A synergic mixed combustion process and system of gasification furnace slag and pulverized coal boiler

Abstract

This invention discloses a co-firing process and system for gasification slag and pulverized coal boilers. The process includes the following steps: collecting basic attribute data of gasification slag and raw coal, and extracting corresponding historical combustion condition datasets from a historical operating database; selecting the target historical operating condition with the highest matching degree; generating a fuel mixing ratio instruction for the current moment; calculating the discrete deviation value between real-time combustion feedback data and standard combustion data in the target historical operating condition; and generating an ash removal control signal. This invention dynamically determines the initial blending ratio by collecting basic fuel attributes and constructing feature vectors, and then performing similarity matching with historical high-efficiency operating conditions. This achieves a shift from relying on fixed rules to data-driven dynamic optimization, improving the stability, economy, and safety of the blending process.

Description

Technical Field

[0001] This invention relates to the field of clean coal technology, and in particular to a co-firing process and system for gasification slag and pulverized coal boilers. Background Technology

[0002] With the promotion of clean coal utilization technologies and the increasing environmental protection requirements, the resource utilization of solid waste generated from coal gasification—gasification slag—is receiving increasing attention. Blending it with raw coal and then feeding it into a pulverized coal boiler for combustion is an effective way to achieve its harmless and resource-based disposal, which is of great significance for reducing enterprises' solid waste disposal costs and improving economic efficiency.

[0003] However, existing technologies for co-firing gasification slag generally suffer from low combustion control precision and poor stability. In related technologies, the co-firing ratio is often set based on a fixed range or operator experience, making it difficult to adapt to frequent fluctuations in the moisture content, calorific value, and other properties of the raw coal and gasification slag. This static or semi-empirical, extensive control mode easily leads to unstable combustion conditions in the furnace, flame center deviation, reduced efficiency, and even safety issues such as coking and flameout. Furthermore, existing methods largely rely on timed soot blowing or passive handling after the absolute value of the flue gas temperature exceeds the limit, resulting in delayed early warnings and an inability to achieve optimized intervention, thus hindering the precise, stable, and proactive control of the gasification slag co-firing process. Summary of the Invention

[0004] This application provides a co-firing process and system for gasified slag and pulverized coal boilers, which solves the problem of mismatch between static experience control and dynamic complex operating conditions in the prior art. It realizes the transformation from relying on fixed rules to data-driven dynamic optimization, and improves the stability, economy and safety of the co-firing process.

[0005] This application provides a co-firing process for gasification slag and pulverized coal boiler, including the following steps: collecting basic attribute data of gasification slag and raw coal, and extracting the corresponding historical combustion condition dataset from the historical operation database;

[0006] Based on the aforementioned basic attribute data, a fuel feature vector for the current batch is constructed. The fuel feature vector is then compared with the sample vectors marked as stable operation in the historical combustion condition dataset using multidimensional spatial similarity calculation. The target historical operating condition with the highest matching degree is then selected. Extract the mixing ratio parameters from the target historical operating conditions, and combine them with the current boiler's real-time load demand data to generate the fuel mixing ratio instruction for the current moment; During the execution of the fuel mixing ratio command, real-time combustion feedback data in the furnace is collected simultaneously, and the discrete deviation value between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions is calculated. Based on the discrete deviation value, a dynamic correction command for the primary and secondary air ratio is generated, and the time series change rate of the flue gas temperature is calculated simultaneously. When the change rate exceeds the preset safety gradient, an ash cleaning control signal is generated.

[0007] Furthermore, the step of constructing the fuel feature vector for the current batch based on the aforementioned basic attribute data and filtering out the target historical operating conditions with the highest matching degree includes: The collected moisture content values ​​of gasification slag and raw coal were normalized to construct a ternary feature coordinate system including moisture, reference calorific value and volatile matter dimensions. Define a search sliding window in the historical operation database, traverse all historical sample vectors within the window, and calculate the Euclidean distance between the current ternary feature coordinates and each historical sample vector; Select the top K historical sample vectors with the smallest Euclidean distance, and statistically analyze the combustion efficiency label values ​​corresponding to these K sample vectors. The sample with the highest combustion efficiency label value is determined as the target historical operating condition, and the fuel pretreatment parameters under this operating condition are extracted as benchmark reference data.

[0008] Furthermore, the step of extracting the mixing ratio parameters from the target historical operating conditions and combining them with the current boiler's real-time load demand data to generate the fuel mixing ratio instruction for the current moment includes: Read the original mixing ratio value from the target historical operating conditions and obtain the target total calorific value from the current boiler's real-time load demand data; The original mixing ratio is adjusted by weighting a correction coefficient derived from the difference between the current fuel feature vector and the target historical operating condition feature vector, as well as the combustion feedback data of the previous time step. The original mixing ratio is adjusted by weighting the correction coefficient to establish a linear balance equation between the fuel mixing ratio and the total target calorific value. The equation is solved to obtain the accurate blending mass ratio that meets the current load requirements. This mass ratio is then converted into a frequency control signal for the feeder.

