions through compartmentalized storage and dynamic proportioning. For example, when it is necessary to reduce sulfur emissions, the system automatically increases the proportion of low-sulfur coal while compensating for calorific value loss with high-calorific-value coal, avoiding the downtime adjustments caused by changing coal types in traditional methods.

[0030] Through the above technical solution, this invention achieves dynamic optimization of coal supply, solving the problem of low combustion efficiency caused by insufficient fuel flexibility. The synergistic blending of multiple coal types reduces pollutant emission concentrations, while the optimized control module 153 reduces the proportion of high-priced coal types used through cost equations, thereby lowering operating costs while meeting environmental protection requirements.

[0031] In some embodiments, the present invention further proposes a screening module 152 specifically used to normalize the coal quality data of each storage bin 11 to obtain multiple coal quality parameters, use a multi-feature similarity algorithm to determine the first similarity between each coal quality parameter and the corresponding demand index in the current demand matrix, obtain a comprehensive similarity score by weighted summation, and select the coal type corresponding to the comprehensive similarity score that is greater than the first preset similarity score as the target coal type.

[0032] Normalization refers to converting coal quality parameters with different dimensions into standardized values ​​with a unified dimension. This can be achieved using the range method or standard deviation method. For example, mapping calorific value parameters to the 0-1 interval and volatile matter parameters to the 0-10 interval eliminates the influence of unit differences on the calculation results. Multi-feature similarity algorithms quantify the degree of matching between coal quality parameters and demand indicators using mathematical methods. This can be achieved using cosine similarity or Euclidean distance algorithms. For example, the absolute value of the difference between the measured sulfur content and the maximum allowable sulfur content in the demand matrix can be used as a similarity index. The weights in the weighted summation are coefficients reflecting the importance of different coal quality parameters under operating conditions. These can be determined using the analytic hierarchy process (AHP) or entropy weighting method. For example, when environmental requirements are strict, the weight coefficient for sulfur content parameters can be increased to 0.3; when load fluctuations are severe, the weight coefficient for combustion stability indicators can be increased to 0.4.

[0033] Specifically, coal quality parameters include indicators such as calorific value, volatile matter, and sulfur content. These parameters are converted into values ​​within the 0-1 range using the range method. For each parameter, the matching degree between its measured value and the corresponding indicator in the demand matrix is ​​calculated. For example, the similarity between a measured sulfur content of 0.8% and the maximum allowable sulfur content of 1.0% is 0.8 / 1.0 = 0.8. The similarities of each parameter are multiplied by their corresponding weights and then summed. For example, the calorific value similarity is weighted at 0.7 × 0.2, and the volatile matter similarity at 0.9 × 0.1, to obtain a comprehensive score. When the comprehensive score exceeds a preset threshold, the coal type is deemed to meet boiler requirements. The weighting coefficients are dynamically adjusted according to the boiler's operating conditions. For example, under low-load conditions, the weight of the ash fusion point parameter is increased to 0.25 to ensure combustion stability.

[0034] This invention eliminates the difference in parameter dimensions through normalization, establishes a multi-dimensional similarity evaluation model, and dynamically adjusts parameter weights in conjunction with operating conditions, making the coal selection process more in line with actual operating requirements.

[0035] Through the above technical solutions, this invention achieves accurate matching of multi-dimensional coal quality parameters, solving the evaluation bias problem caused by differences in parameter dimensions in traditional methods. The dynamic weighting mechanism can adapt to changes in parameter importance under different operating conditions; for example, it automatically increases the evaluation weight of sulfur content during environmental inspections, ensuring that the selected coal types meet both combustion requirements and emission standards. The comprehensive similarity threshold determination mechanism avoids the misrepresentation of high-quality coal types due to a single parameter failing to meet standards, thus improving fuel utilization efficiency.

[0036] In some embodiments, the present invention further proposes that if there exists a unique historical demand matrix corresponding to multiple target coal types, a second similarity between the unique historical demand matrix and the current demand matrix is ​​calculated; if the second similarity is greater than a second preset similarity, the historical proportions of the multiple target coal types corresponding to the unique historical demand matrix are used as the initial proportions; if the second similarity is less than or equal to the second preset similarity, the difference value of each demand indicator in the unique historical demand matrix and the current demand matrix is ​​determined, the proportion adjustment coefficient of each target coal type is determined according to the difference value, and the initial proportions are obtained according to the proportion adjustment coefficients and the historical proportions of the multiple target coal types.

