Combustion air flow control system and method for a biomass burner

Abstract

The application discloses a combustion air volume control system and method for a biomass burner, which firstly filters and differentially maps original data to extract a hot state characteristic vector, and then peels off external heat supply and inversely derives an instantaneous volatile release rate based on the energy conservation principle, solving the problem that the combustion state in the low-temperature blind area is unobservable. On this basis, a benchmark target air volume that is subject to safety constraints is generated through stoichiometric proportioning and blowout prevention index attenuation mechanism, solving the problem that a preset air volume curve cannot adaptively match volatile fluctuations and blow out the fire nucleus. Meanwhile, a furnace temperature second-order derivative polarity discrimination and perturbation compensation mechanism is introduced to inversely correct the anoxic attenuation trend, and through adaptive inhibition of the convection heat dissipation penalty factor and the Lorenz attenuation kernel, the problem that when the compensation air has the dual identities of combustion-supporting oxygen source and convection cold source, the linear superposition leads to the out-of-control of the compensation proportion is solved, and finally a safe and reliable air distribution control instruction is generated through engineering mapping and amplitude truncation.

Description

Technical Field

[0001] This application relates to the field of combustion control technology, and more specifically, to a combustion air volume control system and method for a biomass burner. Background Technology

[0002] Biomass energy, as a carbon-neutral renewable energy source, has broad application prospects in the field of co-firing in coal-fired power units. As the core device for achieving in-furnace co-firing, the precise control of the combustion air volume in biomass burners directly determines combustion efficiency and operational safety. Among the various operating conditions of biomass burners, the cold furnace ignition stage is the most vulnerable and difficult to control. At this time, the furnace temperature is extremely low, and the release of biomass volatiles exhibits a strong nonlinear characteristic. The dynamic matching relationship between its release rate and the air supply rate is extremely sensitive. Once the air distribution becomes unbalanced, it can easily lead to sudden flameout, severely restricting the ignition success rate and operational reliability of biomass burners.

[0003] Existing combustion airflow control technologies have fundamental limitations in addressing this stage: Firstly, conventional solutions rely on tail-end flue oxygen sensors for feedback air distribution. However, in the low-temperature environment of a cold furnace, these sensors have not yet reached their effective operating temperature, and there is a significant physical delay in flue gas transmission, preventing the control system from sensing the actual volatile matter release status within the furnace in real time. Secondly, some solutions employ open-loop control using a preset airflow curve that increases linearly over time. This fails to adaptively adjust based on fluctuations in volatile matter release caused by batch-to-batch differences in fuel moisture content. When the airflow exceeds the heat balance threshold that the fire core can withstand for self-sustaining combustion, a large amount of unheated cold air plunders heat from the fire core through forced convection, resulting in blow-out flameout. Furthermore, even with feedforward compensation mechanisms to slightly perturb the baseline airflow and alleviate local oxygen deficiency, the compensation air, upon entering the low-temperature furnace, simultaneously serves as both a combustion oxygen source and a convective cold source. A nonlinear antagonistic coupling exists between these two entities, and simple linear superposition cannot quantify this heat dissipation penalty effect. The fire core may still be submerged by cold due to uncontrolled compensation ratios.

[0004] Therefore, an optimized combustion air volume control scheme for biomass burners is desired. Summary of the Invention

[0005] To address the aforementioned technical problems, this application is proposed. Embodiments of this application provide a combustion airflow control system and method for a biomass burner.

[0006] According to one aspect of this application, a method for controlling the combustion air volume of a biomass burner is provided, comprising: S1, During the cold furnace ignition stage, the original sensor data stream is filtered, extracted, and differentially mapped to obtain the thermal feature vector; S2, perform energy stripping and inversion processing on the thermal feature vector to obtain the instantaneous volatile release rate; S3, estimate the benchmark target air volume under the anti-blowout safety constraint based on the instantaneous volatile release rate and the hot characteristic vector to obtain the benchmark target air volume; S4, polarity discrimination and perturbation compensation processing are performed on the baseline target air volume and thermal characteristic vector to obtain the compensated target air volume; S5 performs engineering mapping and amplitude limiting truncation on the compensated target air volume to obtain the air distribution actuator control command. The air distribution actuator control command is output to the variable frequency drive device of the combustion fan to complete the dynamic air volume regulation during the ignition stage.

[0007] According to another aspect of this application, a combustion airflow control system for a biomass burner is provided, comprising: The filtering extraction and differential mapping module is used to filter, extract and differentially map the raw sensor data stream during the cold furnace ignition stage to obtain the thermal feature vector; The energy stripping and inversion module is used to perform energy stripping and inversion processing on the thermal feature vector to obtain the instantaneous volatile release rate. The benchmark target air volume estimation module is used to estimate the benchmark target air volume under the anti-blowout safety constraint based on the instantaneous volatile release rate and the hot characteristic vector to obtain the benchmark target air volume; The polarity discrimination and perturbation compensation module is used to perform polarity discrimination and perturbation compensation processing on the baseline target air volume and thermal characteristic vector to obtain the compensated target air volume; The engineering mapping and limiting truncation module is used to perform engineering mapping and limiting truncation processing on the compensated target air volume to obtain the air distribution actuator control command. The air distribution actuator control command is output to the variable frequency drive device of the combustion fan to complete the dynamic air volume regulation during the ignition stage.

[0008] Compared with existing technologies, this invention proposes a combustion airflow control method for biomass burners. It abandons the reliance on feedback signals from tail-end flue gas oxygen sensors and preset time-varying airflow curves, instead utilizing the first and second time derivatives of the furnace transient temperature, combined with igniter power, to perform energy stripping inversion and estimate the instantaneous release rate of biomass volatiles in real time. This estimate serves as the feedforward driving source for air distribution decisions, making the air supply a dynamic follower of the volatile release state. Furthermore, an exponential approximation decay function with the absolute value of furnace temperature as the independent variable is introduced to construct anti-blowout safety constraints, ensuring that the amount of cold air entering the furnace is strictly suppressed at extremely low furnace temperatures. Simultaneously, the polarity of the second derivative of furnace temperature is used to activate oxygen deficiency perturbation compensation. Furthermore, considering the antagonistic coupling characteristics of the compensating air acting as both a combustion oxygen source and a convective cold source, a nonlinear quadratic heat dissipation penalty factor based on the proportion of compensating air volume is constructed. The compensation amplitude is then adaptively and flexibly suppressed using a Lorentz-type decay kernel function. This achieves a continuous and smooth dynamic balance between the two opposing constraints of preventing blowout and preventing asphyxiation, thereby improving the robustness of combustion air volume control and ignition success rate during the cold furnace ignition stage. Attached Figure Description

[0009] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0010] Figure 1 This is a flowchart of a combustion air volume control method for a biomass burner according to an embodiment of this application; Figure 2 This is a schematic diagram of the data flow of a combustion air volume control method for a biomass burner according to an embodiment of this application; Figure 3 This is a flowchart illustrating the process of energy stripping and inversion of the thermal feature vector to obtain the instantaneous volatile release rate in the combustion air volume control method for a biomass burner according to embodiments of this application. Figure 4 This is a flowchart illustrating the baseline target air volume estimation based on the instantaneous volatile release rate and thermal characteristic vector under the anti-blowout safety constraint of the combustion air volume control method for biomass burners according to embodiments of this application. Figure 5 This is a flowchart illustrating the method for controlling combustion air volume in a biomass burner according to an embodiment of this application, which involves polarity discrimination and perturbation compensation processing of a baseline target air volume and a thermal characteristic vector to obtain a compensated target air volume. Figure 6This is a flowchart illustrating the process of superimposing a reference target air volume and a perturbation air volume compensation value in a set of variables to be superimposed to obtain a compensated target air volume according to the combustion air volume control method for a biomass burner in the embodiments of this application. Figure 7 This is a block diagram of a combustion air volume control system for a biomass burner according to an embodiment of this application. Detailed Implementation

[0011] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0012] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0013] While this application makes various references to certain modules in the systems according to embodiments of this application, any number of different modules can be used and run on the central control room monitoring host of a distributed control system, an industrial-grade programmable logic controller in the burner local control cabinet, a field edge computing gateway, and / or a cloud platform. The modules described are merely illustrative, and different aspects of the systems and methods may use different modules.