[0009] Furthermore, the step of simultaneously acquiring real-time combustion feedback data within the furnace and calculating the discrete deviation between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions includes: The current oxygen distribution data and flame center temperature data are acquired in real time through the furnace sensor array to form a real-time status matrix; The generated fuel mixing ratio instruction is used as an index key to search the historical database for the standard state matrix that should be theoretically achieved under this specific ratio. The oxygen distribution deviation and temperature center offset are obtained by performing a difference operation on the real-time state matrix and the standard state matrix, and these two deviations are combined into a comprehensive discrete deviation index that characterizes combustion instability.

[0010] Furthermore, the step of generating dynamic correction instructions for the primary and secondary air distribution ratio based on the discrete deviation values ​​includes: The comprehensive discrete deviation index is input into a preset graded adjustment rule base, which contains air volume gain coefficients corresponding to different deviation ranges; Based on the current range of the comprehensive discrete deviation index, match the corresponding air volume gain coefficient, and calculate the adjustment step size of the primary air pressure and the opening compensation value of the secondary damper. The adjustment step size and opening compensation value are encapsulated into a fluidization velocity optimization package and sent to the fan actuator.

[0011] Furthermore, the steps of encapsulating the adjustment step size and aperture compensation values ​​into a fluidization rate optimization package include: The particle residence time parameter in the furnace is monitored and calculated by the ratio of furnace height to real-time fluidizing velocity. Determine whether the current particle residence time parameter meets the time threshold requirement for the burnout of residual carbon in the gasification slag. If not, add a flow field damping coefficient to the fluidization velocity optimization package.

[0012] Furthermore, when the rate of change exceeds a preset safety gradient, the step of generating a dust accumulation cleaning control signal includes: The exhaust temperature values ​​of the tail flue were continuously recorded at a fixed sampling frequency to construct a temperature time series dataset; The temperature time series dataset is processed by first-order differentiation to obtain the instantaneous rise slope of flue gas temperature. Combined with the proportion of gasification slag in the current fuel mixing ratio instruction, the ash accumulation sensitivity factor is calculated. The product of the instantaneous rise slope and the ash accumulation sensitivity factor is defined as the coking risk value. When the coking risk value continues to be higher than the preset alarm threshold, the execution action of increasing the frequency of steam soot blowing and the intensity of soot blowing is triggered.

[0013] Furthermore, the steps following the triggering of the actions to increase the frequency of steam soot blowing and the intensity of sonic soot blowing include: Continuously monitor the flue gas temperature drop curve after the action is performed; Calculate the response time required for the exhaust gas temperature to drop back to the baseline temperature, mark this response time as the cleaning effect feedback value, and store this feedback value in the historical operation database.

[0014] Further steps in collecting basic attribute data of gasification furnace slag and raw coal include: The digital interface of the online moisture analyzer is connected to read the real-time conductivity data of the gasification slag, and the total moisture content is obtained by inversion through the preset conductivity and moisture conversion model. The total moisture content is compared with the preset drying threshold. When the total moisture content is higher than the threshold, a scheduling instruction is generated for the turning machine in the drying yard until the data collected again meets the requirements.

[0015] This application provides a co-firing system for gasification furnace slag and pulverized coal boilers, used to realize a co-firing process of gasification furnace slag and pulverized coal boilers, including: Data acquisition module, working condition screening module, instruction generation module, deviation value calculation module, instruction correction module; The data acquisition module is used to collect basic attribute data of gasification slag and raw coal, and extract the corresponding historical combustion condition dataset from the historical operation database. The operating condition screening module is used to construct the fuel feature vector of the current batch based on the basic attribute data, perform multi-dimensional spatial similarity calculation between the fuel feature vector and the sample vector marked as stable operation in the historical combustion operating condition dataset, and screen out the target historical operating condition with the highest matching degree. The instruction generation module is used to extract the mixing ratio parameters from the target historical operating conditions, and combine them with the real-time load demand data of the current boiler to generate the fuel mixing ratio instruction for the current moment. The deviation calculation module is used to simultaneously collect real-time combustion feedback data in the furnace during the execution of the fuel mixing ratio command, and calculate the discrete deviation value between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions. The instruction correction module is used to generate a dynamic correction instruction for the primary and secondary air ratio based on the discrete deviation value, and simultaneously calculate the time series change rate of the flue gas temperature. When the change rate exceeds the preset safety gradient, an ash cleaning control signal is generated.

