Search

CN-121971996-A - Self-adaptive ammonia spraying device for flue gas denitration

CN121971996ACN 121971996 ACN121971996 ACN 121971996ACN-121971996-A

Abstract

A smoke denitration self-adaptive ammonia spraying device relates to the technical field of smoke denitration environment protection and is used for improving denitration efficiency. In the device, the flue gas denitration self-adaptive ammonia spraying device converts complex on-line mechanical adjustment into external fluid control by constructing a cooperative control frame of static hardware matching and dynamic software adjustment, and the problem of blocking and abrasion under a high-temperature high-dust environment is avoided by utilizing a fixed gap unit without a movable component, so that the long-term operation reliability of the flue gas denitration self-adaptive ammonia spraying device is improved. Meanwhile, by combining the fusion algorithm of the AI model on the multi-source data and the double-level intelligent diagnosis early warning mechanism, double self-adaption of the ammonia spraying strategy on space unevenness and time change is realized, and further the denitration efficiency is improved on the premise of improving the stability of the flue gas denitration self-adaption ammonia spraying device.

Inventors

  • YU MEINA
  • LIU XIANZHUO
  • Fu Qianshi

Assignees

  • 天津国环节能科技有限公司

Dates

Publication Date
20260505
Application Date
20260312

Claims (10)