[0014] Flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0015] In existing biomass burners, during the cold ignition stage, the oxygen sensor suffers from severe measurement delays and malfunctions at low temperatures. Furthermore, the preset open-loop airflow curve cannot adapt to the nonlinear volatile matter precipitation caused by fuel fluctuations, making the flame core highly susceptible to extinguishing due to excessive cold air intrusion. Simultaneously, existing feedforward compensation mechanisms simply linearly superimpose compensation air, ignoring the nonlinear antagonistic coupling relationship where cold air serves as both a combustion oxygen source and a convective cold source. This often leads to uncontrolled heat dissipation penalties, causing the flame core to be submerged by cold air and abruptly extinguish. Therefore, this application proposes a combustion airflow control method for biomass burners. Through multi-dimensional feature extraction, soft measurement of volatile matter, and antagonistic adaptive compensation, a closed-loop dynamic airflow control path is constructed for the entire cold ignition process. Specifically, this method first performs deburring filtering and differential mapping on the transient furnace temperature, igniter power, and feed rate to extract thermal feature vectors containing the first and second derivatives of the furnace temperature. Then, based on the law of energy conservation, it removes the external forced heating from the total heat accumulation rate, thereby accurately reflecting the true instantaneous volatile release rate within the low-temperature sensing blind zone. On this basis, combined with the stoichiometric air-fuel ratio and the anti-blowout index decay mechanism, a benchmark target air volume strictly limited by furnace temperature safety is generated to avoid sudden cold air intrusion. Subsequently, the second derivative of the furnace temperature is extracted for polarity discrimination to capture the oxygen deficiency and asphyxiation trend. When activating perturbation compensation, a convective heat dissipation penalty factor reflecting the compensation ratio is innovatively constructed, and the Lorentz decay kernel is used to dynamically and adaptively suppress the compensation amplitude, effectively resolving the antagonistic contradiction between combustion support and heat dissipation. Finally, the compensated air volume is mapped by an electromechanical transfer function and truncated by dual extreme value limiting to generate a safe and reliable air distribution control command, which is then sent to the frequency converter, comprehensively ensuring the combustion stability and success rate during the ignition stage.

[0016] Figure 1 This is a flowchart of a combustion air volume control method for a biomass burner according to an embodiment of this application. Figure 2 This is a schematic diagram of the data flow in a combustion airflow control method for a biomass burner according to an embodiment of this application. Figure 1 and Figure 2As shown, the combustion air volume control method for a biomass burner according to an embodiment of this application includes the following steps: S1, during the cold furnace ignition stage, filtering, extracting, and differentially mapping the original sensor data stream to obtain a thermal feature vector; S2, performing energy stripping and inversion processing on the thermal feature vector to obtain the instantaneous volatile matter release rate; S3, estimating the benchmark target air volume under the anti-blowout safety constraint on the instantaneous volatile matter release rate and the thermal feature vector to obtain the benchmark target air volume; S4, performing polarity discrimination and perturbation compensation processing on the benchmark target air volume and the thermal feature vector to obtain the compensated target air volume; S5, performing engineering mapping and amplitude limiting truncation processing on the compensated target air volume to obtain the air distribution actuator control command, which is output to the variable frequency drive device of the combustion fan to complete the dynamic air volume regulation during the ignition stage.

[0017] Specifically, in step S1, during the cold furnace ignition stage, the original sensor data stream is filtered, extracted, and differentially mapped to obtain a thermal feature vector. It should be noted that during the initial cold furnace ignition stage, electric sparks and unburned particles can cause pulse interference to the sensor, and the weak thermodynamic evolution trend cannot be directly quantified. Furthermore, the misaligned, fragmented data does not meet the matrix operation requirements for subsequent energy inversion. Therefore, the technical solution of this application first filters, extracts, and differentially maps the original sensor data stream during the cold furnace ignition stage to obtain a thermal feature vector. Through this processing, pulse noise in the original signal is eliminated, and the dynamic evolution gradient of furnace thermal energy accumulation is simultaneously captured, providing structured input features for subsequent volatile matter soft measurement.

[0018] More specifically, in a concrete example of this application, the transient furnace temperature, igniter power, and initial biomass feed rate contained in the original sensor data stream are subjected to deburring median filtering smoothing to obtain a smoothed feature set, which includes smoothed furnace temperature, smoothed power, and smoothed feed rate. For each collected discrete sampled value, a sliding time window centered on the current sampling time and including look-ahead and historical data is constructed. All sampled values ​​within the window are sorted by numerical value, and the median value is taken as the smoothed signal at the current time, achieving the extraction of the true signal baseline without causing phase delay. The relevant calculations conform to the following relationship: in, This indicates that any basic element, after filtering, appears at time [time value missing]. The smoothing value, The mathematical operator that takes the median of a set is denoted as . This represents the historical, current, and look-ahead discrete sampled values ​​within the sliding window. This represents the single-sided step radius of the sliding window. This represents the discrete sampling period. Next, the transient temperature differential evolution gradient is calculated for the smoothed furnace temperature in the smoothed feature element set. The first and second derivatives of the furnace temperature are then converged with the smoothed furnace temperature, smoothed electrical power, and smoothed feed rate from the smoothed feature element set to obtain the enhanced feature element set. After separating the smoothed furnace temperature, a backward difference method is used to calculate the first time derivative of the furnace temperature, representing the heating rate, based on the temperature difference between the current time and the previous sampling period. Then, the calculated first derivative sequence is used again for difference differentiation to obtain the second time derivative of the furnace temperature, reflecting the volatile matter deflagration trend. The difference calculation conforms to the following relationship: in, This represents the first derivative of the furnace temperature. This represents the second derivative of the furnace temperature. This indicates the current smoothed furnace temperature. This represents the smoothed furnace temperature of the previous sampling period. Finally, the smoothed furnace temperature, smoothed electrical power, smoothed feed rate, first derivative of furnace temperature, and second derivative of furnace temperature in the enhanced feature set are tensor-quantized and encapsulated to obtain the thermal feature vector. Based on a pre-calibrated dimensionality mask, the smoothed basic state variables and evolution gradient variables are mapped and combined into a one-dimensional column vector according to the order of physical dimensions, eliminating the fragmentation of data during memory transfer and meeting the format requirements for simultaneous input of diverse heterogeneous data. The tensor-quantized encapsulation structure conforms to the following relationship: in, This represents the final encapsulated composite state vector. For the smoothed electrical power of the mapping, The smooth feed rate is the mapped rate.

[0019] Specifically, in step S2, the thermal characteristic vector is subjected to energy stripping and inversion processing to obtain the instantaneous volatile release rate. It should be noted that, given that the oxygen sensor in the tail flue has not yet reached its effective operating temperature and there is a significant physical delay during the cold furnace ignition stage, it is impossible to directly measure the nonlinear volatile release state of up to 60% to 70% within the biomass fuel. Furthermore, the overall heat accumulation in the initial stage of the furnace is actually a mixture of forced heating heat release from the igniter and weak spontaneous combustion heat release from the volatiles. Based on this, the technical solution of this application further performs energy stripping and inversion processing on the thermal characteristic vector to obtain the instantaneous volatile release rate. Through the above processing, external forced heating interference is effectively eliminated from the total heat accumulation rate of the system based on the principle of energy conservation. Within the low-temperature blind zone where the sensor fails, the pure thermodynamic evolution gradient is reverse-converted across the domain into a physical release rate with real mass properties, providing direct data support for subsequent accurate matching of aerodynamic loads.