[0016] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: By matching the current fuel properties with historical high-efficiency operating conditions, the most suitable operating parameter template was found for the dynamically changing fuel quality, solving the problems of inaccurate experience ratios and poor combustion state caused by coal quality fluctuations, and realizing the initial accurate setting of the blending ratio.

[0017] Furthermore, by using the combustion result residual carbon burnout rate as a feedback coefficient to calculate according to the ratio, and using the current execution ratio as a benchmark index for combustion state evaluation, the fuel supply and air volume distribution can be dynamically optimized based on the real-time feedback of combustion effect, timely suppressing combustion deviations and improving the boiler's adaptability and operational stability under co-firing conditions.

[0018] Furthermore, by analyzing the changing trends of flue gas temperature rather than its absolute value, and coupling this with the coking tendency of the fuel itself, i.e., the proportion of slag, a forward-looking early warning model for ash accumulation and coking was established. This transforms the cleaning process from post-remedial action to pre-emptive prevention, avoiding irreversible decline in the heat exchange efficiency of the heating surfaces. While ensuring the long-term safe operation of the boiler, it also improves the overall energy utilization efficiency and economy. Attached Figure Description

[0019] Figure 1 A process flow diagram of co-firing of gasification slag and pulverized coal boiler provided in this application embodiment; Figure 2 This is a schematic diagram of a co-firing system for gasification slag and pulverized coal boiler provided in an embodiment of this application. Detailed Implementation

[0020] This application provides a co-firing process and system for gasified slag and pulverized coal boilers, solving the problem of mismatch between static experience control and dynamic complex operating conditions in existing technologies. By collecting basic fuel attributes and constructing feature vectors, similarity matching with historical high-efficiency operating conditions is performed to determine the initial co-firing ratio. During execution, the ratio is corrected in real time based on the residual carbon burnout rate feedback from the previous moment, and the deviation between real-time combustion data and standard values ​​is calculated simultaneously using the current ratio as an index, thereby dynamically adjusting the primary and secondary air ratios to stabilize combustion. Simultaneously, the risk of ash accumulation is predicted by calculating the flue gas temperature change rate and combining it with the slag ratio, triggering cleaning in advance. This achieves a shift from relying on fixed rules to data-driven dynamic optimization, improving the stability, economy, and safety of the co-firing process.

[0021] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0022] like Figure 1 The diagram shown is a process flow chart of a co-firing process of gasification slag and pulverized coal boiler provided in an embodiment of this application. The process is applied to a co-firing system of gasification slag and pulverized coal boiler and includes the following steps: collecting basic attribute data of gasification slag and raw coal, and extracting the corresponding historical combustion condition dataset from the historical operation database. The basic attribute data includes moisture content and calorific value. Based on the aforementioned basic attribute data, a fuel feature vector for the current batch is constructed. The fuel feature vector is then compared with the sample vectors marked as stable operation in the historical combustion condition dataset using multidimensional spatial similarity calculation. The target historical operating condition with the highest matching degree is then selected. Extract the mixing ratio parameters from the target historical operating conditions, combine them with the current real-time load demand data of the boiler, generate the fuel mixing ratio instruction for the current moment through interpolation calculation, and send the fuel mixing ratio instruction to the coal feeding control unit. During the execution of the fuel mixing ratio command, real-time combustion feedback data in the furnace is collected simultaneously, and the discrete deviation value between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions is calculated. Based on the discrete deviation value, a dynamic correction command for the primary and secondary air ratio is generated, and the time series change rate of the flue gas temperature is calculated simultaneously. When the change rate exceeds the preset safety gradient, an ash cleaning control signal is generated.

[0023] By collecting and analyzing historical data, a stable combustion historical condition that best matches the current fuel characteristics is determined as a benchmark. Then, the fuel mix ratio under this benchmark condition is dynamically adjusted and implemented based on the current boiler load demand. During implementation, the combustion status is monitored in real time and compared with the expected state of the benchmark condition. Based on the deviation from the comparison, the air supply system is adjusted in real time to stabilize combustion, while the flue gas temperature change affected by ash accumulation is independently monitored, and soot blowing is automatically triggered when the risk of coking is too high. The entire process achieves coordinated control of fuel blending, combustion optimization, and coking prevention.

