Air-powder coupling adjusting system and method based on combustion stability index driving
By constructing a combustion stability index-based air-coal coupling regulation system, multi-source signals are collected in real time and coal is supplied quickly to compensate for fuel, and primary air volume is dynamically matched. This solves the problem of short-cycle instability during combustion and improves the robustness and safety of combustion control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUAZHONG UNIV OF SCI & TECH
- Filing Date
- 2026-02-04
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies cannot detect and respond quickly to short-cycle instability during combustion in real time, making combustion fluctuations difficult to control and potentially causing safety accidents. Furthermore, the lack of a coordinated wind-coal regulation mechanism makes it easy to introduce new sources of fluctuations.
A combustion stability index-based air-coal coupling regulation system is constructed. Multi-source signals are collected in real time through the furnace status sensing module to construct the combustion stability index. The small coal silo rapid coal supply module compensates for fuel in seconds. Combined with the air-coal coordinated regulation module, the primary air volume is dynamically matched to ensure that the air-coal balance is not disrupted during the regulation process.
It achieves accurate identification and rapid response to combustion stability, suppresses short-cycle fluctuations, improves the robustness and safety of the combustion process, and avoids the introduction of new fluctuation sources due to rapid control.
Smart Images

Figure CN122015118A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of combustion control and operation optimization technology of coal-fired boilers, specifically relating to a coal-fired air coupling regulation system and method driven by combustion stability index. Background Technology
[0002] Against the backdrop of "dual carbon" targets and the construction of new power systems, the deep participation of coal-fired power units in peak shaving has become the norm. Frequent load changes, wide-load operation, and variable coal quality make the combustion process in the furnace prone to dynamic instability phenomena such as short-cycle pressure oscillations, flame flashing, and localized flameout. Traditional combustion control systems mainly rely on slow parameters such as oxygen content and main steam temperature for steady-state regulation, lacking the ability to perceive and quickly suppress combustion fluctuations in real time, making it difficult to cope with disturbances on the order of seconds or even sub-seconds. Once combustion instability is not suppressed in time, it may lead to serious accidents such as deflagration, fire extinguishing, and overheating of heating surfaces, threatening unit safety and grid stability. To address combustion instability, Chinese patent CN111678167B discloses an online air-coal pulverized coal control system for ultra-supercritical units. Through online monitoring of coal powder concentration, fineness, primary / secondary air, CO, and fly ash carbon content, it uses an air-coal balancing distributor to automatically adjust the coal powder distribution angle of each pipeline, achieving static balance in air-coal delivery and improving combustion uniformity. Chinese patent CN115962478A proposes a full-condition adaptive combustion optimization control method for coal-fired power units. Based on historical operating data, it constructs soft measurement models for boiler efficiency, CO, and NOx, and generates optimal combustion parameters (such as coal feed rate and damper opening) through multi-objective optimization, which are then fed into a closed-loop DCS for execution. Its core is data-driven steady-state optimization, suitable for economical and environmentally friendly operation under different loads.
[0003] However, in practical applications, existing technical solutions still have some shortcomings: First, existing technologies do not integrate dynamic signals such as high-frequency fluctuations in flame and furnace pressure, and cannot construct a unified criterion for combustion stability. Therefore, they cannot distinguish between slow operating condition deviations and dangerous combustion oscillations (such as 0.5–3Hz pressure pulsations), resulting in inaccurate detection and judgment of short-cycle instability. Second, the optimization commands of existing technologies act on the main coal mill system, with slow response (minutes) or only adjusting the distribution ratio without changing the total coal supply. They cannot dynamically compensate for global fuel shortages or excesses within seconds, making it difficult to suppress combustion fluctuations caused by sudden disturbances. Third, if only the coal supply is increased without coordinating the primary air volume, it is easy to cause local coal-rich areas with insufficient air or dilute phase transport, which will exacerbate combustion instability. Existing technologies lack a coordinated mechanism for rapid fuel correction and synchronous air volume adjustment, which may introduce new sources of fluctuation due to rapid control. Summary of the Invention
[0004] To address the shortcomings of the existing technologies, the technical problem to be solved by this invention is to provide a combustion stability index-driven air-coal coupling regulation system and method, which constructs a quantifiable combustion stability index, utilizes an independent rapid coal supply channel to achieve second-level compensation on the fuel side, and introduces an air-coal coordination mechanism to dynamically match the primary air volume while rapidly adjusting the coal, ensuring that the air-coal balance is not disrupted during the regulation process.