[0020] Figure 3This is a flowchart illustrating the process of energy stripping and inversion of a thermal characteristic vector to obtain the instantaneous volatile matter release rate in a combustion airflow control method for a biomass burner according to an embodiment of this application. Figure 3 As shown, step S2 includes: S21, obtaining a heat analysis balance group by performing dimensional analysis and addressing extraction on the thermal feature vector; S22, removing the external heat source from the total heat accumulation rate and igniter power in the heat analysis balance group to obtain the virtual net combustion heat release; S23, based on the preset biomass volatile calorific value constant and system calibration mapping coefficient, performing physicochemical parameter division mapping and release rate inversion calculation on the virtual net combustion heat release to obtain the instantaneous volatile release rate.

[0021] In step S21, a heat analysis equilibrium group is obtained by performing dimensionality analysis and addressing extraction on the thermal feature vector. It should be noted that since the encapsulated feature vector contains multi-dimensional physical parameters, directly using it for energy inversion would introduce computational interference from irrelevant data, and a single temperature change rate cannot equivalently represent the true energy accumulation state of the furnace in terms of physical dimensions. Therefore, the technical solution of this application further obtains a heat analysis equilibrium group by performing dimensionality analysis and addressing extraction on the thermal feature vector. Through the above processing, the extracted temperature evolution gradient is effectively transformed into the overall heat absorption dynamic rate of the system, providing an accurate data input boundary for removing the external forced heating source.

[0022] More specifically, in a concrete example of this application, firstly, dimensionality mask parsing is performed on the input thermal feature vector to selectively extract the first-order time derivative of the furnace transient temperature, reflecting the current furnace heating rate, and the igniter power term, representing the external energy injection conditions. Based on this, a system equivalent comprehensive heat capacity coefficient representing the specific cold furnace physical structure is introduced. The extracted furnace temperature first-order time derivative is multiplied by this heat capacity coefficient, transforming the pure temperature change rate into a physical dynamic rate representing the overall heat absorption and accumulation of the furnace system at the current moment. The relevant heat conversion calculation conforms to the following relationship: in, This represents the total heat accumulation rate calculated at the current moment. This represents the system's equivalent comprehensive heat capacity curing reference value, obtained by pre-calibrating the thermophysical constants of the water-cooled walls and internal media of the industrial furnace. This represents the first-order time derivative of the furnace temperature at the current moment, extracted from the thermal characteristic vector. After completing the above state transformation calculation, the obtained total heat accumulation rate data and the synchronously extracted igniter power data are tightly coupled, packaged, and recombined to generate a heat analysis balance group.

[0023] In step S22, the total heat accumulation rate and igniter power in the heat analysis balance group are separated from the external heat source to obtain the virtual net combustion heat release. It should be noted that, in the cold furnace ignition and anti-blowout scenario, the overall heat accumulation in the furnace is essentially a superposition of the forced baking heat release from the external igniter and the spontaneous combustion heat release from the initial volatile matter of biomass. Directly using the total heat accumulation rate cannot measure the true combustion state of the fuel itself. Therefore, the technical solution of this application further separates the total heat accumulation rate and igniter power in the heat analysis balance group from the external heat source to obtain the virtual net combustion heat release. Through the above processing, the interference of the external forced heat source is effectively eliminated based on the law of conservation of energy, and the net heat generation data fed back from the weak ignition of the fuel itself is obtained.

[0024] More specifically, in a concrete example of this application, the input heat analysis balance group is first analyzed to read the total heat accumulation rate and igniter power. A heat stripping model is established based on the law of energy conservation, using the total heat accumulation rate as the benchmark data for measuring the total heat absorption of the furnace system. Considering the heat conversion loss in the electric heating process, the igniter power is multiplied by the static electrothermal conversion efficiency calibration constant to calculate the actual effective external heating power injected into the furnace. Subsequently, by subtracting this effective external heating power from the total heat accumulation rate, the externally imposed heat source is dynamically stripped away. The remaining heat value is determined to be the spontaneous combustion heat release from early volatile matter precipitation at low landing point temperatures, thus generating virtual net combustion heat release. The external heat source stripping calculation conforms to the following relationship: in, This represents the virtual net heat release from combustion, calculated using heat stripping techniques at the current moment. This represents the extracted total heat accumulation rate. This represents the static electrothermal conversion efficiency calibration constant of the igniter, and its value is greater than 0 and less than or equal to 1. This indicates the igniter's electrical power.

[0025] In step S23, based on the preset biomass volatile matter calorific value constant and system calibration mapping coefficient, the virtual net combustion heat release is mapped by physicochemical parameters and the release rate is inverted to obtain the instantaneous volatile matter release rate. It should be noted that since pure heat values ​​belong to the thermodynamic domain, they cannot be directly stoichiometrically matched with the inlet air volume at the fluid dynamics level, and the non-adiabatic heat dissipation attenuation at extremely low furnace temperatures will cause deviations between the theoretical calculation value and the actual fuel release mass. Therefore, the technical solution of this application further uses the preset biomass volatile matter calorific value constant and system calibration mapping coefficient to map the virtual net combustion heat release by physicochemical parameters and the release rate is inverted to obtain the instantaneous volatile matter release rate. Through the above processing, the cross-domain physical conversion from the heat domain to the mass domain is effectively completed, accurately obtaining dynamic fuel release indicators with true mass attributes while offsetting multiple boundary deviation effects.

[0026] More specifically, in a particular example of this application, a division operation is performed on the obtained virtual net combustion heat release, dividing it by the empirical volatile matter unit calorific value constant for the currently selected biomass species to obtain the preliminary volatile matter mass quotient under ideal conditions. To offset the real-time radiative heat dissipation attenuation and convective deviation effects faced by the non-adiabatic furnace in low-temperature environments, a system calibration mapping coefficient is introduced, which integrates physical conditions such as sensor measurement point position attenuation and pipeline leakage. The calculated preliminary volatile matter mass quotient is multiplied by this calibration mapping coefficient for dynamic compensation correction, accurately inverting the heat parameters into the instantaneous biomass volatile matter release mass flow rate in real space, and then outputting the instantaneous volatile matter release rate. This physicochemical parameter inversion calculation conforms to the following relationship: in, This represents the instantaneous volatile release rate generated at the current moment. This represents the system comprehensive calibration mapping coefficients. This represents the input virtual net heat release from combustion. This represents the standardized lower heating value constant of the volatile components in the target biomass fuel.

[0027] Specifically, in step S3, the instantaneous volatile release rate and thermal characteristic vector are used to estimate the benchmark target air volume under the anti-blowout safety constraint to obtain the benchmark target air volume. It should be noted that, given that during the cold furnace ignition stage, if air distribution is based solely on stoichiometric equilibrium without considering the low furnace temperature environment, the influx of a large amount of unpreheated cold air will cause convective heat dissipation to exceed the meager heat release from combustion of volatiles, thereby disrupting the thermal balance of the initial fire core and triggering blowout-type fire suppression. Based on this, the technical solution of this application further estimates the benchmark target air volume under the anti-blowout safety constraint based on the instantaneous volatile release rate and thermal characteristic vector to obtain the benchmark target air volume. Through the above processing, the volatile mass parameter is effectively converted into theoretical air volume, and a temperature-based thermodynamic safety boundary limit is forcibly applied to this theoretical air distribution, ensuring that cold air supply is suppressed at extremely low absolute furnace temperatures, avoiding the risk of sudden fire core extinguishing due to the sudden influx of cold air.