[0024] Furthermore, the step of constructing the fuel feature vector for the current batch based on the aforementioned basic attribute data and filtering out the target historical operating conditions with the highest matching degree includes: The collected moisture content values ​​of gasification slag and raw coal were normalized to construct a ternary feature coordinate system including moisture, reference calorific value and volatile matter dimensions. The construction steps include: Mix gasification slag and raw coal at an initial estimated ratio (e.g., 1:9), and calculate the weighted average moisture content of the mixed fuel. Weighted average lower heating value Weighted average dry ash-free volatile matter Composed of ternary feature coordinates This coordinate system comprehensively represents the combustion characteristics of the current batch of fuel.

[0025] Define a search sliding window in the historical operation database, traverse all historical sample vectors within the window, and calculate the Euclidean distance between the current ternary feature coordinates and each historical sample vector; The calculation process for Euclidean distance is as follows: In a historical database (e.g., the last 30 days) of stable operating records, each record has its corresponding historical feature vector. ; h is the number of the historical operating condition sample.

[0026] The Euclidean distance is derived using the Euclidean distance calculation formula: ; distance The smaller the value, the more similar the characteristics of the two. Select the K historical samples with the smallest distance (e.g., K=5), and then choose the record with the highest combustion efficiency at that time as the target historical operating condition. This ensures that not only are the fuel characteristics similar, but also the reference template with the best operating performance is found.

[0027] Select the top K historical sample vectors with the smallest Euclidean distance, and statistically analyze the combustion efficiency label values ​​corresponding to these K sample vectors. The sample with the highest combustion efficiency label value is determined as the target historical operating condition, and the fuel pretreatment parameters under this operating condition are extracted as benchmark reference data.

[0028] Furthermore, the step of extracting the mixing ratio parameters from the target historical operating conditions and combining them with the current boiler's real-time load demand data to generate the fuel mixing ratio instruction for the current moment includes: Read the original mixing ratio value from the target historical operating conditions and obtain the target total calorific value from the current boiler's real-time load demand data; The original mixing ratio is adjusted by weighting a correction coefficient derived from the difference between the current fuel feature vector and the target historical operating condition feature vector, as well as the combustion feedback data of the previous time step. The correction factor is calculated as follows: First, obtain the attribute matching factor, and then calculate the current fuel feature vector. With the target historical working condition feature vector cosine similarity (The closer the value is to 1, the more similar the similarity): ; Attribute matching factor Defined as: ; Next, obtain the combustion feedback factor: Based on the combustion performance of the previous cycle, calculate the combustion feedback factor. : ; In the formula, This is the average measured residual carbon content of fly ash / bottom ash from the previous cycle. The standard residual carbon content expected to be achieved under the target historical operating condition ratio, when combustion is complete ( )hour, This indicates that co-combustion can be maintained or increased; when combustion is incomplete ( )hour, This indicates that blending should be reduced.

[0029] Finally, calculate the overall correction factor: ; when Time: This indicates that the current fuel properties are significantly different from the historical best operating conditions, or the combustion effect in the previous cycle was poor, and the slag blending ratio will be reduced.

[0030] when Time: This indicates that the current fuel is highly similar to historical operating conditions, and combustion was good in the previous cycle, so the slag blending ratio can be maintained or appropriately increased.

[0031] The original mixing ratio is adjusted by weighting the correction coefficient to establish a linear balance equation between the fuel mixing ratio and the total target calorific value. The equation is solved to obtain the accurate blending mass ratio that meets the current load requirements. This mass ratio is then converted into a frequency control signal for the feeder.

[0032] The steps for establishing and solving equilibrium equations include: First, calculate the corrected baseline ratio. According to the principle of energy conservation, the total heat power input to the boiler should equal the load demand. Establish the equation... .in, and These are the required mass flow rates of raw coal and gasification slag, respectively. and These are their lower heating values. It's about time. Simultaneously satisfying... The solution can be obtained immediately by connecting the two. and This leads to the final precise co-firing mass ratio. .

[0033] Furthermore, the step of simultaneously acquiring real-time combustion feedback data within the furnace and calculating the discrete deviation between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions includes: The current oxygen distribution data and flame center temperature data are acquired in real time through the furnace sensor array to form a real-time status matrix; The steps involved in real-time state matrix analysis include: Data is read in real time from oxygen and temperature sensors (such as thermocouples or infrared thermometers) arranged at multiple points in the furnace to form a real-time status matrix. For example, the oxygen content data vector is The temperature data vector is .

[0034] The generated fuel mixing ratio instruction is used as an index key to search the historical database for the standard state matrix that should be theoretically achieved under this specific ratio. The standard state matrix lookup steps include: Based on the currently implemented fuel blending instructions Use the key index to retrieve data from the historical database. Find all historically used indexes related to... Data records of identical or very similar (e.g., differences within ±5%) proportions during stable operation periods were used to calculate the average of oxygen and temperature data from these records, which served as the standard state matrix. .

