Coal blending combustion comprehensive management method and system based on machine learning

Abstract

The invention discloses a fire coal blending combustion comprehensive management method and system based on machine learning, and the method comprises the steps: obtaining the historical blending combustion data and historical combustion efficiency data of fire coal, and analyzing and determining the blending combustion characteristics affecting the combustion efficiency; constructing a prediction model based on the blending combustion characteristics and a preset neural network model to predict a combustion efficiency prediction value; a combustion efficiency required value of fire coal is obtained, a combustion efficiency difference value is determined based on the combustion efficiency required value and the combustion efficiency predicted value, and a combustion efficiency difference evaluation value is obtained through analysis and evaluation; performing fitting analysis on the historical blending combustion data and the historical combustion efficiency data to determine a comprehensive influence coefficient of the blending combustion characteristics on the combustion efficiency; and determining an adjustment coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and adjusting the current blending combustion data to manage the coal blending combustion. According to the method, the combustion efficiency is automatically optimized and adjusted by introducing machine learning to comprehensively analyze and process data, and the fire coal combustion process can be finely managed.

Description

Technical Field

[0001] The present invention relates to the technical field of coal blending combustion, and particularly to a comprehensive management method and system for coal blending combustion based on machine learning. Background Art

[0002] In modern industrial production, coal, as an important energy source, is widely used in power generation, heating, and industrial production processes. However, the waste gas emissions generated during the coal combustion process have a serious impact on the environment and human health. In order to improve the efficiency of the coal combustion process and reduce the pollutants emitted, the comprehensive management of coal blending combustion has become a key task. The comprehensive management method of coal blending combustion aims to maximize the coal combustion efficiency and optimize the energy utilization by deeply analyzing, predicting, and regulating the coal blending combustion process.

[0003] However, traditional methods usually rely on empirical models or rules, unable to accurately predict the coal combustion efficiency, difficult to cope with the complex changes and uncertainties in the combustion process, unable to achieve precise combustion efficiency optimization. Moreover, traditional methods usually require manual intervention and adjustment, lacking automation and intelligence, unable to achieve real-time monitoring, automatic adjustment, and intelligent optimization, with low efficiency and response speed. And traditional methods are often limited to single indicators or local optimization, lacking the ability of overall comprehensive optimization, unable to fully consider the mutual influence and comprehensive effect among various factors, resulting in limited management effects in the coal blending combustion process. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a comprehensive management method and system for coal blending combustion based on machine learning, including: Obtain the historical coal blending data and historical combustion efficiency data of coal, and perform data analysis on the historical coal blending data and historical combustion efficiency data to determine the blending characteristics affecting the combustion efficiency; Construct a combustion efficiency prediction model based on the blending characteristics and a preset neural network model, and perform prediction based on the combustion efficiency prediction model and the current coal blending data to obtain a combustion efficiency prediction value; Obtain the required value of the coal combustion efficiency, determine the combustion efficiency difference value based on the required value of the combustion efficiency and the combustion efficiency prediction value, and analyze and evaluate the combustion efficiency difference value to obtain a combustion efficiency difference evaluation value; Perform fitting analysis on the historical coal blending data and historical combustion efficiency data, and determine the comprehensive influence coefficient of the blending characteristics on the combustion efficiency according to the fitting analysis result; Determine an adjustment coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and adjust the current coal blending data according to the adjustment coefficient to comprehensively manage the coal blending combustion.

[0005] Further, obtaining historical blending data and historical combustion efficiency data of coal, and performing data analysis on the historical blending data and historical combustion efficiency data to determine blending characteristics affecting combustion efficiency, including: Obtaining historical blending data of coal, and extracting several candidate blending characteristics from the historical blending data; Determining the data corresponding to each candidate blending characteristic, and analyzing the correlation between the data corresponding to each candidate blending characteristic and combustion efficiency; Selecting candidate blending characteristics with a correlation greater than a preset value, and for each candidate blending characteristic, determining the change amount of the combustion efficiency corresponding to this candidate blending characteristic when other candidate blending characteristics are the same; Selecting candidate blending characteristics with a change amount greater than a preset value, and determining them as the blending characteristics affecting combustion efficiency in the historical blending data.