[0028] Figure 4 This is a flowchart illustrating the estimation of the baseline target airflow under anti-blowout safety constraints using the instantaneous volatile matter release rate and thermal characteristic vector in the combustion airflow control method for biomass burners according to embodiments of this application, to obtain the baseline target airflow. Figure 4 As shown, step S3 includes: S31, separating the smooth furnace temperature by unpacking and addressing the thermal feature vector, and calculating the theoretical air volume ratio based on the instantaneous volatile release rate to obtain the theoretical air demand, and then encapsulating the theoretical air demand and the smooth furnace temperature to obtain the air distribution basic data pair; S32, determining the anti-blowing attenuation factor based on the smooth furnace temperature in the air distribution basic data pair, and then recombining the anti-blowing attenuation factor and the theoretical air demand in the air distribution basic data pair to obtain the restricted air distribution combination; S33, truncating the theoretical air demand and the anti-blowing attenuation factor in the restricted air distribution combination to obtain the benchmark target air volume.

[0029] In step S31, the smooth furnace temperature is separated by unpacking and addressing the thermal characteristic vector, and the theoretical air volume ratio is calculated based on the instantaneous volatile release rate to obtain the theoretical air volume requirement. The theoretical air volume requirement and the smooth furnace temperature are then encapsulated to obtain the basic air distribution data pair. It should be noted that since the instantaneous volatile release rate is a mass flow rate parameter, it cannot be directly mapped to the volumetric flow rate parameter required by the actuator, and the subsequent construction of the anti-blowout safety boundary needs to rely on the current absolute thermodynamic state. Based on this, the technical solution of this application further separates the smooth furnace temperature by unpacking and addressing the thermal characteristic vector, and calculates the theoretical air volume requirement based on the instantaneous volatile release rate. The theoretical air volume requirement and the smooth furnace temperature are then encapsulated to obtain the basic air distribution data pair. Through the above processing, the cross-domain conversion of volatile mass to the air volume required for ideal combustion is effectively completed, and a close coupling relationship is established between the theoretical air distribution requirement and the current thermal characteristics.

[0030] More specifically, in a concrete example of this application, an addressing and unpacking operation is performed on the incoming thermal feature vector to extract the single element representing the current absolute thermodynamic level, namely the smoothed furnace temperature. This temperature element is temporarily stored for downstream use, while the memory space occupied by the remaining elements in the vector is safely released to prevent data detachment. Subsequently, the incoming instantaneous volatile matter release rate with mass flow properties is multiplied by a pre-set stoichiometric ideal nominal air-fuel ratio, which is pre-calibrated based on the oxidation equilibrium equation of a specific mixture of biomass volatile matter. Through the above multiplication operation, the theoretical air volume limit required to support complete ideal combustion at the current order of magnitude of volatile matter is calculated. This theoretical air volume ratio calculation conforms to the following relationship: in, This represents the theoretical air demand calculated using stoichiometry at the current moment. This represents the instantaneous volatile release rate of the input. This represents the pre-calibrated ideal air-fuel ratio constant. Finally, the calculated theoretical air demand is packaged with the previously separated and temporarily stored smoothed furnace temperature to generate a back-to-back aggregated object and output the air distribution basic data pair.

[0031] In step S32, based on the smoothed furnace temperature in the air distribution baseline data, the anti-blowing attenuation factor is determined. Then, the anti-blowing attenuation factor is recombinated with the theoretical air demand in the air distribution baseline data to obtain a restricted air distribution combination. It should be noted that, since the furnace absolute temperature is extremely low during the initial cold furnace ignition stage, directly supplying cold air according to the theoretical air demand would cause convective heat loss to far exceed the heat released by the weak fire core, triggering a blow-out-type fire suppression. Therefore, the technical solution of this application further determines the anti-blowing attenuation factor based on the smoothed furnace temperature in the air distribution baseline data, and then recombines the anti-blowing attenuation factor with the theoretical air demand in the air distribution baseline data to obtain a restricted air distribution combination. Through the above processing, a nonlinear safety boundary based on furnace temperature is effectively constructed, forcibly suppressing cold air intrusion at low furnace temperatures and smoothly releasing the air supply restriction as temperature accumulates.

[0032] More specifically, in a concrete example of this application, the input air distribution baseline data is decomposed to separate the smoothed furnace temperature and the theoretical required air volume. Addressing the phenomenon that cold air can easily extinguish weak fire cores, the separated smoothed furnace temperature is used as the independent variable to construct and introduce an exponential approximation decay function based on the natural logarithm base. The underlying control logic of this function is configured such that when the furnace absolute temperature is low, the calculation result approaches 0 to strongly suppress cold air intrusion; as the fire bed temperature accumulates and rises, the calculation result non-linearly and smoothly approaches 1 to gradually relax the air volume restriction. This anti-extinguishing exponential decay calculation conforms to the following relationship: in, This represents the dimensionless nonlinear parameter, i.e., the anti-extinguishing attenuation factor, calculated based on the real-time furnace temperature at the current moment. Its value range is dynamically constrained to be between 0 and 1. The base of the natural logarithm. This represents the control factor for the sensitivity to blowout attenuation. This factor is calibrated by the interaction between the intensity of cold air convection heat absorption and the geometric dimensions of the furnace space, and is used to adjust the slope of the exponential curve. This represents the smoothed furnace temperature at the current moment. After the above function calculation is completed, the generated anti-exhaustion attenuation factor and the theoretical air demand transmitted upstream are repackaged to generate a restricted air distribution combination that is transmitted backward.

[0033] In step S33, the theoretical air demand and the anti-blowout attenuation factor in the constrained air distribution combination are truncated to a benchmark target air volume to obtain the benchmark target air volume. It should be noted that since the theoretical air demand only meets the chemical combustion balance requirements, if applied directly to a low-furnace-temperature ignition environment without constraint, there is still a risk of disrupting the thermal balance of the weak fire core due to excessive cold air intrusion. Based on this, the technical solution of this application further truncates the theoretical air demand and the anti-blowout attenuation factor in the constrained air distribution combination to obtain the benchmark target air volume. Through the above processing, the safety attenuation mechanism for anti-blowout is effectively forced onto the pure chemical air supply demand, generating the maximum permissible basic air volume that meets the thermodynamic safety boundary in the initial stage of ignition.

[0034] More specifically, in a concrete example of this application, the input constrained air distribution combination is deconstructed to extract the theoretical air demand and the pre-calculated anti-blowout attenuation factor. The theoretically sufficient rigid air intake to handle the current volatile matter level, i.e., the theoretical air demand, is directly multiplied by the anti-blowout attenuation factor, which acts as a safety boundary threshold. Through this multiplication truncation operation, the theoretical air supply, which originally only considered chemical equilibrium, is forcibly subject to thermodynamic safety constraints to prevent sudden cooling and flameout, thereby generating the maximum safe baseline air volume allowed to be introduced into the combustion chamber during the low furnace temperature stage of ignition. This baseline target air volume truncation calculation conforms to the following relationship: in, This represents the baseline target airflow output after being truncated through the safety attenuation mechanism at the current moment. This represents the theoretical air volume required for unpacking a self-constrained air distribution combination. This represents the blowout attenuation factor of the unconfined air distribution combination. After completing the truncation calculation, the final output baseline target air volume is generated, providing a basic reference base for subsequent feedforward compensation to address the trend of oxygen deficiency and asphyxiation.