[0035] This ensures that the benchmark for comparison is matched with current fuel conditions.

[0036] The oxygen distribution deviation and temperature center offset are obtained by performing a difference operation on the real-time state matrix and the standard state matrix, and these two deviations are combined into a comprehensive discrete deviation index that characterizes combustion instability.

[0037] The deviation calculation process is as follows: Calculate the root mean square error of oxygen distribution: This reflects the uniformity deviation of the oxygen field inside the furnace and calculates the absolute deviation of the average temperature. This reflects the shift in overall combustion intensity.

[0038] Where n is the total number of oxygen sensor monitoring points, i is the number of oxygen measurement points, traversing from 1 to n, m is the total number of temperature sensor monitoring points, and j is the number of temperature measurement points, traversing from 1 to m.

[0039] Generation of the comprehensive dispersion index: due to and Different weights are assigned to the effects on combustion stability, and a comprehensive discrete deviation index is generated by weighted summation. Weighting coefficients and Determined by the boiler's design characteristics and operating experience, For example, for tangential combustion boilers that are particularly sensitive to oxygen distribution, It can be set to 0.6. Set it to 0.4.

[0040] Furthermore, the step of generating dynamic correction instructions for the primary and secondary air distribution ratio based on the discrete deviation values ​​includes: The comprehensive discrete deviation index is input into a preset graded adjustment rule base, which contains air volume gain coefficients corresponding to different deviation ranges; Based on the current range of the comprehensive discrete deviation index, match the corresponding air volume gain coefficient, and calculate the adjustment step size of the primary air pressure and the opening compensation value of the secondary damper. The adjustment step size and opening compensation value are encapsulated into a fluidization velocity optimization package and sent to the blower actuator to change the gas-solid two-phase flow field distribution in the furnace, forcing the real-time combustion feedback data to converge to the standard combustion data.

[0041] The calculated comprehensive dispersion index The mapping is used to adjust the primary and secondary air systems, forming a rule-based torch shape and flow field optimization layer outside of PID control.

[0042] A pre-defined hierarchical adjustment rule base is essentially a piecewise function. For example: when hour, (No adjustment); when hour, ;when hour, Piecewise function By analyzing a large amount of historical operating data, the comprehensive dispersion index was obtained. The relationship between the changes in air volume and the successfully adjusted air volume was determined through regression analysis.

[0043] Based on the current comprehensive dispersion index Found Calculate the wind pressure adjustment step size. .

[0044] in This is the baseline primary air pressure under the current operating conditions. Calculate the secondary damper opening compensation value. .in This is the reference opening degree for each secondary air damper. It is a structural coefficient used to scale the adjustment amount according to the burner arrangement and wind box structure of different boilers. It is provided by the boiler manufacturer or calibrated through cold aerodynamic field tests.

[0045] The calculated primary wind pressure adjustment step size and secondary damper opening compensation value The data is packaged into a fluidization rate optimization package and sent to the primary air fan inverter and the secondary damper actuator. For example, the new control target is: primary air pressure. Specific floor secondary air damper opening .

[0046] Furthermore, the steps of encapsulating the adjustment step size and aperture compensation values ​​into a fluidization rate optimization package include: The particle residence time parameter in the furnace is monitored and calculated by the ratio of furnace height to real-time fluidizing velocity. Determine whether the current particle residence time parameter meets the time threshold requirement for the burnout of residual carbon in the gasification slag. If not, add a flow field damping coefficient to the fluidization velocity optimization package to extend the particle reaction time by reducing the local flow velocity.

[0047] Among them, the particle residence time parameter According to the furnace height Compared with the currently calculated real-time fluidization velocity The estimation formula is as follows: .

[0048] It is calculated based on the total primary air volume, furnace cross-sectional area, and temperature correction.

[0049] Preset a threshold for residual carbon burnout time This threshold was obtained through laboratory combustion experiments on typical gasification furnace slag particles. If the calculated... This indicates that the wind speed is too high.

[0050] At this point, a flow field damping coefficient is introduced before sending the fluidization velocity optimization packet. For example, take 0.7. Adjust the step size of the originally calculated primary wind pressure. Revised to .

[0051] This means that even if an increase in airflow is needed to stabilize combustion, the increase will be actively suppressed; or if the original instruction is to reduce air pressure, the reduction will be smaller; thus prioritizing ensuring that the particles have sufficient reaction time and reducing mechanical incomplete combustion losses.

