Intelligent control and waste heat recovery system for cooling air of a squaring and drying machine

Abstract

The application discloses a kind of cooling wind intelligent regulation and control and waste heat recovery system of scale hopper dry deslagging machine, and it relates to cooling intelligent regulation and control technical field.The cooling wind intelligent regulation and control and waste heat recovery system of scale hopper dry deslagging machine includes: data acquisition unit is responsible for gathering unit operation data set and dry deslagging machine measurement data set and current bottom slag share, virtual slag amount calculation unit relies on operating parameter to calculate virtual slag amount, slag temperature inversion unit is reversed with heat transfer model slag temperature and accounts for slag carrying heat, optimization control unit is fed back with slag carrying heat and virtual slag amount, obtains optimization control instruction under multiple constraints, safety control unit generates virtual temperature rise rate and completes layered protection determination, the present application is preferably guaranteed safety control by instruction coordination unit, maximizes dry slag waste heat recovery in safety boundary, and make cooling wind regulation and control in dynamic change operating environment continuously fit actual heat transfer demand.

Description

Technical Field

[0001] This invention relates to the field of intelligent cooling control technology, specifically to an intelligent control system for cooling air and waste heat recovery for a scale bucket dry slag discharge machine. Background Technology

[0002] During the operation of a coal-fired power plant boiler, high-temperature slag is continuously discharged from the bottom of the furnace. To safely transport and cool this high-temperature slag, a scale-type dry slag discharge machine is commonly used. Its working principle is as follows: the high-temperature slag falls from the slag well onto the slowly moving scale-type conveyor belt. The negative pressure of the boiler furnace is used to draw in ambient air, which enters from the tail of the dry slag discharge machine and flows in the opposite direction along the conveyor belt to exchange heat with the high-temperature slag. The heated air eventually enters the boiler furnace to participate in combustion, recovering the physical heat of the slag into the boiler's thermal cycle.

[0003] In the above process, the adjustment of cooling air volume is directly related to the slag cooling effect and waste heat recovery efficiency. Excessive cooling air volume will cause excessive cold air to enter the furnace, reducing boiler efficiency and increasing fan energy consumption; insufficient cooling air volume will make it difficult to cool the slag, causing equipment such as the bucket conveyor belt and chain to operate in a high-temperature environment, which poses a safety hazard.

[0004] Based on the above solutions, the limitations of the existing technology include at least the following problems: the adjustment of the cooling air volume of the existing scale bucket dry slag discharge machine mainly relies on the slag discharge temperature measuring point arranged at the tail end of the slag discharge machine as the feedback basis. This measuring point reflects the final temperature of the material after the complete heat exchange cycle. From the change of the slag discharge condition at the bottom of the furnace to the change of the response of this tail measuring point, it is necessary to go through the entire process of the material slowly moving on the scale bucket conveyor belt.

[0005] During variable load operation or coal quality switching, the amount of slag fed into the furnace and the heat carried by the slag fluctuate significantly in a short period of time, while the tail temperature measuring point has not yet reflected the above changes. During this period, the cooling air volume remains at the opening corresponding to the previous operating conditions. A mismatch occurs between the actual heat exchange state inside the slag discharger and the cooling air supply capacity. When the amount of slag suddenly increases, the cooling air volume is insufficient to remove the new heat load in the lag period. The thermal stress on the hopper and conveyor chain rises rapidly. In order to avoid the risk of equipment overheating in the lag period, the cooling damper is usually kept at a large opening to reserve sufficient safety margin. This conservative air volume setting for a long time results in a considerable proportion of cold air entering the furnace without sufficient heat exchange, which can easily dilute the expected benefits of waste heat recovery by flue gas heat loss and the power consumption of the blower. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an intelligent control system for cooling air and waste heat recovery in a scale bucket dry slag discharge machine. This system solves the problem in existing technologies where delayed tail temperature feedback forces conservative airflow adjustment, diluting the benefits of waste heat recovery.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent control system for cooling air and waste heat recovery of a hopper dry slag discharger, comprising: a data acquisition unit for acquiring the unit operation dataset and the dry slag discharger measurement dataset of the hopper dry slag discharger, as well as the current bottom slag share; a virtual slag quantity calculation unit for calculating the virtual slag quantity entering the hopper dry slag discharger based on the unit operation dataset and the current bottom slag share; and a slag temperature inversion unit for iteratively inverting and outputting the bottom slag discharge temperature using a countercurrent heat transfer model based on the virtual slag quantity and the dry slag discharger measurement dataset, and calculating the heat carried by the slag to be recovered in conjunction with the virtual slag quantity.

[0008] The optimization control unit is used to generate candidate cooling air volume as decision variables based on the measured value of cooling air volume in the dry slag discharger measurement dataset, with the heat carried by slag and virtual slag volume as feedforwards. Under the constraints of preset scale bucket safety temperature threshold, tail slag discharge safety temperature threshold and furnace oxygen prediction constraints, the optimal cooling air volume is solved as the optimization control command with the goal of maximizing the net benefit of waste heat recovery.

[0009] The safety control unit is used to perform heat balance processing based on the heat carried by the slag and the measurement data of the dry slag discharger, generate a virtual temperature rise rate on the surface of the hopper, and output a safety control command after confidence assessment and hierarchical protection in combination with the measurement data of the slag discharger; the command coordination output unit is used to coordinate the safety control command with priority over the economic control command to generate the final cooling air volume control command, so as to maximize the recovery of waste heat of dry slag within the safety boundary.

[0010] Further, the specific steps for calculating the virtual slag quantity entering the hopper dry slag discharger are as follows: Obtain the operating status signal and instantaneous coal quantity of each coal feeder from the unit operation data set; only accumulate the instantaneous coal quantity of the coal feeders in operation status to obtain the total coal feed quantity; obtain the received ash content of the coal entering the furnace from the unit operation data set; comprehensively process the total coal feed quantity, the received ash content of the coal entering the furnace, and the current bottom slag share to obtain the initial virtual slag quantity for the current time; merge the initial virtual slag quantity for the current time with the historical initial virtual slag quantities calculated and stored in previous consecutive calculations to form an initial virtual slag quantity sequence, and perform a first-order low-pass filter to obtain the virtual slag quantity entering the hopper dry slag discharger.

[0011] Furthermore, the specific construction method of the counter-current heat transfer discrete model is as follows: the counter-current heat transfer channel between the slag and cooling air in the slag dry discharge machine is discretized into multiple nodes along the slag flow direction; for each node, slag-side heat balance equation and air-side heat balance equation are established respectively, wherein the slag-side heat balance equation includes the axial variation term of the heat carried by the slag flow and the heat release term from the slag to the air, and the air-side heat balance equation includes the axial variation term of the heat absorbed by the air flow and the term of heat absorption from the slag; adjacent nodes are coupled through the slag flow temperature transfer and the air flow temperature transfer to form a counter-current heat transfer discrete model.

[0012] Further, the specific steps for iteratively inverting and outputting the bottom slag discharge temperature using the countercurrent heat transfer model are as follows: Obtain historical values ​​of the bottom slag discharge temperature that have converged in previous iterations, and take the moving average of these historical values ​​as the iteration starting point for the current estimated bottom slag discharge temperature; in each iteration, substitute the current estimated bottom slag discharge temperature into the countercurrent heat transfer discrete model to obtain the corresponding predicted cooling air outlet temperature, calculate the current deviation between this predicted value and the measured cooling air outlet temperature in the dry slag discharger measurement dataset; and apply a preset small perturbation to the current estimated bottom slag discharge temperature, determine the gradient based on the ratio of the change in the current deviation value before and after the perturbation to the amount of the small perturbation, and update the estimated bottom slag discharge temperature in the opposite direction of the gradient with a preset step size. ;

[0013] in, This is the estimated value of the current bottom slag discharge temperature in the (n+1)th iteration. This is the estimated value of the current bottom slag discharge temperature in the nth iteration. Let be the gradient of the nth iteration. Let n be the step size of the nth iteration; and With the absolute value of the current deviation Decrease and gradually shrink, adaptively adjusting as follows: ;in, Based on the step size factor, The value is a preset positive decimal; the iteration process is repeated until the current deviation value of the consecutive preset number of iterations is less than the preset convergence threshold, and the estimated value of the bottom slag discharge temperature of the last iteration is taken as the bottom slag discharge temperature.