[0035] Specifically, in step S4, the baseline target air volume and thermal characteristic vector are subjected to polarity discrimination and perturbation compensation processing to obtain the compensated target air volume. It should be noted that, given that during the cold furnace ignition stage, the extremely conservative anti-blowing baseline air volume in the early stages often leads to oxygen deficiency in the local ignition zone, causing the initial fire core, although present, to fall into a state of oxygen deficiency and suffocation. Simultaneously, the perturbation compensation air, after entering the low-temperature furnace, serves as both a combustion oxygen source and a convective cold source. Its simple linear superposition can lead to uncontrolled compensation ratios due to unpredictable quadratic nonlinear heat dissipation penalties, causing the fire core to be submerged by cold and suddenly extinguished. Based on this, the technical solution of this application further performs polarity discrimination and perturbation compensation processing on the baseline target air volume and thermal characteristic vector to obtain the compensated target air volume. Through the above processing, the smoldering decay trend can be effectively captured by determining the polarity of combustion thermodynamic acceleration. At the same time as activating feedforward perturbation, a Lorentz decay core with quantified heat dissipation penalty intensity is innovatively introduced to achieve adaptive flexible suppression of compensation amplitude. This ensures that the perturbation operation to prevent asphyxiation is strictly constrained within the safe envelope of the self-sustaining combustion of the fire core, completely eliminating the risk of sudden extinction caused by cold wind submerging the fire core.

[0036] Figure 5 This is a flowchart illustrating the method for controlling combustion airflow in a biomass burner according to embodiments of this application, which involves polarity discrimination and perturbation compensation processing of a baseline target airflow and a thermal characteristic vector to obtain a compensated target airflow. Figure 5 As shown, step S4 includes: S41, extracting the second derivative of furnace temperature by unpacking and addressing the thermal feature vector, and performing unilateral threshold polarity discrimination and negative asphyxiation feature extraction on the second derivative of furnace temperature to obtain a compensation feedforward quantity packet; S42, based on a preset sensitivity gain coefficient, calculating the linear gain of oxygen starvation counter-current compensation intensity on the negative acceleration amplitude in the compensation feedforward quantity packet to obtain the perturbation air volume compensation value, and then combining the perturbation air volume compensation value with the benchmark target air volume in the compensation feedforward quantity packet to obtain the variable group to be superimposed; S43, superimposing the benchmark target air volume and the perturbation air volume compensation value in the variable group to be superimposed with a feedforward envelope to obtain the compensated target air volume.

[0037] In step S41, the second derivative of the furnace temperature is extracted by unpacking and addressing the thermal feature vector. Then, a one-sided threshold polarity discrimination and negative suffocation feature extraction process is performed on the second derivative of the furnace temperature to obtain a compensation feedforward packet. It should be noted that, under a conservative baseline air distribution strategy in the early stages of ignition, local areas are prone to oxygen depletion, leading to delayed decay of the nascent flame core. This oxygen-deficient smoldering state cannot be detected and measured in a timely manner using a single absolute temperature value. Therefore, the technical solution of this application further extracts the second derivative of the furnace temperature by unpacking and addressing the thermal feature vector, and performs one-sided threshold polarity discrimination and negative suffocation feature extraction on the second derivative of the furnace temperature to obtain a compensation feedforward packet. Through the above processing, the negative change in combustion thermodynamic acceleration is effectively utilized to dynamically capture the suffocation decay trend of the flame core, providing a precise logical triggering benchmark for subsequent perturbation compensation.

[0038] More specifically, in a specific example of this application, an unpacking and addressing operation is performed on the thermal feature vector accompanying the data stream, extracting the second derivative of the furnace temperature, which characterizes the combustion thermodynamic acceleration, and simultaneously releasing the remaining elements within the vector to avoid data redundancy. Subsequently, the extracted second derivative of the furnace temperature is subjected to a one-sided threshold logic for polarity determination. If and only if the combustion acceleration is negative, it is determined that the local ignition zone has depleted oxygen due to conservative air intake in the early stages, and the flame core has entered a state of suffocation and hysteresis decay. Based on this, the feature capture mechanism is activated, and its absolute value is extracted as the negative acceleration amplitude. If the combustion acceleration is positive or equal to 0, its value is masked and set to zero, indicating that the flame core has not decayed and no intervention is required. The negative suffocation feature extraction calculation conforms to the following relationship: in, This indicates that the negative acceleration amplitude extracted at the current moment is... This represents the second derivative of the furnace temperature extracted from the unpacked thermal eigenvector. This represents a mathematical operator that takes the absolute value of the second derivative with respect to the furnace temperature. After completing the polarity feature extraction, the generated negative acceleration amplitude and the pre-transmitted reference target air volume are repackaged into a data package, and finally a compensation feedforward quantity packet is generated and output, providing a variable carrier containing logical trigger instructions for calculating the perturbation compensation air volume.

[0039] In step S42, based on a preset sensitivity gain coefficient, the negative acceleration amplitude in the compensation feedforward package is linearly gained to calculate the oxygen starvation counter-current compensation intensity to obtain the perturbation airflow compensation value. Then, the perturbation airflow compensation value is combined with the benchmark target airflow in the compensation feedforward package to obtain the set of variables to be superimposed. It should be noted that, since the extracted negative acceleration amplitude, after capturing the asphyxiation decay trend of the fire core, only reflects the rate of temperature hysteresis change at the thermodynamic level, it cannot be directly used for air volume intervention at the fluid dynamics level. Furthermore, intervention without proportional constraints can easily lead to the airflow exceeding the thermodynamic safety boundary. Therefore, the technical solution of this application further calculates the perturbation airflow compensation value based on a preset sensitivity gain coefficient, using linear gain to calculate the oxygen starvation counter-current compensation intensity. Then, the perturbation airflow compensation value is combined with the benchmark target airflow in the compensation feedforward package to obtain the set of variables to be superimposed. Through the above processing, the attenuation characterization parameters in the thermodynamic dimension are effectively converted across domains into the standardized air volume correction increment required to break the oxygen-deficient smoldering state, providing a quantitative fluid parameter basis for subsequent command superposition.

[0040] More specifically, in a concrete example of this application, the incoming compensation feedforward packet is unpacked and decoupled to obtain the negative acceleration amplitude and the baseline target airflow generated in the historical steps. The obtained negative acceleration amplitude is input into the proportional compensation logic specifically designed for the hypoxic suffocation condition. By multiplying it by a sensitivity gain coefficient with engineering experience attributes, it is converted and mapped into the perturbation air volume increment needed at the current moment to break the local suffocation state. The perturbation compensation intensity calculation conforms to the following relationship: in, This represents the micro-disturbance airflow compensation value calculated at the current moment, whose physical dimensions are converted to standardized airflow. This represents the hysteresis sensitivity gain multiplier, which is pre-calibrated based on the mechanical disturbance capability of the burner's swirl distribution plate. This represents the negative acceleration amplitude obtained from the decoupling of the upstream data packet body. After the calculation is completed and the specific airflow compensation value is obtained, the result is combined and packaged with the baseline target airflow retained by the decoupling at the data level to generate and output the variable group to be superimposed. The basic computing power skeleton and the micro-correction parameters to be fed forward are seamlessly encapsulated and retained for direct use by the downstream superposition control logic.

[0041] In step S43, the baseline target airflow and the compensation value of the perturbation airflow in the variable set to be superimposed are superimposed with a feedforward envelope to obtain the compensated target airflow. It should be noted that, given that step S4 has pointed out that the compensation air has the antagonistic coupling characteristics of both a combustion oxygen source and a convective cold source, simple linear superposition carries the risk of uncontrolled compensation ratio. Therefore, a constraint mechanism for quantifying the heat dissipation penalty intensity needs to be introduced before the final superposition. Based on this, the technical solution of this application further superimposes the baseline target airflow and the compensation value of the perturbation airflow in the variable set to be superimposed with a feedforward envelope to obtain the compensated target airflow. Through the above processing, a quadratic convective heat dissipation penalty factor based on the proportion is effectively introduced, and the Lorentz attenuation kernel is used to adaptively smooth the amplitude suppression of the original perturbation airflow, so that the anti-suffocation compensation amount is flexibly and dynamically constrained within the thermodynamic safety envelope of the self-sustaining combustion of the fire core, eliminating the risk of sudden extinguishing due to cold air flooding.