[0037] The historical demand matrix refers to the set of boiler demand parameters recorded by the system in past operating cycles, including indicators such as calorific value range, volatile matter range, and sulfur content upper limit. It can be stored in a database and indexed using a time series for rapid retrieval. The second similarity metric quantifies the degree of matching between historical and current demands, and can be calculated using Euclidean distance or cosine similarity algorithms. Weighting is dynamically adjusted based on boiler operating priorities. The difference value refers to the numerical deviation of the same demand indicator between the historical and current demand matrices. For example, if the sulfur content upper limit is adjusted from a historical value of 1.5% to a current value of 1.3%, the difference value is -0.2%, which can be quantified through difference calculation or percentage change. The proportional adjustment coefficient is used to correct historical proportions. Its generation process combines the contribution of each coal type to the demand indicator with the interaction of the difference value, and can be implemented through matrix operations or linear programming models.

[0038] Specifically, when the screening module 152 detects that there is a unique historical demand matrix corresponding to the current target coal type, it first calculates the similarity between the historical demand matrix and the current demand matrix. For example, the cosine similarity algorithm is used to calculate the similarity of the index vectors of the two matrices. If the result is higher than a preset threshold (e.g., 85%), the proportion data corresponding to the historical demand matrix is ​​directly used as the initial value for optimization calculation. If the similarity is lower than the threshold, the numerical differences of each demand index are compared item by item. For example, the current sulfur content upper limit is reduced by 0.2% compared to the historical value, and the calorific value requirement is increased by 200 kcal / kg. Subsequently, the system generates a proportion adjustment coefficient based on the direction of the difference (e.g., sulfur content needs to be reduced) and the magnitude (e.g., 0.2%), combined with the actual contribution of each coal type to sulfur content, calorific value, and other indicators under the historical proportion. For example, if a certain coal type has a high contribution to sulfur content in the historical proportion and the current sulfur content needs to be reduced, its adjustment coefficient will be assigned a negative value to reduce the proportion of that coal type. Finally, the initial proportion is obtained by linearly superimposing the historical proportion and the adjustment coefficient, providing a starting point closer to the optimal solution for subsequent optimization calculations.

[0039] This invention introduces a historical demand matrix similarity judgment mechanism, which can effectively identify historical data similar to the current operating conditions and use verified proportioning schemes as the starting point for optimization, significantly reducing the number of iterations. Simultaneously, when demand changes significantly, a dynamic adjustment mechanism driven by the difference value ensures that the initial proportions both retain historical operating experience and adapt to new demands, avoiding optimization failure due to excessive deviation in initial values.

[0040] Through the above technical solutions, this invention accelerates the solution process of multivariate optimization equations, shortening the response time for adjusting the feed ratio by approximately 30%-50%. It also improves the matching degree between the initial ratio and the optimal solution, reducing repeated iterations caused by unreasonable initial values ​​during the optimization process. Furthermore, the ratio adjustment mechanism, combining difference value analysis and contribution calculation, ensures that the system can quickly generate a suitable initial ratio when boiler demand parameters undergo local adjustments, avoiding a decrease in combustion efficiency or excessive pollutant emissions due to demand fluctuations.

[0041] In some embodiments, the present invention further proposes that when the screening module 152 determines the proportion adjustment coefficient of each target coal type based on the difference value, it can be implemented in the following manner: determining the demand adjustment information for each coal quality parameter based on the difference value, the demand adjustment information including the adjustment direction and adjustment range of each coal quality parameter; determining the contribution of each target coal type to each coal quality parameter under the historical proportion of multiple target coal types; and determining the proportion adjustment coefficient of each target coal type based on the demand adjustment information and the contribution.

[0042] Demand adjustment information refers to determining the direction and specific magnitude of adjustment for each coal quality parameter based on the difference between the current demand matrix and the historical demand matrix. Specifically, a difference algorithm can be used, for example, converting the sulfur content difference into a percentage adjustment, and using the sign to distinguish the direction of increase or decrease. This feature is used to quantify the adjustment targets for different coal quality parameters, providing a data foundation for proportional adjustments.

