Boiler low load combustion control method and system based on acoustic early warning and coal quality feedforward

Abstract

A kind of based on acoustic early warning and coal quality feedforward boiler low load combustion control method and system, including obtaining real-time medium-high frequency pressure pulsation signal in the state of boiler low load combustion, real-time coal quality data and real-time operating parameter;Real-time medium-high frequency pressure pulsation signal is extracted after feature to obtain combustion stability characteristic value, and CSDI is calculated based on combustion stability characteristic value;According to real-time coal quality data, the coal quality feedforward compensation corresponding to the basic combustion control instruction is calculated;When CSDI is greater than or equal to a first level early warning threshold, enter dynamic anti-disturbance mode, to carry out fuzzy control based on CSDI and its corresponding rate of change, coal quality feedforward compensation and real-time operating parameter, generate anti-disturbance fine tuning;Through coal quality feedforward compensation, anti-disturbance fine tuning and basic combustion control instruction, anti-disturbance control instruction is generated to realize the control of boiler low load combustion, to overcome the hysteresis and passivity of existing boiler low load combustion control system.

Description

Technical Field

[0001] This application relates to the field of automatic boiler combustion control technology, specifically to a method and system for low-load combustion control of boilers based on acoustic early warning and coal quality feedforward. Background Technology

[0002] With the rapid growth of new energy installed capacity, coal-fired power units need to frequently participate in deep grid peak shaving and operate at low load (usually below the boiler's stable combustion load without oil injection) for a long time. Under this condition, the furnace temperature is low and the combustion stability drops sharply, making them extremely sensitive to disturbances such as fuel quality and air distribution changes. This can easily lead to combustion fluctuations, violent fluctuations in furnace pressure, and even fire extinguishing and other safety accidents. At the same time, the generation and control of pollutants such as NOx are more difficult under low load.

[0003] In related technologies, current boiler combustion control systems (such as burner management systems (BMS) and coordinated control systems (CCS) mostly rely on conventional parameters such as air-coal ratio, oxygen content, and furnace negative pressure to provide feedback regulation for low-load combustion in the furnace. However, this control method has significant lag: when a signal of combustion deterioration (such as large fluctuations in negative pressure) is detected, the disturbance has often already occurred and amplified. In other words, the control system is in a "passive response" state, resulting in untimely adjustments and a tendency to over-adjust, exacerbating instability. Therefore, developing a method that can detect combustion instability risks in advance and proactively and forward-lookingly adjust control commands is of great significance for ensuring the safe, environmentally friendly, and efficient operation of the unit during deep peak shaving. Summary of the Invention

[0004] This application provides a method and system for controlling low-load combustion in a boiler based on acoustic early warning and coal quality feedforward, which can solve the technical problem of poor timeliness and accuracy of adjustment caused by the passive method of feedback regulation for low-load combustion in the prior art.

[0005] In a first aspect, embodiments of this application provide a method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward, the method comprising: Acquire real-time medium- and high-frequency pressure pulsation signals, real-time coal quality data, and real-time operating parameters corresponding to the boiler's low-load combustion state; Feature extraction is performed on real-time medium- and high-frequency pressure pulsation signals to obtain combustion stability feature values, and the combustion stability deviation index (CSDI) is calculated based on the combustion stability feature values. The coal quality feedforward compensation amount corresponding to the basic combustion control command is calculated based on the real-time coal quality data. When CSDI is greater than or equal to the first-level warning threshold, the system enters dynamic anti-disturbance mode to perform fuzzy control based on the CSDI and its corresponding rate of change, the coal quality feedforward compensation amount, and the real-time operating parameters, thereby generating anti-disturbance fine-tuning amount. The anti-disturbance control command is generated by the coal quality feedforward compensation amount, the anti-disturbance fine-tuning amount, and the basic combustion control command, so as to realize the control of low-load combustion of the boiler through the anti-disturbance control command.

[0006] In conjunction with the first aspect, in one embodiment, the method further includes: When CSDI is less than the first-level warning threshold, it enters steady-state feedforward mode to generate feedforward combustion control commands based on coal quality feedforward compensation and basic combustion control commands. Low-load combustion control of the boiler is achieved through feedforward combustion control commands.

[0007] In conjunction with the first aspect, in one embodiment, the method further includes: When CSDI is greater than or equal to the Level 2 warning threshold, the system enters the safety linkage mode. Based on the dynamic anti-disturbance mode control, the control intensity or response priority is further increased, and the preset emergency combustion stabilization plan is triggered. The Level 2 warning threshold is greater than the Level 1 warning threshold.

[0008] In conjunction with the first aspect, in one embodiment, the step of extracting features from the real-time mid-to-high frequency pressure pulsation signal to obtain combustion stability feature values ​​includes: Filtering and spectrum analysis are performed on real-time medium- and high-frequency pressure pulsation signals to obtain real-time sound pressure level energy, real-time main frequency offset, and real-time sound signal non-uniformity index under a preset specific frequency band. Real-time sound pressure level energy, real-time dominant frequency offset, and real-time acoustic signal non-uniformity index are used as characteristic values ​​of combustion stability.

[0009] In conjunction with the first aspect, in one embodiment, determining the combustion stability deviation index (CSDI) based on combustion stability characteristic values ​​includes: The real-time sound pressure level energy, real-time main frequency offset, and real-time acoustic signal non-uniformity index are compared with the standard sound pressure level energy, standard main frequency offset, and standard acoustic signal non-uniformity index in the preset stable combustion acoustic fingerprint benchmark library to obtain the energy difference, offset difference, and index difference. The CSDI is obtained by weighting the energy difference, offset difference, and exponential difference. Among them, the standard feature values ​​in the preset stable combustion acoustic fingerprint benchmark library are established when the boiler is under stable operating conditions.

[0010] In conjunction with the first aspect, in one embodiment, the real-time coal quality data includes dry ash-free volatile matter, as-received moisture content, and as-received lower heating value. The step of calculating the coal quality feedforward compensation amount corresponding to the basic combustion control command based on the real-time coal quality data includes: Based on the real-time coal quality data, the target parameters of the current coal quality are calculated, including the theoretical ignition heat, the stable combustion index, and the optimal combustion air distribution range. The coal quality feedforward compensation amount corresponding to the basic combustion control command is calculated based on the target parameters and the preset design parameter values.

[0011] In conjunction with the first aspect, in one embodiment, the disturbance rejection control command includes commands to adjust the fuel distribution of each layer and / or to adjust the secondary air ratio while maintaining a constant total input energy.