[0014] Furthermore, the specific steps for solving the optimal cooling air volume as the optimization control command are as follows: Based on the measured value of the cooling air volume, generate multiple candidate cooling air volumes within the preset search range and at a preset step size; use the heat carried by the slag and the virtual slag volume as feedforward information, and combine each candidate cooling air volume to call the countercurrent heat transfer discrete model to predict the cooling air outlet temperature that can be achieved under the candidate cooling air volume.

[0015] Calculate the net waste heat recovery revenue corresponding to each candidate cooling air volume using the following formula. : ;in, The candidate cooling air volumetric flow rate, To use the counter-current heat transfer discrete model to predict the cooling air outlet temperature, To obtain the ambient temperature, It is the preset air density. The specific heat capacity of air is preset. This refers to the heat price coefficient stored in the database, which converts recovered heat into revenue. This represents the power consumption of the cooling fan at this candidate airflow rate, and , For pre-stored at rated air volume Power consumption at that time This refers to the electricity price coefficient;

[0016] Within the feasible region comprised of all candidate cooling air volumes that satisfy the preset safety thresholds for the hopper, the tail-end slag discharge, and the furnace oxygen prediction constraints, the selected cooling air volume is... The largest candidate cooling airflow is selected as the optimal cooling airflow.

[0017] Furthermore, the furnace oxygen prediction constraint is achieved as follows: Under stable boiler combustion conditions, a step disturbance is applied to the cooling air volume, and the response curve of the furnace outlet oxygen volume is recorded to identify a first-order transfer function model with pure time delay; the transfer function model is converted into a discrete difference equation form oxygen prediction model, using the measured furnace outlet oxygen volume in the dry ash remover measurement dataset as the current state, and the difference between the candidate cooling air volume and the current measured cooling air volume as the input increment to recursively calculate the oxygen prediction value sequence within a preset time period; the maximum value in the oxygen prediction value sequence is taken as the future oxygen peak corresponding to the candidate cooling air volume; a preset furnace outlet oxygen upper limit is obtained, and the maximum candidate cooling air volume that does not cause the future oxygen peak to exceed the furnace outlet oxygen upper limit is determined as the dynamic air volume upper limit, serving as the constraint boundary of the feasible region.

[0018] Further, the specific steps for generating the virtual temperature rise rate of the scale bucket surface are as follows: using the measured values ​​of cooling air volume and cooling air outlet temperature in the dry slag discharge machine measurement dataset, calculate the actual heat recovered by the cooling air; subtract the heat carried by the slag from the actual heat recovered by the cooling air to obtain the net heat load; perform temperature rise conversion processing between the net heat load and the preset equivalent heat capacity to obtain the virtual temperature rise rate of the scale bucket surface.

[0019] Further, the specific steps for outputting the safety control command are as follows: Obtain multi-point temperature measurement signals from the scale bucket bearing section of the dry slag discharger measurement dataset, and take the maximum temperature value of each measurement point as the measured value of the scale bucket surface temperature; perform sliding window fitting on the measured value of the scale bucket surface temperature to obtain the measured temperature rise rate and determine the fitting confidence level; compare the fitting confidence level with a preset confidence threshold, and determine the current judgment criterion based on the comparison result; if the current judgment criterion exceeds the preset warning threshold but does not exceed the preset safety threshold, output a preventive air increase command; if the current judgment criterion exceeds the preset safety threshold, output a safety air volume command; combine the preventive air increase command and the safety air volume command to constitute the safety control command.

[0020] Furthermore, the specific steps for determining the current judgment criteria based on the comparison results are as follows:

[0021] When the confidence level of the fit is higher than the preset confidence threshold, the measured temperature rise rate is determined to be reliable and is used as the current judgment basis. The difference between the measured temperature rise rate and the virtual temperature rise rate is calculated as the bias and the virtual temperature rise rate is corrected. When the confidence level of the fit is lower than the preset confidence threshold, the measured temperature rise rate is determined to be unreliable and the corrected virtual temperature rise rate is used as the current judgment basis.

[0022] Furthermore, it also includes calibrating the current bottom slag share: maintaining a first-in-first-out timestamp queue, recording the virtual slag volume and the current bottom slag share for each time period with a preset time granularity; acquiring the real-time speed of the bucket conveyor belt and the preset fixed length of the carrying section, and determining the slag discharge time corresponding to the current slag being discharged into the tail; retrieving the record whose timestamp is closest to the slag discharge time from the timestamp queue, extracting the virtual slag volume in the record, and using it as predictive data causally aligned with the measured value of the tail slag discharge volume in the current dry slag discharger measurement dataset; performing causal calibration processing on the predictive data in the preset calibration window after causal alignment with the measured value of the tail slag discharge volume to obtain the calibrated bottom slag share, and feeding back to update the current bottom slag share.

[0023] The present invention has the following beneficial effects:

[0024] (1) The intelligent control and waste heat recovery system for the cooling air of the scale bucket dry slag discharge machine incorporates the coal feeder's operating status signal and instantaneous coal quantity into the calculation link through the virtual slag quantity calculation unit. It only accumulates the coal quantity data of the coal feeder during operation and integrates the ash content of the coal entering the furnace and the dynamically calibrated bottom slag ratio. The virtual slag quantity entering the dry slag discharge machine is obtained before the material falls into the slag discharge machine. Combined with the slag temperature inversion unit, which uses the measured value of the cooling air outlet temperature as a reference, the system iteratively approximates the bottom slag discharge temperature using a countercurrent heat transfer discrete model. The combination of the two makes the heat carried by the slag more efficient. The acquisition is no longer limited by the long time delay characteristics of the tail slag discharge temperature measurement point. The optimized control unit uses the heat information and virtual slag volume obtained in advance as feedforward to search for candidate air volume near the current measured value of cooling air volume. Within the feasible region defined by the constraints of safe temperature and furnace oxygen content prediction, the optimal output is selected based on the criterion of maximizing net benefit. The decision point of air volume adjustment is shifted from after receiving the tail temperature signal to the same time period when the slag discharge condition changes, thereby providing greater adjustment space for waste heat recovery while maintaining the safety margin of equipment operation.

[0025] (2) The intelligent control and waste heat recovery system for cooling air of the scale bucket dry slag discharge machine, through the safety control unit, compares the difference between the heat carried by the slag and the actual heat recovered by the cooling air to obtain the net heat load currently accumulated on the scale bucket body and the chain of the bearing section. Then, it is combined with the equivalent heat capacity to convert it into the virtual temperature rise rate of the scale bucket surface. This virtual temperature rise rate is generated independently of the infrared temperature measurement signal. It can reflect the heat imbalance trend when the temperature measurement point on the scale bucket surface has not changed significantly due to thermal inertia. At the same time, the confidence of the linear fitting result of the infrared measured temperature sliding window is evaluated. Only when the determination coefficient meets the threshold condition is the measured temperature rise rate used as the current judgment basis. Otherwise, the corrected virtual temperature rise rate is used instead. The two signals are backups of each other and are kept aligned under the correction mechanism. This avoids false triggering and missed triggering caused by infrared signal jump or short-term failure in high temperature dust environment. This ensures that the safety intervention action of the cooling air volume is neither excessive nor sluggish, and provides a reliable guarantee for the continuous operation of the optimization control unit within the safety boundary.

