A method and system for targeted deep desorption control of adsorption beds for exhaust gas treatment

Abstract

The application discloses a kind of adsorption bed targeted depth desorption control method and system for waste gas treatment, it is related to waste gas treatment technical field, comprising the following steps: obtaining the organic matter concentration curve of desorption end period;Exponential decay fitting is carried out to the attenuation section of organic matter concentration curve, and the difference between organic matter concentration curve and fitting curve is regarded as the tailing section of slow precipitation of corresponding high-boiling-point organic matter;According to the concentration value of the tailing section at different desorption time, the residual amount of the high-boiling-point organic matter at different depths in the adsorption bed is calculated by inversion, and a residual amount spatial distribution map is generated;When the spatial distribution map shows that the residual amount of a local area exceeds a preset threshold, targeted depth desorption is carried out on the local area in the next adsorption-desorption cycle.The application effectively avoids the risk of equipment thermal runaway, significantly extends the service life of the adsorption material while significantly reducing the desorption energy consumption.

Description

Technical Field

[0001] This invention relates to the field of waste gas treatment technology, and more specifically, to a method and system for targeted deep desorption control of adsorption beds for waste gas treatment. Background Technology

[0002] In the field of volatile organic compound (VOCs) treatment, for large volumes of low to medium concentration VOCs, the "spray pretreatment + adsorption concentration + catalytic combustion" integrated process is currently widely used in industry. This process enriches pollutants through an adsorption bed and uses a hot gas flow to strip the pollutants during the desorption stage, sending them to the catalytic combustion chamber for oxidation and decomposition.

[0003] However, in actual engineering operations, the following technical problems exist: 1. Accumulation of high-boiling-point organic matter leads to adsorbent failure. Industrial waste gas often contains small amounts of high-boiling-point, large-molecule organic matter. In the conventional desorption stage, the existing desorption temperature and time are insufficient to completely expel these substances from the adsorbent (such as activated carbon or zeolite). With the increase of adsorption-desorption cycles, high-boiling-point organic matter gradually accumulates inside the adsorption bed, continuously occupying adsorption sites, eventually leading to irreversible "poisoning" failure of the adsorbent and a significant decrease in treatment efficiency.

[0004] 2. The residual state inside the adsorption bed is imperceptible. Existing monitoring methods are mostly based on detecting the total concentration at the inlet and outlet of the adsorption bed, which can only reflect the overall treatment effect and cannot accurately perceive the real-time residual state at different depths and in different areas inside the bed. Operators often only realize that the adsorbent has been damaged when they find a significant drop in treatment efficiency, which has obvious lag and cannot provide early warning and intervention.

[0005] 3. Fixed desorption methods result in energy waste and incomplete regeneration. Traditional desorption processes typically use fixed durations and air volumes, a "one-size-fits-all" approach. They lack targeted treatment methods for localized areas where high-boiling-point substances accumulate. This not only leads to energy waste (overheating of already cleaned areas) but also fails to completely eliminate localized hazards, shortening the maintenance cycle of the entire system.

[0006] To address the above problems, this invention proposes a solution. Summary of the Invention

[0007] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a targeted deep desorption control method and system for adsorption beds in waste gas treatment. By collecting the organic matter concentration curve at the end of desorption, the system identifies the tailing segment of slowly precipitating high-boiling-point organic matter through exponential decay fitting. Based on the tailing segment, the spatial distribution of residual amount is inverted, and targeted deep desorption is performed on the local areas exceeding the standard. This solves the problems of adsorbent failure caused by the gradual accumulation of high-boiling-point organic matter in the adsorption bed, as well as energy waste and incomplete regeneration caused by fixed desorption.

[0008] To achieve the above objectives, the present invention provides the following technical solution: A targeted deep desorption control method for adsorption beds in waste gas treatment includes the following steps: obtaining an organic matter concentration curve at the end of desorption; performing exponential decay fitting on the decay segment of the organic matter concentration curve, and using the difference between the organic matter concentration curve and the fitted curve as the tailing segment for the slow precipitation of high-boiling-point organic matter; calculating the residual amount of the high-boiling-point organic matter at different depths within the adsorption bed based on the concentration value of the tailing segment at different desorption times, and generating a spatial distribution map of the residual amount; when the spatial distribution map shows that the residual amount in a local area exceeds a preset threshold, targeted deep desorption is performed on that local area in the next adsorption-desorption cycle.