[0006] Further, constructing a combustion efficiency prediction model based on the blending characteristics and a preset neural network model, and performing prediction based on the combustion efficiency prediction model and current blending data to obtain a combustion efficiency prediction value, including: Constructing a data set based on the blending characteristics and corresponding historical blending data, and inputting the data set into the preset neural network model to construct an initial combustion efficiency prediction model; Dividing the data set into a training set and a test set according to a certain ratio, and inputting the training set and the test set into the initial combustion efficiency prediction model; Training and testing the initial combustion efficiency prediction model until the initial combustion efficiency prediction model meets the preset convergence condition to obtain a combustion efficiency prediction model; Obtaining current blending data of coal, and inputting the current blending data into the combustion efficiency prediction model, and performing prediction by the combustion efficiency prediction model to obtain a combustion efficiency prediction value.

[0007] Further, obtaining a required value of combustion efficiency of coal, determining a combustion efficiency difference value based on the required value of combustion efficiency and the combustion efficiency prediction value, and analyzing and evaluating the combustion efficiency difference value to obtain a combustion efficiency difference evaluation value, including: Obtaining the required value of combustion efficiency of coal, subtracting the combustion efficiency prediction value from the required value of combustion efficiency for calculation, and determining the calculated difference value as the combustion efficiency difference value; Obtaining a preset standard deviation of combustion efficiency, calculating the ratio of the combustion efficiency difference value to the standard deviation of combustion efficiency, and evaluating this ratio to obtain a combustion efficiency difference evaluation value.

[0008] Further, performing fitting analysis on the historical blending data and historical combustion efficiency data, and determining a comprehensive influence coefficient of the blending characteristics on combustion efficiency according to the analysis results, including: Determine the data corresponding to each blending feature in the historical blending data, and fit the data corresponding to each blending feature with the historical combustion efficiency data to construct a blending feature - combustion efficiency curve; Determine the peak and valley values in each blending feature - combustion efficiency curve, and calculate the difference between the peak and the valley value to obtain the difference amplitude corresponding to each blending feature; Determine the correlation between the data corresponding to each candidate blending feature and the combustion efficiency, and use the correlation as the weight corresponding to each blending feature; Perform weighted addition calculation on the difference amplitude corresponding to each blending feature and the weight to obtain the comprehensive influence coefficient of the blending feature on the combustion efficiency.

[0009] Further, determining a regulation coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and regulating the current blending data according to the regulation coefficient to comprehensively manage the coal blending, including: Calculate according to the comprehensive influence coefficient and the combustion efficiency difference evaluation value to obtain a coal blending regulation value, and determine the regulation coefficient according to the coal blending regulation value; Preset the corresponding relationship between the regulation coefficient - coal blending regulation value interval, wherein for each coal blending regulation value interval in the corresponding relationship between the regulation coefficient - coal blending regulation value interval, a corresponding regulation coefficient is associated; Obtain the coal blending regulation value, and based on the mapping relationship of the coal blending regulation value interval to which the coal blending regulation value belongs in the corresponding relationship between the regulation coefficient - coal blending regulation value interval, select the regulation coefficient corresponding to the coal blending regulation value interval as the corresponding regulation coefficient; Regulate the data corresponding to each blending feature in the current blending data according to the regulation coefficient, and comprehensively manage the coal blending according to the regulated blending data.

[0010] Further, the calculation formula of the coal blending regulation value is: K = α * L + β * P, where K is the coal blending regulation value, α is the first conversion coefficient, L is the comprehensive influence coefficient, β is the second conversion coefficient, and P is the combustion efficiency difference evaluation value.

[0011] The present invention also provides a coal blending comprehensive management system based on machine learning, including: An acquisition module, configured to acquire the historical blending data and historical combustion efficiency data of coal, and perform data analysis on the historical blending data and historical combustion efficiency data to determine the blending features affecting the combustion efficiency; A prediction module, configured to construct a combustion efficiency prediction model based on the blending features and a preset neural network model, and perform prediction based on the combustion efficiency prediction model and the current blending data to obtain a combustion efficiency prediction value; An evaluation module, configured to obtain a required value of the combustion efficiency of coal for combustion, determine a combustion efficiency difference value based on the required value of the combustion efficiency and a predicted value of the combustion efficiency, and analyze and evaluate the combustion efficiency difference value to obtain a combustion efficiency difference evaluation value; A determination module, configured to perform fitting analysis on historical co-firing data and historical combustion efficiency data, and determine a comprehensive influence coefficient of the co-firing characteristics on the combustion efficiency according to the fitting analysis result; A management module, configured to determine an adjustment coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and adjust the current co-firing data according to the adjustment coefficient to perform comprehensive management of coal co-firing.

