Coal gas waste heat recovery system and recovery method

Abstract

The invention belongs to the technical field of coal gas waste heat recovery, and provides a coal gas waste heat recovery system and method, and the method comprises the steps: building a recovery operation database based on the recovery operation data of the coal gas waste heat recovery system during each coal gas waste heat recovery operation in a plurality of historical cycles, respectively extracting the recovery resistance and the waste heat recovery amount in a plurality of historical periods to obtain a correlation diagnosis value, if the correlation degree is close, performing stability analysis on the change trend of the recovery resistance in the plurality of historical periods, evaluating whether the change trend of the recovery resistance is stable or not, and if the change trend of the recovery resistance is stable, determining the waste heat recovery amount according to the correlation degree between the recovery resistance and the waste heat recovery amount. According to the method, the correlation change curve is obtained, whether the fitted correlation fitting line is matched with the correlation change curve or not is analyzed, if yes, the correlation fitting equation corresponding to the correlation fitting line serves as the waste heat recovery prediction model, the prediction precision of the constructed waste heat recovery prediction model is improved, and the problem that it is difficult to determine the recovery resistance adjustment value according to the target waste heat recovery amount is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of coal gas waste heat recovery, and in particular relates to a coal gas waste heat recovery system and a recovery method. Background Art

[0002] In the industrial production processes of steel, chemical industry, etc., a large amount of waste heat is generated after gas combustion or treatment. At present, most gas waste heat recovery systems have problems of low recovery efficiency and high energy consumption.

[0003] In existing technologies, it is impossible to accurately understand the intrinsic relationship between recovery resistance and waste heat recovery volume, making it difficult to effectively analyze and predict the recovery process based on historical data. This results in a lack of scientific optimization and adjustment basis for waste heat recovery operations, low recovery efficiency, and difficulty meeting the current industrial production needs for energy conservation, emission reduction, and efficient energy utilization. Therefore, this application obtains the waste heat recovery amount and recovery resistance of the gas waste heat recovery system during each gas waste heat recovery operation over multiple historical periods, and analyzes the correlation between the recovery resistance and the waste heat recovery amount over multiple historical periods. This not only reflects the stability of the correlation between the waste heat recovery amount and the recovery resistance over multiple different historical periods, but also predicts the waste heat recovery amount after adjusting the recovery resistance, thereby improving the optimization adjustment efficiency during the gas recovery operation. If the correlation between the waste heat recovery amount and the recovery resistance is relatively stable in multiple different historical periods, the stability of the recovery resistance change trend in multiple historical periods is further evaluated. Under the condition that the recovery resistance change trend is determined to be stable, a waste heat recovery prediction model is constructed to obtain the required adjustment amount of the recovery resistance. This not only improves the prediction accuracy of the constructed waste heat recovery prediction model, but also solves the problem of difficulty in determining the recovery resistance adjustment value based on the target waste heat recovery amount.

[0004] To this end, the present invention provides a gas waste heat recovery system and recovery method. Summary of the Invention

[0005] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0006] The technical solution adopted by the present invention to solve its technical problem is: In a first aspect, a method for recovering waste heat from coal gas comprises the following steps: The gas waste heat recovery system is used to obtain the recovery resistance of each gas waste heat recovery operation over multiple historical periods, as well as the waste heat recovery volume after the operation. These data are then analyzed and integrated to build a recovery operation database. Extract the recovery resistance and waste heat recovery amount from the recovery operation database, perform correlation analysis, and evaluate the correlation between the recovery resistance and waste heat recovery amount; If the correlation is close, then a stability analysis of the recovery resistance change trend over multiple historical periods is conducted to assess whether the recovery resistance change trend is stable; If it is stable, a waste heat recovery prediction model is constructed based on the correlation between the recovery resistance and the waste heat recovery amount, and the recovery resistance prediction amount is obtained.

[0007] Preferably, a historical period is randomly selected and divided equally into several historical periods; The recovery resistance and waste heat recovery amount in each historical period are combined to obtain multiple recovery data groups, which are sorted according to the time series of the historical period to build a recovery operation database.

[0008] Preferably, the collection volume is subjected to correlation difference analysis, and the process is as follows: In the recovery operation database, the recovery resistance and waste heat recovery amount in each recovery data group are extracted respectively, and the output is obtained through the Pearson distance improvement model to obtain the correlation degree difference value.