[0042] Figure 6 This is a flowchart illustrating the process of superimposing a reference target airflow and a perturbation airflow compensation value in a set of variables to be superimposed, according to an embodiment of this application, to obtain a compensated target airflow. Figure 6 As shown, step S43 includes: S431, calculating the ratio of the perturbation airflow compensation value and the baseline target airflow in the variable group to be superimposed and constructing a nonlinear heat dissipation penalty to obtain the convective heat dissipation penalty factor; S432, performing adversarial adaptive suppression on the perturbation airflow compensation value and the convective heat dissipation penalty factor to obtain the constraint compensation value; S433, performing controlled superposition of the baseline target airflow and the constraint compensation value to obtain the compensated target airflow.

[0043] In step S431, the proportion of the perturbation airflow compensation value in the variable set to be superimposed to the benchmark target airflow is calculated, and a nonlinear heat dissipation penalty is constructed to obtain the convective heat dissipation penalty factor. It should be noted that, given that during the cold furnace ignition stage, the extent of convective heat dissipation damage to the fire core after the upstream outputs the perturbation airflow compensation value depends on its volume ratio relative to the current benchmark target airflow. A larger ratio means a higher proportion of cold air that the fire core needs to absorb in the extremely low furnace temperature environment, resulting in a more severe convective heat dissipation penalty. Direct superposition can easily lead to the fire core being engulfed by the cold air. Based on this, the technical solution of this application further calculates the proportion of the perturbation airflow compensation value in the variable set to be superimposed to the benchmark target airflow and constructs a nonlinear heat dissipation penalty to obtain the convective heat dissipation penalty factor. Through the above processing, the intensity of this heat dissipation penalty is effectively quantified before the final superposition is performed.

[0044] More specifically, in a specific example of this application, an unpacking operation is performed on the set of variables to be superimposed, separating the encapsulated baseline target airflow and the perturbation airflow compensation value, calculating the volume ratio between the two, and applying a quadratic power operation to this ratio based on the Newtonian approximation that heat dissipation power is approximately proportional to the square of the airflow velocity in forced convection heat transfer, multiplying it by a pre-calibrated heat dissipation sensitivity coefficient obtained by fitting an empirical curve of the difference between ambient temperature and fire core temperature under initial cold furnace conditions, to construct a dimensionless factor reflecting the intensity of convection heat dissipation penalty at the current moment. The relevant convection heat dissipation penalty factor calculation conforms to the following relationship: in, In order to be in The convective heat dissipation penalty factor obtained from the time-varying construction is a dimensionless scalar. The convective heat dissipation sensitivity calibration coefficient is obtained by fitting an empirical curve of the temperature gradient between the ambient temperature and the core temperature under initial cold furnace conditions. This is the compensation value for micro-disturbance airflow from the unpacked input variable group. This is the baseline target airflow unpacked from the input variable set. After this calculation, the calculated convective heat dissipation penalty factor, along with the transparent baseline target airflow and the perturbation airflow compensation value, are encapsulated together as a penalty enhancement variable set. In the scenario of airflow control for cold ignition of biomass burners, the cold source identity of the compensating air is quantified for the first time using a nonlinear quadratic penalty model. In specific scenario examples, when the proportion of the compensation amount to the baseline airflow is small, the penalty factor approaches zero, indicating that the heat dissipation effect is negligible compared to the combustion-supporting effect. However, when the proportion increases, the penalty factor rises rapidly on a quadratic basis, accurately depicting the physical reality that adding a little more airflow in industrial-grade cold ignition scenarios can lead to the sudden extinguishing of the fire core. For example, the baseline target airflow calculated under the low-temperature conditions at the initial stage of ignition is too small. At this time, even if a small amount of micro-disturbance airflow compensation value is input, its volume ratio will increase rapidly. After quadratic power calculation and nonlinear amplification of the heat dissipation sensitivity coefficient, the final output convective heat dissipation penalty factor will show a rapidly increasing numerical performance, accurately quantifying the heat dissipation threat posed by the current compensation cold air entry to the weak fire core, and providing a quantitative boundary for subsequent dynamic constraint air distribution operations.

[0045] In step S432, the perturbation airflow compensation value and the convective heat dissipation penalty factor are subjected to antagonistic adaptive suppression to obtain a constraint compensation value. It should be noted that after obtaining the penalty factor quantifying the convective heat dissipation intensity, the perturbation airflow compensation value needs to be adaptively compressed based on this, rather than simply and crudely truncated. Hard truncation, i.e., directly reducing the compensation amount to zero when the penalty factor exceeds a certain threshold, would lead to a sudden interruption of oxygen supply in engineering, causing the already suffocating fire core to jump from slow decay to instantaneous extinguishing, which is also unacceptable. Based on this, the technical solution of this application further applies antagonistic adaptive suppression to the perturbation airflow compensation value and the convective heat dissipation penalty factor to obtain a constraint compensation value. Through the above processing, a smooth, continuous, and physically interpretable decay function is effectively introduced, allowing the compensation amount to flexibly self-suppress according to the severity of the heat dissipation penalty.

[0046] More specifically, in a specific example of this application, based on the adversarial adaptive compensation suppression using a Lorentz attenuation kernel, the penalized enhancement variable set is unwrapped to obtain the perturbation airflow compensation value and the convective heat dissipation penalty factor. The perturbation airflow compensation value is then multiplied by a Lorentz-type attenuation kernel function driven by the convective heat dissipation penalty factor. The relevant adversarial adaptive suppression calculation conforms to the following relationship: in, In order to be in The constraint compensation value generated after adversarial adaptive suppression has the same physical dimensions as the original compensation value. This is the compensation value for micro-disturbance airflow from the unpacked input variable group. The convection heat dissipation penalty factor is used to unpack the input variable set. This refers to the Lorentz-type decay kernel function, whose value range is strictly constrained to the interval between 0 and 1. In actual control processes, the Lorentz-type decay kernel is chosen instead of the traditional step cutoff or linear decay based on the following scenario considerations: The inherent long-tail characteristic of the Lorentz function ensures that even under extreme operating conditions with large heat dissipation penalties, a very small compensation path is still retained, avoiding irreversible asphyxiation and extinction caused by the complete interruption of oxygen supply due to hard cutoff. At the same time, the function remains continuous and monotonically decreasing throughout the entire process of the penalty factor changing from 0 to positive infinity, without any abrupt jump points, thereby avoiding step impact on the variable frequency actuator of the combustion fan. For example, when the initial combustion chamber temperature is low, resulting in a high calculated convective heat loss penalty factor, the multiplier coefficient will be smoothly reduced to a tiny fraction between 0 and 1 after the transformation and mapping of the Lorentz-type decay kernel function. This allows the micro-disturbance airflow compensation value to be reduced proportionally and transformed into a constraint compensation value, rather than being directly set to 0. This reduces the convective heat loss caused by the cold air intrusion while maintaining the most basic micro-oxygen delivery channel at the bottom layer, ensuring that the fire core receives enough oxidant to breathe and does not die instantly from oxygen deficiency. This achieves a dynamic physical balance between convective heat loss suppression and combustion support maintenance.