[0026] (3) The intelligent control and waste heat recovery system for the cooling air of the scale bucket dry slag discharge machine optimizes the control unit, takes the slag quantity and the heat carried by the slag material as the feedforward basis, and solves the optimal cooling air volume with the goal of maximizing the net benefit of waste heat recovery within the range of multiple safety boundary constraints. It accurately matches the optimal cooling air flow corresponding to the working condition. It also corrects the calculation error by calibrating the bottom slag share time sequence and aligning the working condition causality, so that the virtual slag quantity and heat exchange calculation results fit the actual working conditions on site, ensuring that the air volume optimization results are continuously accurate. Thus, under the premise of ensuring equipment safety and furnace stability, it reasonably reduces redundant cooling air volume, reduces the ineffective energy consumption of the blower, and maximizes the utilization of dry slag waste heat. In this way, the cooling air control continuously fits the actual heat exchange needs in the dynamically changing operating environment, taking into account both equipment safety and waste heat recovery.

[0027] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0028] Figure 1 This is a block diagram of an intelligent control system for cooling air and waste heat recovery for a scale bucket dry slag discharge machine according to the present invention.

[0029] Figure 2 This is a flowchart illustrating the specific steps involved in the intelligent control and waste heat recovery system for the cooling air of a scale bucket dry slag discharge machine according to the present invention, which iteratively inverts the output of the bottom slag discharge temperature using a countercurrent heat exchange model.

[0030] Figure 3 This is a flowchart illustrating the steps for implementing the furnace oxygen content prediction constraint in the intelligent control of cooling air and waste heat recovery system of a scale bucket dry slag discharge machine according to the present invention. Detailed Implementation

[0031] Please see Figure 1 This invention provides a technical solution: an intelligent control system for cooling air and waste heat recovery of a hopper dry ash discharger, comprising: a data acquisition unit, used to acquire the unit operation dataset and the dry ash discharger measurement dataset of the hopper dry ash discharger, as well as the current bottom ash share. The unit operation dataset includes boiler operating parameters and coal quality parameters. The dry ash discharger measurement dataset includes measured values ​​of cooling air outlet temperature, cooling air volume, tail ash discharge volume, hopper surface temperature, and furnace outlet oxygen content. The initial value of the current bottom ash share is a preset value, which is obtained by calibration and feedback during subsequent operation.

[0032] The virtual slag volume calculation unit is used to calculate the virtual slag volume entering the scale bucket dry slag discharger based on the unit operation dataset and the current bottom slag share.

[0033] The slag temperature inversion unit is used to iteratively invert the output of the bottom slag discharge temperature based on the virtual slag quantity and the dry slag discharger measurement dataset through a countercurrent heat transfer model, and to calculate the heat carried by the slag to be recovered by combining the virtual slag quantity.

[0034] The optimization control unit is used to generate candidate cooling air volume as decision variables based on the measured value of cooling air volume in the dry slag discharger measurement dataset, with the heat carried by slag and virtual slag volume as feedforwards. Under the constraints of preset scale bucket safety temperature threshold, tail slag discharge safety temperature threshold and furnace oxygen prediction constraints, the optimal cooling air volume is solved as the optimization control command with the goal of maximizing the net benefit of waste heat recovery.

[0035] The safety control unit is used to perform heat balance processing based on the heat carried by the slag and the measurement data of the dry slag discharger, generate a virtual temperature rise rate on the surface of the slag bucket, and output a safety control command after combining the measurement data of the slag discharger with confidence assessment and hierarchical protection.

[0036] The instruction coordination output unit is used to coordinate safety control instructions over economic control instructions to generate the final cooling airflow control instruction, so as to maximize the recovery of waste heat from dry slag within the safety boundary. Specifically:

[0037] ;

[0038] in, To preset a safe airflow, This is the preset offset airflow.

[0039] The specific steps for calculating the heat carried by the slag to be recovered are as follows:

[0040] Obtain ambient temperature And the preset specific heat capacity of slag And by combining the virtual slag volume with the slag discharge temperature at the bottom of the furnace, the heat carried by the slag is calculated comprehensively:

[0041] ;

[0042] in, For virtual waste quantity, This is the inverted bottom slag discharge temperature.

[0043] Specifically, the steps for calculating the virtual slag volume entering the hopper dry slag discharger are as follows:

[0044] The operating status signal and instantaneous coal quantity of each coal feeder are obtained from the unit operation data set. Only the instantaneous coal quantity of the coal feeders that are in operation is accumulated to obtain the total coal feed, which is as follows:

[0045] Suppose there are N coal feeders in total, and the operating status signal of the u-th coal feeder is: A value of 1 indicates operation, and 0 indicates shutdown. The instantaneous coal feeder capacity is... ( In each calculation cycle (e.g., 1 second), all coal feeders are traversed to calculate the total coal feed rate. The formula is:

[0046] ;

[0047] The received ash content of the coal is obtained from the unit operation data set. The received ash content of the coal remains unchanged from the most recently entered value or online measurement value during periods when the coal type is not changed. Upon detecting a coal type change signal, it is automatically updated to the ash content value corresponding to the new coal type. Specifically:

[0048] Read the base ash content of the coal received from the database or the operator's input interface. Logically, this value is maintained as the most recent valid value: if no coal type switching signal is detected, the most recent valid value (such as the test value from the previous shift or the online ash analyzer reading from the previous moment) is always used; once the control logic detects a coal type switching signal (such as a "coal feeding completed" command issued by the operator or a change signal in the pulverizing system combination), an update is immediately triggered, and the value is changed accordingly. Update the test values ​​to match the new coal type to ensure that the coal quality data is consistent with the current combustion conditions;

[0049] The total coal feed rate, the ash content of the coal received into the furnace, and the current bottom ash content are comprehensively processed to obtain the initial virtual ash quantity for the current batch, which is as follows:

[0050] The initial virtual waste amount of the current time Calculated by the following formula:

[0051] ;

[0052] in, This represents the total coal feed for the current operation. The ash content of the currently effective coal received into the furnace is based on... The current bottom ash share represents the mass proportion of the total ash from the coal fed into the furnace that is discharged in the form of bottom ash.

[0053] The current initial virtual slag quantity is merged with the previously calculated and stored historical initial virtual slag quantities to form an initial virtual slag quantity sequence. This sequence is then subjected to a first-order low-pass filter to obtain the virtual slag quantity entering the hopper dry slag discharger. Specifically:

[0054] Maintain a first-in-first-out queue of length L (e.g., 10) in memory to store the most recent L cycles (including the current one). Values ​​are used to form an initial virtual slag quantity sequence, and the filtered virtual slag quantity is... The algorithm is given by a first-order low-pass filter:

[0055] ;

[0056] in, The filter coefficients are based on the time constant. (e.g., 10 seconds) and calculation cycle (e.g., 1 second) Calculated, , This is the virtual slag quantity after the previous filtering, which aims to eliminate high-frequency spurious fluctuations caused by the instantaneous coal drop impact of the coal feeder and measurement noise.

[0057] In this implementation plan, the effective coal quantity is screened by the operation signal to complete the total quantity statistics. Combined with the ash content data updated synchronously with the coal type switching, and the bottom ash share, the real-time ash quantity is initially calculated. The instantaneous calculation results are smoothed by time-series queue and first-order filtering to reduce the data fluctuations caused by coal fluctuations. The system continuously outputs stable and reliable ash quantity values, providing stable front-end input for subsequent heat exchange model inversion and air volume optimization, and eliminating the control deviation caused by the lag in tail ash discharge measurement.

[0058] Specifically, the specific construction method of the countercurrent heat transfer discrete model is as follows:

[0059] The heat exchange channel between the slag and cooling air in the scale bucket dry slag discharge machine is discretized into multiple nodes along the slag flow direction, specifically as follows:

[0060] Total length of effective heat exchange section The slag is divided into M nodes at equal intervals along the slag's direction of travel. Node indices j range from 0 to M, where j=0 represents the hot slag inlet (slag discharge port) and j=M represents the cold slag outlet (slag discharge port). Cooling air flows in the opposite direction to the slag (counter-current), from j=M to j=0. The distance between adjacent nodes... ;

[0061] In one embodiment, M can be 5 to balance model accuracy with controller computational load;

[0062] For each node, slag-side heat balance equations and air-side heat balance equations are established. The slag-side heat balance equation includes an axial variation term for the heat carried by the slag flow and a term for heat release from the slag to the air. The air-side heat balance equation includes an axial variation term for the heat absorbed by the air flow and a term for heat absorption from the slag. Specifically:

[0063] Using a first-order explicit discretization scheme, for internal nodes ,exist The temperature at time k is calculated from the parameters at time k.