[0012] Secondly, embodiments of this application provide a low-load combustion control system for a boiler based on acoustic early warning and coal quality feedforward, the control system comprising: The data sensing layer is used to acquire real-time medium- and high-frequency pressure pulsation signals, real-time coal quality data, and real-time operating parameters corresponding to the boiler's low-load combustion state. The intelligent analysis layer is used to extract features from real-time medium- and high-frequency pressure pulsation signals to obtain combustion stability feature values, and calculate the combustion stability deviation index (CSDI) based on the combustion stability feature values; and calculate the coal quality feedforward compensation amount corresponding to the basic combustion control command based on the real-time coal quality data. The execution control layer is used to enter the dynamic disturbance rejection mode when the CSDI is greater than or equal to the first-level warning threshold. It performs fuzzy control based on the CSDI and its corresponding rate of change, the coal quality feedforward compensation amount, and the real-time operating parameters to generate disturbance rejection fine-tuning amount. It generates disturbance rejection control command through the coal quality feedforward compensation amount, the disturbance rejection fine-tuning amount, and the basic combustion control command to achieve low-load combustion control of the boiler through the disturbance rejection control command.

[0013] In conjunction with the second aspect, in one implementation, the execution control layer is further configured to: When CSDI is less than the first-level warning threshold, it enters steady-state feedforward mode to generate feedforward combustion control commands based on coal quality feedforward compensation and basic combustion control commands. Low-load combustion control of the boiler is achieved through feedforward combustion control commands.

[0014] In conjunction with the second aspect, in one implementation, the execution control layer is further configured to: When CSDI is greater than or equal to the Level 2 warning threshold, the system enters the safety linkage mode. Based on the dynamic anti-disturbance mode control, the control intensity or response priority is further increased, and the preset emergency combustion stabilization plan is triggered. The Level 2 warning threshold is greater than the Level 1 warning threshold.

[0015] In conjunction with the second aspect, in one implementation, the intelligent analysis layer is specifically used for: Filtering and spectrum analysis are performed on real-time medium- and high-frequency pressure pulsation signals to obtain real-time sound pressure level energy, real-time main frequency offset, and real-time sound signal non-uniformity index under a preset specific frequency band. Real-time sound pressure level energy, real-time dominant frequency offset, and real-time acoustic signal non-uniformity index are used as characteristic values ​​of combustion stability.

[0016] In conjunction with the second aspect, in one implementation, the intelligent analysis layer is further used for: The real-time sound pressure level energy, real-time main frequency offset, and real-time acoustic signal non-uniformity index are compared with the standard sound pressure level energy, standard main frequency offset, and standard acoustic signal non-uniformity index in the preset stable combustion acoustic fingerprint benchmark library to obtain the energy difference, offset difference, and index difference. The CSDI is obtained by weighting the energy difference, offset difference, and exponential difference. Among them, the standard feature values ​​in the preset stable combustion acoustic fingerprint benchmark library are established when the boiler is under stable operating conditions.

[0017] In conjunction with the second aspect, in one embodiment, the real-time coal quality data includes dry ash-free volatile matter, as-received moisture content, and as-received lower heating value; the intelligent analysis layer is further specifically used for: The target parameters of the current coal quality are calculated based on real-time coal quality data. The target parameters include theoretical ignition heat, stable combustion index, and optimal combustion air distribution range. The coal quality feedforward compensation amount corresponding to the basic combustion control command is calculated based on the target parameters and the preset design parameter values.

[0018] In conjunction with the second aspect, in one embodiment, the disturbance rejection control command includes commands to adjust the fuel distribution of each layer and / or to adjust the secondary air ratio while maintaining a constant total input energy.

[0019] The beneficial effects of the technical solutions provided in this application include: This study acquires real-time mid-to-high frequency pressure pulsation signals, real-time coal quality data, and real-time operating parameters under low-load combustion conditions in boilers. Since the mid-to-high frequency pressure pulsations within the boiler furnace directly originate from the intensity of combustion chemical reactions and aerodynamic processes such as flame turbulence and vortex shedding, when combustion becomes unstable, the heat release rate oscillates periodically and flame oscillation intensifies. This directly leads to regular changes in the amplitude, dominant frequency, and spectral structure of pressure pulsations at specific locations within the furnace. These changes precede significant fluctuations in the large-scale furnace negative pressure (low-frequency, average value signal). Therefore, feature extraction from the real-time mid-to-high frequency pressure pulsation signals yields characteristic indicators that characterize the combustion state (i.e., combustion stability characteristic values). Based on these combustion stability characteristic values, the Combustion Stability Deviation Index (CSDI) can be determined to achieve acoustic early warning, providing a "time lead time." This addresses the problem of perceiving "early changes in combustion state caused by unknown or residual disturbances," achieving time / process-side lead time. Simultaneously, based on real-time coal quality data… The data calculates the coal quality feedforward compensation amount corresponding to the basic combustion control command, that is, the most important combustion disturbance source (fuel change) is identified in real time, and the air-coal ratio is adjusted in advance in the form of feedforward to offset the disturbance at the source, which greatly improves the adaptability of the control system to coal quality fluctuations. It solves the problem of compensation for "known input disturbance", that is, to achieve spatial / input-side foresight. When CSDI is greater than or equal to the first-level warning threshold, it enters the dynamic disturbance rejection mode. After fuzzy control based on CSDI and its corresponding rate of change, coal quality feedforward compensation amount and real-time operating parameters, disturbance rejection fine adjustment amount is generated. Finally, disturbance rejection control command is generated by coal quality feedforward compensation amount, disturbance rejection fine adjustment amount and basic combustion control command to realize the control of boiler low-load combustion. It can be seen that this application deeply integrates acoustic warning (reflecting the state of combustion result) and coal quality feedforward (reflecting the root cause of disturbance input) information to form spatiotemporal synergy, so as to make more accurate and forward-looking decisions under low-load combustion of boiler, effectively improving the timeliness and accuracy of adjustment. Attached Figure Description

[0020] Figure 1 This is a schematic flowchart of an embodiment of the acoustic early warning and coal quality feedforward boiler low-load combustion control method of this application; Figure 2 This is a schematic diagram of the functional modules of an embodiment of the acoustic early warning and coal quality feedforward boiler low-load combustion control system of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0023] In a first aspect, embodiments of this application provide a method for controlling low-load combustion in a boiler based on acoustic early warning and coal quality feedforward.

[0024] In one embodiment, reference is made to Figure 1 , Figure 1 This is a schematic flowchart illustrating an embodiment of the acoustic early warning and coal quality feedforward boiler low-load combustion control method of this application. Figure 1 As shown, the low-load combustion control method for boilers based on acoustic early warning and coal quality feedforward includes: Step S10: Acquire the real-time medium- and high-frequency pressure pulsation signals, real-time coal quality data, and real-time operating parameters corresponding to the boiler's low-load combustion state.