[0012] Compared with the prior art, the beneficial effects of a method and system for comprehensive management of coal co-firing based on machine learning according to an embodiment of the present invention are as follows: By constructing a combustion efficiency prediction model based on historical data and a neural network model, the present invention can accurately predict the combustion efficiency of coal for combustion, and improve the prediction accuracy of the combustion efficiency through analysis of historical data and model training, helping to achieve more effective management of coal co-firing; By comparing the predicted value and the required value of the combustion efficiency, the present invention obtains a difference evaluation of the combustion efficiency, helps to evaluate the efficiency level of the current coal co-firing, can timely detect the deviation of the combustion efficiency, and take corresponding measures for adjustment to ensure the high efficiency and environmental protection of the combustion process; By performing fitting analysis and determining the influence coefficient, the present invention can deeply understand the influence degree and direction of each co-firing characteristic on the combustion efficiency, which helps to optimize the co-firing data, focus on the characteristics with the greatest influence on the combustion efficiency, and achieve more accurate control of the combustion efficiency; By determining the adjustment coefficient and adjusting the current co-firing data, the present invention can optimize the management of coal co-firing according to the actual situation, improve the combustion efficiency and environmental protection level, realize the comprehensive management and optimization of the coal co-firing process, and ensure that the combustion efficiency reaches the expected target. Description of the Drawings

[0013] Figure 1 is a schematic flow structure diagram of a method for comprehensive management of coal co-firing based on machine learning in an embodiment of the present invention; Figure 2 is a schematic composition diagram of a system for comprehensive management of coal co-firing based on machine learning in an embodiment of the present invention. Detailed Embodiments

[0014] The following further describes in detail the specific embodiments of the present application with reference to the drawings and embodiments. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0015] In the description of the present application, it should be understood that the orientation or positional relationship indicated by terms such as "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the platform or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present application.

[0016] The terms "first", "second" are only used for descriptive purposes and cannot be construed as indicating or implying a relative importance coefficient or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first", "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0017] In the description of the present application, it should be noted that unless otherwise clearly specified and limited, the terms "installed", "connected", "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0018] As Figure 1 shown, in the embodiments of the present application, a comprehensive management method for coal blending combustion based on machine learning is provided, including: S100: Obtain the historical blending data and historical combustion efficiency data of coal, and perform data analysis on the historical blending data and historical combustion efficiency data to determine the blending characteristics affecting the combustion efficiency; S200: Construct a combustion efficiency prediction model based on the blending characteristics and a preset neural network model, and perform prediction based on the combustion efficiency prediction model and the current blending data to obtain a combustion efficiency prediction value; S300: Obtain the required value of the combustion efficiency of coal, determine the combustion efficiency difference value based on the required value of the combustion efficiency and the combustion efficiency prediction value, and analyze and evaluate the combustion efficiency difference value to obtain a combustion efficiency difference evaluation value; S400: Perform fitting analysis on the historical blending data and historical combustion efficiency data, and determine the comprehensive influence coefficient of the blending characteristics on the combustion efficiency according to the fitting analysis result; S500: Determine an adjustment coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and adjust the current blending data according to the adjustment coefficient to comprehensively manage the coal blending combustion.

[0019] Furthermore, the present invention can accurately predict the combustion efficiency of coal by constructing a combustion efficiency prediction model based on historical data and a neural network model. Through the analysis of historical data and model training, the prediction accuracy of combustion efficiency is improved, which helps to achieve more effective coal blending management. By comparing the predicted value and the required value of combustion efficiency, the present invention obtains the difference evaluation of combustion efficiency, which helps to evaluate the efficiency level of current coal blending, can timely detect the deviation of combustion efficiency, and take corresponding measures for adjustment to ensure the high efficiency and environmental protection of the combustion process. Through fitting analysis and determination of influence coefficients, the present invention can deeply understand the influence degree and direction of each blending characteristic on combustion efficiency, which helps to optimize blending data, focus on the characteristics with the greatest influence on combustion efficiency, and achieve more accurate control of combustion efficiency. By determining the adjustment coefficient and adjusting the current blending data, the present invention can optimize coal blending management according to actual conditions, improve combustion efficiency and environmental protection level, realize the comprehensive management and optimization of the coal blending process, and ensure that the combustion efficiency reaches the expected target.