  1. 1. The utility model provides a flue gas denitration self-adaptation spouts ammonia device which characterized in that includes: The data acquisition module is used for acquiring running state data of the denitration reactor, wake flow evaluation feedback data and regional concentration statistical average values used for representing spatial distribution differences of the flue; The AI calculation module is used for taking the regional concentration statistical average value as a static space characteristic, taking the running state data as a dynamic time sequence characteristic, inputting the dynamic time sequence characteristic into a preset depth neural network model to perform multi-source characteristic fusion calculation, and obtaining target ammonia spraying distribution quantity corresponding to each fixed gap unit, wherein the fixed gap units are assembled on a linear grid body, the inside of the linear grid body is not provided with a mechanical adjusting component which slides relatively, and the linear grid body is horizontally arranged in an inlet flue of the denitration reactor along the smoke flow direction; The control and regulation module is used for receiving the target ammonia injection distribution amount, converting the target ammonia injection distribution amount into a variable frequency control signal aiming at the total ammonia supply fluid pressure and a partition flow control signal aiming at each independent flow regulation branch of the multichannel flow distributor, wherein the multichannel flow distributor is provided with a plurality of independent flow regulation branches, and each independent flow regulation branch is respectively butted to the fixed gap unit through an ammonia injection pipeline made of corrosion-resistant materials; The ammonia spraying execution module is used for adjusting the total ammonia supply fluid pressure based on the variable frequency control signal, driving the multichannel flow distributor according to the partition flow control signal, and guiding the ammonia gas into each fixed gap unit for spraying according to the target ammonia spraying distribution; The data visualization module is used for associating and displaying the execution state of the target ammonia injection allocation amount and the wake flow evaluation feedback data, generating a hardware reconstruction early warning when the wake flow evaluation feedback data is judged to generate denitration deviation exceeding a preset performance degradation threshold, and the hardware reconstruction early warning is used for indicating to reacquire the regional concentration statistical average value and replace the preloaded fixed gap unit.
  2. 2. The flue gas denitration self-adaptive ammonia spraying device according to claim 1, wherein the data acquisition module specifically comprises: The spatial concentration difference pre-detection unit is used for arranging a sensor array in a grid installation area in an inlet flue of the denitration reactor, continuously collecting area nitrogen oxide concentration distribution data in a preset period, performing outlier removal pre-treatment on the area nitrogen oxide concentration distribution data, and calculating to obtain an area concentration statistical average value corresponding to each detection point; The operation condition real-time acquisition unit is used for synchronously reading the flue gas flow data, the flue gas temperature data, the flue gas pressure data and the actual execution ammonia spraying amount of the current denitration device flowing through the denitration reactor through a multidimensional sensor group, and integrating the operation state data input into the preset depth neural network model; And the wake efficiency feedback unit is used for monitoring the state of the downstream outlet flue gas after the catalyst reaction, acquiring a real-time nitrogen oxide concentration deviation value and an ammonia escape concentration value after denitration, extracting the transient change rate of the unit load and the fluctuation amplitude of the device inlet working condition parameter, and compounding to form wake evaluation feedback data.
  3. 3. The flue gas denitration self-adaptive ammonia spraying device according to claim 2, wherein the matching of the fixed void unit and the area concentration statistical average value is realized by a preset hardware optimizing rule, and specifically comprises: A concentration gradient dividing unit, configured to pre-establish a plurality of concentration gradient intervals from low to high, and allocate a cylindrical detachable structure corresponding to a standard aperture size to each concentration gradient interval as an alternative fixed void unit; The static mapping pairing unit is used for comparing the regional concentration statistical average value corresponding to each detection point with the concentration gradient interval one by one, judging a target gradient interval to which the regional concentration statistical average value belongs, and extracting the standard aperture size bound by the target gradient interval; The hardware preassembly guiding unit is used for outputting an assembly map containing coordinate positions of all detection points and corresponding standard aperture sizes so as to instruct offline screwing-in installation of the fixed gap unit in the grid installation area.
  4. 4. The flue gas denitration self-adaptive ammonia spraying device according to claim 1, wherein the preset depth neural network model specifically comprises: the multi-scale sequence decomposition unit is used for carrying out multi-scale decomposition on the nitrogen oxide concentration time sequence characteristics in the running state data by adopting wavelet transformation to obtain a high-frequency fluctuation characteristic vector representing the smoke mutation points; The time sequence dependency capturing unit is used for inputting the high-frequency fluctuation feature vector and the static space feature into a branch of a gate control circulating unit built in a model, and calculating evolution correlation features between the device load and the smoke parameters through a long-term and short-term memory mechanism; And the autoregressive multi-step prediction unit is used for combining a residual jump connection structure set by a model based on the evolution correlation characteristics to prevent the gradient of the deep network from disappearing, performing long-short-time combined rolling iterative algorithm, and outputting the target ammonia injection allocation amount corresponding to each fixed gap unit at the next control moment.
  5. 5. The flue gas denitration self-adaptive ammonia spraying device according to claim 1, wherein the control and regulation module specifically comprises: The global pressure supply amount calculating unit is used for carrying out mathematical summation on the target ammonia injection distribution amounts corresponding to all the independent flow regulating branches to obtain a total target ammonia injection amount, and mapping the total target ammonia injection amount into the variable frequency control signal according to a hydrodynamic conversion function; The independent channel proportion analysis unit is used for calculating the dynamic distribution proportion of the target ammonia injection distribution amount of each fixed gap unit in the total target ammonia injection amount, and converting the dynamic distribution proportion into the partition flow control signals for controlling the opening of the corresponding branch electric valve.
  6. 6. The flue gas denitration adaptive ammonia injection device of claim 1, wherein the AI computation module further comprises: And the multisource dimension eliminating unit is used for calling a Z-score standardization algorithm to normalize the running state data with different physical dimension attributes, and eliminating gradient offset caused by extremum difference among heterogeneous data to the multisource feature fusion calculation process.
  7. 7. The adaptive ammonia spraying device for flue gas denitration according to claim 5, wherein the ammonia spraying execution module specifically comprises: The global variable-frequency pressurizing unit is used for receiving the variable-frequency control signal, and performing global pressure adjustment on ammonia supply fluid from an ammonia storage tank through a variable-frequency ammonia pump so as to meet the dry pipe dynamic pressure requirement corresponding to the total target ammonia injection amount; The flow guiding and limiting unit is used for receiving the partition flow regulating control signals, respectively regulating the fluid sectional areas of all independent flow regulating branches in the multichannel flow distributor, splitting pressurized ammonia gas and respectively injecting the split pressurized ammonia gas into all corresponding fixed gap units; And the forced flow injection unit is used for forced injection of the input ammonia through the static aperture physically preset in the fixed gap unit.
  8. 8. The flue gas denitration adaptive ammonia spraying device according to claim 1, wherein the data visualization module specifically comprises: the efficiency state association mapping unit is used for synchronously comparing and displaying an execution curve of the target ammonia injection distribution quantity aligned with a time axis with an actual denitration index curve in the wake flow evaluation feedback data on a visual interface; The software elastic fine adjustment triggering unit activates a model online learning mechanism when the denitration deviation falls into a preset safety regulation fuzzy domain, does not change hardware configuration, and automatically corrects the target ammonia injection distribution amount by means of the preset depth neural network model; and when the denitration deviation exceeds the preset safety regulation fuzzy domain and reaches the preset performance degradation threshold, fusing the model online learning mechanism and forcibly triggering and issuing the hardware reconstruction early warning.
  9. 9. The adaptive ammonia spraying device for flue gas denitrification according to claim 8, wherein the triggering logic of the data visualization module is: When the positive and negative deviation amount of the real-time nitrogen oxide concentration continuously deviating from a set value after denitration exceeds a safety tolerance or the ammonia escape concentration value exceeds an early warning line and the duration of the out-of-limit state exceeds a preset time window, judging that the ammonia escape concentration value falls into the preset safety regulation fuzzy domain, and triggering the software elastic fine adjustment triggering unit; When the average absolute error of the predicted ammonia injection amount and the actual ammonia injection amount of the model online learning mechanism continuously exceeds a preset generalization tolerance, or the transient change rate of the unit load in the wake flow evaluation feedback data and the fluctuation amplitude of the device inlet working condition parameters exceed the design response extremum of the denitration device, the model online learning mechanism is judged to reach the preset performance degradation threshold, and the hardware out-of-limit fusing alarm unit is triggered.
  10. 10. The smoke denitration self-adaptive ammonia spraying device according to claim 1, wherein a plurality of mounting holes for assembling the fixed gap units are uniformly distributed on the straight-line grid body, the fixed gap units are of cylindrical detachable structures and are screwed into the corresponding mounting holes through outer wall threaded connection respectively, the side walls of the fixed gap units are provided with physically preset static apertures, and fluororubber sealing elements are embedded at the matched connection parts of the fixed gap units and the mounting holes to prevent ammonia leakage.