[0025] It should be understood, as an example, that the energy released by fuel combustion in a boiler furnace generates pressure waves. The audible high-frequency portion is combustion noise, while low-frequency pressure fluctuations affect the furnace negative pressure. It is noteworthy that pressure pulsations within the boiler furnace (especially in the mid-to-high frequency range, such as 50-300Hz) directly originate from the intensity of the combustion chemical reaction and aerodynamic processes such as flame turbulence and vortex shedding. When combustion becomes unstable, the heat release rate oscillates periodically, and flame oscillation intensifies, directly causing regular changes in the amplitude, dominant frequency, and spectral structure of pressure pulsations at specific locations within the furnace. These changes precede significant fluctuations in the large-scale furnace negative pressure (such as low-frequency, average-value signals). In general, early signs of combustion instability often manifest first in the mid-to-high frequency pulsations of the local flame, which then gradually couple and amplify into low-frequency pressure oscillations throughout the entire furnace.

[0026] However, traditional furnace negative pressure sensors can only measure extremely low-frequency average pressure changes. This embodiment, however, uses an array of acoustic sensors (such as high dynamic response pressure sensors or sound pressure sensors) arranged in the boiler burner area and on the four upper walls of the furnace to collect combustion noise and mid-to-high frequency pressure pulsation signals under low-load combustion conditions. The mid-to-high frequency pressure pulsation signals can reflect the detailed characteristics of combustion instability earlier, and together with the "combustion noise," constitute the complete set of "acoustic signals." This forms the physical basis for achieving "early warning" functionality and is the core technological foundation for upgrading from "passive feedback" to "active warning and anti-disturbance." It should be noted that the pressure pulsation and combustion sound within the furnace are essentially the same physical phenomenon manifested at different frequency bands, with different sensing methods and different analytical focuses; they can be understood as a continuous spectrum of "pressure fluctuations."

[0027] Furthermore, the real-time changes in the quality of coal entering the furnace are one of the most significant sources of disturbance affecting combustion stability. However, traditional systems lack online and forward-looking perception and utilization of coal quality parameters (such as volatile matter, moisture, and calorific value). In this embodiment, it is preferable to use an online coal quality analyzer installed at the coal feeder outlet or on the coal conveyor belt to detect real-time coal quality data, including volatile matter, moisture, and calorific value, of the coal entering the furnace, and to provide a data basis for coal quality feedforward calculation.

[0028] Understandably, combustion stability is the most critical issue when a boiler operates under low load conditions, and operating parameters such as load, fuel quantity, air volume, oxygen content, and negative pressure are the core factors affecting combustion stability. Among these, fuel quantity determines the furnace heat load and pulverized coal concentration, forming the basis of combustion stability; load parameters reflect the overall system operating state and determine the selection of control strategies; air volume parameters directly affect the oxygen supply and aerodynamic field during combustion, serving as a key means of regulating the combustion state; oxygen content refers to the volume percentage of oxygen in the flue gas, a key indicator reflecting the excess air coefficient during combustion; negative pressure refers to the negative value of the furnace pressure relative to atmospheric pressure, and its fluctuations have amplified impact on the aerodynamic field under low load conditions. Therefore, this embodiment will utilize conventional monitoring instruments such as furnace negative pressure sensors, oxygen sensors, fuel quantity measuring devices, air volume measuring devices, and secondary damper opening feedback devices to collect real-time operating parameters, providing a data foundation for boiler low-load combustion control and ensuring safe, stable, and efficient boiler operation under low load conditions.

[0029] Step S20: Extract features from the real-time medium- and high-frequency pressure pulsation signal to obtain combustion stability feature values, and calculate the combustion stability deviation index CSDI based on the combustion stability feature values.

[0030] As an example, it is understandable that the real-time mid-to-high frequency pressure pulsation signal collected by the acoustic sensor array is a raw data stream. Therefore, it needs to be processed by filtering (filtering out mechanical noise unrelated to combustion) and spectrum analysis to transform it into characteristic indicators that can characterize the combustion state, so as to extract combustion stability characteristic values ​​(including but not limited to specific frequency band energy, dominant frequency offset, and non-uniformity index). Then, the combustion stability deviation index CSDI can be calculated based on the combustion stability characteristic values. CSDI is the bridge connecting "sensing" and "control", which quantifies the degree to which the current combustion state deviates from the ideal stable state.

[0031] Step S30: Calculate the coal quality feedforward compensation amount corresponding to the basic combustion control command based on the real-time coal quality data.

[0032] It should be noted that the "basic combustion control commands" mentioned in this embodiment refer to the basic control commands generated by the boiler main coordinated control system (CCS) after calculation by conventional control logic (such as fuel main control and air volume main control) based on the unit load command. These commands are the basic command set for controlling combustion when the acoustic early warning and coal quality feedforward compensation mechanism described in this embodiment is not introduced. They typically include, but are not limited to, total fuel quantity commands, total air volume commands, and primary air pressure commands. In this embodiment, based on the combustion dynamics model and the collected real-time coal quality data, the coal quality feedforward compensation amount for compensating and adjusting the above-mentioned basic combustion control commands will be calculated.

[0033] Step S40: When CSDI is greater than or equal to the first-level early warning threshold, enter the dynamic anti-disturbance mode, and perform fuzzy control based on the CSDI and its corresponding rate of change, the coal quality feedforward compensation amount and the real-time operating parameters to generate anti-disturbance fine-tuning amount.

[0034] As an example, in this embodiment, the warning level will be determined based on the relationship between CSDI and the warning threshold, so as to select different combustion control modes based on the warning level. The warning threshold includes a primary warning threshold and a secondary warning threshold. The primary warning threshold is used to characterize the boiler as being in a critical state of stable combustion, while the secondary warning threshold is used to characterize the boiler as being in a critical state of extremely unstable combustion. It should be noted that the specific values ​​of the primary and secondary warning thresholds can be determined according to actual needs and are not limited here, as long as the primary warning threshold is less than the secondary warning threshold.

[0035] Based on this, when CSDI < Level 1 warning threshold, it indicates that the boiler is in a relatively stable combustion state. In this case, only feedforward compensation is needed, and no warning is required. The warning level can be set to 0 or set to no warning state. When Level 1 warning threshold ≤ CSDI < Level 2 warning threshold, it indicates that the boiler is in an unstable combustion state. In this case, not only feedforward compensation but also active disturbance rejection is needed. In other words, a warning is needed. The warning level can be set to 1 or set to warning state. When CSDI ≥ Level 2 warning threshold, it indicates that the boiler is in an extremely unstable combustion state. In this case, not only feedforward compensation and active disturbance rejection are needed, but also emergency combustion stabilization intervention is needed. In other words, an alarm is needed. The warning level can be set to 2 or set to alarm state.