[0009] In a preferred embodiment, obtaining the organic matter concentration curve at the end of desorption includes: extracting the peak organic matter concentration at the current desorption stage, and using the product of the peak concentration and a preset attenuation coefficient as a dynamic trigger threshold; when the organic matter concentration drops to its dynamic trigger threshold, initiating high-frequency sampling to obtain a concentration-time series from initiation to the end of desorption; and performing filtering and noise reduction processing on the concentration-time series to obtain the organic matter concentration curve.

[0010] In a preferred embodiment, the exponential decay fitting of the decay segment of the organic matter concentration curve includes: calculating the first derivative sequence of the organic matter concentration curve, and extracting a continuous time interval in which the absolute value is greater than a preset rate threshold as a reference decay segment; performing exponential function least squares fitting on the reference decay segment to obtain a decay function; and extrapolating the decay function in the time domain to the high-frequency sampling end node of the current desorption stage to generate a fitting curve.

[0011] In a preferred embodiment, the calculation of the first derivative sequence of the organic matter concentration curve specifically includes: setting a sliding time window, using the Savitzky-Golay filtering method to perform local polynomial fitting on the discrete data points of the organic matter concentration curve within the window; performing analytical differentiation on the fitted polynomial, and extracting the first derivative value of the center node of the sliding time window; sliding the time window to traverse the organic matter concentration curve to obtain the first derivative sequence.

[0012] In a preferred embodiment, the step of using the difference between the organic concentration curve and the fitted curve as the tailing segment corresponding to the slow precipitation of high-boiling-point organic matter includes: calculating the residual between the organic concentration curve and the fitted curve at the same time to obtain a residual sequence; extracting the standard deviation of the residual within the baseline decay segment and constructing a dynamic error threshold by combining it with a preset confidence interval; traversing the residual sequence in the forward direction along the time axis and taking the first time node that continuously exceeds the dynamic error threshold as the tailing start point; and extracting the residual sequence between the tailing start point and the high-frequency sampling end node of the current desorption stage as the tailing segment.

[0013] In a preferred embodiment, generating the spatial distribution map of residual amounts includes: obtaining the apparent velocity of the desorption gas flow and the porosity of the adsorption bed, and constructing a time-space mapping coefficient; calculating the lag time relative to the tailing start point at each moment within the tailing segment, and converting the lag time into depth coordinates within the adsorption bed using the time-space mapping coefficient; performing discrete integration on the concentration values ​​at each moment of the tailing segment in combination with the volumetric flow rate of the desorption gas flow to obtain the residual amount of high-boiling-point organic matter corresponding to each depth coordinate; and matching the depth coordinates with the corresponding residual amounts to generate a spatial distribution map of residual amounts.

[0014] In a preferred embodiment, obtaining the apparent velocity of the desorption gas flow and the porosity of the adsorption bed includes: collecting the actual pressure difference between the inlet and outlet of the adsorption bed at the current desorption stage, extracting the initial pressure difference of the adsorption bed under the same desorption gas flow rate; calculating the difference between the actual pressure difference and the initial pressure difference, and back-calculating the current porosity change based on the difference; and correcting the initial porosity of the adsorption bed with the porosity change to obtain the corrected porosity.

[0015] In a preferred embodiment, before calculating the lag time of each moment within the tailing segment relative to the tailing start point, the method further includes: obtaining the axial diffusion coefficient of the desorbed gas flow within the adsorption bed; establishing a transfer function of the one-dimensional convection-diffusion equation of the desorbed gas flow based on the axial diffusion coefficient and the apparent velocity; performing deconvolution calculation on the residual sequence using the transfer function; and extracting the convergent residual sequence corresponding to the deconvolution calculation as the corrected tailing segment.

[0016] In a preferred embodiment, the targeted depth desorption includes: increasing the temperature of the desorbed gas entering the local region, and / or increasing the residence time of the desorbed gas in the local region.

[0017] A targeted deep desorption control system for adsorption beds in waste gas treatment includes: a concentration curve acquisition module for acquiring the organic matter concentration curve at the end of desorption; a tailing segment identification module for performing exponential decay fitting on the decay segment of the organic matter concentration curve, and taking the difference between the organic matter concentration curve and the fitted curve as the tailing segment for the slow precipitation of the corresponding high-boiling-point organic matter; a residual amount inversion module for inverting and calculating the residual amount of the high-boiling-point organic matter at different depths in the adsorption bed based on the concentration value of the tailing segment at different desorption times, and generating a residual amount spatial distribution map; and a regeneration control module for performing targeted deep desorption on the local area in the next adsorption-desorption cycle when the residual amount in a local area shown in the spatial distribution map exceeds a preset threshold.