[0005] The technical solution adopted in this invention is as follows:
[0006] A combustion stability index-driven air-coal coupling regulation system includes a furnace state sensing module, a combustion stability index construction module, a coupling control decision module, a small coal silo rapid coal supply execution module, an air-coal coordinated regulation module, and a safety degradation module.
[0007] The furnace status sensing module includes a flame detector, a furnace pressure or negative pressure sensor, a burner outlet temperature sensor, a furnace temperature sensor, and a flue gas analyzer; it is used to collect multi-source combustion status signals in real time, and to perform time alignment and synchronization with a sampling period of 0.1–1s, providing raw data input for the construction of combustion stability indicators;
[0008] The combustion stability index construction module includes a signal preprocessing unit, a sliding window feature extraction unit, and a multi-source feature fusion algorithm execution unit; it is used to extract furnace pressure fluctuation energy, flame variation coefficient, and CO / O2 deviation features, fuse them to generate a single scalar or vector form of combustion stability index, and divide the stable zone, early warning zone, and strong intervention zone accordingly.
[0009] The coupled control decision module includes a feedback controller, a feedforward compensator, a constraint management unit, and an anti-saturation processing logic unit; it is used to output the rapid powder supply correction amount for the small powder hopper, output the primary air or air distribution fine-tuning command, and ensure that the control actions are executed within the safety boundary.
[0010] The small pulverized coal silo rapid pulverized coal supply execution module includes a small pulverized coal silo, a metering pulverized coal feeder, a branch regulating valve, a branch quick-opening valve, a check valve, a branch pipeline, and a mixing point. The mixing point is located before the primary air main pipe enters the burner distributor. The compensation pulverized coal supplied by the small pulverized coal silo is mixed with the primary air at the mixing point and then enters the burner. This is used to rapidly inject compensation pulverized coal and suppress short-cycle combustion fluctuations.
[0011] The air-powder coordinated adjustment module includes a primary air regulating valve or variable frequency fan control system, a secondary air distribution damper actuator, and an air-powder matching calculation unit; it is used to dynamically adjust the primary air volume to maintain air-powder matching, fine-tune the local secondary air distribution, and suppress local fluctuations.
[0012] The safety degradation module includes a sensor health diagnosis unit, an actuator status monitoring unit, and a constraint touch detection logic unit; it is used to evaluate sensor reliability and actuator availability in real time, trigger degradation in case of failure or exceeding limits, maintain the normal operation of the basic DCS control loop, and ensure the safety baseline.
[0013] The working principle of the components of the above-mentioned air-powder coupling regulation system is as follows:
[0014] The entire system uses combustion stability indicators as the core driving signal, forming a closed-loop intelligent adjustment chain: the furnace status sensing module collects multi-dimensional signals, which are pre-processed and sent to the combustion stability indicator construction module to output the stability quantification value in real time; the coupled control decision module generates a rapid coal supply correction amount based on the indicator deviation, which is then injected into the small coal silo execution module in seconds to compensate for the coal powder; the air-coal coordinated adjustment module synchronously calculates the equivalent air-coal matching index, and if the coal supply correction amount causes air-coal mismatch, it finely adjusts the primary air volume or secondary air distribution to maintain coal carrying stability; the safety degradation module monitors the system health status throughout the process, and automatically degrades the control intensity once an abnormality is detected to prevent malfunctions from amplifying disturbances; all modules are integrated with the unit's DCS through standard I / O channels to achieve millisecond-level data interaction and control output, without the need for external systems.
[0015] A combustion stability index-driven air-coal coupling regulation method, implemented based on the aforementioned combustion stability index-driven air-coal coupling regulation system, includes the following steps:
[0016] Step S1: System initialization and parameter tuning;
[0017] Preferably, step S1 includes:
[0018] Set the sliding window length and step size: the sliding time window length is fixed at 30s, and the update step size is fixed at 1s;
[0019] Configure the filter frequency band;
[0020] Define the threshold for stable metric partitions;
[0021] Set the maximum correction amount for the small powder hopper, the upper and lower limits for airflow adjustment, and inventory constraints.