[0020] In an embodiment of the present application, a comprehensive management method for coal blending based on machine learning is provided. The method includes obtaining historical blending data and historical combustion efficiency data of coal, and performing data analysis on the historical blending data and historical combustion efficiency data to determine the blending characteristics affecting combustion efficiency, including: obtaining the historical blending data of coal, and extracting several candidate blending characteristics from the historical blending data; determining the data corresponding to each candidate blending characteristic, and analyzing the correlation between the data corresponding to each candidate blending characteristic and combustion efficiency; selecting the candidate blending characteristics with a correlation greater than a preset value, and for each candidate blending characteristic, determining the change amount of the combustion efficiency corresponding to the candidate blending characteristic when other candidate blending characteristics are the same; selecting the candidate blending characteristics with a change amount greater than the preset value, and determining them as the blending characteristics affecting combustion efficiency in the historical blending data.

[0021] Specifically, historical data during coal blending combustion is obtained, including various parameter characteristics during the combustion process; several candidate blending combustion characteristics are selected from the historical data, and these characteristics may include factors such as fuel type, combustion temperature, oxygen content, fuel ratio, etc.; the correlation between each candidate blending combustion characteristic and combustion efficiency is analyzed, and the correlation degree between them is measured through correlation analysis technical means; the candidate blending combustion characteristics with a correlation degree greater than a preset threshold are selected, and these characteristics have a significant impact on combustion efficiency; for each candidate blending combustion characteristic with a relatively high correlation degree, when other candidate blending combustion characteristics remain unchanged, the change amount of the combustion efficiency corresponding to this characteristic is determined; the candidate blending combustion characteristics with a change amount greater than the preset threshold are selected, and these characteristics are determined as the key blending combustion characteristics affecting combustion efficiency in the historical blending data. This step identifies the blending combustion characteristics that have an important impact on combustion efficiency in the historical coal blending data; establishes a correlation model of the blending combustion characteristics that have a greater impact on combustion efficiency, which helps to optimize the management and control of the combustion process; improves the understanding and analysis ability of the coal blending combustion process, and provides strong support for the optimization and improvement of combustion efficiency In an embodiment of the present application, a comprehensive management method for coal blending combustion based on machine learning is provided. A combustion efficiency prediction model is constructed based on blending combustion characteristics and a preset neural network model, and prediction is performed based on the combustion efficiency prediction model and current blending data to obtain a combustion efficiency prediction value, including: constructing a data set based on blending combustion characteristics and corresponding historical blending data, and inputting the data set into the preset neural network model to construct an initial combustion efficiency prediction model; dividing the data set into a training set and a test set according to a certain ratio, and inputting the training set and the test set into the initial combustion efficiency prediction model; training and testing the initial combustion efficiency prediction model until the initial combustion efficiency prediction model meets the preset convergence condition to obtain a combustion efficiency prediction model; obtaining the current blending data of coal, and inputting the current blending data into the combustion efficiency prediction model, and predicting by the combustion efficiency prediction model to obtain a combustion efficiency prediction value.

[0022] Specifically, a dataset is constructed based on candidate co-firing characteristics and corresponding historical co-firing data, including the characteristics and corresponding combustion efficiency data; the dataset is input into a preset neural network model to construct an initial combustion efficiency prediction model, which is used to predict the combustion efficiency; the constructed dataset is divided into a training set and a test set according to a certain ratio to evaluate the performance and generalization ability of the model; the training set and the test set are respectively input into the initial combustion efficiency prediction model for training and testing the model; the initial combustion efficiency prediction model is trained and tested, and by continuously adjusting the model parameters and optimization algorithms until the model meets the preset convergence conditions, the final combustion efficiency prediction model is obtained; the current co-firing data of the coal is obtained and input into the trained combustion efficiency prediction model, and the prediction value of the current combustion efficiency is obtained by the combustion efficiency prediction model, and this prediction value can help optimize the combustion process and improve the efficiency. The neural network model constructed based on historical data in this step can predict the combustion efficiency and achieve precise management and optimization of the coal combustion process; by training and testing the model, the accuracy and generalization ability of the combustion efficiency prediction model are improved, enabling it to adapt to different combustion situations; applying the prediction model to the current combustion data can predict the combustion efficiency in real time, helping to achieve real-time monitoring and optimization of the combustion process, and improving production efficiency and energy conservation and emission reduction effects.