[0018] This invention precisely identifies the tailing segment representing the slow precipitation of high-boiling-point organic matter by performing exponential decay fitting on the organic matter concentration curve at the end of desorption and extracting the difference. This allows for the generation of a residual spatial distribution map, enabling precise three-dimensional digital localization of hidden high-boiling-point pollutants within the adsorption bed. This effectively overcomes the technical blind spot of traditional waste gas treatment equipment, which cannot detect the depth of localized "poisoning" within the bed. Furthermore, by performing targeted deep desorption on specific localized areas in the next cycle when the residual amount exceeds a preset threshold, this invention completely changes the crude "global blind desorption" model of traditional processes. It can precisely remove stubborn localized residues like surgery, effectively extending the overall lifespan of the adsorption material, while significantly avoiding ineffective operations on areas that do not exceed the threshold, thereby significantly reducing the system's operating energy consumption. Attached Figure Description

[0019] Figure 1 This is a schematic flowchart of an adsorption bed targeted deep desorption control method for waste gas treatment according to the present invention. Figure 2 This is a schematic diagram of the structure of an adsorption bed targeted deep desorption control system for waste gas treatment according to the present invention; Figure 3 This is a graph showing the concentration curve of organic matter at the end of desorption and the dynamic trigger threshold determination in an embodiment of the present invention. Figure 4 This is an example of an embodiment of the invention showing the identification of the exponential decay fitting curve and the tail segment of high-boiling-point organic compounds. Figure 5 This is a comparison chart of the average residual amount of the bed after multiple adsorption-desorption cycles in the embodiments of the present invention; Figure 6 This is a comparison chart of the residual amount of the bed at different depths before and after targeted depth desorption in an embodiment of the present invention; Figure 7 This is a spatial distribution diagram of the residual amount of high-boiling-point organic matter in the adsorption bed in an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1, Figure 1 This invention discloses a method for targeted deep desorption control of an adsorption bed for waste gas treatment, comprising the following steps: S1, obtain the organic matter concentration curve at the end of desorption; In this embodiment, obtaining the organic matter concentration curve at the end of desorption specifically involves: S11, the system monitors the concentration changes throughout the desorption process in real time using an online VOCs monitor (preferably a PID photoionization sensor or an FID flame ionization sensor with fast response characteristics) installed at the adsorption bed outlet pipeline. The control unit continuously extracts the peak concentration of organic matter that appears in the current desorption stage. .

[0022] Considering the significant differences in initial load and composition of different batches of waste gas in industrial sites, using a fixed concentration value as the benchmark for determining tailing is prone to misjudgment. Therefore, this embodiment introduces an adaptive dynamic triggering mechanism: the peak concentration value is used as the triggering factor. With preset attenuation coefficient Multiply to calculate the dynamic trigger threshold for the current batch in real time. Wherein, the preset attenuation coefficient The preferred value range is 5% to 15%, and in this embodiment, it is preferably 10%.

[0023] S12, when the VOCs online monitoring instrument detects that the concentration of organic matter in the desorption gas flow continues to decrease and first drops to the dynamic trigger threshold. At that time, the control system automatically issues a command to instantly switch the sensor’s data acquisition mode from the conventional low-frequency inspection mode (e.g., sampling frequency of 0.1Hz to 1Hz) to the high-frequency sampling mode (e.g., sampling frequency of 10Hz to 50Hz), and acquires a high-density concentration-time series from the start time to the end of the current desorption cycle.

[0024] S13. Considering the noise interference from pipeline vibration and the start-up and shutdown of electrical equipment in industrial settings, the directly measured concentration-time series may exhibit spikes and jumps. The system performs filtering and denoising on the high-frequency sampled concentration-time series, removing outliers and outputting a smooth organic concentration curve. This provides accurate input data for subsequent exponential decay fitting and inversion calculations. It should be noted that the filtering and denoising here is a preliminary denoising step in the data preprocessing stage; before performing numerical calculations such as derivatives on the concentration curve, further conformal filtering algorithms such as Savitzky-Golay can be used to balance smoothness and derivative calculation accuracy.