[0009] Preferably, the volume is analyzed for correlation trend differences, and the process is as follows: Based on the recovery resistance and waste heat recovery amount in each recovery data group, a recovery resistance change curve and a waste heat recovery change curve are constructed; The recovery resistance change curve and the waste heat recovery change curve are divided respectively, and a plurality of resistance sub-curves and a plurality of waste heat sub-curves are obtained accordingly; Obtain the slope of each resistance sub-curve and each residual heat sub-curve, output them through the Pearson distance improvement model, and obtain the correlation trend difference value; The correlation degree difference value and the correlation trend difference value are summed to calculate the period correlation degree.

[0010] Preferably, the correlation between the recovery amounts is as follows: Calculate the standard deviation of the period correlation corresponding to all historical periods to obtain the correlation diagnosis value; If the correlation diagnosis value is less than or equal to the correlation diagnosis threshold, a close correlation signal is generated.

[0011] Preferably, the stability of the recovery resistance change trend is analyzed as follows: Combining the resistance sub-curve slopes corresponding to adjacent resistance sub-curves on the recovery resistance change curve to obtain multiple resistance sub-slope groups, inputting them into the Euclidean calculation formula, and outputting the sub-slope difference value; Obtain the distance between the coordinates of adjacent peak points and the distance between the coordinates of adjacent trough points on the recovery resistance change curve, mark them as peak width and trough width respectively, and output them using the Euclidean calculation formula to obtain the width difference value; The wave width difference value and the sub-slope difference value are summed to obtain the period stability value.

[0012] Preferably, the process of evaluating whether the recovery resistance change trend is stable is as follows: Calculate the standard deviation of the periodic stability values corresponding to the recovery resistance in all historical periods to obtain the stability assessment value; If the stability evaluation value is less than or equal to the period stability threshold, a recovery resistance stability signal is generated.

[0013] Preferably, the construction process of the waste heat recovery prediction model is as follows: The recovery resistance and waste heat recovery amount are combined to construct a correlation change curve, and two coordinate points are arbitrarily selected as training sets. The least squares method is used for fitting to obtain the correlation fitting line and the corresponding correlation fitting equation; The remaining coordinates on the correlation change curve are used as the validation set. Any coordinates from the validation set are input into the correlation fitting equation, and the fitting Y coordinates are output. These coordinates are combined with the corresponding original Y coordinates to obtain multiple fitting analysis groups. The fitting analysis groups are then processed using the Euclidean calculation formula to output the correlation fitting matching value. If the correlation fitting match value is greater than the correlation fitting match threshold, re-fitting is required until the correlation fitting match value is less than the correlation fitting match threshold; If the correlation fitting matching value is less than or equal to the correlation fitting matching threshold, the correlation fitting equation corresponding to the constructed correlation fitting line is used as the waste heat recovery prediction model.

[0014] Preferably, the recovery resistance prediction is obtained according to the waste heat recovery prediction model, and the process is as follows: The current recovery resistance corresponding to the gas waste heat recovery system is input into the waste heat recovery prediction model, and the recovery resistance prediction value is output.

[0015] A gas waste heat recovery system includes the following modules: Operation data integration module: This module uses the gas waste heat recovery system to obtain the recovery resistance during each gas waste heat recovery operation over multiple historical periods, as well as the waste heat recovery volume after the operation, and analyzes and integrates these data to build a recovery operation database. Data association analysis module: extracts recovery resistance and waste heat recovery amount from the recovery operation database, performs correlation analysis, and evaluates the correlation between recovery resistance and waste heat recovery amount; Resistance change diagnosis module: If the correlation is close, the module will conduct a stability analysis on the recovery resistance change trend over multiple historical periods to assess whether the recovery resistance change trend is stable. Waste heat recovery prediction module: If stable, a waste heat recovery prediction model is constructed based on the correlation between the recovery resistance and the waste heat recovery amount, and the recovery resistance prediction amount is obtained.