[0023] In an embodiment of the present application, a comprehensive management method for coal co-firing based on machine learning is provided. The method includes obtaining the required value of the combustion efficiency of the coal, determining the combustion efficiency difference value based on the required value of the combustion efficiency and the predicted value of the combustion efficiency, and analyzing and evaluating the combustion efficiency difference value to obtain the combustion efficiency difference evaluation value, including: obtaining the required value of the combustion efficiency of the coal, calculating the difference between the required value of the combustion efficiency and the predicted value of the combustion efficiency, and determining the calculated difference as the combustion efficiency difference value; obtaining the preset standard deviation of the combustion efficiency, calculating the ratio of the combustion efficiency difference value to the standard deviation of the combustion efficiency, and evaluating this ratio to obtain the combustion efficiency difference evaluation value.

[0024] Specifically, obtain the required value of the combustion efficiency of coal, which is the efficiency value desired during the combustion process; calculate the difference between the required value of the combustion efficiency and the predicted value of the combustion efficiency obtained through the prediction model to obtain the difference value of the combustion efficiency, that is, the gap between the actual efficiency and the required efficiency; obtain the pre-set standard deviation of the combustion efficiency, which is a measure of the change in the combustion efficiency, and calculate the ratio of the difference value of the combustion efficiency to the standard deviation of the combustion efficiency. This ratio reflects the degree of deviation between the actual efficiency and the required efficiency; compare this ratio with the pre-set standard, and evaluate according to the size of the ratio to obtain the evaluation value of the combustion efficiency difference, which is used to evaluate the stability and efficiency of the combustion process. This step realizes the accurate evaluation of the coal combustion efficiency, helps to monitor the deviation between the actual efficiency and the expected efficiency during the combustion process; provides a quantitative evaluation method. Through the evaluation value of the combustion efficiency difference, problems and optimization space in the combustion process can be discovered in time; it helps to optimize the management and control of the combustion process, improve the combustion efficiency, reduce energy consumption and emissions, so as to achieve energy conservation, emission reduction and improvement of production efficiency.

[0025] In an embodiment of the present application, a comprehensive management method for coal blending combustion based on machine learning is provided. The fitting analysis of historical blending data and historical combustion efficiency data is performed, and the comprehensive influence coefficient of the blending characteristics on the combustion efficiency is determined according to the analysis results, including: determining the data corresponding to each blending characteristic in the historical blending data, and fitting the data corresponding to each blending characteristic with the historical combustion efficiency data to construct a blending characteristic-combustion efficiency curve; determining the peak and valley values in each blending characteristic-combustion efficiency curve, and calculating the difference between the peak and valley values to obtain the difference amplitude corresponding to each blending characteristic; determining the correlation degree between the data corresponding to each candidate blending characteristic and the combustion efficiency, and using the correlation degree as the weight corresponding to each blending characteristic; performing a weighted summation calculation of the difference amplitude corresponding to each blending characteristic and the weight to obtain the comprehensive influence coefficient of the blending characteristics on the combustion efficiency.

[0026] Specifically, determine the data corresponding to each co-firing feature in the historical co-firing data, such as the coal blending ratio, combustion temperature, etc.; fit the data corresponding to each co-firing feature with the historical combustion efficiency data to construct a co-firing feature - combustion efficiency curve, revealing the relationship between different co-firing features and combustion efficiency; determine the peaks and valleys in each co-firing feature - combustion efficiency curve, calculate their differences, and obtain the difference amplitude corresponding to each co-firing feature, that is, the fluctuation range of combustion efficiency; determine the correlation degree between each co-firing feature and combustion efficiency, and use the correlation degree as a weight to reflect the influence degree of this feature on combustion efficiency; perform weighted summation calculation on the difference amplitude corresponding to each co-firing feature and the weight to obtain the comprehensive influence coefficient of the co-firing feature on combustion efficiency, and this coefficient reflects the influence of all co-firing features on the overall combustion efficiency. This step reveals the relationship between different co-firing features and combustion efficiency, helping to understand the influence mechanism of each co-firing feature on combustion efficiency; by calculating the difference amplitude and weight, the influence degree of each co-firing feature on combustion efficiency is quantified, which helps to determine the strategy for optimizing co-firing data; the comprehensive influence coefficient provides an index for comprehensively considering the influence of each co-firing feature, which can be used to guide the control and optimization of the combustion process, improve combustion efficiency and reduce energy consumption.