[0022] Step S2: Real-time acquisition of furnace status signals;
[0023] Preferably, in step S2, the furnace state sensing module collects and outputs the following five types of signals, which constitute a fixed input set for the combustion stability closed loop:
[0024] (1) Flame stabilization signal F s (t), which is output by the flame detector and forms a stable metric through its internal algorithm;
[0025] (2) Furnace pressure signal P(t), which is furnace pressure or furnace negative pressure;
[0026] (3) The furnace temperature signal T(t) is the representative point temperature of the furnace;
[0027] (4) Flue gas oxygen content signal O2(t);
[0028] (5) Carbon monoxide signal CO(t) in flue gas;
[0029] The sampling period is set as follows: 0.2s for flame stability signal and furnace pressure signal, and 1s for furnace temperature, oxygen content and CO signal; all signals are aligned with a unified timestamp to form a sliding time window input sequence.
[0030] Step S3: Construct combustion stability indicators;
[0031] Preferably, in step S3, the combustion stability index construction module calculates four characteristic quantities within a sliding time window, and calculates the combustion fluctuation index CBI and the combustion stability index CSI accordingly, and then performs index partitioning; specifically including:
[0032] 3.1 Characteristic Quantities:
[0033] (1) Furnace pressure fluctuation intensity Ep:
[0034] ,
[0035] Among them, P k This represents the k-th pressure sample value within the sliding window. is the average pressure within the sliding window, and N is the number of samples within the sliding window;
[0036] (2) Flame fluctuation coefficient CV f :
[0037] ,
[0038] in, The standard deviation of the flame stability signal within the sliding window. This represents the average value of the flame stability signal within the sliding window. To prevent constants with a denominator of zero;
[0039] (3) CO risk quantity R CO :
[0040] ,
[0041] Among them, CO avg The average CO value within the sliding window. K represents the average rate of change of CO within the sliding window. CO This is the risk gain coefficient for CO;
[0042] (4) Oxygen deviation D O2 :
[0043] ,
[0044] Among them, O 2,avg The average O2 value within the sliding window, O 2,set Set the oxygen content as a setpoint (dO2 / dt). avg K represents the average rate of change of oxygen within the sliding window. O2 This is the oxygen content deviation gain coefficient;
[0045] 3.2 Normalization:
[0046] Perform normalization for each feature x:
[0047] ,
[0048] Where, x min With x max This serves as the tuning boundary for the corresponding characteristic quantity;
[0049] 3.3 Combustion Fluctuation Index and Combustion Stability Index:
[0050] ,
[0051] ,
[0052] Wherein, CBI is the combustion fluctuation index, CSI is the combustion stability index, and w1, w2, w3, and w4 are fixed weight parameters;
[0053] Output deviation:
[0054] ,
[0055] CSI tar The target stability index;
[0056] 3.4 Indicator Zones:
[0057] Establish stable zones, early warning zones, and strong intervention zones:
[0058] Stable region: CSI >= CSI G ;
[0059] Warning Zone: CSI R <=CSI <CSI G ;
[0060] High Intervention Zone: CSI <CSI R ;
[0061] Among them, CSIG With CSI R The threshold is set and written into the DCS parameter table.
[0062] Step S4: Determine the stable state and trigger control;
[0063] Preferably, in step S4,
[0064] If CSI is in the stable region: maintain current control;
[0065] If CSI is in the warning zone: activate feedforward compensation;
[0066] If CSI is in the strong intervention zone: activate feedback inhibition control;
[0067] Among them, feedforward compensation It consists of disturbance quantities, and the calculation formula is as follows:
[0068] ,
[0069] in The term representing the rate of change of load, For a wind disturbance term, For oxygen deviation; V PA For primary air volume, V PA,set k is the primary air volume setpoint. L K PA k O2 These are the tuning parameters;
[0070] Feedback quantity in feedback suppression control Using PI, the calculation formula is as follows:
[0071] ,
[0072] Among them, K p K i These are the tuning parameters.
[0073] Step S5: Calculate the rapid powder supply correction amount;
[0074] Preferably, in step S5, the small powder hopper quickly adjusts the powder supply amount. The calculation formula is:
[0075] ,
[0076] Constraint shaping is fixed and executed.
[0077] Limit:
[0078] Limited to ,
[0079] Speed limit:
[0080] ,
[0081] Inventory constraint: When the availability duration of the small pink warehouse is less than the inventory threshold, ΔQ will be... max Compress to a conservative upper limit and then proceed with smooth recycling.
[0082] Anti-saturation: When the output exceeds the limit for 5 seconds, the integral term is frozen and the parameter group is switched to conservative.
[0083] Where, ΔQ min ΔQ max These are the tuning parameters.