[0043] The contribution rate refers to the actual impact of each target coal type on the various coal quality parameters of the mixed coal under historical supply ratios. Specifically, a weighted contribution calculation method can be used; for example, multiplying the sulfur content of coal type A by its historical ratio yields the contribution ratio of that coal type to the total sulfur content of the mixed coal. This feature is used to assess the contribution weight of each coal type to the overall coal quality parameters, providing a basis for calculating the adjustment coefficient.

[0044] The proportion adjustment coefficient refers to the correction factor for the supply ratio of each coal type, calculated comprehensively based on the direction and contribution of demand adjustment. Specifically, a linear weighted algorithm can be used, for example, multiplying the contribution of a certain coal type to sulfur reduction with its adjustment magnitude, and then combining this with weight parameters to generate the adjustment coefficient. This feature links the demand adjustment target with the characteristics of the coal type, achieving scientific and reasonable proportion optimization.

[0045] Specifically, when a difference is detected between the current demand matrix and historical demand, the system first analyzes the adjustment requirements for each coal quality parameter. For example, if the maximum allowable sulfur content decreases by 0.2%, the system sets the sulfur content adjustment direction to negative and the adjustment magnitude to 0.2%. Next, it calculates the contribution of each target coal type to sulfur content under historical supply ratios; for example, coal type A contributes 60% of the total sulfur content, and coal type B contributes 40%. Based on the adjustment direction, the system applies greater adjustment pressure to coal types with higher contributions. Finally, a weighted adjustment coefficient is generated. For example, because coal type A has a high sulfur content contribution and needs to reduce sulfur content, its adjustment coefficient will guide a reduction in the supply ratio of that coal type.

[0046] This invention establishes a contribution calculation model that directly links coal type characteristics with dynamic demand, enabling precise, data-driven proportional adjustments. For example, in scenarios requiring simultaneous adjustments to sulfur content and calorific value, existing technologies struggle to balance the conflicting adjustments of different parameters, while this invention automatically generates the optimal adjustment strategy through contribution weight allocation.

[0047] Through the above technical solution, this invention can accurately calculate the adjustment amount of the feeding ratio of each type of coal according to real-time demand changes, effectively solving the problem of multi-parameter collaborative optimization. When boiler operating parameters or environmental protection requirements change dynamically, the system can automatically identify key adjustment parameters and generate a scientific adjustment plan by combining the historical contribution data of each type of coal, ensuring that the mixed coal always meets the requirements of multiple dimensions such as calorific value and pollutant emissions, avoiding the problem of response lag or parameter imbalance caused by manual adjustment.

[0048] In some embodiments, the present invention further proposes that when the screening module 152 determines the proportional adjustment coefficient of each target coal type according to the demand adjustment information and the contribution degree, it can be implemented in the following manner: calculate the product of the contribution degree of the j-th coal quality parameter in the i-th target coal type and the adjustment range of the j-th coal quality parameter, assign a positive or negative sign according to the adjustment direction of the j-th coal quality parameter, and add them together. Multiply the sum by the weight of the j-th coal quality parameter to obtain the influence factor of the i-th target coal type; use the ratio of the influence factor of the i-th target coal type to the influence factors of all target coal types as the proportional adjustment coefficient of the i-th target coal type; i=1,2,…,N, N is the number of multiple target coal types, j=1,2,…,M, M is the number of coal quality parameters.

[0049] Here, contribution refers to the proportion of the actual contribution of the target coal type to a specific coal quality parameter of the mixed coal type under historical ratios, relative to the total contribution. This can be achieved using an actual contribution ratio algorithm. This parameter quantifies the influence weight of different coal types on the mixed coal quality. Adjustment magnitude refers to the specific numerical change required for coal quality parameters in the current demand matrix. This can be achieved using a difference calculation method. This parameter determines the direction and intensity of coal type ratio adjustments. Weight refers to the importance coefficient of different coal quality parameters in comprehensive optimization. This can be dynamically adjusted using expert experience methods or machine learning models. This parameter reflects the priority differences of different environmental protection indicators or combustion performance indicators.