[0025] Figure 3 This is an example of an embodiment of the invention showing the organic matter concentration curve at the end of desorption and the dynamic trigger threshold determination diagram, such as... Figure 3 As shown, the horizontal axis represents desorption time, and the vertical axis represents the outlet organic matter concentration (ppm). The solid line represents the measured organic matter concentration curve, and the dashed line represents the dynamic trigger threshold (10% of the peak concentration, i.e., 45 ppm, in this embodiment). When the measured concentration curve decreases and crosses the dynamic trigger threshold, the high-frequency sampling mode is triggered, and the high-frequency sampling area is marked by a light-colored shaded area in the figure. Through this adaptive dynamic triggering mechanism, the sampling start point can be automatically adjusted according to the actual load of each batch of exhaust gas, avoiding false triggering or missed triggering caused by a fixed threshold, and ensuring accurate capture of the concentration signal of slowly precipitating high-boiling-point organic matter at the end of desorption.

[0026] S2, perform exponential decay fitting on the decay segment of the organic matter concentration curve, and take the difference between the organic matter concentration curve and the fitted curve as the tail segment of the slow precipitation of the corresponding high-boiling-point organic matter. In this embodiment, the exponential decay fitting of the decay segment of the organic matter concentration curve is specifically as follows: S21. In a real exhaust gas desorption site, the raw concentration curves collected by the sensors inevitably have wind pressure fluctuations and electromagnetic white noise superimposed on them. If the rate of change (i.e., the first derivative) is calculated directly using the conventional adjacent point difference method, this tiny high-frequency noise will be drastically amplified by mathematical differential operations, resulting in severe spikes in the derivative curve, and thus making it impossible to accurately extract the continuous decay interval.

[0027] To overcome the above problems, this embodiment innovatively employs the SG filtering algorithm for processing. Specifically, the system sets a sliding time window (preferably with a window width of [value missing]) on the concentration-time axis. (Up to 21 discrete sampling points). During each window slide, the algorithm does not directly calculate the difference, but instead uses the least squares method to fit the discrete data points within the window into a local polynomial, preferably a second- or third-order polynomial, for example:

[0028] Furthermore, the local fitting polynomial is directly differentiated analytically to extract the high-fidelity first derivative value of the center node of the sliding time window. As the time window traverses the entire organic concentration curve, the system outputs a smooth, continuous first derivative sequence that filters out high-frequency oscillations.

[0029] S22, after obtaining the first derivative sequence, the system calculates its absolute value sequence and compares it with a preset rate threshold. A comparison is made. Among them, the preset rate threshold... This is used to distinguish between the rapid desorption phase of conventional VOCs and the slow precipitation phase of high-boiling-point organic compounds. In this embodiment, The values ​​are determined experimentally based on the adsorbent type and typical desorption conditions. For example, the derivative values ​​corresponding to a concentration decrease rate of 1.0–5.0 mg / (m³·s) are used. The system extracts the derivative sequence whose absolute values ​​are consistently greater than [a certain value]. The longest continuous time interval is used to extract the organic matter concentration curve corresponding to this interval as the baseline decay segment.

[0030] S23. After extracting the baseline decay segment, the system identifies the parameters of this segment based on a classic first-order analytical kinetic model from chemical engineering. Specifically, a least-squares fitting algorithm is used to construct the ideal decay function:

[0031] in, This is the theoretical desorption concentration. For time variables, This is the initial concentration coefficient. The resolution rate constant is used to characterize the resolution properties of low-boiling-point substances.

[0032] Solve the above by minimizing the sum of squared residuals. and The optimal estimates of the two parameters are obtained. After obtaining the ideal decay function, the system extrapolates this function in the time domain to the high-frequency sampling end node of the current desorption stage, generating a complete fitting curve.

[0033] In this embodiment, the difference between the organic concentration curve and the fitted curve is used as the tailing segment corresponding to the slow precipitation of high-boiling-point organic matter, specifically as follows: S24, calculate the residuals between the organic matter concentration curve and the fitted curve at the same time point to obtain the residual sequence:

[0034] in, It is a residual sequence. Represents the organic matter concentration curve. This represents the fitted curve.