[0016] The beneficial effects of the present invention are as follows: 1. The present invention uses the recovery operation data of each gas waste heat recovery operation of the gas waste heat recovery system over multiple historical periods as a basis to construct a recovery operation database. The recovery resistance and waste heat recovery amount in each of these historical periods are extracted to obtain correlation diagnostic values. This not only reflects the stability of the correlation between the waste heat recovery amount and the recovery resistance over multiple different historical periods, but also predicts the waste heat recovery amount after adjusting the recovery resistance, thereby improving the optimization and adjustment efficiency of the gas recovery operation. 2. If the correlation is close, the present invention performs a stability analysis on the recovery resistance change trend in multiple historical periods to evaluate whether the recovery resistance change trend is stable. If it is stable, the correlation between the recovery resistance and the waste heat recovery amount is analyzed, and whether the fitted correlation fitting line is consistent with the correlation change curve. If it is consistent, the correlation fitting equation corresponding to the correlation fitting line is used as the waste heat recovery prediction model, thereby not only improving the prediction accuracy of the constructed waste heat recovery prediction model, but also solving the problem of difficulty in determining the recovery resistance adjustment value according to the target waste heat recovery amount. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below with reference to the accompanying drawings.

[0018] Figure 1 This is a flow chart of the steps of a method for recovering waste heat from coal gas according to the present invention; Figure 2 It is a schematic diagram of a gas waste heat recovery system of the present invention. DETAILED DESCRIPTION