[0047] In step S433, the baseline target airflow and the constraint compensation value are controlled superimposed to obtain the compensated target airflow. It should be noted that, given that after the progressive constraints of the two upstream improved sub-steps, the perturbation airflow compensation value has been adaptively compressed by the Lorentz decay kernel function based on the real-time convective heat dissipation penalty intensity, the final airflow superposition operation can be safely performed. However, unconstrained direct superposition poses a risk of disrupting the combustion thermal balance. Therefore, the technical solution of this application further controls the superposition of the baseline target airflow and the constraint compensation value to obtain the compensated target airflow. Through the above processing, the controlled superposition output under the thermodynamic safety envelope is effectively completed. This ensures that the addition operation here is constrained by the thermodynamic safety envelope by the adversarial penalty mechanism of the two upstream improved sub-steps, guaranteeing that the superimposed compensated target airflow will not cause the convective heat dissipation power to exceed the self-sustaining combustion heat release power of the fire core due to the influx of compensation at any time.

[0048] More specifically, in a particular example of this application, the final superimposed variable group is unwrapped to obtain the baseline target airflow and constraint compensation values. The two are then combined by positive addition, and the relevant controlled superposition output calculation conforms to the following relationship: in, The final compensated target air volume is the output at time t. To determine the baseline target airflow from the input variable set, This is to unwrap the constraint compensation value of the input variable group. Unlike the direct linear superposition without constraints in the first embodiment, in a specific scenario, when the biomass burner faces low furnace temperature and oxygen deficiency during the initial ignition stage, the pre-calculation stage has compressed the original perturbation compensation amount into a numerically reduced constraint compensation value based on the high risk of convective heat dissipation. At this time, by performing positive addition and merging, the baseline target air volume for maintaining basic thermal safety is added to this numerically reduced constraint compensation value. The final output compensated target air volume only provides the fire core with a small amount of supplementary oxygen required to break the suffocation decay, avoiding excessive cut-in of redundant cold air and ensuring that the entire ignition and air supply process proceeds smoothly within the dual thermodynamic safety boundaries of preventing sudden cooling and suffocation.

[0049] Specifically, in step S5, the compensated target airflow is subjected to engineering mapping and amplitude limiting truncation processing to obtain the air distribution actuator control command. The air distribution actuator control command is output to the variable frequency drive device of the combustion fan to complete the dynamic airflow regulation during the ignition stage. It should be noted that, since the underlying variable frequency drive device cannot directly identify the volumetric flow parameters in the fluid dynamics dimension, and the unclamped electrical commands are prone to exceeding the hardware safe operating range, it may cause the motor to overheat and stop, resulting in oxygen depletion or excessive cold air blowing out the flame core. Based on this, the technical solution of this application further performs engineering mapping and amplitude limiting truncation processing on the compensated target airflow to obtain the air distribution actuator control command. The air distribution actuator control command is output to the variable frequency drive device of the combustion fan to complete the dynamic airflow regulation during the ignition stage. Through the above processing, the cross-domain conversion from the fluid domain to the electrical control domain is effectively completed, and strict hardware safety boundary restrictions are imposed on the terminal execution operation to ensure the safe and stable operation of the burner ignition process.

[0050] More specifically, in a concrete example of this application, the compensated target airflow is first mapped using an electromechanical transfer function to obtain a frequency reference signal. The electromechanical transfer function in the underlying hardware frequency converter library is then called to multiply the compensated target airflow by a frequency conversion mapping coefficient that has been tested and calibrated. This coefficient comprehensively considers mechanical characteristics such as the fan impeller displacement ratio, the number of motor pole pairs, and the transmission slip ratio, thereby accurately converting the air volume flow rate parameter into an electrical frequency signal that can be directly controlled by the frequency converter drive. The relevant engineering mapping conversion calculation conforms to the following relationship: in, This represents the frequency-given signal obtained at the current moment through conversion. This represents the input target air volume after compensation. This represents the frequency conversion mapping coefficient. Subsequently, based on the preset lower and upper limits of the fan speed for ignition conditions, the frequency command signal undergoes dual extreme value limiting and saturation truncation processing to obtain the air distribution actuator control command. To address the physical hazard of the fan's extremely low speed during the initial ignition phase causing motor overheating and shutdown, thus completely cutting off the oxygen supply, a lower speed limit clamp is applied. To address the hazard of excessive air intrusion inducing the weak flame core to extinguish, an upper speed limit clamp is applied. A logic comparison operator with both minimum and maximum constraints is used to perform forced upper and lower boundary saturation truncation of the frequency command signal. This truncation processing logic conforms to the following relationship: in, This indicates the final generated and issued air distribution actuator control command. This indicates the minimum threshold speed of the fan during ignition. This indicates the maximum speed limit threshold. and These represent the logical comparison truncation operators for obtaining the larger and smaller values, respectively. After completing the above calculations, the generated air distribution actuator control commands, which are within the fault-tolerant safety range and have clear physical execution meaning, are directly output to the variable frequency drive of the combustion fan, thereby driving the underlying fan to operate and smoothly complete the dynamic air distribution task of the entire ignition stage.

[0051] In summary, the combustion airflow control method for biomass burners according to the embodiments of this application is explained. First, it filters, smooths, and performs differential mapping on raw sensor data such as transient furnace temperature, igniter power, and initial biomass feed rate to extract a thermal feature vector including the temperature evolution gradient. Then, based on the principle of energy conservation, it performs external heat source stripping and physicochemical parameter inversion on this vector, softly measuring the instantaneous volatile matter release rate within the low-temperature blind zone where the oxygen sensor fails, thus solving the problem of unobservable combustion state during the cold furnace ignition stage. Furthermore, it generates a safety-constrained benchmark target airflow through stoichiometric ratio calculation and a nonlinear anti-blowout index decay mechanism, solving the problem of cold air blowing out the flame core due to the preset airflow curve's inability to adaptively match volatile matter fluctuations. Simultaneously, the polarity discrimination of the second derivative of furnace temperature and the perturbation compensation mechanism are introduced to reverse the asphyxiation attenuation trend caused by local oxygen deficiency. Through the construction of the convective heat dissipation penalty factor and the adaptive suppression of the Lorentz attenuation kernel, the problem of compensation ratio runaway caused by linear superposition when the compensation air plays the dual role of combustion oxygen source and convective cold source is solved. Finally, safe and reliable air distribution actuator control commands are generated through engineering mapping and amplitude limiting truncation.

[0052] Furthermore, a combustion air volume control system for a biomass burner is also provided.

[0053] Figure 7This is a block diagram of a combustion airflow control system for a biomass burner according to an embodiment of this application. Figure 7 As shown, the combustion air volume control system 100 for a biomass burner according to an embodiment of this application includes: a filtering extraction and differential mapping module 110, used to filter, extract, and differentially map the original sensor data stream during the cold furnace ignition stage to obtain a thermal characteristic vector; an energy stripping and inversion module 120, used to perform energy stripping and inversion processing on the thermal characteristic vector to obtain an instantaneous volatile matter release rate; a reference target air volume estimation module 130, used to estimate the reference target air volume under the anti-blowout safety constraint based on the instantaneous volatile matter release rate and the thermal characteristic vector to obtain a reference target air volume; a polarity discrimination and perturbation compensation module 140, used to perform polarity discrimination and perturbation compensation processing on the reference target air volume and the thermal characteristic vector to obtain a compensated target air volume; and an engineering mapping and amplitude limiting truncation module 150, used to perform engineering mapping and amplitude limiting truncation processing on the compensated target air volume to obtain a control command for the air distribution actuator, which is output to the variable frequency drive device of the combustion fan to complete the dynamic air volume regulation during the ignition stage.