[0027] In an embodiment of the present application, a comprehensive management method for coal co-firing based on machine learning is provided. The adjustment coefficient is determined based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and the current co-firing data is adjusted according to the adjustment coefficient to comprehensively manage coal co-firing, including: calculating based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value to obtain a coal co-firing adjustment value, and determining an adjustment coefficient according to the coal co-firing adjustment value; presetting the corresponding relationship between the adjustment coefficient - coal co-firing adjustment value interval, wherein for each coal co-firing adjustment value interval in the corresponding relationship between the adjustment coefficient - coal co-firing adjustment value interval, a corresponding adjustment coefficient is associated; obtaining the coal co-firing adjustment value, and based on the mapping relationship of the coal co-firing adjustment value interval to which the coal co-firing adjustment value belongs in the corresponding relationship between the adjustment coefficient - coal co-firing adjustment value interval, selecting the adjustment coefficient corresponding to the coal co-firing adjustment value interval as the corresponding adjustment coefficient; adjusting the data corresponding to each co-firing feature in the current co-firing data according to the adjustment coefficient, and comprehensively managing coal co-firing according to the adjusted co-firing data.

[0028] Specifically, according to the comprehensive influence coefficient of the previous coal blending characteristics and the evaluation value of the combustion efficiency difference, the coal blending adjustment value is calculated, which reflects the degree of adjustment required for the coal blending ratio; the adjustment coefficient is determined based on the coal blending adjustment value, and this coefficient can be used to map the coal blending adjustment value to a specific adjustment amount; the corresponding relationship between the adjustment coefficient - coal blending adjustment value interval is preset in advance to ensure that each coal blending adjustment value interval has a corresponding adjustment coefficient; the actual coal blending adjustment value is obtained, and according to the corresponding relationship between the adjustment coefficient - coal blending adjustment value interval, it is mapped to the corresponding adjustment coefficient; according to the adjustment coefficient, the data corresponding to each blending characteristic in the current blending data is adjusted to adjust the coal blending ratio to optimize the combustion efficiency. This step realizes the function of automatically adjusting the coal blending ratio according to the influence degree of coal blending characteristics and the difference in combustion efficiency, improves the combustion efficiency and energy conservation and emission reduction; establishes the mapping relationship between the adjustment coefficient and the coal blending adjustment value, making the adjustment process more intelligent and accurate; through the comprehensive management of coal blending, the combustion process can be optimized, the production efficiency can be improved, the energy consumption and emissions can be reduced, so as to achieve sustainable development and environmental protection goals.

[0029] In an embodiment of the present application, a comprehensive management method for coal blending based on machine learning is provided, and the calculation formula for the coal blending adjustment value is: K = α * L + β * P, where K is the coal blending adjustment value, α is the first conversion coefficient, L is the comprehensive influence coefficient, β is the second conversion coefficient, and P is the evaluation value of the combustion efficiency difference.

[0030] As Figure 2 shown, in an embodiment of the present application, a comprehensive management system for coal blending based on machine learning is provided, including: an acquisition module for acquiring the historical blending data and historical combustion efficiency data of coal, and performing data analysis on the historical blending data and historical combustion efficiency data to determine the blending characteristics affecting the combustion efficiency; a prediction module for constructing a combustion efficiency prediction model based on the blending characteristics and a preset neural network model, and making a prediction based on the combustion efficiency prediction model and the current blending data to obtain a combustion efficiency prediction value; an evaluation module for obtaining the required value of the combustion efficiency of coal, determining the combustion efficiency difference value based on the required value of the combustion efficiency and the combustion efficiency prediction value, and analyzing and evaluating the combustion efficiency difference value to obtain an evaluation value of the combustion efficiency difference; a determination module for performing fitting analysis on the historical blending data and historical combustion efficiency data, and determining the comprehensive influence coefficient of the blending characteristics on the combustion efficiency according to the fitting analysis result; a management module for determining an adjustment coefficient based on the comprehensive influence coefficient and the evaluation value of the combustion efficiency difference, and adjusting the current blending data according to the adjustment coefficient to comprehensively manage the coal blending.