[0064] Slag side temperature The update equation is:

[0065]

[0066] ;

[0067] in, To calculate the time step ( ), It is the transport time of slag between nodes ( ), It is the transport speed of slag ( The speed of the scale belt is determined by the speed of the scale belt. It is the product of the total heat transfer coefficient of node j and the heat transfer area. ), It is virtual waste quantity ( ), It has the specific heat capacity of slag. The second term on the right side of the equation represents the heat flowing into the upstream node, and the third term represents the heat released to the air.

[0068] Air side temperature The update equation is:

[0069]

[0070] ;

[0071] in, It is the transport time of air between nodes. ), It is the air velocity ( ), determined by cooling airflow The cross-sectional area of ​​the flow path is calculated. It is the mass flow rate of cooling air ( ), It is the specific heat capacity of air. The second term on the right side of the equation is the heat flowing into the downstream node (due to countercurrent), and the third term is the heat absorbed from the slag side.

[0072] Adjacent nodes are coupled through slag flow temperature transfer and air flow temperature transfer, forming a counter-current heat transfer discrete model that describes the distribution of slag and air temperatures along the flow path at each node. Specifically:

[0073] The coupling between the slag-side equations and the air-side equations is reflected in the heat transfer terms in their respective equations. The heat release on the slag side is equal to the heat absorption on the air side, forming a closed loop of heat transfer. By solving this set of simultaneous discrete equations node by node, the temperature distribution of slag and air at all nodes at any time can be obtained.

[0074] The countercurrent heat transfer discrete model takes the measured values ​​of virtual slag volume and cooling air volume as input parameters, uses the bottom slag discharge temperature as the slag-side inlet boundary condition, and the ambient temperature as the air-side inlet boundary condition. The output is a predicted value for the cooling air outlet temperature, specifically:

[0075] The boundary conditions of the model are set as follows:

[0076] ;

[0077] ;

[0078] in, This is an estimated value for the slag discharge temperature at the bottom of the furnace. The ambient temperature is measured at the inlet of the blower.

[0079] Input virtual waste quantity and cooling air volume Then, the model starts from the boundary conditions, iterates until steady state or meets the preset time, and finally outputs the air outlet ( Temperature of nodes As a predicted value of cooling air outlet temperature .

[0080] like Figure 2 As shown, the specific steps for iteratively inverting and outputting the bottom slag discharge temperature using a countercurrent heat transfer model are as follows:

[0081] Obtain historical values ​​of the bottom ash discharge temperature that have converged in previous iterations, and take the moving average of these historical values ​​as the iteration starting point for the current iteration's estimated bottom ash discharge temperature. If there are no historical converged values, then the boiler design ash discharge temperature is used as the iteration starting point, specifically:

[0082] At the start of each new inversion round, the historical convergence value queue is checked. If a recent convergence value exists (e.g., within the previous 5 minutes), the moving average of these values ​​is calculated as the starting point for the iteration. ;

[0083] If the system has just been put into operation or has not converged for a long time (historical queue is empty), the boiler design slag discharge temperature (e.g., 850℃) will be used as the default starting point.

[0084] In each iteration, the current estimated bottom slag discharge temperature is substituted into the countercurrent heat transfer discrete model as the slag-side inlet boundary condition. The model is run to obtain the corresponding predicted cooling air outlet temperature. The current deviation between this predicted value and the measured cooling air outlet temperature in the dry slag discharger measurement dataset is calculated, specifically as follows:

[0085] Let the current iteration be the nth iteration, and the estimated bottom slag discharge temperature be... Use it as a boundary condition Input a discrete countercurrent heat transfer model, and the model outputs predicted values. The current deviation value is defined as the difference between the measured value and the predicted value:

[0086] ;

[0087] A preset small perturbation is applied to the current estimated bottom slag discharge temperature, and the model is run again to obtain the predicted cooling air outlet temperature after the perturbation. The gradient is determined based on the ratio of the change in the current deviation value before and after the perturbation to the amount of the small perturbation. The estimated bottom slag discharge temperature is updated in the opposite direction of the gradient with a preset step size. The preset step size gradually decreases as the absolute value of the current deviation value decreases. Specifically:

[0088] For the current estimate Apply a small perturbation (e.g., +1℃), to obtain Run the model again to obtain the perturbed predicted values. and deviation value The numerical gradient can then be approximated as:

[0089] ;

[0090] Updated estimate Adjust in the opposite direction of the gradient:

[0091] ;

[0092] Among them, step size factor With the absolute value of the deviation Decrease and gradually shrink, adaptively adjusting as follows:

[0093] ;

[0094] in, The base step size factor typically ranges from 50 to 200. This is a preset positive decimal number, ranging from 0.1 to 1.0, used to prevent the denominator from being zero. This mechanism ensures that when the deviation is large, the step size is larger for rapid convergence; when the deviation is small, the step size is smaller for fine-grained searching, avoiding oscillations near the convergence point.

[0095] During the iteration process, if the updated estimated value Exceeding the preset temperature limit Then force it to be set to If it is lower than the preset lower temperature limit Then force it to be set to , The maximum slag discharge temperature corresponding to the boiler under full load conditions is usually taken as 900℃ to 950℃; The minimum slag discharge temperature corresponding to the boiler's lowest load condition is usually taken as 700℃ to 750℃;

[0096] The iterative process is repeated until the current deviation value of each of the preset number of iterations is less than the preset convergence threshold. The estimated bottom slag discharge temperature of the last iteration is then used as the bottom slag discharge temperature for online inversion. Specifically:

[0097] Repeat the iterative process of perturbation, prediction, deviation calculation, and state update. To prevent spurious convergence caused by noise from a single measurement, a continuous convergence criterion is adopted: convergence is achieved only when the deviation value obtained from multiple consecutive iterations (e.g., 3 times) is... All are less than the preset convergence threshold The iteration is considered to have converged only when the temperature reaches 0.5℃ (e.g., 0.5℃).

[0098] After convergence, the estimated bottom ash discharge temperature from the last iteration is used as the online inverted bottom ash discharge temperature. Output the value and store it in the historical convergence value queue for use in calculating the iteration starting point in the next inversion. If the number of iterations exceeds the preset maximum number of iterations and still fails to converge, the bottom slag discharge temperature of the previous inversion will remain unchanged.

[0099] In this implementation plan, a countercurrent heat exchange node model is built in segments to reduce the entire process of countercurrent heat exchange between slag and air. The internal heat transfer process of the equipment is fully reproduced by relying on the two-way heat balance equation. The heat exchange outlet temperature is output with boundary conditions, and the iteration starting point is determined by relying on historical convergence data. The gradient is obtained by using numerical perturbation to adaptively update the slag temperature estimation value. With temperature range clamping and continuous convergence judgment, the calculation fluctuation caused by measurement noise is reduced, and the actual furnace bottom slag discharge temperature is accurately reproduced.