[0035] S25, extract the standard deviation of the residuals within the baseline attenuation range, and construct a dynamic error threshold by combining it with a pre-set confidence interval: Extract the residual data corresponding to the baseline attenuation range and calculate its standard deviation. Since the reference decay range is within the pure period before high-boiling-point substances precipitate, this standard deviation physically quantifies the system's background white noise under the current operating conditions. Combined with a preset signal interval (preferably using...) Criteria), constructing dynamic error thresholds :

[0036] in, The mean residual value within the baseline decay range. is the confidence coefficient.

[0037] Furthermore, the residual sequence is traversed forward along the time axis to determine whether the residual value exceeds the dynamic error threshold. To prevent single-point electromagnetic pulse interference, the system will... 1 time series node (preferred) The time-series nodes that all exceed the dynamic error threshold are marked as tailing start points. Finally, the residual sequence between the tailing start point and the high-frequency sampling end point of the current desorption stage is extracted as the tailing segment corresponding to the slow precipitation of high-boiling-point organic matter.

[0038] Figure 4 This is an example of an exponential decay fitting curve and a high-boiling-point organic tail segment identification diagram in an embodiment of the present invention, such as... Figure 4 As shown, the horizontal axis represents desorption time (s), and the vertical axis represents organic concentration (ppm). The solid line in the figure represents the measured concentration curve, the dashed line represents the fitted ideal exponential decay function and its extrapolated portion, and the bold solid line and shaded area represent the identified high-boiling-point organic tailing segment. Within the baseline decay range, the fitted curve and the measured curve are in high agreement, indicating that the ideal exponential decay model can accurately describe the rapid desorption process of low-boiling-point organics. As time progresses into the extrapolation range, the measured curve is significantly higher than the fitted curve; this difference represents the tailing signal formed by the slow precipitation of high-boiling-point organics due to their high boiling point and large desorption activation energy. By separating the tailing segment from the original concentration curve, decoupling analysis of the desorption processes of high-boiling-point and low-boiling-point organics is achieved.

[0039] S3. Based on the concentration values ​​of the tailing segment at different desorption times, the residual amount of the high-boiling-point organic matter at different depths in the adsorption bed is calculated by inversion, and a spatial distribution map of the residual amount is generated. S31, in this embodiment, obtaining the apparent flow rate of the desorption gas stream and the porosity of the adsorption bed specifically involves: It should be noted that after multiple adsorption-desorption cycles, the gradual accumulation of high-boiling-point organic matter in the micropores of the adsorption bed leads to a decrease in local porosity. If the initial porosity of the adsorption bed in its clean state is consistently used for spatiotemporal calculations, a systematic bias will occur. Therefore, a pressure sensor is installed at both the inlet and outlet of the adsorption bed to collect the pressure value during the current desorption stage in real time and calculate the actual pressure difference between the inlet and outlet. Simultaneously, the control system retrieves the initial pressure difference measured under the same desorption gas flow rate conditions when the adsorption bed is in a clean state from the historical database. .

[0040] Furthermore, the difference between the actual pressure drop and the initial pressure drop is calculated. Based on the Ergun drag equation in porous media fluid mechanics, the surge in bed pressure drop and the decrease in porosity exhibit a strict physical negative correlation. To facilitate real-time high-frequency calculations by the industrial control system, an engineering linearized drag model is adopted, and the current porosity change is inferred from the difference. :

[0041] in, The empirical resistance sensitivity coefficient is calibrated in advance through fluid dynamics experiments. Its value is related to the particle size, bulk density and aerodynamic viscosity of the adsorbent material, and the preferred range is 0.15 to 0.35.

[0042] The porosity change The initial porosity of the adsorption bed Dynamic reduction and correction are performed to obtain the corrected porosity. That is, using the formula:

[0043] S32, obtain the apparent velocity of the desorbed airflow under the current operating conditions from the central control unit or flow meter. The apparent velocity refers to the average theoretical velocity of the airflow through the empty bed cross-section, assuming no adsorbent obstruction within the adsorption bed. Simultaneously, the porosity of the current adsorption bed (i.e., the corrected porosity) is retrieved. ).