[0050] Specifically, during boiler operation, when it is detected that the sulfur content needs to be reduced and the calorific value needs to be increased in the current demand matrix, the screening module 152 first calculates the actual contribution of each target coal type to the sulfur content and calorific value under historical blending conditions. For example, coal type A contributed 57.1% of the mixed coal sulfur content in historical blending, while coal type B contributed 42.9%. Subsequently, the adjustment range of sulfur content is multiplied by the contribution, and a positive or negative sign is assigned according to the direction of adjustment; for example, a negative sign is assigned when the sulfur content needs to be reduced. Finally, the weighted calculation results of each coal quality parameter are summed to obtain the influence factor of the i-th target coal type, and the proportion adjustment coefficient of each coal type is obtained through normalization processing, thereby generating a new feed ratio. That is, the ratio of the influence factor of the i-th target coal type to the influence factors of all target coal types is used as the proportion adjustment coefficient of the i-th target coal type.

[0051] This invention establishes a mathematical correlation model to transform abstract requirements such as environmental protection indicators and combustion efficiency into calculable adjustment coefficients, thereby achieving precise dynamic optimization of coal type ratios.

[0052] Through the above technical solution, this invention can automatically adjust the feeding ratio of different coal types according to real-time environmental protection requirements and boiler operating conditions, effectively solving the problem of mixed coal quality parameters deviating from the target range. For example, when it is necessary to reduce sulfur emissions, the system automatically increases the feeding weight of low-sulfur coal types while maintaining stable calorific value, ensuring that both combustion efficiency and environmental protection indicators meet the standards.

[0053] In some embodiments, the present invention further proposes that when the screening module 152 determines the contribution of each target coal type to each coal quality parameter under the historical proportion of the plurality of target coal types, it can be implemented in the following manner: calculate the product of the content of the j-th coal quality parameter in the i-th target coal type and the historical proportion of the i-th target coal type to obtain the actual contribution of the j-th coal quality parameter in the i-th target coal type; and take the ratio of the actual contribution of the j-th coal quality parameter to all actual contributions of the j-th coal quality parameter as the contribution of the j-th coal quality parameter.

[0054] The actual contribution refers to the contribution of a single coal type to a specific coal quality parameter of the mixed coal under historical supply ratios. This can be calculated by multiplying the coal type content by the historical ratio, and is used to quantify the influence of each coal type on the final mixed coal quality. The contribution ratio refers to the proportion of the actual contribution of a single coal type to a specific coal quality parameter to the total contribution of all coal types to that parameter. This can be calculated by the actual contribution ratio, and is used to reflect the weighting relationship of each coal type in the mixing ratio. Coal quality parameters include key indicators such as calorific value, volatile matter, and sulfur content, which can be measured using coal quality testing instruments and are used to characterize the combustion characteristics and environmental performance of different coal types.

[0055] Specifically, when adjusting the parameters of a blended coal type, the contribution of each coal type to the target parameter is first calculated based on historical feed ratios. For example, to reduce sulfur content, the contributions of coal type A and coal type B to the total sulfur content under historical blending ratios are calculated separately. By comparing the contribution ratios of the two, the adjustment direction is determined. If coal type A has a higher sulfur contribution, its feed ratio needs to be reduced. The contribution calculation uses the ratio of the actual contribution to the total contribution to ensure that the adjustment is based on objective quantification. This process is achieved through matrix operations: the coal type content matrix is ​​multiplied by the historical ratio vector to obtain the actual contribution matrix, and then the column vectors of the matrix are normalized to obtain the contribution distribution.

[0056] This invention establishes a mathematical model of actual contribution amount and contribution degree, and accurately decomposes the adjustment requirements of mixing parameters to each coal type, thereby achieving scientific decision-making for blending optimization.

[0057] Through the above technical solution, this invention can accurately identify the influence weight of each coal type on the quality of the blended coal, and quickly determine the optimal adjustment direction when demand changes. By quantifying the contribution index, it avoids subjective errors from manual experience-based judgment, improving the accuracy and response speed of the blending ratio adjustment. This technical approach effectively solves the problem of insufficient dynamic adjustment capability of traditional coal blending systems, ensuring that the parameters of the blended coal always meet the boiler combustion requirements and environmental standards.