[0052] Furthermore, when the rate of change exceeds a preset safety gradient, the step of generating a dust accumulation cleaning control signal includes: The exhaust temperature values ​​of the tail flue were continuously recorded at a fixed sampling frequency to construct a temperature time series dataset; The temperature time series dataset is processed by first-order differentiation to obtain the instantaneous rise slope of flue gas temperature. Combined with the proportion of gasification slag in the current fuel mixing ratio instruction, the ash accumulation sensitivity factor is calculated. The product of the instantaneous rise slope and the ash accumulation sensitivity factor is defined as the coking risk value. When the coking risk value continues to be higher than the preset alarm threshold, the execution action of increasing the frequency of steam soot blowing and the intensity of soot blowing is triggered.

[0053] The calculation process for the rate of change is as follows: The flue gas temperature time series collected at fixed periods... Perform a first-order numerical differential to calculate the instantaneous slope of ascent. g represents the sampling count index; This refers to the time point corresponding to the g-th sampling. This refers to the time point corresponding to the (g-1)th sampling.

[0054] Based on the fact that gasifier slag ash typically has a high tendency to coke, the ash accumulation sensitivity factor was obtained. .

[0055] in, This refers to the mass percentage of gasification slag in the current fuel mix. It is an empirical coefficient, determined based on the characteristics of the gasification slag used, such as ash melting point and alkali metal content. For example, a larger value is taken for slag with a low ash melting point.

[0056] The coking risk value is obtained based on the instantaneous rise slope of the flue gas temperature and the ash accumulation sensitivity factor of the fuel itself. .

[0057] Set an alarm threshold. For example, 0.5. When the coking risk value consecutively exceeding A period (e.g., 5 minutes) exceeding the alarm threshold If a risk of accelerated dust accumulation is detected, a dust cleaning control signal is automatically generated.

[0058] The ash removal control signal will notify the soot blowing control system to temporarily insert an additional steam soot blowing or increase the intensity of the sonic soot blower in a specific area.

[0059] Furthermore, the steps following the triggering of the actions to increase the frequency of steam soot blowing and the intensity of sonic soot blowing include: Continuously monitor the flue gas temperature drop curve after the action is performed; Calculate the response time required for the flue gas temperature to drop back to the baseline temperature, mark this response time as the cleaning effect feedback value, and store this feedback value in the historical operation database to update the correction weight when calculating the coking risk value next time.

[0060] That is, after triggering the soot blowing action, continuously monitor the flue gas temperature drop curve. Record the temperature from the start of soot blowing. The flue gas temperature dropped to the historical baseline temperature under this load and ratio. Response time experienced Response time The shorter the time, the better the dust removal effect and the less serious the dust accumulation; conversely, the longer the time, the more stubborn the dust accumulation may be or the warning may be delayed.

[0061] Response time As a feedback value of cleaning effect The characteristics of the event, along with the specific details of the event, are stored in the historical database. These characteristics include fuel ratio, load, and batch of gasification slag used.

[0062] In subsequent calculations of the ash accumulation sensitivity factor At that time, the empirical coefficients used It will be fine-tuned and updated based on historical feedback. The update rules include: if the cleaning effect feedback value under the most recent N similar operating conditions... A longer duration indicates more severe dust accumulation or a later warning; therefore, the intensity should be increased slightly according to a certain proportion. This means that when a similar situation occurs again, the coking risk value will reach the threshold earlier, thus triggering more timely dust removal.

[0063] Further steps in collecting basic attribute data of gasification furnace slag and raw coal include: The digital interface of the online moisture analyzer is connected to read the real-time conductivity data of the gasification slag, and the total moisture content is obtained by inversion through the preset conductivity and moisture conversion model. The total moisture content is compared with the preset drying threshold. When the total moisture content is higher than the threshold, a scheduling instruction is generated for the turning machine in the drying yard until the data collected again meets the requirements.

[0064] The specific steps for online moisture detection include: using an online microwave or infrared moisture analyzer to measure the electrical conductivity of the gasification furnace slag sample in real time. Through a conductivity-moisture conversion model The total moisture content was obtained through inversion. .

[0065] Among them, parameters and Moisture content of slag samples with different moisture contents was determined by laboratory standard drying method at the production site, and their electrical conductivity was measured simultaneously. Then, curve fitting is performed to obtain the result.

[0066] The total moisture content value detected online. Compared with the preset drying qualification threshold Comparison, for example: It is 15%.

[0067] If the moisture content exceeds the standard, an instruction will be automatically generated to notify the fuel management unit to increase the frequency of turning the drying yard or extend the drying time until it passes the re-inspection.