[0100] Specifically, the steps for determining the optimal cooling airflow as the optimization control command are as follows:

[0101] Based on the measured cooling airflow value, multiple candidate cooling airflow values ​​are generated within a preset search range at a preset step size, specifically as follows:

[0102] Define the current measured cooling airflow as The lower limit of the search range is the minimum allowable air volume. (Determined by the constraint of preventing flue gas leakage, that is, under the rated negative pressure condition of the boiler, the cooling air volume is gradually reduced, and when the signs of flue gas leakage begin to appear at the ash discharge port or observation hole of the dry ash discharge machine, the cooling air volume at this time is recorded as the critical leakage air volume.) ,Pick ,in For safety factors, the value ranges from 1.1 to 1.3, with the upper limit being the maximum physical air volume of the fan. Within this range Within, with a preset step size (e.g., 0.5 Nm³ / s) uniformly generated Candidate cooling air volume ,in ;

[0103] Using the heat carried by the slag and the virtual slag volume as feedforward information, and combining them with each candidate cooling airflow, the counter-current heat transfer discrete model is invoked to predict the cooling air outlet temperature achievable under that candidate cooling airflow. Specifically:

[0104] For each candidate air volume Compare it with the current virtual waste volume Ambient temperature As input, the countercurrent heat transfer discrete model is invoked (using the latest inverted model). (Based on the boundary conditions at the slag side inlet), a steady-state prediction calculation is performed.

[0105] The model will output the result under this operating condition, when the cooling airflow becomes At that time, the new predicted value of the cooling air outlet temperature will be reached. ;

[0106] Meanwhile, the countercurrent heat transfer discrete model provides the slag temperature at each node. The maximum value of the slag temperature at each node is taken as the predicted value of the scale surface temperature. ,Pick Slag temperature at the node As the predicted value of tail-end slag discharge temperature ;

[0107] when Exceeding the preset safe temperature threshold for the scales, or When the preset tail discharge slag safety temperature threshold is exceeded, the candidate cooling air volume is judged to be in violation of safety constraints.

[0108] Calculate the net waste heat recovery benefit for each candidate cooling airflow using the following formula:

[0109] ;

[0110] in, Candidate cooling air volume flow rate ( ), To use the counter-current heat transfer discrete model to predict the cooling air outlet temperature, For ambient temperature, It is air density ( ), The specific heat capacity of air ( ), The heat price factor for converting recovered heat into economic benefits ( ), The power consumption of the cooling fan at this candidate airflow rate ( ), Electricity price coefficient ( The two terms in the formula calculate the instantaneous net profit on the same time scale;

[0111] The current standard coal unit price can be obtained in the following ways: by reading the data entered by operators or provided in the power plant information. (yuan / kg) and boiler thermal efficiency Calculate automatically using the following formula:

[0112] ;

[0113] In the formula, The lower heating value of standard coal (kJ / kg) is a preset constant, taken as 29307 kJ / kg according to national standards. It will be updated when the price of standard coal or boiler operating parameters change. and The value will be automatically recalculated. ;

[0114] It can be obtained in the following ways:

[0115] Read the current on-grid electricity price or plant power unit price entered by operators or provided by the power plant information. (Yuan / kWh), automatically calculated using the following formula:

[0116] ;

[0117] Convert the unit from "yuan / kWh" to "yuan / kJ" to maintain consistency with the dimensions of the revenue item. Update the unit when the electricity price is adjusted. Get the value and update automatically. ;

[0118] The first item of the net revenue from waste heat recovery represents the revenue from coal combustion replaced by cooling air supplied to the boiler furnace, and the second item represents the cost of electricity consumed by the cooling fan. The difference between the two reflects the economic trade-off between the increase in recovered heat and the increase in fan power consumption when the cooling air volume increases.

[0119] Wind power consumption The air volume can be represented as a cubic function of the fan's characteristic curve:

[0120] ;

[0121] Rated air volume Reduce power consumption and increase airflow While it can recover more heat and increase revenue, it also increases the power consumption of the wind turbine. Cubic growth increases costs, and the optimal airflow should maximize the difference between the two.

[0122] Within the feasible region comprised of all candidate cooling air volumes that satisfy the preset safe temperature threshold for the hopper, the preset safe temperature threshold for tail-end slag discharge, and the furnace oxygen prediction constraint, the selected cooling air volume is... The largest candidate cooling airflow is selected as the optimal cooling airflow, specifically:

[0123] From all candidate air volumes, those options that violate safety constraints are eliminated. The criteria for determining the safety temperature constraints of the hopper and the tail discharge safety temperature are as follows: if the predicted surface temperature of the hopper obtained in the preceding steps... Exceeding the preset safe temperature threshold of the hopper, or the predicted temperature of the tail discharge slag. If the preset tail discharge slag safety temperature threshold is exceeded, the candidate air volume is determined to violate safety constraints.

[0124] The criterion for determining oxygen constraints is: if the candidate air volume corresponds to a future peak oxygen level... Exceeding the upper limit of oxygen content at the furnace outlet If the candidate airflow violates the oxygen constraint, the retained candidate airflow constitutes the feasible region. Calculate the feasible region one by one. The objective function value corresponding to each candidate air volume Take the middle one The largest candidate airflow is determined as the optimal cooling airflow. :

[0125] ;

[0126] If none of the candidate cooling airflows meet the constraints, the candidate cooling airflow closest to the safety constraint on the feasible region boundary is selected as the optimal cooling airflow to ensure safety is prioritized. Specifically:

[0127] Under extreme operating conditions, if all candidate air volumes If at least one constraint is violated, i.e. the feasible domain is empty, then the economic optimization objective temporarily gives way to safety.

[0128] In practice, this means directly selecting the maximum physical air volume of the fan. This is because the airflow represents the maximum cooling capacity that the system can provide, and can prioritize safety when the equipment is at risk of overheating (in which case a short-term breach of oxygen constraints is allowed).

[0129] like Figure 3 As shown, the furnace oxygen content prediction constraint is achieved in the following way:

[0130] Under stable boiler combustion conditions, a step disturbance is applied to the cooling air volume, and the response curve of the oxygen quantity at the furnace outlet is recorded. The parameters of the first-order transfer function model with pure time delay are identified, including static gain, time constant, and time delay, which are as follows:

[0131] When the boiler load, coal quantity and other combustion conditions are stable, a step disturbance of the cooling damper or frequency converter is triggered automatically or manually (such as rapidly increasing the opening by 10%).

[0132] Simultaneously record the increase in cooling air volume and furnace outlet oxygen content Based on the response data, a first-order transfer function model with pure time delay is fitted using identification algorithms such as least squares.

[0133] ;

[0134] in, This is static gain (representing the final change in oxygen content caused by a unit change in air volume, such as...). ), It is the time constant ( ), It is pure time delay ( This set of parameters describes the dynamic relationship between changes in cooling airflow and the oxygen response in the furnace.

[0135] The transfer function model is transformed into a discrete difference equation form for oxygen prediction. Using the measured oxygen content at the furnace outlet from the dry ash removal machine's measurement dataset as the current state, and the difference between the candidate cooling air volume and the current measured cooling air volume as the input increment, a sequence of predicted oxygen values ​​for a future preset time period is recursively calculated. Specifically:

[0136] Set the number of lag steps Based on the current measured cooling airflow. With candidate air volume Obtain step increment Define discrete time k as the current time, and the measured oxygen content as... ,make For the m-th step prediction ( The recursive formula is:

[0137] ;

[0138] in, , , Input items: (Oxygen level baseline at the current moment) Values ​​are determined based on lag relationships:

[0139] ;

[0140] This yields the oxygen content prediction sequence. ;

[0141] The maximum value in the predicted oxygen content sequence is taken as the future peak oxygen content corresponding to the candidate cooling airflow. Take the maximum value of the corresponding predicted sequence. This is the maximum oxygen shock that the candidate air volume may cause.

[0142] The preset upper limit of oxygen content at the furnace outlet is obtained. The maximum candidate cooling air volume that will not cause the future oxygen peak to exceed the upper limit of oxygen content at the furnace outlet is determined as the dynamic air volume upper limit for the current control cycle, and serves as one of the constraint boundaries of the feasible region. Specifically, it is as follows:

[0143] Read the upper limit of oxygen content at the furnace outlet set by the operator. (e.g., 3.5%), perform a one-dimensional search among all candidate air volumes to find the one that meets the condition. Maximum air volume ;

[0144] This value This is the dynamic airflow safety limit for the current control cycle. In subsequent economic optimization, all... All candidate air volumes were deemed to violate oxygen constraints and were eliminated, resulting in an upper limit for the effective air volume in the feasible region. ;

[0145] If the future peak oxygen level corresponding to a candidate cooling airflow exceeds the upper limit of the oxygen level at the furnace outlet, then that candidate cooling airflow is excluded from the feasible region, specifically as follows:

[0146] If a candidate air volume Calculated future peak oxygen levels If the candidate air volume is not feasible, it will be directly judged as "infeasible" and removed from the feasible region of optimization, and will not participate in the net profit. A comparison.