[0058] The following example illustrates the process of determining the contribution of each target coal type to each coal quality parameter under the historical proportions of the multiple target coal types; and determining the proportion adjustment coefficient for each target coal type based on the demand adjustment information and the contribution: Suppose that the current demand of a boiler is adjusted as follows: sulfur content decreases by 0.2% (adjustment amount ΔS=-0.2%), calorific value increases by 200 kcal / kg (adjustment amount ΔQ=+200), the target coal types are A and B, and the historical ratio is A:B=40%:60%.

[0059] Step 1: Calculate the actual contribution: Coal type A: Sulfur content 2%, calorific value 5000 kcal / kg; Contribution to sulfur: 40% × 2% = 0.8%; Contribution to calorific value: 40% × 5000 = 2000 kcal / kg.

[0060] Coal type B: Sulfur content 1%, calorific value 4000 kcal / kg; Contribution to sulfur: 60% × 1% = 0.6%; Contribution to calorific value: 60% × 4000 = 2400 kcal / kg.

[0061] Historical parameters of the blended coal: sulfur = 1.4%, calorific value = 4400 kcal / kg.

[0062] Step 2: Calculate the contribution (as a percentage of the total contribution): A's contribution to sulfur: 0.8% / 1.4%≈57.1%; B's contribution to sulfur: 0.6% / 1.4%≈42.9%.

[0063] A's contribution to heat generation: 2000 / 4400≈45.5%; B's contribution to heat generation: 2400 / 4400≈54.5%.

[0064] Step 3: Calculate the impact factors (assuming that the weights for sulfur and calorific value are both 0.5): Influencing factors of coal type A: (57.1%×(-0.2%) + 45.5%×200)×0.5 ≈ (-0.114% + 91)×0.5≈45.44 (Note: Standardized units are required in practice; this is simplified to a relative value).

[0065] Influencing factors of coal type B: (42.9%×(-0.2%) + 54.5%×200)×0.5≈(-0.086% + 109)×0.5≈54.46.

[0066] Step 4: Determine the adjustment coefficient (normalization): Total factor = 45.44 + 54.56 = 100; The adjustment factor for A is approximately 0.45 (45.44 / 100); the adjustment factor for B is approximately 0.55 (54.56 / 100).

[0067] In some embodiments, the present invention further proposes, when there are multiple historical demand matrices corresponding to multiple target coal types, to calculate the third similarity between the multiple historical demand matrices and the current demand matrix; if the maximum third similarity is greater than the second preset similarity, the historical proportion of the multiple target coal types corresponding to the historical demand matrix with the maximum third similarity is used as the initial proportion; if the maximum third similarity is less than or equal to the second preset similarity, the historical proportions of the multiple target coal types in all historical demand matrices are weighted and averaged to obtain the average proportion value as the initial proportion.

[0068] The third similarity refers to the degree of matching between the current demand matrix and the historical demand matrix. Specifically, it can be calculated using Euclidean distance or cosine similarity algorithms to determine the differences in various coal quality parameters. After normalization, a similarity score of 0-100% is obtained, which is used to select the closest historical operating conditions. The weighted average is a calculation method that linearly superimposes multiple historical proportions according to preset rules. Specifically, a time decay model or similarity association model can be used to determine the weight coefficients of each historical proportion, making recent data or highly similar data have a greater impact on the initial proportion.

[0069] Specifically, when the system detects multiple historical matching records in the current demand matrix, it first iterates through all historical demand matrices and calculates their third similarity with the current matrix. For example, if three historical matrices have third similarities of 85%, 78%, and 60%, and the second preset similarity is set to 70%, then the historical proportion corresponding to the 85% similarity is selected as the initial proportion. If the third similarity of all historical matrices does not exceed 70%, weights are assigned based on the similarity values ​​of each historical matrix. For example, the similarity values ​​are normalized using weight coefficients, and the three historical proportions are weighted to obtain an average proportion. This initial proportion serves as the starting point for the multivariate optimization equation, which can accelerate the solution process and improve the stability of material supply control.

[0070] This invention, through similarity screening and weighted averaging mechanisms, can automatically select the most suitable historical ratio or generate a comprehensive ratio, significantly improving the rationality of the initial value and optimizing the convergence speed.

[0071] Through the above technical solution, the present invention solves the problem of inaccurate selection of initial feeding ratio under multiple historical scenarios, ensures that the optimization algorithm quickly converges to the optimal solution that meets the dynamic needs of the boiler, reduces the risk of control fluctuations caused by initial value deviation, and enhances the stability and response efficiency of the mixed feeding process.