[0068] like Figure 2The diagram shown is a structural schematic of a co-firing system for gasification slag and pulverized coal boiler provided in an embodiment of this application. The co-firing system for gasification slag and pulverized coal boiler provided in an embodiment of this application includes: a data acquisition module, an operating condition screening module, an instruction generation module, a deviation value calculation module, and an instruction correction module. The data acquisition module is used to collect basic attribute data of gasification slag and raw coal, and extract the corresponding historical combustion condition dataset from the historical operation database. The operating condition screening module is used to construct the fuel feature vector of the current batch based on the basic attribute data, perform multi-dimensional spatial similarity calculation between the fuel feature vector and the sample vector marked as stable operation in the historical combustion operating condition dataset, and screen out the target historical operating condition with the highest matching degree. The instruction generation module is used to extract the mixing ratio parameters from the target historical operating conditions, and combine them with the real-time load demand data of the current boiler to generate the fuel mixing ratio instruction for the current moment. The deviation calculation module is used to simultaneously collect real-time combustion feedback data in the furnace during the execution of the fuel mixing ratio command, and calculate the discrete deviation value between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions. The instruction correction module is used to generate a dynamic correction instruction for the primary and secondary air ratio based on the discrete deviation value, and simultaneously calculate the time series change rate of the flue gas temperature. When the change rate exceeds the preset safety gradient, an ash cleaning control signal is generated.

[0069] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the specific device can be divided into different functional modules to complete all or part of the functions described above.

[0070] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0071] In the embodiments covered by this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the division of modules is merely a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connections shown or discussed may be indirect coupling or communication connections through some interfaces, devices, or modules, or they may be electrical, mechanical, or other forms of connection.

[0072] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0073] Furthermore, the functional modules in the various embodiments of this application can be implemented either in hardware or as software functional modules. If these functional modules are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a computer program product, which includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program product is stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0074] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. A co-firing process for gasification furnace slag and pulverized coal boiler, characterized in that, Includes the following steps: Collect basic attribute data of gasification furnace slag and raw coal, and extract the corresponding historical combustion condition dataset from the historical operation database; Based on the aforementioned basic attribute data, a fuel feature vector for the current batch is constructed. The fuel feature vector is then compared with the sample vectors marked as stable operation in the historical combustion condition dataset using multidimensional spatial similarity calculation. The target historical operating condition with the highest matching degree is then selected. Extract the mixing ratio parameters from the target historical operating conditions, and combine them with the current boiler's real-time load demand data to generate the fuel mixing ratio instruction for the current moment; During the execution of the fuel mixing ratio command, real-time combustion feedback data in the furnace is collected simultaneously, and the discrete deviation value between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions is calculated. Based on the discrete deviation value, a dynamic correction command for the primary and secondary air ratio is generated, and the time series change rate of the flue gas temperature is calculated simultaneously. When the change rate exceeds the preset safety gradient, an ash cleaning control signal is generated.

2. The co-firing process of gasification slag and pulverized coal boiler as described in claim 1, characterized in that: The steps for constructing the fuel feature vector for the current batch based on the aforementioned basic attribute data and filtering out the target historical operating conditions with the highest matching degree include: The collected moisture content values ​​of gasification slag and raw coal were normalized to construct a ternary feature coordinate system including moisture, reference calorific value and volatile matter dimensions. Define a search sliding window in the historical operation database, traverse all historical sample vectors within the window, and calculate the Euclidean distance between the current ternary feature coordinates and each historical sample vector; Select the top K historical sample vectors with the smallest Euclidean distance, and statistically analyze the combustion efficiency label values ​​corresponding to these K sample vectors. The sample with the highest combustion efficiency label value is determined as the target historical operating condition, and the fuel pretreatment parameters under this operating condition are extracted as benchmark reference data.

3. The co-firing process of gasification slag and pulverized coal boiler as described in claim 1, characterized in that: The steps for extracting the mixing ratio parameters from the target historical operating conditions and combining them with the current boiler's real-time load demand data to generate the fuel mixing ratio instruction for the current moment include: Read the original mixing ratio value from the target historical operating conditions and obtain the target total calorific value from the current boiler's real-time load demand data; The original mixing ratio is adjusted by weighting a correction coefficient derived from the difference between the current fuel feature vector and the target historical operating condition feature vector, as well as the combustion feedback data of the previous time step. The original mixing ratio is adjusted by weighting the correction coefficient to establish a linear balance equation between the fuel mixing ratio and the total target calorific value. The equation is solved to obtain the accurate blending mass ratio that meets the current load requirements. This mass ratio is then converted into a frequency control signal for the feeder.