[0147] In this implementation plan, multiple sets of cooling air volumes are generated in batches to complete full-condition traversal calculations. Based on the heat exchange model, the changes in equipment temperature and furnace oxygen content corresponding to each air volume are predicted. The waste heat gain and fan energy consumption corresponding to each set of air volumes are calculated simultaneously. The effective air volume range is selected by taking into account the equipment and furnace operation requirements. The air volume with the highest benefit is selected for output. When extreme risks occur in the operating conditions, the system automatically switches to the maximum cooling air volume allowed by the equipment. The entire process takes into account equipment protection and combustion stability, and simultaneously improves waste heat recovery and reduces fan energy consumption.

[0148] Specifically, the steps for generating the virtual temperature rise rate on the surface of the scales are as follows:

[0149] Using the measured values ​​of cooling air volume and cooling air outlet temperature from the dry slag remover measurement dataset, the actual heat recovered by the cooling air is calculated, specifically as follows:

[0150] The actual heat power recovered by the cooling air ( The results are calculated from air volume, temperature rise, and physical property parameters.

[0151] ;

[0152] in, It is the air density (taken as 1.2 kg / m³). This is the actual measured value of cooling airflow. It is the actual measured cooling air outlet temperature. This is the ambient temperature, which represents the heat that has been effectively recovered and sent into the boiler furnace through the cooling air;

[0153] The difference between the heat carried by the slag and the actual heat recovered by the cooling air is used to obtain the net heat load borne by the scale bucket body and the supporting chain section at the current moment, which is as follows:

[0154] Net heat load ( The power is defined as the portion of the total heat released by the slag that is not carried away by the cooling air in time and is temporarily accumulated on the body of the slag bucket equipment.

[0155] ;

[0156] in, It is the total heat power carried by the slag as it cools from its initial temperature to ambient temperature;

[0157] when When the heat released by the slag exceeds the heat carried away by the cooling air, the scale bucket body and the chain bearing section are absorbing the excess heat and heating up; when When the cooling air carries away more heat than the slag releases, the scale bucket body and the chain supporting the load are releasing stored heat and cooling down; when When the time is right, it indicates that the heat balance is achieved;

[0158] The net heat load is converted into the equivalent heat capacity of the scale body and the supporting chain section by temperature rise calculation to obtain the virtual temperature rise rate of the scale surface, which is as follows:

[0159] Virtual temperature rise rate of scale surface ( The calculation is as follows:

[0160] ;

[0161] in, ( ) is the equivalent heat capacity of the scale body and the supporting chain section. This is a comprehensive parameter calibrated through design data and on-site temperature rise test, which represents the amount of heat that the entire equipment needs to absorb to increase its temperature by 1°C.

[0162] This virtual rate directly reflects the severity of the current heat imbalance and is generated independently of the measured surface temperature of the sluice box. This is important when the measured surface temperature of the sluice box has not yet changed significantly due to thermal inertia. It can already reflect the trend and extent of heat imbalance in advance.

[0163] The specific steps for outputting safety control commands are as follows:

[0164] Multi-point temperature measurement signals of the scale bucket bearing section were obtained from the measurement data of the dry slag discharge machine. The maximum temperature value of each measurement point was taken as the measured value of the scale bucket surface temperature. Specifically:

[0165] Real-time reading of signals from at least three infrared thermometers or thermocouples installed at different locations on the scale bucket bearing section. The maximum value is taken as the measured value of the scale surface temperature and used for subsequent protection logic:

[0166] ;

[0167] Selecting the maximum value ensures that the protection logic is triggered based on the worst-case scenario, preventing local overheating from being missed.

[0168] A sliding window fitting was performed on the measured surface temperature of the sac to obtain the measured temperature rise rate, and the fitting confidence level was determined, as follows:

[0169] Maintain a memory with a length of A sliding window (e.g., 50 sampling points, corresponding to 5 seconds of data) stores the most recent data. The sequence performs a least-squares linear fit on the data within the window every preset number of milliseconds (e.g., 200ms). ,in Calculate the coefficient of determination of the fit in minutes. As a confidence level for the fit:

[0170] ;

[0171] in, This represents the w-th measured temperature value within the sliding window. Let w be the temperature value at the w-th sampling point on the least squares fitted line. Within the sliding window The arithmetic mean of the measured temperature values;

[0172] The value range is from 0 to 1. The closer the value is to 1, the better the linear fit of the temperature data points, and the higher the reliability of the measured temperature rise rate. The closer the value is to 0, the more dispersed the temperature data points are, the less reliable the linear fitting is, and the measured temperature rise rate is severely affected by noise.

[0173] The fit confidence level is compared with a preset confidence threshold, and the current judgment criterion is determined based on the comparison result;

[0174] Set warning thresholds and safety thresholds. The safety threshold is higher than the warning threshold. Specifically, the judgment is based on a set of thresholds, regardless of whether the temperature rise rate is virtual or measured. (Set warning thresholds.) (e.g., 4.0℃ / min) and safety threshold (e.g., 6.0℃ / min), and ;

[0175] If the current judgment exceeds the preset warning threshold but does not exceed the preset safety threshold, a preventive wind increase command will be output, specifically:

[0176] When the rate value used for the current determination satisfy If the system determines that the equipment temperature is rising too quickly but is not yet in a dangerous state, it will output a "preventive airflow increase command" to adjust the final control airflow to the optimized command. Slightly increase from the original level:

[0177] ;

[0178] in, A preset safety bias airflow (e.g., 0.5 Nm³ / s) is used to actively increase cooling capacity, suppressing the temperature rise trend in its early stages.

[0179] If the current judgment criterion exceeds the preset safety threshold, a forced safety airflow command will be output, specifically as follows:

[0180] when At this point, the system determines that the equipment is at risk of overheating. The optimized control command is immediately bypassed, and a "forced safe airflow command" is output, setting the control airflow to a preset high safety value. (e.g., 80% of the fan's rated air volume):

[0181] ;

[0182] This command has the highest priority, ensuring that the device quickly returns to a safe state;

[0183] The preventive ventilation increase command and the mandatory safe air volume command together constitute the safety control command, which is as follows:

[0184] The two types of commands triggered based on different levels of temperature rise rate, with safety as the primary objective, are collectively referred to as safety control commands. In the final output arbitration stage of the control system, safety control commands have higher priority than optimization control commands.

[0185] The specific steps for determining the current judgment criteria based on the comparison results are as follows:

[0186] When the confidence level of the fit is higher than the preset confidence threshold, the measured temperature rise rate is determined to be reliable, and the measured temperature rise rate is used as the current determination basis.

[0187] Simultaneously, the difference between the measured temperature rise rate and the virtual temperature rise rate is calculated as a bias, and the virtual temperature rise rate is corrected to ensure that the corrected virtual temperature rise rate has higher accuracy when used as a judgment criterion when the measured temperature rise rate is unreliable. Specifically:

[0188] like The current measured temperature rise rate Trustworthy, and used as the current criterion for entering the layered protection logic. ;

[0189] Simultaneously, the difference between the measured temperature rise rate and the virtual temperature rise rate is calculated as a bias, and the virtual temperature rise rate is corrected to ensure that the corrected virtual temperature rise rate has higher accuracy when used as a judgment criterion when the measured temperature rise rate is unreliable. Specifically:

[0190] Utilizing this "golden moment" when both are reliable, calculate a real-time bias. Calibrate the virtual model online:

[0191] ;

[0192] Corrected virtual temperature rise rate for:

[0193] ;

[0194] This bias It includes all steady-state errors caused by model simplification (such as inaccurate heat capacity) and unmodeled factors (such as changes in environmental heat dissipation). The significance of the correction is that the current measurement is reliable, so it is used to make the judgment; but the next measurement may become unreliable due to dust obstruction. At this time, the virtual rate has been corrected, the deviation is smaller, and the replacement judgment is more accurate, forming an interactive correction closed loop of "current calibration, future benefit".