[0084] Step S6: Execute powder supply from the small powder hopper;
[0085] Preferably, in step S6, the small pulverized coal silo outputs pulverized coal through the metering and feeding device, and then sends it to the mixing position on the burner side via a branch pipeline, so as to achieve rapid dynamic adjustment of the fuel side. The sequence of actions for supplying pulverized coal to the small pulverized coal silo is as follows:
[0086] Hot standby: The regulating valve is opened to the pre-open position u pre The quick-opening valve is closed, and the powder feeder is in the start-up permission state;
[0087] Rapid vibration suppression: When the powder feeder starts, the powder feeding rate is set according to the slope R. up From 0 to Q ps,set Regulating valve tracking control;
[0088] Strong intervention: The quick-opening valve opens to the emergency opening degree u emg The amount of powder fed is based on the slope R. strong Climbing uphill to Q ps,emg;
[0089] Smooth recovery: powder feed rate according to slope R down When the temperature drops back to 0, the quick-opening valve closes, the regulating valve returns to the pre-open position, and the powder feeder stops.
[0090] Downgrade exit: Branch valves are closed, and the powder feeder stops.
[0091] Step S7: Air-powder coordinated adjustment;
[0092] Preferably, in step S7, the wind-coal coordinated adjustment module uses an equivalent wind-coal matching index R. e For collaborative control objects:
[0093]
[0094] Q base The main pulverized coal supply system is given the fuel quantity, ∆Q s V provides rapid powder supply correction for the small powder hopper. PAThis refers to the primary air volume;
[0095] The system sets the allowable range for the equivalent air-powder matching index [R]. e,low ,R e,high [and the lower limit of primary wind safety V] PA,min ;
[0096] The collaboration rules are as follows:
[0097] When R e >R e,high Output primary airflow increase command ΔV PA , making R e Return to the allowed range;
[0098] When R e <R e,low And V PA >V PA,min Output primary airflow reduction command ΔV PA , making R e Return to the allowed range;
[0099] When V PA =V PA,min : Lock the wind in place so that it will no longer drop, and Q ps The upper limit is compressed to a conservative limit to avoid secondary fluctuations caused by mismatch between wind and powder;
[0100] Secondary wind distribution correction uses a fixed trigger condition: R CO Exceeding the CO risk threshold and lasting for 3 seconds, or F s (t) continuously decreases for 3 seconds; after triggering, the secondary damper performs a balance correction, the correction range is limited to keep the total air volume unchanged.
[0101] Step S8: Security monitoring and downgrade processing;
[0102] Preferably, in step S8, the downgrade criterion is one of the following conditions:
[0103] (1) The flame stability signal does not refresh for more than 2 seconds;
[0104] (2) The furnace pressure signal does not refresh for more than 2 seconds;
[0105] (3) The deviation between the powder feeder's feedback and the command exceeds the threshold and lasts for 5 seconds;
[0106] (4) The material level in the small powder silo is lower than the lower limit of the material level and remains so for 10 seconds;
[0107] (5) The branch differential pressure exceeds the blockage threshold and remains so for 5 seconds;
[0108] The relegation actions include:
[0109] (1) Entering a downgraded exit state;
[0110] (2) Close the quick-opening valve and the regulating valve;
[0111] (3) Stop the powder feeder;
[0112] (4) Exit primary wind coordination and secondary wind allocation correction;
[0113] (5) Maintain the conventional combustion control circuit;
[0114] (6) Generate alarms and record events;
[0115] Event logs include: trigger criterion code, trigger time, maximum CBI value, minimum CSI value, cumulative compensation amount, downgrade duration, and cumulative action amount of primary wind coordination.
[0116] Step S9: Execute repeatedly;
[0117] Repeat steps S2–S8 according to the set time period to form a closed-loop regulation.
[0118] The beneficial effects obtained by adopting the above technical solution are as follows:
[0119] (1) The furnace state sensing module of the present invention collects flame stability signal, furnace pressure signal, furnace temperature signal, flue gas oxygen signal and flue gas carbon monoxide signal. The combustion stability index construction module calculates four characteristic quantities within the sliding time window: furnace pressure fluctuation intensity, flame fluctuation coefficient, CO risk quantity and oxygen deviation quantity. Based on these, it calculates the combustion fluctuation index CBI and the combustion stability index CSI, and then divides the index into regions. This realizes the engineering quantification and discrimination of combustion stability, and can accurately identify short-cycle dangerous oscillation (strong intervention zone), early warning fluctuation (early warning zone) and stable operating condition (stable zone), providing accurate input for rapid control.