[0019] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0020] Example 1

[0021] See also Figure 1 As shown, a method for recovering waste heat from coal gas according to an embodiment of the present invention comprises the following steps: Step 1: Obtain the recovery resistance of each gas waste heat recovery operation over multiple historical periods, as well as the waste heat recovery volume after the operation, through the gas waste heat recovery system. Analyze and integrate these data to build a recovery operation database. Those skilled in the art will understand that: in the context of coal gas waste heat recovery, the historical cycle refers to the cycle in which the coal gas waste heat recovery system performs multiple waste heat recovery operations; It should be noted that the recovery operation data includes recovery resistance and waste heat recovery amount; In a preferred embodiment, a historical period is arbitrarily selected; Divide the historical cycle into several equal historical periods, and the time intervals of each historical period are equal; It is understood that the historical period refers to the period of time during which the gas waste heat recovery system performs waste heat recovery operations. In each historical period, the gas waste heat recovery system performs only one gas waste heat recovery operation, and during each gas waste heat recovery operation, the recovery resistance and waste heat recovery amount are recorded. The recovery resistance of each gas waste heat recovery operation is obtained as follows: Install pressure sensors at the inlet and outlet of the gas recovery pipeline respectively, measure the pipeline inlet pressure and outlet pressure, perform differential processing, and calculate the ratio with the pipeline inlet pressure to output the recovery resistance; Among them, the waste heat recovery amount can be directly recorded by the coal gas waste heat recovery system and calculated by comparing it with the waste heat recovery standard amount corresponding to the coal gas waste heat recovery system; It should be noted that the standard amount of waste heat recovery corresponding to the coal gas waste heat recovery system is a theoretical waste heat recovery amount based on the waste heat recovery logic carried by the coal gas waste heat recovery system; The recovery resistance and the amount of waste heat recovered during each waste heat recovery operation of the gas waste heat system are combined to obtain multiple recovery data groups, and the data are sorted according to the time series of the historical period to obtain a recovery data group sequence; It can be understood that the recycling data series corresponds to the recycling operation data within a historical period; Sort the historical periods corresponding to the recycling data series according to the time series, sort the recycling data series corresponding to all historical periods, and build a recycling operation database; Step 2: Based on the recovery operation database, the recovery resistance and waste heat recovery amount in multiple historical periods are extracted, and correlation analysis is performed to evaluate the correlation between the recovery resistance and the waste heat recovery amount; In a preferred embodiment, a recycling data set sequence is arbitrarily extracted from the recycling operation database; In the recovery data group sequence, the recovery resistance and waste heat recovery amount in each recovery data group are extracted respectively; All the recovery resistance after extraction and the waste heat recovery amount are analyzed through the Pearson distance improved model. The process is as follows: A1, averages all the extracted recovery resistances and outputs the average recovery resistance; A2, averages all the extracted waste heat recovery amounts and outputs the waste heat recovery mean; A3, based on the Pearson distance formula, improves the model through Pearson distance: , output the correlation degree difference value ,in, Expressed as the mean recovery resistance, Expressed as the mean value of waste heat recovery, Expressed as The recovery resistance value in the historical period, Expressed as The waste heat recovery amount in a historical period, n is expressed as the total number of historical periods, and the total number of recovery resistance values is the same as the total number of waste heat recovery amounts; With the X-axis representing time and the Y-axis representing recovery resistance, all extracted recovery resistances are used as coordinate points to construct a recovery resistance change curve; On the recovery resistance change curve, the local recovery resistance change curve between adjacent coordinate points is used as a resistance sub-curve to obtain multiple resistance sub-curves; Extract the coordinates of the endpoints of any resistance sub-curve, connect them, and calculate them using the slope calculation formula to obtain the slope of the resistance sub-curve; It should be noted that the endpoint coordinates of the resistance sub-curve are the starting point coordinates and the ending point coordinates on the resistance sub-curve respectively; Similarly, with the X-axis as time and the Y-axis as recovery resistance, all the waste heat recovery amounts after extraction are used as coordinate points to construct the waste heat recovery change curve; On the waste heat recovery variation curve, the local waste heat recovery variation curve between adjacent coordinate points is taken as a waste heat sub-curve to obtain multiple waste heat sub-curves; Extract the endpoint coordinates of any residual heat sub-curve, connect them, and calculate them using the slope calculation formula to obtain the slope of the residual heat sub-curve; It should be noted that the endpoint coordinates of the residual heat sub-curve are the starting point coordinate and the ending point coordinate on the residual heat sub-curve respectively; The slopes of multiple resistance sub-curves are combined with the slopes of the residual heat sub-curve and processed using the Pearson distance improvement model. The process is as follows: B1, average the slopes of multiple resistance sub-curves and output the average slope of the resistance sub-curves; B2, averaging the slopes of multiple residual heat sub-curves, and outputting the average slope of the residual heat sub-curves; B3, based on the Pearson distance formula, improves the model through Pearson distance: , calculate the correlation trend difference value ,in, It is represented by the first The slope of the resistance curve, It is represented by the first The slope of the residual heat sub-curve, It is expressed as the average slope of all resistance sub-curves on the recovery resistance change curve. It is represented by the mean slope of all waste heat sub-curves on the waste heat recovery curve, and m is represented by the total number of resistance sub-curve slopes; It should be noted that the total number of slopes of the resistance sub-curve is the same as the total number of slopes of the residual heat sub-curve; It is understandable that the purpose of using the Pearson distance formula to improve model processing is to: The improved Pearson distance model is essentially based on the idea of the Pearson correlation coefficient, and analyzes from two different dimensions: degree change and trend change. It quantifies the degree correlation and trend change correlation between the waste heat recovery volume and the recovery resistance in the historical period. This reflects not only whether the degree correlation between the waste heat recovery volume and the recovery resistance is close in the historical period, but also whether the trend change correlation is close. This provides data support for adjusting the recovery resistance in the future to improve the waste heat recovery volume. The correlation degree difference value and the correlation trend difference value are summed to calculate the period correlation degree; Calculate the standard deviation of the period correlation corresponding to all historical periods to obtain the correlation diagnosis value; It can be understood that the meaning of the correlation diagnostic value is that it can reflect the stability of the correlation between the waste heat recovery amount and the recovery resistance in multiple different historical periods. Specifically, if the correlation diagnostic value is large, it means that the correlation between the waste heat recovery amount and the recovery resistance has fluctuated greatly in multiple different historical periods. If the correlation diagnostic value is small, it means that the correlation between the waste heat recovery amount and the recovery resistance has fluctuated less in multiple different historical periods and has stronger stability. Therefore, it can not only reflect the stability of the correlation between the waste heat recovery amount and the recovery resistance in multiple different historical periods, but also predict the waste heat recovery amount after adjusting the recovery resistance, thereby improving the optimization adjustment efficiency during the gas recovery operation. The correlation diagnosis value is compared with the correlation diagnosis threshold value as follows: If the correlation diagnosis value is greater than the correlation diagnosis threshold, it means that the correlation between the waste heat recovery amount and the recovery resistance fluctuates greatly in multiple different historical periods, generating a non-tight correlation signal; If the correlation diagnosis value is less than or equal to the correlation diagnosis threshold, it means that the correlation between the waste heat recovery amount and the recovery resistance has a small fluctuation in multiple different historical periods, and a close correlation signal is generated; It should be noted that the associated diagnostic threshold is set by those skilled in the art; The specific scheme of this embodiment is as follows: based on the recovery operation data of each gas waste heat recovery operation of the gas waste heat recovery system in multiple historical periods, a recovery operation database is constructed, and the recovery resistance and waste heat recovery amount in multiple historical periods are respectively extracted to obtain correlation diagnostic values. This can not only reflect the stability of the correlation between the waste heat recovery amount and the recovery resistance in multiple different historical periods, but also predict the waste heat recovery amount after adjusting the recovery resistance, thereby improving the optimization adjustment efficiency during the gas recovery operation.