[0072] In some embodiments, the present invention further proposes a data acquisition device 14, which is also used to acquire the particle size distribution and density uniformity of the mixed coal; the dynamic mixing device 13 is also used to automatically adjust the stirring speed, stirring blade angle and mixing time of the dynamic mixing device 13 according to the particle size distribution data and density uniformity data, combined with a preset mixing uniformity model, so as to make the uniformity of the mixed coal reach the preset uniformity.

[0073] Among these, particle size distribution refers to the proportion of particles of different sizes in the mixed coal, which can be measured in real time using a laser particle size analyzer or sieving method. The data is used to determine whether the dispersion of coal powder particles meets combustion requirements. Density uniformity refers to the degree of density difference of the mixed coal in different regions, which can be detected using a gamma-ray densitometer or ultrasonic sensor. The data is used to assess the physical homogeneity of the coal powder mixture. The mixing uniformity model is a coal powder mixing quality evaluation system established based on statistical methods. Specifically, it can use analysis of variance or a coefficient of variation calculation model, and determine whether the mixing meets the standards by setting a threshold. Stirring speed, blade angle, and mixing time are used as adjustment parameters, which can be achieved through a variable frequency motor driving the stirring shaft, a hydraulically adjustable blade tilt device, and a timer linkage control, used to dynamically optimize the mechanical action intensity of the mixing process.

[0074] Specifically, during system operation, the data acquisition device 14 acquires the particle size distribution curve and density distribution thermogram of the mixed coal in real time, and inputs the raw data into the mixing uniformity model for dispersion calculation. If the model output value exceeds the preset uniformity threshold, the dynamic mixing device 13 initiates a parameter adjustment program: increasing the stirring speed to 800-1200 rpm if the particle size distribution is too wide, and adjusting the blade angle to 15-25 degrees to enhance shear force; if the area with a density difference exceeding 5% accounts for more than 10%, the mixing time is extended by 10-15 minutes. After all adjustment parameters are executed, the system re-acquires data for secondary verification until the model output value meets the preset standard.

[0075] This invention solves the problem of uneven mixing caused by differences in the physical properties of coal by using closed-loop control of online detection and dynamic adjustment, and avoids the lag defect of manual sampling and detection required in traditional methods.

[0076] Through the above technical solution, this invention achieves online monitoring and automatic optimization of the physical properties of mixed coal types, ensuring sufficient contact and stable combustion of pulverized coal during combustion. When the system detects that the mixing uniformity is not up to standard, it promptly triggers an alarm to prompt operators to intervene, effectively preventing problems such as decreased combustion efficiency or excessive pollutant emissions caused by uneven mixing.

[0077] In some embodiments, the present invention further proposes that the intelligent control device 15 also includes an intelligent replenishment reminder module. The intelligent replenishment reminder module is communicatively connected to an external supplier information system and a logistics scheduling system. The intelligent replenishment reminder module is used to dynamically predict the replenishment demand time for each type of coal based on the real-time inventory of each storage bin 11, the coal consumption rate, and a preset safety stock threshold. Combining real-time traffic data, GPS location information of transport vehicles, and real-time inventory information of coal suppliers, the optimal replenishment transportation route is calculated through a path optimization algorithm. The path optimization algorithm considers factors such as path length, estimated transportation time, and real-time fuel prices to generate the lowest-cost transportation plan. When the inventory of any type of coal is detected to be lower than the preset replenishment threshold, a replenishment reminder message is automatically pushed to the management personnel terminal, along with the generated lowest-cost transportation plan, a recommended supplier list, and the estimated arrival time, to guide timely and economical coal procurement and transportation.

[0078] The real-time inventory refers to the coal inventory data continuously collected by weight sensors or level gauges installed in the storage silo 11. Specifically, this can be achieved using digital weighing sensors or radar level gauges, reflecting the remaining material level in each silo in real time. The coal consumption rate refers to the amount of coal output per unit time corresponding to the opening degree of the lower gate 12 of each silo. This can be calculated using flow sensors combined with time-series data analysis, used to predict future coal consumption trends. The path optimization algorithm refers to a transportation path planning method based on graph theory and operations research models. Specifically, it can be implemented using Dijkstra's algorithm or genetic algorithms combined with real-time traffic data, used to find the lowest-cost delivery solution under various transportation constraints.