[0036] In addition, this embodiment also provides multi-mode active control decision-making: (1) Steady-state feedforward mode: When there is no warning, the coal quality feedforward compensation amount is smoothly superimposed on the original air and coal command issued by the boiler main control system (CCS) to realize "static" active adjustment based on coal quality changes and compensate for coal quality disturbances; (2) Dynamic anti-disturbance mode: When an acoustic warning signal is received, the dynamic anti-disturbance algorithm is immediately activated, that is, the gain of the feedforward compensation amount is dynamically adjusted with CSDI as the controlled variable to actively suppress the instability trend. This mode has a higher priority than the steady-state feedforward mode; (3) Safety interlocking and linkage (i.e., safety linkage mode): When an acoustic alarm signal is received, a combustion deterioration chain signal can be sent to the boiler safety monitoring system (FSSS) and the preset emergency stable combustion strategy (such as automatic operation of micro oil gun) can be triggered. It can be seen that this embodiment calculates the optimal countermeasure strategy in real time according to the severity and development trend of the warning. It should be understood that when implementing the above-mentioned drive mode switching, the value of CSDI and its rate of change are the sole decision-making basis for the system to switch between "steady-state feedforward mode", "dynamic disturbance rejection mode" and "safety linkage mode".

[0037] Therefore, when CSDI is greater than or equal to the first-level warning threshold, it indicates that the system is currently in a warning state. In this case, the system needs to enter a dynamic disturbance rejection mode. Using CSDI and its rate of change, coal quality feedforward compensation, and real-time operating parameters as inputs, a set of optimal fine-tuning values ​​(i.e., disturbance rejection fine-tuning values, or combustion enhancement commands used to quickly suppress CSDI) is dynamically calculated through fuzzy inference rules or model predictive control algorithms. This aims to rapidly suppress instability trends and reshape a stable aerodynamic field. It should be noted that the fuzzy inference rule can preferably be fuzzy PID (Proportional-Integral-Derivative Control), and the model predictive control algorithm can preferably be MPC (Model Predictive Control).

[0038] Specifically, when CSDI ≥ Level 1 warning threshold, the system will switch from "steady-state feedforward mode" to "dynamic disturbance rejection mode". At this time, the control objective changes from "smooth compensation" to "rapid suppression of instability trend". In this mode, the system essentially constitutes a fast closed-loop auxiliary control system with CSDI as the controlled variable, and the setpoint of this closed loop is the safety threshold (or zero) of CSDI. It should be understood that the feedback input is the real-time CSDI and its rate of change (i.e., derivative), the feedforward input is the benchmark feedforward compensation amount (i.e., coal quality feedforward compensation amount) calculated by coal quality feedforward, which serves as a bias or baseline for the control output, while the disturbance input is the real-time operating parameters such as the current total fuel quantity, load, and air volume.

[0039] Based on this, the fuzzy PID algorithm dynamically calculates an additional set of control increments (i.e., anti-disturbance fine-tuning) online, using two key inputs: "how high is the current CSDI" (i.e., error) and "how fast is the CSDI deteriorating" (i.e., error rate of change). For example, if the CSDI is high and rising rapidly, it may output a large "upper-level fuel bias increment" and a "lowering of the upper-level secondary damper opening." It should be noted that the specific working methods and principles of the fuzzy PID algorithm are common knowledge in this field, and for the sake of simplicity, they will not be elaborated upon here.

[0040] The MPC algorithm, based on a simplified combustion dynamics model, predicts the CSDI (Chemical Combustion Displacement) trajectory under different control strategies (such as adjusting the fuel distribution ratio in each layer) over several future steps. It then selects the optimal control sequence that allows the CSDI to fall back to the safe zone most quickly and smoothly. This sequence includes specific, fine-tuned adjustments to the fuel and damper settings for each layer to mitigate disturbances. It should be noted that the specific working methods and principles of the MPC algorithm are common knowledge in the field and will not be elaborated upon here for the sake of brevity.

[0041] It is understood that the aforementioned anti-disturbance fine-tuning, when executed, specifically manifests as the generation of refined adjustment commands while maintaining the total boiler input energy (i.e., total fuel quantity) and total air volume essentially unchanged. These commands include: adjustments to the fuel distribution ratio of each burner layer, and / or adjustments to the opening degree of each secondary air damper (i.e., secondary air ratio). Through this fine-tuning strategy of "total conservation and internal distribution optimization," the combustion state inside the furnace can be quickly reshaped and instability trends suppressed while minimizing disruption to the balance of the main control system.

[0042] Step S50: Generate an anti-disturbance control command using the coal quality feedforward compensation amount, the anti-disturbance fine-tuning amount, and the basic combustion control command, so as to achieve low-load combustion control of the boiler through the anti-disturbance control command.

[0043] In this exemplary embodiment, the final output disturbance rejection control command consists of coal quality feedforward compensation, disturbance rejection fine-tuning, and basic combustion control command, i.e., disturbance rejection control command = basic combustion control command + coal quality feedforward compensation (Δ) + disturbance rejection fine-tuning (δ). This disturbance rejection control command is then sent to the corresponding actuators such as the coal feeder, primary air fan, forced draft fan, and various dampers to achieve active disturbance rejection control for low-load boiler combustion, improving the timeliness and accuracy of feedback regulation. It is understood that this embodiment can also monitor the changing trends of parameters such as CSDI, furnace negative pressure, and oxygen content over a period of time after adjustment. Based on the monitoring results, the control effect can be evaluated, and the early warning threshold and control algorithm parameters can be adaptively optimized online, further improving the timeliness and accuracy of feedback regulation.

[0044] In summary, this embodiment can achieve at least the following beneficial effects: (1) Early warning, turning passive into active: Utilizing the characteristic that acoustic signals are extremely sensitive to changes in combustion state, the combustion instability trend can be detected tens of seconds to several minutes in advance before conventional parameters such as furnace negative pressure and oxygen content fluctuate significantly, thus gaining valuable time for control intervention; (2) Root cause disturbance resistance, improving adaptability: Through online coal quality analysis, the most important combustion disturbance source (fuel change) can be identified in real time, and the air-coal ratio can be adjusted in advance in a feedforward manner to offset the disturbance from the source, greatly improving the control system's adaptability to coal quality. Adaptability to fluctuations; (3) Intelligent integration and precise control: Deeply integrate acoustic early warning (reflecting the state of combustion results) with coal quality feedforward (reflecting the root cause of disturbance input) information to make more accurate and forward-looking decisions, avoid misjudgment of single signals and frequent ineffective adjustments; (4) Ensure safety and promote environmental protection: Effectively reduce the risk of low-load fire extinguishing and improve the safety of peak-shaving operation; By maintaining combustion stability in the optimal range, it helps to suppress CO and incomplete combustion losses caused by oxygen deficiency or uneven temperature, and at the same time creates stable conditions for achieving low NOx combustion under low load.