[0022] Example 2

[0023] See also Figure 1 As shown, a method for recovering waste heat from coal gas according to an embodiment of the present invention comprises the following steps: Step 3: If the correlation is close, conduct a stability analysis on the trend of recycling resistance changes over multiple historical periods to assess whether the trend of recycling resistance changes is stable; In a preferred embodiment, a recovery resistance change curve corresponding to any historical period is selected; Extract the slopes of all resistance sub-curves on the recovery resistance change curve and calculate them using the Euclidean formula. The process is as follows: Combining the resistance sub-curve slopes corresponding to adjacent resistance sub-curves on the recovery resistance change curve to obtain a plurality of resistance sub-slope groups; Input multiple resistance sub-slope groups into the Euclidean calculation formula: , calculate the sub-slope difference value , where m represents the total number of resistance sub-slope groups on the recovery resistance change curve and m-1 represents the total number of resistance sub-slope groups. It is expressed as the slope of the j-1th resistance sub-curve in the m-1th resistance sub-slope group. Represented as the m-1th resistance slope group The slope of the resistance curve; Extract the coordinates of all peak points and trough points on the recovery resistance change curve; The distance between the coordinates of adjacent peak points and the distance between the coordinates of adjacent trough points are obtained by the coordinate distance formula, and marked as peak width and trough width; All peak widths and trough widths are processed using the Euclidean calculation formula. The process is as follows: The peak widths and trough widths corresponding to adjacent peaks and troughs are combined to obtain multiple groups of wave width analysis groups; Input multiple sets of bandwidth analysis groups into the Euclidean calculation formula: , calculate the difference in wave width ,in, Expressed as the total number of wave width analysis groups, Expressed as The peak width within the wave width analysis group, Expressed as The trough width within each wave width analysis group; It is understandable that the purpose of using the Euclidean distance formula is to provide a unified quantitative standard for measuring the differences between different characteristics (the slope of the resistance sub-curve and the width of the peaks and troughs). Whether analyzing the stability of local changes or the stability of overall fluctuations, the same calculation method can be used to obtain the corresponding difference values, thus comprehensively reflecting that the change trend of the retracement resistance over the historical period is stable. The wave width difference value and the sub-slope difference value are summed to obtain the period stability value; Calculate the standard deviation of the periodic stability values corresponding to the recovery resistance in all historical periods and output the stable evaluation value; It is understood that the stability assessment value means that it is obtained by calculating the standard deviation of the periodic stability values corresponding to the recovery resistance in all historical periods, thereby reflecting the fluctuation of the stability of the recovery resistance change trend in multiple historical periods. It can also be used as a reference to predict whether the waste heat recovery capacity can be provided to the required standard after the recovery resistance is adjusted in the future based on the analyzed fluctuation. The cycle stability value is compared with the cycle stability threshold value as follows: If the cycle stability value is greater than the cycle stability threshold, it means that the trend of the recovery resistance in each historical cycle is quite different and the stability is weak, thus generating a recovery resistance fluctuation signal; If the cycle stability value is less than or equal to the cycle stability threshold, it means that the trend of the recovery resistance in each historical cycle is relatively small and the stability is relatively strong, and a recovery resistance stability signal is generated; Step 4: If the recovery resistance change trend is stable, then based on the correlation between the recovery resistance and the waste heat recovery amount, a waste heat recovery prediction model is constructed, and the recovery resistance prediction amount is obtained to complete the prediction work; In a preferred embodiment, the recovery resistance and waste heat recovery amount in any historical period are obtained respectively; With the X-axis representing the recovery resistance and the Y-axis representing the waste heat recovery amount, a correlation change curve is constructed; Randomly select any two coordinates on the correlation change curve as the training set, and fit them using the least squares method to obtain the correlation fitting line and the corresponding correlation fitting equation: ; It should be noted that The slope of the correlation fitting equation is obtained by inputting any two coordinates on the correlation change curve into the slope calculation formula. is a constant; The remaining coordinates on the correlation change curve are used as the validation set. Any coordinates from the validation set are input into the correlation fitting equation, and the fitting Y coordinates are output. These coordinates are combined with the corresponding original Y coordinates to obtain multiple fitting analysis groups. The fitting analysis groups are then processed using the Euclidean calculation formula to output the correlation fitting matching value. Specifically, the Euclid calculation formula: , calculate the associated fitting matching value ,in, Expressed as the original Y coordinate within the t-th fitting analysis group, is the fitting Y coordinate in the t-th fitting analysis group, It is expressed as the total number of fitted analysis groups; The association fit match value is compared with the association fit match threshold as follows: If the correlation fitting match value is greater than the correlation fitting match threshold, it means that the constructed correlation fitting line has a low matching degree compared with the remaining coordinate points on the correlation change curve, and refitting is required until the correlation fitting match value is less than the correlation fitting match threshold; If the correlation fitting match value is less than or equal to the correlation fitting match threshold, it means that the constructed correlation fitting line has a higher matching degree than the remaining coordinate points on the correlation change curve, and the correlation fitting equation corresponding to the constructed correlation fitting line is used as the waste heat recovery prediction model; Input the waste heat recovery amount required by the current gas waste heat recovery system into the waste heat recovery prediction model to predict the recovery resistance that needs to be adjusted as the recovery resistance prediction amount; The technical solution of this embodiment is specifically as follows: if the correlation is close, a stability analysis is performed on the recovery resistance change trend in multiple historical periods to evaluate whether the recovery resistance change trend is stable. If it is stable, the correlation between the recovery resistance and the waste heat recovery amount is analyzed, and whether the fitted correlation fitting line is consistent with the correlation change curve. If it is consistent, the correlation fitting equation corresponding to the correlation fitting line is used as the waste heat recovery prediction model, thereby not only improving the prediction accuracy of the constructed waste heat recovery prediction model, but also solving the problem of difficulty in determining the recovery resistance adjustment value according to the target waste heat recovery amount.