[0079] Specifically, the weight sensor inside storage silo 11 uploads inventory data to the intelligent replenishment reminder module every five minutes. This module analyzes the correlation between historical consumption data and the current opening degree of the lower gate 12, establishing a linear regression model to predict the coal consumption for the next three days. When the prediction shows that the inventory of a certain type of coal will fall below the safety threshold within the next 24 hours, the system automatically triggers the replenishment process. The route optimization algorithm calls the map API to obtain real-time traffic flow data, combines the current location of the transport vehicle with the coordinates of the supplier's warehouse, calculates the weighted value of fuel cost and time cost for five candidate routes, and selects the transportation plan with the lowest overall cost. The generated replenishment reminder information is pushed to the mobile terminal of the management personnel through the enterprise WeChat interface, and at the same time, the contract fulfillment rate data of the recommended supplier is presented in the form of a visual chart.

[0080] In a specific implementation, the safety stock threshold is set at 1.5 times the average daily consumption, based on the coal procurement cycle. A replenishment alert is triggered when the stock level falls below this threshold. Fuel price data in the route optimization algorithm is updated in real-time through access to a daily fuel price database published by relevant platforms. The predicted transportation time is corrected using a machine learning model combined with historical road condition data for the same time period. The supplier recommendation list prioritizes qualified suppliers within 300 kilometers of the target power plant and with no recent quality complaints.

[0081] This invention, through automated data collection and intelligent algorithms, reduces replenishment decision-making time from eight hours in a manual mode to within five minutes, and improves transportation route planning efficiency by more than sixty times. Compared with a fixed-cycle replenishment model, the dynamic inventory forecasting model can reduce the frequency of emergency procurement by 20%.

[0082] Through the above technical solutions, this invention effectively solves the risk of supply interruption in coal-fired systems caused by untimely manual replenishment, and avoids resource waste caused by high empty load rates of transport vehicles. By integrating real-time inventory data from suppliers, suppliers with sufficient inventory can be prioritized, reducing order response time by 40%. The route optimization algorithm comprehensively considers multiple cost factors, reducing the average cost of a single transportation trip by 15%, while ensuring timely coal delivery to guarantee production continuity.

[0083] In some embodiments, the intelligent control device 15 is further configured to: after stopping mixing, based on the particle size distribution data and the density uniformity data, and in conjunction with a preset mixing uniformity model, generate an alarm message if the uniformity of the mixed coal is detected to be less than the preset uniformity.

[0084] In some embodiments, the intelligent control device 15 further includes: The human-computer interaction module is used to display the operation information of the intelligent raw coal silo distribution system and to receive control information from the operator.

[0085] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent raw coal bunkering system, characterized in that, include: Multiple independent storage bins, each used to store a specific type of coal; Multiple lower gates, each located at the bottom of each storage bin, are used to control the discharge rate of each type of coal. A dynamic mixing device is used to mix different types of coal to obtain a mixed coal mixture; The data acquisition device is used to collect coal quality data for each storage bin, temperature field distribution inside the boiler furnace, flue gas concentration, and boiler output load. Intelligent control device, including: The generation module is used to generate the current demand matrix of the boiler based on the temperature field distribution, the flue gas concentration, the boiler output load, and current environmental protection requirements. The filtering module is used to match the coal quality data of each storage bin with the current demand matrix to obtain multiple target coal types that meet the boiler requirements; The optimization control module is used to solve a multivariate optimization equation based on the current demand matrix to obtain the optimal feeding ratio of the multiple target coal types, so as to control the lower gate corresponding to the multiple target coal types based on the optimal feeding ratio; wherein, the multivariate optimization equation is an equation constructed with the objectives of maximizing combustion efficiency, minimizing pollutant emissions, and optimizing coal consumption cost.

2. The intelligent raw coal bunkering system according to claim 1, characterized in that, The filtering module is specifically used for: The coal quality data of each storage bin is normalized to obtain multiple coal quality parameters for each storage bin; A multi-feature similarity algorithm is used to determine the first similarity between each coal quality parameter and the corresponding demand index in the current demand matrix, and the comprehensive similarity score is obtained by weighted summation; wherein, the weight of each coal quality parameter is determined according to the working condition requirements; The coal type corresponding to the comprehensive similarity score greater than the first preset similarity score is selected as the target coal type.