[0045] Furthermore, in one embodiment, the method further includes: When CSDI is less than the first-level warning threshold, it enters steady-state feedforward mode to generate feedforward combustion control commands based on coal quality feedforward compensation and basic combustion control commands. Low-load combustion control of the boiler is achieved through feedforward combustion control commands.

[0046] In this exemplary embodiment, if CSDI < Level 1 warning threshold is detected, it indicates that the current warning level is a no-warning state, meaning the boiler is in a relatively stable combustion state. At this time, only feedforward compensation is needed, without dynamic disturbance rejection. Therefore, the control system operates in steady-state feedforward mode, superimposing the coal quality feedforward compensation amount calculated in step S30 onto the basic combustion control command to form a feedforward combustion control command for output. This allows the actuators such as the coal feeder, primary air fan, forced draft fan, and various dampers to perform feedforward control of the boiler's low-load combustion based on the feedforward combustion control command. It should be noted that when superimposing the coal quality feedforward compensation amount onto the basic combustion control command, a rate limit can also be applied to the coal quality feedforward compensation amount to achieve stable control of the boiler's low-load combustion.

[0047] Furthermore, in one embodiment, the method further includes: When CSDI is greater than or equal to the Level 2 warning threshold, the system enters the safety linkage mode. Based on the dynamic anti-disturbance mode control, the control intensity or response priority is further increased, and the preset emergency combustion stabilization plan is triggered. The Level 2 warning threshold is greater than the Level 1 warning threshold.

[0048] In this exemplary embodiment, if CSDI ≥ the secondary threshold, it indicates that the current warning level is an alarm state, meaning the boiler is in a highly unstable combustion state. In this case, not only is feedforward compensation and active disturbance rejection necessary, but emergency combustion stabilization intervention is also required. Therefore, the control system enters a safety linkage mode to further enhance the control strength or response priority based on the dynamic disturbance rejection mode control, and sends an alarm to a higher-level safety system to trigger and execute the preset emergency combustion stabilization plan. It should be noted that the setting of control strength and response priority can be determined according to actual needs and is not limited here. For example, the response priority can be adjusted from the primary level to the highest level. In addition, the specific scheme of the emergency combustion stabilization plan can be determined according to actual needs and is not limited here. For example, the emergency combustion stabilization plan can be set to automatically activate the micro oil gun.

[0049] Furthermore, in one embodiment, the step of extracting features from the real-time mid-to-high frequency pressure pulsation signal to obtain combustion stability feature values ​​includes: Filtering and spectrum analysis are performed on real-time medium- and high-frequency pressure pulsation signals to obtain real-time sound pressure level energy, real-time main frequency offset, and real-time sound signal non-uniformity index under a preset specific frequency band. Real-time sound pressure level energy, real-time dominant frequency offset, and real-time acoustic signal non-uniformity index are used as characteristic values ​​of combustion stability.

[0050] As an example, it is understandable that early signs of combustion instability often manifest as a significant increase in pulsating energy in a specific frequency band (e.g., a frequency related to a burner swirl number or furnace acoustic mode), or a shift in the dominant frequency. Therefore, the specific value of the preset specific frequency band (i.e., a specific frequency range) can be determined according to actual needs, such as setting it to 50-300Hz. The pressure pulsation in this frequency band directly "encodes" information about combustion stability (e.g., an increase in pulsating energy in this frequency band corresponds to ignition fluctuations, or a change in the spectrum reflects a switching of combustion modes), and its changes occur earlier than the significant changes in traditional furnace negative pressure gauges, thus providing a valuable early warning time window.

[0051] Based on this, after acquiring real-time mid-to-high frequency pressure pulsation signals through an acoustic sensor array (which is extremely sensitive to pressure changes at frequencies up to hundreds of Hz), the signals will be filtered and subjected to spectral analysis to analyze the spectrum (energy distribution at different frequencies) of these signals, rather than their absolute volume, so as to transform them into characteristic indicators that can characterize the combustion state, such as real-time sound pressure level energy at 50-300Hz, real-time dominant frequency offset, and real-time acoustic signal non-uniformity index, etc., combustion stability characteristic values, and use combustion stability characteristic values ​​as the data basis for early warning; among them, specific frequency band energy refers to the sudden increase in pressure pulsation energy at a specific frequency that often accompanies combustion instability; dominant frequency offset refers to the relatively fixed pulsation dominant frequency during stable combustion, which will drift before instability; non-uniformity index refers to the determination of whether the oscillation is local or global through spatial correlation analysis of array signals. It should be noted that for filtering, the Kalman filter algorithm can be used, or other algorithms capable of signal filtering can be selected. There is no limitation here, and the working principle of the filtering algorithm is common knowledge in this field. For the sake of simplicity, it will not be elaborated here. For spectrum analysis, FFT (Fast Fourier Transform) can be used, or other algorithms capable of spectrum analysis can be selected. There is no limitation here, and the working principle of spectrum analysis algorithm is common knowledge in this field. For the sake of simplicity, it will not be elaborated here.

[0052] It is worth noting that providing a "time lead" through mid-to-high frequency pressure pulsation signals is key to achieving "active" disturbance rejection. Traditional control systems rely on furnace negative pressure signals, which only trigger an alarm after the disturbance has amplified, resulting in a passive system response. However, the CSDI corresponding to pressure pulsation can issue an early warning in the early stages of combustion deterioration (when the flame has just begun to flicker and oscillate, but before it causes a significant fluctuation in the overall negative pressure), giving the control system valuable intervention time (tens of seconds to several minutes). Thus, this embodiment uses "acoustics" (essentially pressure pulsation analysis in a specific frequency band, i.e., the capture and analysis of combustion-related mid-to-high frequency pressure pulsation signals) as an early warning method, which precisely captures the earliest and most direct physical characteristic signal of combustion instability, thereby achieving a leap from "processing result" to "intervention process" and effectively improving the timeliness of feedback regulation.

[0053] Further, in one embodiment, the calculation of the combustion stability deviation index (CSDI) based on combustion stability characteristic values ​​includes: The real-time sound pressure level energy, real-time main frequency offset, and real-time acoustic signal non-uniformity index are compared with the standard sound pressure level energy, standard main frequency offset, and standard acoustic signal non-uniformity index in the preset stable combustion acoustic fingerprint benchmark library to obtain the energy difference, offset difference, and index difference. The CSDI is obtained by weighting the energy difference, offset difference, and exponential difference. Among them, the standard feature values ​​in the preset stable combustion acoustic fingerprint benchmark library are established when the boiler is under stable operating conditions.