Description

Self-adaptive ammonia spraying device for flue gas denitration Technical Field The application relates to the technical field of flue gas denitration and environmental protection, in particular to a self-adaptive ammonia spraying device for flue gas denitration. Background In industrial flue gas denitration treatment, a Selective Catalytic Reduction (SCR) technology is widely applied due to the advantages of high denitration efficiency, good stability and the like. The core flow of the technology is that ammonia gas as a reducing agent is sprayed into the flue gas through an ammonia spraying grid arranged in an inlet flue of the reactor, so that the ammonia gas and the flue gas are mixed and then undergo a reduction reaction with nitrogen oxides (NO x) in the flue gas under the action of a catalyst to generate harmless nitrogen and water. In order to adapt to the condition of uneven concentration distribution of nitrogen oxides in the section of the flue gas, the related technology generally adopts a continuously adjustable ammonia injection grid device. This technique provides mechanical adjustment mechanisms (e.g., nested movable inner and outer sleeves or linked fine adjustment valves) at each subunit or nozzle of the ammonia injection grid and operates on-line in conjunction with an external drive module and control system. In the running process of the equipment, the mechanical adjusting mechanism in the grid is driven by the actuating mechanism to generate continuous displacement or rotation, and the effective physical opening area of each nozzle is changed by utilizing the relative sliding among the movable parts, so that different amounts of ammonia gas are distributed to the flue gas areas with different concentrations by changing the physical flow cross section. However, since the flue gas environment of the inlet of the denitration reactor usually contains a large amount of high-concentration fly ash, and is accompanied by high temperature and corrosive volatile substances, the related art relies on mechanical components which move frequently inside and a tiny gap for guaranteeing movement during dynamic adjustment, and under the severe working conditions of high-temperature baking and dust flushing for a long time, the fly ash is easy to invade and harden and deposit in the relative sliding gap of the movable sleeve or valve, so that the denitration efficiency is reduced. Disclosure of Invention The application provides a self-adaptive ammonia spraying device for flue gas denitration, which is used for improving denitration efficiency. The first aspect provides a flue gas denitration self-adaptive ammonia spraying device, which is characterized by comprising a data acquisition module, a first analysis module, a second analysis module and a first analysis module, wherein the data acquisition module is used for acquiring running state data of a denitration reactor, wake flow evaluation feedback data and regional concentration statistical average values used for representing spatial distribution differences of a flue; the AI calculation module is used for taking the regional concentration statistical average value as a static space characteristic, taking the running state data as a dynamic time sequence characteristic, inputting the dynamic time sequence characteristic into the preset depth neural network model to perform multisource characteristic fusion calculation, and obtaining target ammonia spraying distribution quantity corresponding to each fixed gap unit, wherein the fixed gap units are assembled on a linear grid body and have no mechanical adjusting component sliding relatively in the fixed gap units, and the linear grid body is horizontally arranged in an inlet flue of the denitration reactor along the smoke flow direction; the system comprises a control and regulation module, an ammonia spraying execution module, a data visualization module, a data analysis module and a data analysis module, wherein the control and regulation module is used for receiving a target ammonia spraying distribution amount, converting the target ammonia spraying distribution amount into a variable frequency control signal aiming at the total ammonia supply fluid pressure, and a partition flow regulation control signal aiming at each independent flow regulation branch of a multi-channel flow distributor, the multi-channel flow distributor is provided with a plurality of independent flow regulation branches, each independent flow regulation branch is respectively butted to a fixed gap unit through an ammonia spraying pipeline made of corrosion-resistant materials, the ammonia spraying execution module is used for regulating the total ammonia supply fluid pressure based on the variable frequency control signal and driving the multi-channel flow distributor according to the partition flow regulation control signal, guiding ammonia into each fixed gap unit according to the target ammonia spraying distribution amoun