3. The intelligent raw coal bunkering system according to claim 2, characterized in that, The filtering module is also used for: If a unique historical demand matrix exists that corresponds to the multiple target coal types, calculate the second similarity between the unique historical demand matrix and the current demand matrix; If the second similarity is greater than the second preset similarity, the historical proportion of the multiple target coal types corresponding to the unique historical demand matrix is ​​used as the initial proportion; wherein, the initial proportion is used as the starting point for solving the multivariate optimization equation; If the second similarity is less than or equal to the second preset similarity, determine the difference value between the unique historical demand matrix and each demand indicator in the current demand matrix, determine the proportion adjustment coefficient for each target coal type based on the difference value, and obtain the initial proportion based on the proportion adjustment coefficient and the historical proportions of the multiple target coal types.

4. The intelligent raw coal bunkering system according to claim 3, characterized in that, Each demand indicator corresponds to different coal quality parameters. The screening module is also used for: Based on the difference value, the required adjustment information for each coal quality parameter is determined; the required adjustment information includes the adjustment direction and adjustment range for each coal quality parameter; Determine the contribution of each target coal type to each coal quality parameter under the historical proportions of the multiple target coal types; Based on the demand adjustment information and the contribution level, determine the proportion adjustment coefficient for each target coal type.

5. The intelligent raw coal bunkering system according to claim 4, characterized in that, The filtering module is also used for: Calculate the product of the contribution of the j-th coal quality parameter in the i-th target coal type and the adjustment range of the j-th coal quality parameter, assign positive or negative signs according to the adjustment direction of the j-th coal quality parameter, and add them together. Multiply the sum with the weight of the j-th coal quality parameter to obtain the influence factor of the i-th target coal type. The ratio of the influence factor of the i-th target coal type to the influence factors of all target coal types is used as the proportional adjustment coefficient of the i-th target coal type; Where i = 1, 2, ..., N, N is the number of multiple target coal types, and j = 1, 2, ..., M, M is the number of coal quality parameters.

6. The intelligent raw coal bunkering system according to claim 5, characterized in that, The filtering module is also used for: Calculate the product of the content of the j-th coal quality parameter in the i-th target coal type and the historical proportion of the i-th target coal type to obtain the actual contribution of the j-th coal quality parameter in the i-th target coal type. The ratio of the actual contribution of the j-th coal quality parameter to the total actual contribution of the j-th coal quality parameter is taken as the contribution degree of the j-th coal quality parameter.

7. The intelligent raw coal bunkering system according to claim 2, characterized in that, The filtering module is also used for: If multiple historical demand matrices correspond to the multiple target coal types, calculate the third similarity between the multiple historical demand matrices and the current demand matrix; If the maximum third similarity is greater than the second preset similarity, the historical proportion of the multiple target coal types corresponding to the historical demand matrix of the maximum third similarity is used as the initial proportion; If the maximum third similarity is less than or equal to the second preset similarity, the historical proportions of multiple target coal types in all historical demand matrices are weighted and averaged to obtain the average proportion value as the initial proportion.

8. The intelligent raw coal bunkering system according to any one of claims 1-7, characterized in that, The data acquisition device is also used to collect the particle size distribution and density uniformity of the mixed coal. The dynamic mixing device is also used for: Based on the particle size distribution data and the density uniformity data, and combined with the preset mixing uniformity model, the stirring speed, stirring blade angle and mixing time of the dynamic mixing device are automatically adjusted so that the uniformity of the mixed coal reaches the preset uniformity.

9. The intelligent raw coal bunkering system according to claim 8, characterized in that, The intelligent control device is also used for: After mixing stops, based on the particle size distribution data and the density uniformity data, combined with the preset mixing uniformity model, if the uniformity of the mixed coal is not detected to reach the preset uniformity, an alarm message is generated.

10. The intelligent raw coal bunkering system according to any one of claims 1-7, characterized in that, The intelligent control device also includes: The human-computer interaction module is used to display the operation information of the intelligent raw coal silo distribution system and to receive control information from the operator.

Images