4. The co-firing process of gasification slag and pulverized coal boiler as described in claim 1, characterized in that: The steps of synchronously acquiring real-time combustion feedback data in the furnace and calculating the discrete deviation between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions include: The current oxygen distribution data and flame center temperature data are acquired in real time through the furnace sensor array to form a real-time status matrix; The generated fuel mixing ratio instruction is used as an index key to search the historical database for the standard state matrix that should be theoretically achieved under this specific ratio. The oxygen distribution deviation and temperature center offset are obtained by performing a difference operation on the real-time state matrix and the standard state matrix, and these two deviations are combined into a comprehensive discrete deviation index that characterizes combustion instability.

5. The co-firing process of gasification slag and pulverized coal boiler as described in claim 1, characterized in that: The steps for generating dynamic correction commands for the primary and secondary air distribution ratio based on discrete deviation values ​​include: The comprehensive discrete deviation index is input into a preset graded adjustment rule base, which contains air volume gain coefficients corresponding to different deviation ranges; Based on the current range of the comprehensive discrete deviation index, match the corresponding air volume gain coefficient, and calculate the adjustment step size of the primary air pressure and the opening compensation value of the secondary damper. The adjustment step size and opening compensation value are encapsulated into a fluidization velocity optimization package and sent to the fan actuator.

6. The co-firing process of gasification slag and pulverized coal boiler as described in claim 5, characterized in that: The steps to encapsulate the adjustment step size and aperture compensation values ​​into a fluidization rate optimization package include: The particle residence time parameter in the furnace is monitored and calculated by the ratio of furnace height to real-time fluidizing velocity. Determine whether the current particle residence time parameter meets the time threshold requirement for the burnout of residual carbon in the gasification slag. If not, add a flow field damping coefficient to the fluidization velocity optimization package.

7. The co-firing process of gasification slag and pulverized coal boiler as described in claim 1, characterized in that: When the rate of change exceeds the preset safety gradient, the steps for generating the ash accumulation cleaning control signal include: The exhaust temperature values ​​of the tail flue were continuously recorded at a fixed sampling frequency to construct a temperature time series dataset; The temperature time series dataset is processed by first-order differentiation to obtain the instantaneous rise slope of flue gas temperature. Combined with the proportion of gasification slag in the current fuel mixing ratio instruction, the ash accumulation sensitivity factor is calculated. The product of the instantaneous rise slope and the ash accumulation sensitivity factor is defined as the coking risk value. When the coking risk value continues to be higher than the preset alarm threshold, the execution action of increasing the frequency of steam soot blowing and the intensity of soot blowing is triggered.

8. The co-firing process of gasification slag and pulverized coal boiler as described in claim 7, characterized in that: The steps following the triggering of actions to increase the frequency of steam soot blowing and the intensity of acoustic soot blowing include: Continuously monitor the flue gas temperature drop curve after the action is performed; Calculate the response time required for the exhaust gas temperature to drop back to the baseline temperature, mark this response time as the cleaning effect feedback value, and store this feedback value in the historical operation database.

9. The co-firing process of gasification slag and pulverized coal boiler as described in claim 1, characterized in that: The steps for collecting basic attribute data of gasification furnace slag and raw coal include: The digital interface of the online moisture analyzer is connected to read the real-time conductivity data of the gasification slag, and the total moisture content is obtained by inversion through the preset conductivity and moisture conversion model. The total moisture content is compared with the preset drying threshold. When the total moisture content is higher than the threshold, a scheduling instruction is generated for the turning machine in the drying yard until the data collected again meets the requirements.

10. A co-firing system for gasification furnace slag and pulverized coal boiler, used to implement the co-firing process of gasification furnace slag and pulverized coal boiler as described in any one of claims 1-9, characterized in that, include: Data acquisition module, working condition screening module, instruction generation module, deviation value calculation module, instruction correction module; The data acquisition module is used to collect basic attribute data of gasification slag and raw coal, and extract the corresponding historical combustion condition dataset from the historical operation database. The operating condition screening module is used to construct the fuel feature vector of the current batch based on the basic attribute data, perform multi-dimensional spatial similarity calculation between the fuel feature vector and the sample vector marked as stable operation in the historical combustion operating condition dataset, and screen out the target historical operating condition with the highest matching degree. The instruction generation module is used to extract the mixing ratio parameters from the target historical operating conditions, and combine them with the real-time load demand data of the current boiler to generate the fuel mixing ratio instruction for the current moment. The deviation calculation module is used to simultaneously collect real-time combustion feedback data in the furnace during the execution of the fuel mixing ratio command, and calculate the discrete deviation value between the real-time combustion feedback data and the standard combustion data in the target historical operating conditions. The instruction correction module is used to generate a dynamic correction instruction for the primary and secondary air ratio based on the discrete deviation value, and simultaneously calculate the time series change rate of the flue gas temperature. When the change rate exceeds the preset safety gradient, an ash cleaning control signal is generated.

Images