[0120] (2) The coupling control decision module of the present invention generates a rapid powder supply correction amount ΔQ based on the CSI deviation. ps It includes a feedback damping term (to suppress current fluctuations) and a feedforward compensation term (to offset load / airflow disturbances), and shapes the output through constraint management (inventory, rate limit, etc.); the invention controls the metering and feeding device with a small powder silo rapid powder supply execution module independent of the main powder supply system, according to the rapid powder supply correction amount ΔQ. ps Injecting pulverized coal at the mixing point on the burner side increases the coal supply response speed to the second level or even the sub-second level, which can effectively compensate for the instantaneous fuel imbalance caused by sudden changes in coal quality and primary air disturbance, and significantly reduce the combustion fluctuation amplitude.
[0121] (3) The present invention sets up a wind-coal coordinated adjustment module and defines an equivalent wind-coal matching index R. e When the small pink bin outputs the correction amount ΔQps Simultaneously calculate the required primary air volume fine-tuning amount ΔV PA , making R e Maintain within the allowable range; if the airflow is constrained (e.g., has reached the lower limit), automatically reduce ΔQ. ps To prevent sudden changes in the air-coal ratio; to ensure that rapid fuel compensation does not introduce new air-coal mismatch fluctuations, thereby improving the stability of regulation and combustion robustness at the system level. Attached Figure Description
[0122] Figure 1 This is a block diagram of a combustion stability index-driven air-coal coupling regulation system according to the present invention.
[0123] Figure 2 This is a flowchart illustrating the furnace state sensing and combustion stability index construction process of this invention.
[0124] Figure 3 This is a block diagram of the closed-loop control strategy for coupling the small powder silo and the furnace in this invention. Detailed Implementation
[0125] The technical solution of the present invention will now be described more clearly and completely with reference to the accompanying drawings.
[0126] I. System Deployment and Module Connection Methods
[0127] 1. Furnace status sensing module
[0128] This module can connect to at least two types of signal sources:
[0129] Flame-related signals: intensity, flicker rate, stability, etc. output by the flame detector;
[0130] Fluctuation-related signals: high-frequency fluctuations in furnace pressure (or furnace negative pressure), fluctuations in burner outlet temperature, and fluctuations in furnace temperature;
[0131] Reaction-related signals: flue gas oxygen content (O2), CO (optionally also NOx) and their fluctuation characteristics.
[0132] The signal sampling period can be set to 0.1–1s (determined based on the refresh rate of the field instruments and the DCS load), and a unified timestamp and channel alignment are performed.
[0133] 2. Combustion stability index construction module
[0134] The system performs denoising, feature extraction, and fusion of multi-source signals within a sliding time window, outputting combustion stability indicators and deviations. The sliding window length can be 10-60 s; the step size can be 0.5-5 s. It can output a single scalar indicator or a multi-dimensional indicator vector.
[0135] 3. Coupled Control Decision Module
[0136] The module takes the combustion stability index deviation as input and outputs the "rapid correction amount" of the small powder hopper's powder supply, and can also output the primary air / distribution air fine-tuning amount. This module includes feedback control, feedforward compensation, constraint management, and anti-saturation treatment.
[0137] 4. Small powder hopper rapid powder supply execution module
[0138] The small pulverized coal silo is fed by a metering feeding device, such as a metering screw feeder / rotary feeder / weighing feeder. The output pulverized coal is then sent through branch pipelines to the mixing position on the burner side (which can be connected to the primary air main pipe, burner distributor, or a branch pipe of a specific burner layer) to achieve rapid dynamic adjustment of the fuel side. This execution module is primarily used for "short-cycle correction" and does not handle the slow ramp-up task of the unit's total pulverized coal supply.
[0139] 5. Air-powder coordinated adjustment module
[0140] When the system detects a deviation in air-powder matching, it outputs a primary air volume fine-tuning or air distribution fine-tuning to prevent the stability loop from introducing new sources of fluctuation.
[0141] 6. Security Degradation Module
[0142] Perform online diagnostics on sensor reliability, actuator health, and constraint contact status; when degradation is triggered, gradually converge the control action (amplitude limit / speed limit / exit coordination) while keeping the basic control loop unaffected.
[0143] II. Construction of Combustion Stability Indicators
[0144] To achieve engineering adaptability and interpretability, this embodiment uses three sets of features—"fluctuation type + reaction type + flame type"—to construct combustion stability index.