[0024] Example 3

[0025] like Figure 2 As shown, the present application provides a gas waste heat recovery system, which is composed of the following models, specifically including: Operation data integration module: This module uses the gas waste heat recovery system to obtain the recovery resistance during each gas waste heat recovery operation over multiple historical periods, as well as the waste heat recovery volume after the operation, and analyzes and integrates these data to build a recovery operation database. Data association analysis module: Based on the recovery operation database, the recovery resistance and waste heat recovery amount in multiple historical periods are extracted, and correlation analysis is performed to evaluate the correlation between the recovery resistance and the waste heat recovery amount; Resistance change diagnosis module: If the correlation is close, the module will conduct a stability analysis on the recovery resistance change trend over multiple historical periods to assess whether the recovery resistance change trend is stable. Waste heat recovery prediction module: If the recovery resistance change trend is stable, a waste heat recovery prediction model is constructed based on the correlation between the recovery resistance and the waste heat recovery amount, and the recovery resistance prediction amount is obtained to complete the prediction work; The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for recovering waste heat from coal gas, characterized by: include: The gas waste heat recovery system is used to obtain the recovery resistance of each gas waste heat recovery operation over multiple historical periods, as well as the waste heat recovery volume after the operation. These data are then analyzed and integrated to build a recovery operation database. Extract the recovery resistance and waste heat recovery amount from the recovery operation database, perform correlation analysis, and evaluate the correlation between the recovery resistance and waste heat recovery amount; If the correlation is close, then a stability analysis of the recovery resistance change trend over multiple historical periods is conducted to assess whether the recovery resistance change trend is stable; If it is stable, a waste heat recovery prediction model is constructed based on the correlation between the recovery resistance and the waste heat recovery amount, and the recovery resistance prediction amount is obtained.

2. A method for recovering waste heat from coal gas according to claim 1, characterized in that: The process of building a recycling operation database is as follows: Randomly select a historical period and divide it into several equal historical periods; The recovery resistance and waste heat recovery amount in each historical period are combined to obtain multiple recovery data groups, which are sorted according to the time series of the historical period to build a recovery operation database.

3. The method for recovering waste heat from coal gas according to claim 1, characterized in that: The correlation between the recovery resistance and the waste heat recovery amount is analyzed as follows: In the recovery operation database, the recovery resistance and waste heat recovery amount in each recovery data group are extracted respectively, and the output is obtained through the Pearson distance improvement model to obtain the correlation degree difference value.