[0195] When the fitted confidence level is lower than the preset confidence threshold, the measured temperature rise rate is determined to be unreliable. The corrected virtual temperature rise rate is then used as the current criterion, while the bias remains unchanged. Specifically:

[0196] like If the measured data is deemed unreliable (e.g., due to transient contamination of the infrared lens, water mist obstruction, or signal jumps at the measurement point), the judgment criterion should be immediately switched to the calibrated virtual rate.

[0197] ;

[0198] Furthermore, the bias value is stopped being updated during this period. The value of the last measured reliable time remains unchanged.

[0199] In this implementation plan, the virtual temperature rise rate of the hopper is calculated in real time based on the principle of equipment thermal balance. This allows for the early detection of changes in the equipment's thermal imbalance, compensating for the lag in physical temperature detection. By fitting and analyzing the measured temperature data and determining the confidence level, the plan enables intelligent switching between measured and virtual temperature rise. Simultaneously, the virtual model parameters are continuously corrected based on effective operating conditions. Furthermore, a graded airflow control logic is set up, matching corresponding protection commands according to the severity of the temperature rise trend. Even when the detection signal is abnormally affected by the on-site environment, the plan can still maintain reliable judgment capabilities, always prioritizing the stable operation of the equipment and avoiding overheating risks in advance.

[0200] Specifically, this also includes calibrating the current bottom ash content:

[0201] Maintain a first-in-first-out timestamp queue to record the virtual slag quantity and current bottom slag share for each time period with a preset time granularity. Specifically:

[0202] Establish a circular buffer (FIFO queue) with a queue depth sufficient to cover the longest material transport time (e.g., 2 hours, corresponding to 120 records), using a preset fixed record time granularity. (e.g., 60s) Every hour at this granularity, the system will calculate the average virtual slag volume over the past minute. The proportion of bottom ash used when calculating this ash volume. The data, along with the coal feed rate change indicator (stable or abrupt), is packaged into a single data record, along with the current timestamp. Push it to the end of the queue;

[0203] Once the queue length is exceeded, the oldest record at the head of the queue is automatically discarded to ensure that the queue data always reflects the recent operating conditions.

[0204] The real-time speed of the bucket conveyor belt and the preset fixed length of the carrying section are obtained, and the transport time of the material from the slag discharge port to the slag unloading port is calculated. The slag discharge time corresponding to the current slag being unloaded at the tail end is obtained by subtracting the transport time from the current time. Specifically:

[0205] Read the speed encoder signal of the bucket conveyor belt to obtain the belt speed. (m / s). The fixed length of the bearing section of the dry slag discharge machine is known. Based on the inherent design parameters of the dry slag discharge machine, this refers to the lag time required for the material to travel from the slag discharge port to the slag discharge port. for:

[0206] ;

[0207] Current moment The slag discharged at the tail end falls into the slag discharge port of the furnace at the moment... It is deduced as follows:

[0208] ;

[0209] The record whose timestamp is closest to the slag discharge time is retrieved from the timestamp queue. The virtual slag quantity in this record is extracted and used as predictive data causally aligned with the measured tail slag discharge quantity in the current dry slag discharge machine measurement dataset. Specifically:

[0210] Based on the calculated slag discharge time Search within the FIFO queue to find the timestamp and... The historical record with the smallest absolute value of time difference is used to extract the average virtual slag volume data stored in that record and identify it as the theoretical prediction volume when the slag discharged at the tail falls, so as to achieve accurate causal alignment between the predicted slag volume and the measured slag volume in the time dimension.

[0211] The predicted total slag quantity is obtained by summing the virtual slag quantities within the preset calibration window after causal alignment. The measured slag discharge quantities at the tail end within the same calibration window are summed according to the sampling interval to obtain the measured total slag discharge quantity. The ratio of the measured total slag discharge quantity to the predicted total slag quantity is calculated, and after first-order low-pass filtering, it is used as the calibrated bottom slag share. This is then used to update the current bottom slag share. Specifically:

[0212] Set a fixed calibration window (e.g., 1 hour), and statistically analyze the predicted total slag volume within the window at the calibration time. and the actual measured total amount of slag discharged :

[0213] ;

[0214] Let the measured value of the tail discharge volume in the dry slag discharger measurement dataset be the instantaneous slag discharge mass flow rate. The formula for calculating the total amount of slag discharged is:

[0215] ;

[0216] The instantaneous slag discharge mass flow rate can be calculated from the measured parameters of the equipment, specifically as follows: In the formula For bottom ash density, For real-time slag layer height, For the effective width of the scales, For real-time scale bucket speed, The sampling time interval (s) for the tail-end slag discharge parameters.

[0217] Calculation of raw observation values ​​for bottom ash content :

[0218] ;

[0219] To avoid parameter abrupt changes caused by fluctuations in single calibration data, the original observations are filtered using a first-order low-pass filter before updating the global current bottom ash fraction. :

[0220] ;

[0221] in, The bottom ash proportion is calibrated with a filter coefficient ranging from 0.1 to 0.3. This can smooth the parameter update amplitude and suppress the interference of instantaneous operating condition fluctuations. The updated bottom ash proportion is directly fed back to the virtual ash quantity calculation module and participates in the virtual ash quantity iterative calculation of the next control cycle to complete the adaptive closed-loop calibration.

[0222] In this implementation plan, operating data is continuously recorded through a time-series queue. Combined with the characteristics of the bucket conveying system, the corresponding slag discharge time is accurately calculated, achieving time-series matching between the virtual slag quantity prediction data and the actual slag discharge data measured on site. By comparing the theoretical slag quantity with the actual slag discharge quantity through window integration, the bottom slag share parameter is dynamically corrected. Combined with filtering processing, the parameter values ​​are updated smoothly, effectively reducing the calculation deviation caused by operating condition fluctuations. This ensures that the virtual slag quantity calculation results always closely match the actual on-site operating conditions, guaranteeing the accuracy and adaptability of the entire control and operation system.

[0223] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0224] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A smart control system for cooling air and waste heat recovery in a scale bucket dry slag discharge machine, characterized in that, include: The data acquisition unit is used to acquire the unit operation dataset and the dry slag discharger measurement dataset of the scale bucket dry slag discharger, as well as the current bottom slag share; The virtual slag volume calculation unit is used to calculate the virtual slag volume entering the scale bucket dry slag discharger based on the unit operation dataset and the current bottom slag share. The slag temperature inversion unit is used to iteratively invert the output of the bottom slag discharge temperature based on the virtual slag quantity and the dry slag discharger measurement dataset through a countercurrent heat transfer model, and to calculate the heat carried by the slag to be recovered by combining the virtual slag quantity. The optimization control unit is used to generate candidate cooling air volume as decision variables based on the measured value of cooling air volume in the dry slag discharger measurement dataset, with the heat carried by slag and virtual slag volume as feedforwards. Under the constraints of preset scale bucket safety temperature threshold, tail slag discharge safety temperature threshold and furnace oxygen prediction constraints, the optimal cooling air volume is solved as the optimization control command with the goal of maximizing the net benefit of waste heat recovery. The safety control unit is used to perform heat balance processing based on the heat carried by the slag and the measurement data of the dry slag discharger, generate a virtual temperature rise rate on the surface of the slag bucket, and output a safety control command after combining the measurement data of the slag discharger with confidence assessment and hierarchical protection. The instruction coordination output unit is used to coordinate safety control instructions over economic control instructions to generate the final cooling air volume control instruction, so as to maximize the recovery of waste heat from dry slag within the safety boundary.

2. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 1, characterized in that, The specific steps for calculating the virtual slag volume entering the hopper dry slag discharger are as follows: The operating status signal and instantaneous coal quantity of each coal feeder are obtained from the unit operation data. Only the instantaneous coal quantity of the coal feeders that are in operation is accumulated to obtain the total coal feed. And obtain the ash content of the coal received from the furnace from the unit operation data set; The total coal feed, the ash content of the coal received into the furnace, and the current bottom ash content are comprehensively processed to obtain the initial virtual ash quantity for the current cycle. The current initial virtual slag quantity is merged with the historical initial virtual slag quantities calculated and stored in previous consecutive calculations to form an initial virtual slag quantity sequence. A first-order low-pass filter is then applied to obtain the virtual slag quantity entering the slag discharge machine.

3. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 1, characterized in that, The specific construction method of the countercurrent heat transfer discrete model is as follows: The heat exchange channel between the slag and cooling air in the scale bucket dry slag discharge machine is discretized into multiple nodes along the slag flow direction. For each node, slag-side heat balance equations and air-side heat balance equations are established respectively. The slag-side heat balance equation includes the axial variation term of heat carried by the slag flow and the heat release term from the slag to the air. The air-side heat balance equation includes the axial variation term of heat absorbed by the air flow and the term of heat absorption from the slag. Adjacent nodes are coupled through slag flow temperature transfer and air flow temperature transfer, forming a counter-current heat transfer discrete model.

4. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 3, characterized in that, The specific steps for iteratively inverting and outputting the bottom slag discharge temperature using a countercurrent heat transfer model are as follows: Obtain the historical values ​​of the bottom slag discharge temperature that have converged in the previous several iterations, and take the moving average of the historical values ​​as the iteration starting point for the estimated bottom slag discharge temperature of the current iteration; In each iteration, the current estimated value of the bottom slag discharge temperature is substituted into the countercurrent heat exchange discrete model to obtain the corresponding predicted value of the cooling air outlet temperature. The current deviation between the predicted value and the measured value of the cooling air outlet temperature in the dry slag discharge machine measurement dataset is calculated. A preset small perturbation is applied to the current estimated bottom slag discharge temperature. The gradient is determined based on the ratio of the change in the current deviation before and after the perturbation to the amount of the small perturbation. The estimated bottom slag discharge temperature is then updated in the opposite direction of the gradient with a preset step size. ； in, This is the estimated value of the current bottom slag discharge temperature in the (n+1)th iteration. This is the estimated value of the current bottom slag discharge temperature in the nth iteration. Let be the gradient of the nth iteration. Let n be the step size for the nth iteration; and With the absolute value of the current deviation Decrease and gradually shrink, adaptively adjusting as follows: ； in, Basic step size factor, It is a preset positive decimal; The iteration process is repeated until the current deviation value of the consecutive preset number of iterations is less than the preset convergence threshold. The estimated value of the bottom slag discharge temperature of the last iteration is then used as the bottom slag discharge temperature.

5. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 1, characterized in that, The specific steps for determining the optimal cooling airflow as the optimization control command are as follows: Based on the measured value of cooling air volume, multiple candidate cooling air volumes are generated within a preset search range and at a preset step size. Using the heat carried by the slag and the virtual slag volume as feedforward information, and combining each candidate cooling air volume, the counter-current heat transfer discrete model is called to predict the cooling air outlet temperature that can be achieved under the candidate cooling air volume. Calculate the net waste heat recovery revenue corresponding to each candidate cooling air volume using the following formula. : ； in, The candidate cooling air volumetric flow rate, To use the counter-current heat transfer discrete model to predict the cooling air outlet temperature, To obtain the ambient temperature, It is the preset air density. The specific heat capacity of air is preset. This refers to the heat price coefficient stored in the database, which converts recovered heat into revenue. This represents the power consumption of the cooling fan at this candidate airflow rate, and , For pre-stored at rated air volume Power consumption at that time This refers to the electricity price coefficient; Within the feasible region comprised of all candidate cooling air volumes that satisfy the preset safety thresholds for the hopper, the tail-end slag discharge, and the furnace oxygen prediction constraints, the selected cooling air volume is... The largest candidate cooling airflow is selected as the optimal cooling airflow.

6. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 5, characterized in that, The constraints for predicting furnace oxygen levels are achieved in the following way: Under stable boiler combustion conditions, a step disturbance is applied to the cooling air volume, and the response curve of oxygen at the furnace outlet is recorded to identify the first-order transfer function model with pure lag. The transfer function model is converted into a discrete difference equation form of oxygen prediction model. The measured oxygen value at the furnace outlet in the dry ash removal machine measurement dataset is used as the current state. The difference between the candidate cooling air volume and the current measured cooling air volume is used as the input increment. The oxygen prediction value sequence within the future preset time period is recursively calculated. The maximum value in the oxygen prediction value sequence is taken as the future oxygen peak value corresponding to the candidate cooling air volume; The preset upper limit of oxygen content at the furnace outlet is obtained. The maximum candidate cooling air volume that will not cause the future oxygen peak to exceed the upper limit of oxygen content at the furnace outlet is determined as the dynamic air volume upper limit, which serves as the constraint boundary of the feasible region.

7. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 1, characterized in that, The specific steps for generating the virtual temperature rise rate on the surface of the scales are as follows: The actual heat recovered by the cooling air is calculated by using the measured values ​​of the cooling air volume and the cooling air outlet temperature in the data set of the dry slag discharge machine. The net heat load is obtained by subtracting the heat carried by the slag from the actual heat recovered by the cooling air. The net heat load is converted to the preset equivalent heat capacity by temperature rise calculation to obtain the virtual temperature rise rate of the scale surface.

8. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 7, characterized in that, The specific steps for outputting safety control commands are as follows: The temperature signals of multiple points on the scale bucket bearing section were obtained from the measurement data of the dry slag discharge machine, and the maximum temperature of each measuring point was taken as the measured value of the scale bucket surface temperature. The measured surface temperature of the scaly cisterns was fitted using a sliding window to obtain the measured temperature rise rate and determine the fitting confidence level. The fit confidence level is compared with a preset confidence threshold, and the current judgment criterion is determined based on the comparison result; If the current judgment criteria exceed the preset warning threshold but do not exceed the preset safety threshold, output a wind prevention command; If the current judgment criteria exceed the preset safety threshold, a safe airflow command will be output. The preventive airflow increase command and the safe airflow command together constitute the safety control command.

9. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 8, characterized in that, The specific steps for determining the current judgment criteria based on the comparison results are as follows: When the confidence level of the fit is higher than the preset confidence threshold, the measured temperature rise rate is determined to be reliable, and the measured temperature rise rate is used as the current determination basis. The difference between the measured temperature rise rate and the virtual temperature rise rate is calculated as the bias, and the virtual temperature rise rate is corrected accordingly. When the confidence level of the fit is lower than the preset confidence threshold, the measured temperature rise rate is determined to be unreliable, and the corrected virtual temperature rise rate is used as the current basis for judgment.

10. The intelligent control system for cooling air and waste heat recovery of the scale bucket dry slag discharge machine according to claim 1, characterized in that, This also includes calibrating the current bottom ash content: Maintain a first-in-first-out timestamp queue to record the virtual slag quantity and the current bottom slag share for each time period with a preset time granularity; The real-time speed of the bucket conveyor belt and the preset fixed length of the carrying section are obtained, and the slag discharge time corresponding to the slag currently being discharged into the tail section is determined. Retrieve the record whose timestamp is closest to the slag discharge time from the timestamp queue, extract the virtual slag quantity in the record, and use it as predictive data that is causally aligned with the measured value of the tail slag discharge quantity in the current dry slag discharge machine measurement dataset; The predicted data in the preset calibration window after causal alignment is causally calibrated with the measured value of the tail ash discharge to obtain the calibrated bottom ash share, and the current bottom ash share is updated accordingly.

Images