[0031] In summary, the embodiments of the present invention provide a comprehensive management method and system for coal blending combustion based on machine learning, which include: obtaining historical coal blending data and historical combustion efficiency data of coal, and analyzing and determining the blending characteristics affecting the combustion efficiency; constructing a prediction model based on the blending characteristics and a preset neural network model to predict the predicted value of the combustion efficiency; obtaining the required value of the combustion efficiency of coal, determining the combustion efficiency difference value based on it and the predicted value of the combustion efficiency, and obtaining the combustion efficiency difference evaluation value through analysis and evaluation; performing fitting analysis on the historical coal blending data and historical combustion efficiency data to determine the comprehensive influence coefficient of the blending characteristics on the combustion efficiency; determining the adjustment coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and adjusting the current coal blending data to manage the coal blending combustion. The present invention can automatically optimize and adjust the combustion efficiency by introducing machine learning to comprehensively analyze and process data, and can perform refined management on the coal combustion process.

[0032] Finally, it should be noted that: Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

[0033] The above is only one embodiment of the present invention, but it cannot be used to limit the scope of the present invention. Any structural changes made based on the present invention, as long as they do not deviate from the essence of the present invention, should be regarded as falling within the protection scope of the present invention and being restricted. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process and related descriptions of the above-described platform can refer to the corresponding process in the foregoing platform embodiments, and will not be repeated here.

[0034] The term "including" or any other similar term is intended to cover non-exclusive inclusion, so that a process, platform, article, or device / platform including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes the elements inherent in these processes, platforms, articles, or devices / platforms.

[0035] So far, the technical solutions of the present invention have been described in combination with the further embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to closely related technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0036] The above is only the preferred embodiment of the present invention, and is not used to limit the protection scope of the present invention.

Claims

1. A comprehensive management method for coal blending combustion based on machine learning, characterized in that Including: Obtain the historical blending data and historical combustion efficiency data of the coal, perform data analysis on the historical blending data and historical combustion efficiency data, and determine the blending characteristics affecting the combustion efficiency; Construct a combustion efficiency prediction model based on the blending characteristics and a preset neural network model, and perform prediction based on the combustion efficiency prediction model and the current blending data to obtain a combustion efficiency prediction value; Obtain the required value of the combustion efficiency of the coal, determine the combustion efficiency difference value based on the required value of the combustion efficiency and the combustion efficiency prediction value, and analyze and evaluate the combustion efficiency difference value to obtain a combustion efficiency difference evaluation value; Perform fitting analysis on the historical blending data and historical combustion efficiency data, and determine the comprehensive influence coefficient of the blending characteristics on the combustion efficiency according to the fitting analysis result; Determine the adjustment coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and adjust the current blending data according to the adjustment coefficient to comprehensively manage the coal blending.

2. The integrated management method for coal blending combustion based on machine learning according to claim 1, wherein, The obtaining of the historical blending data and historical combustion efficiency data of the coal, performing data analysis on the historical blending data and historical combustion efficiency data, and determining the blending characteristics affecting the combustion efficiency includes: Obtain the historical blending data of the coal, and extract several candidate blending characteristics from the historical blending data; Determine the data corresponding to each candidate blending characteristic, and analyze the correlation between the data corresponding to each candidate blending characteristic and the combustion efficiency; Select the candidate blending characteristics with a correlation greater than the preset value, and for each candidate blending characteristic, determine the change amount of the combustion efficiency corresponding to this candidate blending characteristic when other candidate blending characteristics are the same; Select the candidate blending characteristics with a change amount greater than the preset value and determine them as the blending characteristics affecting the combustion efficiency in the historical blending data.

3. A comprehensive management method for coal blending combustion based on machine learning according to claim 2, characterized in that The constructing of a combustion efficiency prediction model based on the blending characteristics and a preset neural network model, and performing prediction based on the combustion efficiency prediction model and the current blending data to obtain a combustion efficiency prediction value includes: Construct a data set based on the blending characteristics and the corresponding historical blending data, and input the data set into the preset neural network model to construct an initial combustion efficiency prediction model; Divide the data set into a training set and a test set according to a certain ratio, and input the training set and the test set into the initial combustion efficiency prediction model; Train and test the initial combustion efficiency prediction model until the initial combustion efficiency prediction model meets the preset convergence condition to obtain a combustion efficiency prediction model; Obtain the current blending data of the coal, and input the current blending data into the combustion efficiency prediction model, and perform prediction by the combustion efficiency prediction model to obtain a combustion efficiency prediction value.