[0145] 1. Signal preprocessing
[0146] The signals from each channel are processed sequentially as follows: outlier removal; low-pass / band-pass filtering (e.g., retaining the pressure fluctuation component in the 0.2-3 Hz range to characterize the intensity of combustion oscillation; the specific frequency band is adjusted according to the boiler characteristics); and normalization processing (unifying the dimensions and magnitude to the [0,1] range).
[0147] 2. Sliding window feature extraction
[0148] Extract the following within each time window: furnace pressure fluctuation energy; flame intensity variation coefficient; CO fluctuation intensity (or mean deviation); CO fluctuation intensity (or mean deviation); O2 deviation.
[0149] 3. The combustion stability index was obtained through fusion.
[0150] 4. Indicator Targets and Zoning
[0151] Set stable intervals (green zone), early warning intervals (yellow zone), and strong intervention intervals (red zone).
[0152] III. Coupled Control Decision
[0153] In this embodiment, the small powder silo supply is used as a rapid correction channel. Its control output is formed by superimposing three parts: 1) Feedback vibration suppression control, which takes the combustion stability deviation as input and outputs a rapid correction amount; 2) Feedforward compensation, which constructs a feedforward term to offset the fluctuation trend and reduce the overshoot caused by feedback lag when a disturbance source is detected (such as a rapid increase in O2 deviation, a sudden rise in furnace pressure fluctuation or air volume disturbance); 3) Constraint management and shaping, which applies amplitude limit, speed limit, inventory constraint and air volume lower limit constraint.
[0154] IV. Wind-powder synergistic regulation
[0155] When the system enables air-dust coordination, it uses equivalent air-dust indices to maintain "dust-carrying stability," but does not focus on "concentration improvement" as the main narrative; instead, it aims for "matching stability."
[0156] 1. Construct an equivalent matching index:
[0157]
[0158] Q base The main pulverized coal supply system is given the fuel quantity, ∆Q s V provides rapid powder supply correction for the small powder hopper. PA This refers to the primary air volume.
[0159] 2. Collaborative Strategy
[0160] If ∆Q s Increased R e If the upper limit is exceeded, output ∆V. PA Make R e Return to the allowed range.
[0161] If ∆V PA If the fast loop is constrained and unable to move, it will automatically contract ∆Q. s The amplitude is limited to avoid secondary fluctuations caused by wind-powder mismatch.
[0162] 3. Fine-tuning of air distribution
[0163] When abnormal fluctuations in local burners are detected and there are assignable dampers, a fine-tuning amount of secondary air distribution can be output to weaken the local fluctuation source, but the constraint that the total air volume target should not be destroyed must be met.
Claims
1. A combustion stability index-driven air-coal coupling regulation system, characterized in that, It includes a furnace status sensing module, a combustion stability index construction module, a coupled control decision module, a small pulverizer rapid pulverizer supply execution module, an air-pulverizer coordinated adjustment module, and a safety degradation module; The furnace status sensing module includes a flame detector, a furnace pressure or negative pressure sensor, a burner outlet temperature sensor, a furnace temperature sensor, and a flue gas analyzer; it is used to collect multi-source combustion status signals in real time, and to perform time alignment and synchronization with a sampling period of 0.1–1s, providing raw data input for the construction of combustion stability indicators; The combustion stability index construction module includes a signal preprocessing unit, a sliding window feature extraction unit, and a multi-source feature fusion algorithm execution unit; it is used to extract furnace pressure fluctuation energy, flame variation coefficient, and CO / O2 deviation features, fuse them to generate a single scalar or vector form of combustion stability index, and divide the stable zone, early warning zone, and strong intervention zone accordingly. The coupled control decision module includes a feedback controller, a feedforward compensator, a constraint management unit, and an anti-saturation processing logic unit; it is used to output the rapid powder supply correction amount for the small powder hopper, output the primary air or air distribution fine-tuning command, and ensure that the control actions are executed within the safety boundary. The small pulverized coal silo rapid pulverized coal supply execution module includes a small pulverized coal silo, a metering pulverized coal feeder, a branch regulating valve, a branch quick-opening valve, a check valve, a branch pipeline, and a mixing point. The mixing point is located before the primary air main pipe enters the burner distributor. The compensation pulverized coal supplied by the small pulverized coal silo is mixed with the primary air at the mixing point and then enters the burner. This is used to rapidly inject compensation pulverized coal and suppress short-cycle combustion fluctuations. The air-powder coordinated adjustment module includes a primary air regulating valve or variable frequency fan control system, a secondary air distribution damper actuator, and an air-powder matching calculation unit; it is used to dynamically adjust the primary air volume to maintain air-powder matching, fine-tune the local secondary air distribution, and suppress local fluctuations. The safety degradation module includes a sensor health diagnosis unit, an actuator status monitoring unit, and a constraint touch detection logic unit; it is used to evaluate sensor reliability and actuator availability in real time, trigger degradation in case of failure or exceeding limits, maintain the normal operation of the basic DCS control loop, and ensure the safety baseline.