[0054] As described above, the combustion airflow control system 100 for a biomass burner according to embodiments of this application can be implemented in various types of computing devices or control units. For example, it can be deployed in a central control room monitoring host of a distributed control system, or installed in an industrial-grade programmable logic controller within the burner's local control cabinet. In one possible implementation, the combustion airflow control system 100 for a biomass burner according to embodiments of this application can be integrated into the computing device as a software module and / or a hardware module. For example, the system 100 can be a resident combustion control service in the operating system of the computing device, the software module being configured to perform filtering and differential mapping of raw sensor data, inversion of instantaneous volatile release rate based on energy stripping, estimation of the baseline target airflow based on anti-blowout safety constraints, and adversarial adaptive compensation suppression oriented towards convective heat dissipation penalties and Lorentz decay kernels; or it can be a dedicated cold furnace ignition dynamic air distribution control program developed for the computing device. Of course, the system 100 can also be one of the many hardware modules of the computing device or control unit, or it can be embedded in the field programmable gate array circuit to accelerate the analysis of thermal characteristic vectors and the floating-point operation of complex nonlinear compensation algorithms with hardware parallel logic, or it can be a dedicated integrated circuit for dynamic control of combustion air volume for a specific application.

[0055] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for controlling the combustion air volume in a biomass burner, characterized in that, include: S1, During the cold furnace ignition stage, the original sensor data stream is filtered, extracted, and differentially mapped to obtain the thermal feature vector; S2, perform energy stripping and inversion processing on the thermal feature vector to obtain the instantaneous volatile release rate; S3, estimate the benchmark target air volume under the anti-blowout safety constraint based on the instantaneous volatile release rate and the hot characteristic vector to obtain the benchmark target air volume; S4, polarity discrimination and perturbation compensation processing are performed on the baseline target air volume and thermal characteristic vector to obtain the compensated target air volume; S5 performs engineering mapping and amplitude limiting truncation on the compensated target air volume to obtain the air distribution actuator control command. The air distribution actuator control command is output to the variable frequency drive device of the combustion fan to complete the dynamic air volume regulation during the ignition stage.

2. The method for controlling combustion air volume in a biomass burner according to claim 1, characterized in that, The raw sensor data stream includes furnace transient temperature, igniter power, and initial biomass feed rate.

3. The method for controlling combustion air volume in a biomass burner according to claim 2, characterized in that, Step S1 includes: The transient furnace temperature, igniter power and initial biomass feed rate contained in the original sensor data stream are deburred and smoothed by median filtering to obtain a smoothed feature set, which includes smoothed furnace temperature, smoothed power and smoothed feed rate. The transient temperature differential evolution gradient of the smoothed furnace temperature in the smoothed feature element set is calculated, and the first and second derivatives of the furnace temperature are converged with the smoothed furnace temperature, smoothed electric power and smoothed feed rate in the smoothed feature element set to obtain the enhanced feature element set. Tensor quantization is performed on the smoothed furnace temperature, smoothed electric power, smoothed feed rate, first derivative of furnace temperature, and second derivative of furnace temperature in the enhanced feature set to obtain the thermal feature vector.

4. The method for controlling combustion air volume in a biomass burner according to claim 1, characterized in that, Step S2 includes: The thermal analysis equilibrium group is obtained by performing dimensionality analysis and addressing extraction on the thermal feature vector; The total heat accumulation rate and igniter power in the heat analysis balance group are stripped of external heat sources to obtain virtual net combustion heat release. Based on the preset biomass volatile calorific value constant and system calibration mapping coefficient, the virtual net combustion heat release is mapped by physicochemical parameters and the release rate is inverted to obtain the instantaneous volatile release rate.

5. The method for controlling combustion air volume in a biomass burner according to claim 4, characterized in that, To obtain the virtual net combustion heat release, the total heat accumulation rate and igniter power in the heat analysis balance group are stripped of external heat sources. This includes: stripping the total heat accumulation rate and igniter power in the heat analysis balance group of external heat sources using the following formula: in, This is the virtual net heat release of combustion. The total heat accumulation rate, For igniter power, This is the calibration constant for the static electrothermal conversion efficiency of the igniter.

6. The method for controlling combustion air volume in a biomass burner according to claim 1, characterized in that, Step S3 includes: The smooth furnace temperature is separated by unpacking and addressing the thermal feature vector, and the theoretical air volume ratio is calculated by the instantaneous volatile release rate to obtain the theoretical air volume requirement. The theoretical air volume requirement and the smooth furnace temperature are then encapsulated to obtain the basic air distribution data pair. Based on the smoothed furnace temperature aligned with the basic air distribution data, the anti-exhaustion attenuation factor is determined. Then, the anti-exhaustion attenuation factor is recombined with the theoretical air demand aligned with the basic air distribution data to obtain the limited air distribution combination. The theoretical required air volume and the blowout prevention attenuation factor in the restricted air distribution combination are truncated to obtain the baseline target air volume.

7. The method for controlling combustion air volume in a biomass burner according to claim 1, characterized in that, Step S4 includes: The second derivative of furnace temperature is extracted by unpacking and addressing the thermal feature vector, and the second derivative of furnace temperature is processed by one-sided threshold polarity discrimination and negative asphyxiation feature extraction to obtain the compensation feedforward packet. Based on the preset sensitivity gain coefficient, the linear gain of the oxygen starvation counter-current compensation intensity is calculated for the negative acceleration amplitude in the compensation feedforward package to obtain the micro-disturbance air volume compensation value. Then, the micro-disturbance air volume compensation value is combined with the benchmark target air volume in the compensation feedforward package to obtain the set of variables to be superimposed. The baseline target air volume and the perturbation air volume compensation value in the set of variables to be superimposed are superimposed with a feedforward envelope to obtain the compensated target air volume.

8. The method for controlling combustion air volume in a biomass burner according to claim 1, characterized in that, Step S5 includes: Electromechanical transfer function mapping is performed on the compensated target air volume to obtain the frequency command signal; Based on the preset minimum speed lower limit threshold and maximum speed upper limit threshold of the fan under ignition conditions, the frequency command signal is subjected to dual extreme value limiting saturation truncation processing to obtain the control command of the air distribution actuator.

9. The method for controlling combustion air volume in a biomass burner according to claim 7, characterized in that, The baseline target air volume and the perturbation air volume compensation values ​​in the variable set to be superimposed are superimposed using a feedforward envelope to obtain the compensated target air volume, including: The proportion of the micro-perturbation airflow compensation value in the superimposed variable group to the baseline target airflow is calculated and a nonlinear heat dissipation penalty is constructed to obtain the convective heat dissipation penalty factor. The micro-disturbance airflow compensation value and the convective heat dissipation penalty factor are subjected to adversarial adaptive suppression to obtain the constraint compensation value; The baseline target air volume and the constraint compensation value are superimposed in a controlled manner to obtain the compensated target air volume.

10. A combustion air volume control system for a biomass burner, characterized in that, include: The filtering extraction and differential mapping module is used to filter, extract and differentially map the raw sensor data stream during the cold furnace ignition stage to obtain the thermal feature vector; The energy stripping and inversion module is used to perform energy stripping and inversion processing on the thermal feature vector to obtain the instantaneous volatile release rate. The benchmark target air volume estimation module is used to estimate the benchmark target air volume under the anti-blowout safety constraint based on the instantaneous volatile release rate and the hot characteristic vector to obtain the benchmark target air volume; The polarity discrimination and perturbation compensation module is used to perform polarity discrimination and perturbation compensation processing on the baseline target air volume and thermal characteristic vector to obtain the compensated target air volume; The engineering mapping and limiting truncation module is used to perform engineering mapping and limiting truncation processing on the compensated target air volume to obtain the air distribution actuator control command. The air distribution actuator control command is output to the variable frequency drive device of the combustion fan to complete the dynamic air volume regulation during the ignition stage.

Images