4. A comprehensive management method for coal blending combustion based on machine learning according to claim 3, characterized in that, The obtaining of the required value of the combustion efficiency of the coal, determining the combustion efficiency difference value based on the required value of the combustion efficiency and the combustion efficiency prediction value, and analyzing and evaluating the combustion efficiency difference value to obtain a combustion efficiency difference evaluation value includes: Obtain the required value of the combustion efficiency of the coal, perform a subtraction calculation on the required value of the combustion efficiency and the combustion efficiency prediction value, and determine the calculated difference as the combustion efficiency difference value; Obtain the preset standard deviation of combustion efficiency, calculate the ratio of the combustion efficiency difference value to the standard deviation of combustion efficiency, and evaluate this ratio to obtain the combustion efficiency difference evaluation value.

5. A comprehensive management method for coal blending combustion based on machine learning according to claim 4, characterized in that, The fitting analysis of the historical co-firing data and historical combustion efficiency data, and determining the comprehensive influence coefficient of the co-firing characteristics on the combustion efficiency according to the analysis results, includes: Determine the data corresponding to each co-firing characteristic in the historical co-firing data, and fit the data corresponding to each co-firing characteristic with the historical combustion efficiency data to construct a co-firing characteristic - combustion efficiency curve; Determine the peak and valley values in each co-firing characteristic - combustion efficiency curve, and calculate the difference between the peak and valley values to obtain the difference amplitude corresponding to each co-firing characteristic; Determine the correlation degree between the data corresponding to each candidate co-firing characteristic and the combustion efficiency, and use the correlation degree as the weight corresponding to each co-firing characteristic; Perform weighted summation calculation on the difference amplitude corresponding to each co-firing characteristic and the weight to obtain the comprehensive influence coefficient of the co-firing characteristics on the combustion efficiency.

6. A comprehensive management method for coal blending combustion based on machine learning according to claim 5, characterized in that, Based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value to determine the adjustment coefficient, and adjust the current co-firing data according to the adjustment coefficient to comprehensively manage the coal co-firing, including: Calculate according to the comprehensive influence coefficient and the combustion efficiency difference evaluation value to obtain the coal co-firing adjustment value, and determine the adjustment coefficient according to the coal co-firing adjustment value; Preset the corresponding relationship between the adjustment coefficient - coal co-firing adjustment value interval, where for each coal co-firing adjustment value interval, a corresponding adjustment coefficient is associated; Obtain the coal co-firing adjustment value, and based on the mapping relationship of the coal co-firing adjustment value interval to which the coal co-firing adjustment value belongs in the corresponding relationship between the adjustment coefficient - coal co-firing adjustment value interval, select the adjustment coefficient corresponding to the coal co-firing adjustment value interval as the corresponding adjustment coefficient; Adjust the data corresponding to each co-firing characteristic in the current co-firing data according to the adjustment coefficient, and comprehensively manage the coal co-firing according to the adjusted co-firing data.

7. A comprehensive management method for coal blending combustion based on machine learning according to claim 6, characterized in that The calculation formula of the coal co-firing adjustment value is: K = α * L + β * P, where K is the coal co-firing adjustment value, α is the first conversion coefficient, L is the comprehensive influence coefficient, β is the second conversion coefficient, and P is the combustion efficiency difference evaluation value.

8. A comprehensive management system for coal blending combustion based on machine learning, characterized in that, including: An acquisition module, used to acquire the historical co-firing data and historical combustion efficiency data of coal, and perform data analysis on the historical co-firing data and historical combustion efficiency data to determine the co-firing characteristics affecting the combustion efficiency; A prediction module, used to construct a combustion efficiency prediction model based on the co-firing characteristics and a preset neural network model, and perform prediction based on the combustion efficiency prediction model and the current co-firing data to obtain the combustion efficiency prediction value; An evaluation module, used to obtain the required value of the combustion efficiency of coal, determine the combustion efficiency difference value based on the required value of the combustion efficiency and the combustion efficiency prediction value, and analyze and evaluate the combustion efficiency difference value to obtain the combustion efficiency difference evaluation value; A determination module, used to perform fitting analysis on the historical co-firing data and historical combustion efficiency data, and determine the comprehensive influence coefficient of the co-firing characteristics on the combustion efficiency according to the fitting analysis results; The management module is used to determine the adjustment coefficient based on the comprehensive influence coefficient and the combustion efficiency difference evaluation value, and adjust the current co-firing data according to the adjustment coefficient to comprehensively manage the coal co-firing.

Images