[0054] In this exemplary embodiment, the Combustion Stability Deviation Index (CSDI) is calculated by comprehensively analyzing the energy distribution, dominant frequency shift, and spatial non-uniformity of the acoustic signal in the 50-300Hz frequency band. Firstly, during stable boiler operation, historical acoustic signals are collected under different stable operating conditions (i.e., stable operating conditions composed of different loads and coal qualities) to establish a stable combustion acoustic fingerprint benchmark library corresponding to the load and coal quality. This benchmark library contains accurate standard sound pressure level energy, standard dominant frequency shift, and standard acoustic signal non-uniformity index under different stable operating conditions. This data is used to calibrate the real-time sound pressure level energy, real-time dominant frequency shift, and real-time acoustic signal non-uniformity index, thereby generating the CSDI.

[0055] Specifically, the real-time sound pressure level energy, real-time dominant frequency offset, and real-time acoustic signal non-uniformity index are compared with the standard sound pressure level energy, standard dominant frequency offset, and standard acoustic signal non-uniformity index, respectively, to calculate the energy difference between the real-time sound pressure level energy and the standard sound pressure level energy, the offset difference between the real-time dominant frequency offset and the standard dominant frequency offset, and the exponential difference between the real-time acoustic signal non-uniformity index and the standard acoustic signal non-uniformity index. Then, the energy difference, offset difference, and exponential difference are weighted to obtain the CSDI, so that the switching between different control modes can be realized based on the relationship between the CSDI and the warning threshold.

[0056] Further, in one embodiment, the real-time coal quality data includes dry ash-free volatile matter, as-received moisture content, and as-received lower heating value. The step of calculating the coal quality feedforward compensation amount corresponding to the basic combustion control command based on the real-time coal quality data includes: The target parameters of the current coal quality are calculated based on real-time coal quality data. The target parameters include theoretical ignition heat, stable combustion index, and optimal combustion air distribution range. The coal quality feedforward compensation amount corresponding to the basic combustion control command is calculated based on the target parameters and the preset design parameter values.

[0057] In this exemplary embodiment, real-time coal quality data includes, but is not limited to, dry ash-free volatile matter (Vdaf), as-received moisture (Mar), and as-received lower heating value (Qnet,ar). The online coal quality analyzer can calculate target parameters such as the theoretical ignition heat, stable combustion index, and optimal combustion air distribution range of the current coal quality based on the built-in combustion kinetics model and real-time coal quality data such as Vdaf and Mar. The theoretical ignition heat refers to the amount of heat required to heat a unit mass of pulverized coal from its initial temperature to its ignition temperature, and is a key indicator for measuring the ease of ignition of pulverized coal. The stable combustion index is a quantitative indicator that comprehensively reflects the combustion stability of pulverized coal; the larger the value, the more stable the combustion. The optimal combustion air distribution range refers to the range of excess air coefficients that can maintain optimal combustion stability while ensuring combustion efficiency and environmental protection indicators.

[0058] It should be noted that the specific formulas or architectures of the combustion dynamics models corresponding to different target parameters are common knowledge in this field, and will not be elaborated here for the sake of brevity. For example, for the theoretical ignition heat Qign, the corresponding combustion kinetic model is Qign = cp·(Tign - T0) + W·hfg + Qd, where cp represents the specific heat capacity of pulverized coal, which can be calculated using the empirical formula cp = 0.98 + 0.0002·T; Tign represents the ignition temperature of pulverized coal, which is closely related to the coal quality characteristics and can be estimated using the empirical formula Tign = a - b·Vdaf + c·(Aar + Mar), where Aar represents the ash content on an as-received basis, and the empirical coefficients can be a=900℃, b=2.5, and c=1.5; T0 represents the initial temperature of pulverized coal, usually taken as the temperature corresponding to the primary air temperature; W represents the moisture content in pulverized coal, which is equal to Mar / 100; hfg represents the latent heat of vaporization of water, which is approximately 2257 kJ / kg at 100℃; and Qd represents the heat released by the combustion of volatiles in pulverized coal.

[0059] Understandably, design parameter values ​​are a set of benchmark parameters determined during the boiler design phase based on the characteristics of the target coal, combustion process requirements, and environmental standards. They are the "design genes" for the safe, efficient, and environmentally friendly operation of the boiler, and different design coal types have different design parameter values. Based on this, when the actual coal quality deviates from the design value, that is, when one or more of the theoretical ignition heat, stable combustion index, and optimal combustion air distribution range deviate from their corresponding design parameter values, a coal quality feedforward compensation amount (such as the fuel quantity ΔBf adjustment value and the air volume ΔVf adjustment value) will be calculated based on the degree of deviation to dynamically adjust the combustion parameters and maintain stable combustion and environmentally friendly emissions.

[0060] Furthermore, in one embodiment, the disturbance rejection control command includes commands to adjust the fuel distribution of each layer and / or to adjust the secondary air ratio while maintaining the total input energy constant.

[0061] In this exemplary embodiment, control parameters are coordinated under the principle of constant total energy. That is, when calculating the fuel fine-tuning amount for each layer, an equality constraint is applied to ensure that the sum of the fuel increase or decrease for each layer is zero (i.e., maintaining constant total input energy). Thus, without affecting the boiler's main control energy balance, the ignition conditions and flame structure are reshaped solely through internal allocation optimization. Based on this, the anti-disturbance control commands include commands to adjust the fuel allocation for each layer and / or commands to adjust the secondary air ratio (i.e., commands to adjust the opening of the secondary air dampers for each layer to change the aerodynamic field inside the furnace, used to optimize the flame center position inside the furnace) under the constraint of maintaining constant total input energy, aiming to quickly suppress instability trends and reshape a stable aerodynamic field.

[0062] In summary, this embodiment overcomes the lag and passivity of existing boiler low-load combustion control systems, and provides a control method that can provide early warning of combustion instability risks and perform feedforward compensation based on changes in the quality of coal entering the furnace to achieve "active disturbance resistance", thereby significantly improving the stability and safety of boiler combustion under low-load conditions.

[0063] Secondly, this application also provides a low-load combustion control system for a boiler based on acoustic early warning and coal quality feedforward.