4. The method for recovering waste heat from coal gas according to claim 1, characterized in that: The correlation trend difference analysis between recovery resistance and waste heat recovery amount is carried out as follows: Based on the recovery resistance and waste heat recovery amount in each recovery data group, a recovery resistance change curve and a waste heat recovery change curve are constructed; The recovery resistance change curve and the waste heat recovery change curve are divided respectively, and a plurality of resistance sub-curves and a plurality of waste heat sub-curves are obtained accordingly; Obtain the slope of each resistance sub-curve and each residual heat sub-curve, output them through the Pearson distance improvement model, and obtain the correlation trend difference value; The correlation degree difference value and the correlation trend difference value are summed to calculate the period correlation degree.

5. A method for recovering waste heat from coal gas according to claim 4, characterized in that: The correlation between the recovery resistance and the waste heat recovery amount is evaluated as follows: Calculate the standard deviation of the period correlation corresponding to all historical periods to obtain the correlation diagnosis value; If the correlation diagnosis value is less than or equal to the correlation diagnosis threshold, a close correlation signal is generated.

6. A method for recovering waste heat from coal gas according to claim 1, characterized in that: To analyze the stability of the recovery resistance change trend, the process is as follows: Combining the resistance sub-curve slopes corresponding to adjacent resistance sub-curves on the recovery resistance change curve to obtain multiple resistance sub-slope groups, inputting them into the Euclidean calculation formula, and outputting the sub-slope difference value; Obtain the distance between the coordinates of adjacent peak points and the distance between the coordinates of adjacent trough points on the recovery resistance change curve, mark them as peak width and trough width respectively, and output them using the Euclidean calculation formula to obtain the width difference value; The wave width difference value and the sub-slope difference value are summed to obtain the period stability value.

7. A method for recovering waste heat from coal gas according to claim 6, characterized in that: To evaluate whether the trend of reclaim resistance change is stable, the process is as follows: Calculate the standard deviation of the periodic stability values corresponding to the recovery resistance in all historical periods to obtain the stability assessment value; If the stability evaluation value is less than or equal to the period stability threshold, a recovery resistance stability signal is generated.

8. The method for recovering waste heat from coal gas according to claim 1, characterized in that: The construction process of the waste heat recovery prediction model is as follows: The recovery resistance and waste heat recovery amount are combined to construct a correlation change curve, and two coordinate points are arbitrarily selected as training sets. The least squares method is used for fitting to obtain the correlation fitting line and the corresponding correlation fitting equation; The remaining coordinates on the correlation change curve are used as the validation set. Any coordinates from the validation set are input into the correlation fitting equation, and the fitting Y coordinates are output. These coordinates are combined with the corresponding original Y coordinates to obtain multiple fitting analysis groups. The fitting analysis groups are then processed using the Euclidean calculation formula to output the correlation fitting matching value. If the correlation fitting match value is greater than the correlation fitting match threshold, re-fitting is required until the correlation fitting match value is less than the correlation fitting match threshold; If the correlation fitting matching value is less than or equal to the correlation fitting matching threshold, the correlation fitting equation corresponding to the constructed correlation fitting line is used as the waste heat recovery prediction model.

9. A method for recovering waste heat from coal gas according to claim 8, characterized in that: According to the waste heat recovery prediction model, the recovery resistance prediction is obtained. The process is as follows: The current recovery resistance corresponding to the gas waste heat recovery system is input into the waste heat recovery prediction model, and the recovery resistance prediction value is output.

10. A gas waste heat recovery system, performing a gas waste heat recovery according to claims 1 to 9, characterized in that: Includes the following modules: Operation data integration module: This module uses the gas waste heat recovery system to obtain the recovery resistance during each gas waste heat recovery operation over multiple historical periods, as well as the waste heat recovery volume after the operation, and analyzes and integrates these data to build a recovery operation database. Data association analysis module: extracts recovery resistance and waste heat recovery amount from the recovery operation database, performs correlation analysis, and evaluates the correlation between recovery resistance and waste heat recovery amount; Resistance change diagnosis module: If the correlation is close, the module will conduct a stability analysis on the recovery resistance change trend over multiple historical periods to assess whether the recovery resistance change trend is stable. Waste heat recovery prediction module: If stable, a waste heat recovery prediction model is constructed based on the correlation between the recovery resistance and the waste heat recovery amount, and the recovery resistance prediction amount is obtained.

Images