[0044] Because the actual adsorption bed is filled with dense adsorbent material, the desorption gas flow can only travel within microscopic pore channels. According to the laws of mass conservation and fluid continuity, the actual transport velocity of the gas within the pores (i.e., the interstitial velocity) must be much greater than the macroscopic apparent velocity. Based on this physical mechanism, the system reduces the apparent velocity of the desorption gas flow. With respect to the porosity of the adsorption bed Divide to construct the time-space mapping coefficients. :

[0045] S33, before calculating the lag time of each moment within the trailing segment relative to the starting point of the trailing segment, the method further includes: When the desorbed gas flows through a porous medium, axial backmixing (also known as axial diffusion) occurs due to the combined effects of molecular diffusion and convective mixing. This effect "broadens" the concentration signal of high-boiling-point organic compounds released from deeper parts of the bed during transmission—the concentration pulse released instantaneously at a certain depth is stretched into a temporally diffuse signal when transmitted to the outlet by the gas flow. If spatiotemporal inversion is performed directly using the broadened tail segment, the positioning accuracy of the depth coordinates will decrease.

[0046] To eliminate the effects of axial backmixing, this embodiment introduces a deconvolution algorithm from the field of signal processing. Specifically: The axial diffusion coefficient of the desorbed gas flow within the adsorbent medium was obtained through tracer experiments or literature data. Based on this axial diffusion coefficient With apparent flow rate A one-dimensional convection-diffusion equation describing the desorbed gas flow in a porous medium is established. Considering that the axial transport of gas in the adsorption bed is a linear time-invariant (LTI) system, the system undergoes a Laplace transform of this convection-diffusion partial differential equation to establish its transfer function in the complex frequency domain. :

[0047] in, For Laplace frequency domain variables, This represents the axial distance of gas transport.

[0048] Furthermore, the transfer function is used to perform deconvolution calculation on the residual sequence, specifically: The residual sequence is transformed to the frequency domain using either a Fast Fourier Transform (FFT) or a Laplace Transform. And perform division in the frequency domain (or use a Wiener filter algorithm with a regularization parameter to suppress high-frequency noise amplification), that is: calculate the frequency domain form of the true extracted signal:

[0049] Subsequently, it is restored to the time domain through inverse transformation, resulting in the corrected trailing segment.

[0050] S34, calculate the lag time of each moment within the trailing segment relative to the starting point of the trailing segment. By using a time-space mapping coefficient, the lag time is converted into depth coordinates within the adsorption bed. :

[0051] in, Indicates the first The depth coordinates corresponding to each sampling point Indicates the first The relative time interval between each sampling node and the tail start point.

[0052] Furthermore, by combining the volumetric flow rate of the desorbed gas stream, the concentration values ​​at each moment in the tailing stage are discretely integrated:

[0053] in, Representing depth coordinates The amount of high-boiling-point organic matter residue per unit mass of adsorbent within the cross-sectional area. This represents the index number of the integration step within the discrete integration time window, ranging from 0 to... , This represents the total number of discrete sampling points within the preset calculus time window. This indicates the corrected trailing segment at a specific time. discrete concentration values, This represents the overall volumetric flow rate of the system during the desorption phase. This represents the discrete time step in the high-frequency sampling mode of the sensor. The cross-sectional area of ​​the adsorption bed is... The bed depth slice thickness corresponds to the time step. denoted as the bulk density of the adsorbent.

[0054] By matching the depth coordinates with the corresponding residual amount, a spatial distribution map of the residual amount is generated.

[0055] Figure 7 This is a spatial distribution diagram of the residual amount of high-boiling-point organic matter in the adsorption bed in an embodiment of the present invention, such as... Figure 7 As shown, this figure displays the spatial distribution of high-boiling-point organic compound residues within the adsorption bed in three-dimensional voxel format. The X-axis represents the bed width, the Y-axis represents the bed height, and the Z-axis represents the bed depth (i.e., the direction of the desorption gas flow; Z = 0 cm is the inlet, and Z = 30 cm is the outlet). The color of each voxel unit indicates the amount of high-boiling-point organic compound residue (unit: mg / g adsorbent) at that spatial location; a more reddish color indicates a higher residue level. The color scale on the right side of the figure indicates the correspondence between the residue values ​​and the colors. Figure 7In the diagram, the dark red areas (mainly concentrated in the Z=10–20cm depth range and the center of the bed) represent localized high-risk areas where the residual amount exceeds the safety threshold (30mg / g), while the residual amounts at the air inlet (Z=0–5cm) and air outlet (Z=25–30cm) are relatively low. This spatial distribution map allows for a direct and precise location of the depth, lateral area, and severity of irreversible "poisoning" within the adsorption bed, providing clear spatial coordinate guidance for subsequent targeted deep desorption.