[0064] In one embodiment, reference is made to Figure 2 , Figure 2 This is a schematic diagram of the functional modules of an embodiment of the acoustic early warning and coal quality feedforward boiler low-load combustion control system of this application. Figure 2 As shown, the low-load combustion control system for a boiler based on acoustic early warning and coal quality feedforward includes: The data sensing layer is used to acquire real-time medium- and high-frequency pressure pulsation signals, real-time coal quality data, and real-time operating parameters corresponding to the boiler's low-load combustion state. The intelligent analysis layer is used to extract features from real-time medium- and high-frequency pressure pulsation signals to obtain combustion stability feature values, and calculate the combustion stability deviation index (CSDI) based on the combustion stability feature values; and calculate the coal quality feedforward compensation amount corresponding to the basic combustion control command based on the real-time coal quality data. The execution control layer is used to enter the dynamic disturbance rejection mode when the CSDI is greater than or equal to the first-level warning threshold. It performs fuzzy control based on the CSDI and its corresponding rate of change, the coal quality feedforward compensation amount, and the real-time operating parameters to generate disturbance rejection fine-tuning amount. It generates disturbance rejection control command through the coal quality feedforward compensation amount, the disturbance rejection fine-tuning amount, and the basic combustion control command to achieve low-load combustion control of the boiler through the disturbance rejection control command.

[0065] Furthermore, in one embodiment, the execution control layer is also used for: When CSDI is less than the first-level warning threshold, it enters steady-state feedforward mode to generate feedforward combustion control commands based on coal quality feedforward compensation and basic combustion control commands. Low-load combustion control of the boiler is achieved through feedforward combustion control commands.

[0066] Furthermore, in one embodiment, the execution control layer is also used to: when CSDI is greater than or equal to the secondary warning threshold, enter the safety linkage mode, further increase the control intensity or response priority on the basis of dynamic anti-disturbance mode control, and trigger the preset emergency combustion stabilization plan, wherein the secondary warning threshold is greater than the primary warning threshold.

[0067] Furthermore, in one embodiment, the intelligent analysis layer is specifically used for: Filtering and spectrum analysis are performed on real-time medium- and high-frequency pressure pulsation signals to obtain real-time sound pressure level energy, real-time main frequency offset, and real-time sound signal non-uniformity index under a preset specific frequency band. Real-time sound pressure level energy, real-time dominant frequency offset, and real-time sound signal non-uniformity index are used as characteristic values ​​of combustion stability.

[0068] Furthermore, in one embodiment, the intelligent analysis layer is specifically used for: The real-time sound pressure level energy, real-time main frequency offset, and real-time acoustic signal non-uniformity index are compared with the standard sound pressure level energy, standard main frequency offset, and standard acoustic signal non-uniformity index in the preset stable combustion acoustic fingerprint benchmark library to obtain the energy difference, offset difference, and index difference. The CSDI is obtained by weighting the energy difference, offset difference, and exponential difference. Among them, the standard feature values ​​in the preset stable combustion acoustic fingerprint benchmark library are established when the boiler is under stable operating conditions.

[0069] Furthermore, in one embodiment, the real-time coal quality data includes dry ash-free volatile matter, as-received moisture content, and as-received lower heating value; the intelligent analysis layer is further used for: The target parameters of the current coal quality are calculated based on real-time coal quality data. The target parameters include theoretical ignition heat, stable combustion index, and optimal combustion air distribution range. The coal quality feedforward compensation amount corresponding to the basic combustion control command is calculated based on the target parameters and the preset design parameter values.

[0070] Furthermore, in one embodiment, the disturbance rejection control command includes commands to adjust the fuel distribution of each layer and / or to adjust the secondary air ratio while maintaining the total input energy constant.

[0071] In summary, in practical implementation, the control system described in this embodiment operates as an upper-level optimization module of the existing Coordinated Control System (CCS) of the boiler. Specifically, this system receives basic combustion control commands (including total fuel quantity, total air volume, etc.) from the CCS, and superimposes coal quality feedforward compensation and anti-disturbance fine-tuning quantities generated by this system to form the final optimized control commands sent to the actuators. See also Figure 2As shown, the low-load combustion control system for boilers based on acoustic early warning and coal quality feedforward provided in this embodiment includes a data sensing layer, an intelligent analysis layer, and an execution control layer. The data sensing layer includes an acoustic sensor array, an online coal quality analyzer, and conventional monitoring instruments. The acoustic sensor array is arranged in the boiler burner area and on the four upper walls of the furnace to collect combustion noise and pressure pulsation signals within the furnace. The online coal quality analyzer (preferably using neutron activation or laser breakdown spectroscopy) is installed at the coal feeder outlet or on the coal conveyor belt to detect the industrial analysis parameters of the coal entering the furnace in real time (at least including Vdaf, Mar, Qnet,ar). Conventional monitoring instruments include, but are not limited to, furnace negative pressure sensors, oxygen sensors, fuel quantity measuring devices, air volume measuring devices, and feedback devices for the opening of each secondary damper.

[0072] The intelligent analysis layer includes an acoustic early warning module and a coal quality feedforward calculation module. The acoustic early warning module connects to an acoustic sensor array and has a built-in signal processing unit and feature library. It filters and performs spectral analysis on the collected acoustic signals, extracting real-time characteristic indicators related to combustion stability, such as sound pressure level energy (50-300Hz), dominant frequency offset, and acoustic signal non-uniformity index. Then, it calculates the CSDI (Current Stable Combustion Acoustic Fingerprint Index) by comparing the real-time characteristic indicators with a stable combustion acoustic fingerprint benchmark library established based on historical stable operation data. When the CSDI exceeds the first-level early warning threshold, an "early warning" signal is issued; when it exceeds the second-level early warning threshold, an "alarm" signal is issued. The coal quality feedforward calculation module connects to an online coal quality analyzer and has a built-in combustion dynamics model. It calculates the theoretical ignition heat, stable combustion index, and optimal combustion air distribution range of the current coal quality based on real-time coal quality data, compares the calculation results with design parameter values, and generates coal quality feedforward compensation.

[0073] The execution control layer includes the Active Disturbance Control Unit (APCU), which can be implemented by adding functional blocks to an independent PLC (Programmable Logic Controller) or DCS (Distributed Control System). It is worth noting that the APCU is the core controller of this system, connecting the intelligent analysis layer and the execution mechanism. It receives CSDI signals and warning / alarm level signals from the acoustic warning module, as well as coal quality feedforward compensation from the coal quality feedforward calculation module. The working logic of the APCU is as follows: (1) Steady-state feedforward mode: When there is no warning, the APCU smoothly superimposes the coal quality feedforward compensation onto the original air and coal command issued by the boiler main control system (CCS) to realize "static" active adjustment based on coal quality changes and compensate for coal quality disturbances; (2) Dynamic State anti-interference mode: When an acoustic warning signal is received, the APCU immediately activates the dynamic anti-interference algorithm. The algorithm uses CSDI as the controlled variable and dynamically adjusts the gain of the coal quality feedforward compensation to generate fine-tuning instructions for combustion enhancement (such as slightly increasing the upper layer fuel and offsetting the opening of the secondary air damper to reshape the aerodynamic field inside the furnace), actively suppressing the instability trend. This mode has a higher priority than the steady-state feedforward mode; (3) Safety interlock and linkage: When an acoustic alarm signal is received, the APCU can send a combustion deterioration chain signal to the boiler safety monitoring system (FSSS) and trigger the preset emergency stable combustion strategy (such as automatically operating the micro oil gun).