2. A method for air-coal coupling regulation based on combustion stability index, implemented based on the air-coal coupling regulation system of claim 1, characterized in that, Includes the following steps: Step S1: System initialization and parameter tuning; Step S2: Real-time acquisition of furnace status signals; Step S3: Construct combustion stability indicators; Step S4: Determine the stable state and trigger control; Step S5: Calculate the rapid powder supply correction amount; Step S6: Execute powder supply from the small powder hopper; Step S7: Air-powder coordinated adjustment; Step S8: Security monitoring and downgrade processing; Step S9: Execute repeatedly; Repeat steps S2–S8 according to the set time period to form a closed-loop regulation.
3. The air-powder coupling adjustment method according to claim 2, characterized in that, Step S1 includes: Set the sliding window length and step size: the sliding time window length is fixed at 30s, and the update step size is fixed at 1s; Configure the filter frequency band; Define the threshold for stable metric partitions; Set the maximum correction amount for the small powder hopper, the upper and lower limits for airflow adjustment, and inventory constraints.
4. The air-powder coupling adjustment method according to claim 3, characterized in that, In step S2, the furnace status sensing module collects and outputs the following five types of signals: (1) Flame stabilization signal F s (t), which is output by the flame detector and forms a stable metric through its internal algorithm; (2) Furnace pressure signal P(t), which is furnace pressure or furnace negative pressure; (3) The furnace temperature signal T(t) is the representative point temperature of the furnace; (4) Flue gas oxygen content signal O2(t); (5) Carbon monoxide signal CO(t) in flue gas; The sampling period is set as follows: 0.2s for flame stability signal and furnace pressure signal, and 1s for furnace temperature, oxygen content and CO signal; all signals are aligned with a unified timestamp to form a sliding time window input sequence.
5. The air-powder coupling adjustment method according to claim 4, characterized in that, In step S3, the combustion stability index construction module calculates four characteristic quantities within a sliding time window, and calculates the combustion fluctuation index CBI and the combustion stability index CSI accordingly, and then performs index partitioning; specifically including: 3.1 Characteristic Quantities: (1) Furnace pressure fluctuation intensity Ep: , Among them, P k This represents the k-th pressure sample value within the sliding window. is the average pressure within the sliding window, and N is the number of samples within the sliding window; (2) Flame fluctuation coefficient CV f : , in, The standard deviation of the flame stability signal within the sliding window. This represents the average value of the flame stability signal within the sliding window. To prevent constants with a denominator of zero; (3) CO risk quantity R CO : , Among them, CO avg The average CO value within the sliding window. K represents the average rate of change of CO within the sliding window. CO This is the risk gain coefficient for CO; (4) Oxygen deviation D O2 : , Among them, O 2,avg The average O2 value within the sliding window, O 2,set Set the oxygen content as a setpoint (dO2 / dt). avg K represents the average rate of change of oxygen within the sliding window. O2 This is the oxygen content deviation gain coefficient; 3.2 Normalization: Perform normalization for each feature x: , Where, x min With x max This serves as the tuning boundary for the corresponding characteristic quantity; 3.3 Combustion Fluctuation Index and Combustion Stability Index: , , Wherein, CBI is the combustion fluctuation index, CSI is the combustion stability index, and w1, w2, w3, and w4 are fixed weight parameters; Output deviation: , CSI tar The target stability index; 3.4 Indicator Zones: Establish stable zones, early warning zones, and strong intervention zones: Stable region: CSI >= CSI G ; Warning Zone: CSI R <=CSI <CSI G ; High Intervention Zone: CSI <CSI R ; Among them, CSI G With CSI R The threshold is set and written into the DCS parameter table.