[0056] S4. When the spatial distribution map shows that the residual amount in a local area exceeds the preset threshold, targeted deep desorption is performed on that local area in the next adsorption-desorption cycle.

[0057] In this embodiment, the targeted depth desorption includes: After the system completes the spatiotemporal inversion of the entire bed, when the residual spatial distribution map shows that the residual amount of high-boiling-point organic matter in one or more local areas exceeds the preset safety threshold, the system will initiate a targeted deep desorption procedure for the area exceeding the standard in the next adsorption-desorption cycle.

[0058] Figure 6 This is a comparison diagram of the residual amount of the bed at different depths before and after targeted depth desorption in an embodiment of the present invention, such as... Figure 6 As shown, the horizontal axis represents bed depth (cm), and the vertical axis represents the residual amount of high-boiling-point organic matter (mg / g). Darker columns represent the residual amount before targeted desorption, and lighter columns represent the residual amount after targeted desorption. Figure 6 As shown, after targeted desorption, the residual amount in the 10–20 cm depth range decreased significantly from the excessive level (>30 mg / g) to below the safety threshold, while the residual amount in other depth ranges remained basically unchanged or only slightly decreased. This indicates that targeted depth desorption can precisely act on local areas with excessive residual amounts, avoiding over-desorption of already cleaned areas, thereby significantly reducing energy consumption while ensuring regeneration effectiveness.

[0059] The control system uses spatial coordinate mapping to precisely locate the independent gridded air valve unit corresponding to the air inlet of the adsorption bed, and automatically schedules and executes one of the following three desorption enhancement strategies (i.e., the physical implementation of "and / or") based on the severity of pollution in that local area: Strategy 1: When the residual amount exceeds the standard and the substance is a high activation energy component, the system increases the local intake air temperature by adjusting the output power of the proportional-integral-derivative (PID) controller of the electric heater in the corresponding independent air intake branch. The system's built-in target temperature calculation formula is as follows:

[0060] in, This indicates the increased temperature of the target desorbed gas. This indicates the material's highest safe critical temperature. This represents the reference desorption temperature under normal desorption conditions. This indicates the preset safe threshold for the amount of residue allowed in this area. This indicates the preset temperature compensation gain coefficient.

[0061] Strategy Two: When limited by the system's maximum heating power or explosion-proof temperature, the system employs dynamic enhancement measures. Specific actions include: throttling (reducing) the opening of the independent damper corresponding to the local area, lowering the apparent flow velocity, and / or, while dampers in other normal areas are closed and desorption is complete, keeping the damper in this local area open to extend the purging cycle. The formula for calculating the target extended residence time is:

[0062] in, This indicates the increased target dwell time. This represents the baseline residence time under normal desorption conditions. The time compensation weighting coefficient (dimensionless) is preferred to be 0.2 to 0.5.

[0063] Strategy 3: When When the levels are extremely high (e.g., exceeding the safety threshold by more than 3 times), posing a serious risk of "irreversible poisoning," the system simultaneously executes the two instructions mentioned above. That is, it both increases the temperature of the desorbed gas entering the local area and increases the residence time of the desorbed gas in the local area.

[0064] like Figure 5 As shown, the horizontal axis represents the number of adsorption-desorption cycles, and the vertical axis represents the average residual amount in the bed (normalized percentage). The upper curve in the figure shows the residual accumulation trend using the traditional fixed desorption process, and the lower curve shows the residual accumulation trend using the targeted depth desorption process of this invention. The horizontal dashed line represents the failure threshold (set as 85% of the normalized residual amount). Figure 5 As shown, the traditional process reaches the failure threshold after approximately 68 cycles, while the method of this invention extends the failure cycle count to approximately 94 cycles, increasing the adsorption bed lifespan by approximately 38%. Simultaneously, at any given number of cycles, the average residual amount in the bed layer using the method of this invention is significantly lower than that of the traditional process. These results demonstrate that this invention, by periodically monitoring the spatial distribution of residual amounts and performing targeted deep desorption on locally excessive areas, effectively inhibits the accumulation of high-boiling-point organic matter within the bed layer, significantly extending the lifespan of the adsorption material and reducing system operating energy consumption while maintaining treatment efficiency.