[0074] In summary, this embodiment provides a system that can detect combustion instability risks in advance and proactively and forward-lookingly adjust control commands, which is of great significance for ensuring the safe, environmentally friendly, and efficient operation of the unit during deep peak shaving.

[0075] The functions of each part in the acoustic early warning and coal quality feedforward boiler low-load combustion control system mentioned above correspond to the steps in the embodiment of the acoustic early warning and coal quality feedforward boiler low-load combustion control method. Their functions and implementation processes will not be described in detail here.

[0076] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0077] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0078] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0079] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0080] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0081] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0082] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward, characterized in that, The boiler low-load combustion control method based on acoustic early warning and coal quality feedforward includes: Acquire real-time medium- and high-frequency pressure pulsation signals, real-time coal quality data, and real-time operating parameters corresponding to the boiler's low-load combustion state; Feature extraction is performed on real-time medium- and high-frequency pressure pulsation signals to obtain combustion stability feature values, and the combustion stability deviation index (CSDI) is calculated based on the combustion stability feature values. The coal quality feedforward compensation amount corresponding to the basic combustion control command is calculated based on the real-time coal quality data. When CSDI is greater than or equal to the first-level warning threshold, the system enters dynamic anti-disturbance mode to perform fuzzy control based on the CSDI and its corresponding rate of change, the coal quality feedforward compensation amount, and the real-time operating parameters, thereby generating anti-disturbance fine-tuning amount. The anti-disturbance control command is generated by the coal quality feedforward compensation amount, the anti-disturbance fine-tuning amount, and the basic combustion control command, so as to realize the control of low-load combustion of the boiler through the anti-disturbance control command.

2. The method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward as described in claim 1, characterized in that, The method further includes: When CSDI is less than the first-level warning threshold, it enters steady-state feedforward mode to generate feedforward combustion control commands based on coal quality feedforward compensation and basic combustion control commands. Low-load combustion control of the boiler is achieved through feedforward combustion control commands.

3. The method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward as described in claim 1, characterized in that, The method further includes: When CSDI is greater than or equal to the Level 2 warning threshold, the system enters the safety linkage mode. Based on the dynamic anti-disturbance mode control, the control intensity or response priority is further increased, and the preset emergency combustion stabilization plan is triggered. The Level 2 warning threshold is greater than the Level 1 warning threshold.

4. The method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward as described in claim 1, characterized in that, The step of extracting features from real-time medium-to-high frequency pressure pulsation signals to obtain combustion stability feature values ​​includes: Filtering and spectrum analysis are performed on real-time medium- and high-frequency pressure pulsation signals to obtain real-time sound pressure level energy, real-time main frequency offset, and real-time sound signal non-uniformity index under a preset specific frequency band. Real-time sound pressure level energy, real-time dominant frequency offset, and real-time sound signal non-uniformity index are used as characteristic values ​​of combustion stability.

5. The method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward as described in claim 4, characterized in that, The determination of the combustion stability deviation index (CSDI) based on combustion stability characteristic values ​​includes: The real-time sound pressure level energy, real-time main frequency offset, and real-time acoustic signal non-uniformity index are compared with the standard sound pressure level energy, standard main frequency offset, and standard acoustic signal non-uniformity index in the preset stable combustion acoustic fingerprint benchmark library to obtain the energy difference, offset difference, and index difference. The CSDI is obtained by weighting the energy difference, offset difference, and exponential difference. Among them, the standard feature values ​​in the preset stable combustion acoustic fingerprint benchmark library are established when the boiler is under stable operating conditions.

6. The method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward as described in claim 1, characterized in that, The real-time coal quality data includes dry ash-free volatile matter, as-received moisture content, and as-received lower heating value. The calculation of the coal quality feedforward compensation amount corresponding to the basic combustion control command based on the real-time coal quality data includes: Based on the real-time coal quality data, the target parameters of the current coal quality are calculated, including the theoretical ignition heat, the stable combustion index, and the optimal combustion air distribution range. The coal quality feedforward compensation amount corresponding to the basic combustion control command is calculated based on the target parameters and the preset design parameter values.

7. The method for low-load combustion control of a boiler based on acoustic early warning and coal quality feedforward as described in claim 1, characterized in that: The disturbance rejection control commands include commands to adjust the fuel distribution of each layer and / or adjust the secondary air ratio while maintaining the total input energy constant.

8. A low-load combustion control system for a boiler based on acoustic early warning and coal quality feedforward, characterized in that, The low-load combustion control system for boilers based on acoustic early warning and coal quality feedforward includes: The data sensing layer is used to acquire real-time medium- and high-frequency pressure pulsation signals, real-time coal quality data, and real-time operating parameters corresponding to the boiler's low-load combustion state. The intelligent analysis layer is used to extract features from real-time medium- and high-frequency pressure pulsation signals to obtain combustion stability feature values, and calculate the combustion stability deviation index (CSDI) based on the combustion stability feature values; and calculate the coal quality feedforward compensation amount corresponding to the basic combustion control command based on the real-time coal quality data. The execution control layer is used to enter the dynamic disturbance rejection mode when the CSDI is greater than or equal to the first-level warning threshold. It performs fuzzy control based on the CSDI and its corresponding rate of change, the coal quality feedforward compensation amount, and the real-time operating parameters to generate disturbance rejection fine-tuning amount. It generates disturbance rejection control command through the coal quality feedforward compensation amount, the disturbance rejection fine-tuning amount, and the basic combustion control command to achieve low-load combustion control of the boiler through the disturbance rejection control command.

9. The low-load combustion control system for boilers based on acoustic early warning and coal quality feedforward as described in claim 8, characterized in that, The execution control layer is also used for: When CSDI is less than the first-level warning threshold, it enters steady-state feedforward mode to generate feedforward combustion control commands based on coal quality feedforward compensation and basic combustion control commands. Low-load combustion control of the boiler is achieved through feedforward combustion control commands.

10. The low-load combustion control system for boilers based on acoustic early warning and coal quality feedforward as described in claim 8, characterized in that, The execution control layer is also used for: When CSDI is greater than or equal to the Level 2 warning threshold, the system enters the safety linkage mode. Based on the dynamic anti-disturbance mode control, the control intensity or response priority is further increased, and the preset emergency combustion stabilization plan is triggered. The Level 2 warning threshold is greater than the Level 1 warning threshold.

Images