6. The air-powder coupling adjustment method according to claim 5, characterized in that, In step S4 If CSI is in the stable region: maintain current control; If CSI is in the warning zone: activate feedforward compensation; If CSI is in the strong intervention zone: activate feedback inhibition control; Among them, feedforward compensation It consists of disturbance quantities, and the calculation formula is as follows: , in The term representing the rate of change of load, For a wind disturbance term, For oxygen deviation; V PA For primary air volume, V PA,set k is the primary air volume setpoint. L K PA k O2 These are the tuning parameters; Feedback quantity in feedback suppression control Using PI, the calculation formula is as follows: , Among them, K p K i These are the tuning parameters.
7. The air-powder coupling adjustment method according to claim 6, characterized in that, In step S5, the small powder hopper quickly adjusts the powder supply amount. The calculation formula is: , Constraint shaping is fixed and executed. Limit: Limited to , Speed limit: , Inventory constraint: When the availability duration of the small pink warehouse is less than the inventory threshold, ΔQ will be... max Compress to a conservative upper limit and then proceed with smooth recycling. Anti-saturation: When the output exceeds the limit for 5 seconds, the integral term is frozen and the parameter group is switched to conservative. Where, ΔQ min ΔQ max These are the tuning parameters.
8. The air-powder coupling adjustment method according to claim 7, characterized in that, In step S6, the small pulverized coal silo outputs pulverized coal through the metering and feeding device, and sends it to the mixing position on the burner side via a branch pipeline to achieve rapid dynamic adjustment of the fuel side. The sequence of actions for supplying pulverized coal to the small pulverized coal silo is as follows: Hot standby: The regulating valve is opened to the pre-open position u pre The quick-opening valve is closed, and the powder feeder is in the start-up permission state; Rapid vibration suppression: When the powder feeder starts, the powder feeding rate is set according to the slope R. up From 0 to Q ps,set Regulating valve tracking control; Strong intervention: The quick-opening valve opens to the emergency opening degree u emg The amount of powder fed is based on the slope R. strong Climbing uphill to Q ps,emg; Smooth recovery: powder feed rate according to slope R down When the temperature drops back to 0, the quick-opening valve closes, the regulating valve returns to the pre-open position, and the powder feeder stops. Downgrade exit: Branch valves are closed, and the powder feeder stops.
9. The air-powder coupling adjustment method according to claim 8, characterized in that, In step S7, the wind-coal coordinated adjustment module uses the equivalent wind-coal matching index R e For collaborative control objects: ; Q base The main pulverized coal supply system is given the fuel quantity, ∆Q s V provides rapid powder supply correction for the small powder hopper. PA This refers to the primary air volume; The system sets the allowable range for the equivalent air-powder matching index [R]. e,low ,R e,high [and the lower limit of primary wind safety V] PA,min ; The collaboration rules are as follows: When R e >R e,high Output primary airflow increase command ΔV PA , making R e Return to the allowed range; When R e <R e,low And V PA >V PA,min Output primary wind reduction command ΔV PA , making R e Return to the allowed range; When V PA =V PA,min : Lock the wind in place so that it will no longer drop, and Q ps The upper limit is compressed to a conservative limit to avoid secondary fluctuations caused by mismatch between wind and powder; The secondary air distribution correction adopts a fixed trigger condition: RCO exceeds the CO risk threshold and lasts for 3 seconds, or Fs(t) decreases continuously and lasts for 3 seconds; after triggering, the secondary air damper performs a balance correction, and the correction range is limited to keep the total air volume unchanged.
10. The air-powder coupling adjustment method according to claim 9, characterized in that, In step S8, the downgrade criterion is one of the following conditions: (1) The flame stability signal does not refresh for more than 2 seconds; (2) The furnace pressure signal does not refresh for more than 2 seconds; (3) The deviation between the powder feeder's feedback and the command exceeds the threshold and lasts for 5 seconds; (4) The material level in the small powder silo is lower than the lower limit of the material level and remains so for 10 seconds; (5) The branch differential pressure exceeds the blockage threshold and remains so for 5 seconds; The relegation actions include: (1) Entering a downgraded exit state; (2) Close the quick-opening valve and the regulating valve; (3) Stop the powder feeder; (4) Exit primary wind coordination and secondary wind allocation correction; (5) Maintain the conventional combustion control circuit; (6) Generate alarms and record events; Event logs include: trigger criterion code, trigger time, maximum CBI value, minimum CSI value, cumulative compensation amount, downgrade duration, and cumulative action amount of primary wind coordination.