[0065] Example 2, as Figure 2 As shown, this embodiment provides an adsorption bed targeted deep desorption control system for waste gas treatment, including: The concentration curve acquisition module is used to acquire the organic matter concentration curve at the end of desorption. The tailing segment identification module is used to perform exponential decay fitting on the decay segment of the organic matter concentration curve, and take the difference between the organic matter concentration curve and the fitted curve as the tailing segment of the slow precipitation of the corresponding high-boiling-point organic matter. The residual amount inversion module is used to invert and calculate the residual amount of the high-boiling-point organic matter at different depths in the adsorption bed based on the concentration value of the tailing segment at different desorption times, and generate a spatial distribution map of the residual amount. The regeneration control module is used to perform targeted deep desorption of the local area in the next adsorption-desorption cycle when the spatial distribution map shows that the residual amount in a local area exceeds a preset threshold.

[0066] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0067] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0068] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0069] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0070] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0071] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for targeted deep desorption control of an adsorption bed for waste gas treatment, characterized in that, Includes the following steps: Obtain the organic matter concentration curve at the end of desorption; The decay segment of the organic matter concentration curve is fitted with an exponential decay method, and the difference between the organic matter concentration curve and the fitted curve is taken as the tail segment of the slow precipitation of the corresponding high-boiling-point organic matter. Based on the concentration values ​​of the tailing segment at different desorption times, the residual amount of the high-boiling-point organic matter at different depths within the adsorption bed is calculated by inversion, and a spatial distribution map of the residual amount is generated. When the spatial distribution map shows that the residual amount in a local area exceeds the preset threshold, targeted deep desorption is performed on that local area in the next adsorption-desorption cycle.

2. The method according to claim 1, characterized in that, The process of obtaining the organic matter concentration curve at the end of desorption includes: Extract the peak concentration of organic matter in the current desorption stage, and use the product of the peak concentration and the preset attenuation coefficient as the dynamic trigger threshold. When the concentration of organic matter drops to its dynamic trigger threshold, high-frequency sampling is initiated to obtain the concentration-time series from the start to the end of desorption; The concentration-time series was filtered and denoised to obtain the organic matter concentration curve.

3. The method according to claim 2, characterized in that, The process of performing exponential decay fitting on the decay segment of the organic matter concentration curve includes: Calculate the first derivative sequence of the organic matter concentration curve and extract the continuous time interval where the absolute value is greater than the preset rate threshold as the benchmark decay segment; The attenuation function is obtained by performing an exponential function least squares fit on the reference attenuation segment; The decay function is extrapolated in the time domain to the high-frequency sampling end node of the current desorption stage to generate a fitted curve.

4. The method according to claim 3, characterized in that, The sequence of first derivatives for calculating the organic concentration curve specifically includes: A sliding time window was set, and the Savitzky-Golay filtering method was used to perform local polynomial fitting on the discrete data points of the organic matter concentration curve within the window. The first derivative value of the center node of the sliding time window is extracted by analytical differentiation of the fitted polynomial. By sliding the time window through the organic concentration curve, the first derivative sequence is obtained.

5. The method according to claim 4, characterized in that, The method of using the difference between the organic concentration curve and the fitted curve as the tailing segment corresponding to the slow precipitation of high-boiling-point organic matter includes: Calculate the residuals between the organic matter concentration curve and the fitted curve at the same time point to obtain the residual sequence; Extract the standard deviation of the residuals within the baseline attenuation range and construct a dynamic error threshold by combining it with a pre-set confidence interval; Traverse the residual sequence along the time axis in the forward direction and take the first time node that continuously exceeds the dynamic error threshold as the tailing start point; The residual sequence between the starting point of the tail and the high-frequency sampling end node of the current desorption stage is extracted as the tail segment.

6. The method according to claim 5, characterized in that, The generation of the spatial distribution map of the residual amount includes: Obtain the apparent flow rate of the desorption gas and the porosity of the adsorption bed, and construct the time-space mapping coefficient; Calculate the lag time relative to the starting point of the tail segment at each moment, and convert the lag time into depth coordinates within the adsorption bed using the time-space mapping coefficient. By combining the volumetric flow rate of the desorbed gas stream, the concentration values ​​at each moment in the trailing section are discretely integrated to obtain the residual amount of high-boiling-point organic matter corresponding to each depth coordinate. By matching the depth coordinates with the corresponding residual amount, a spatial distribution map of the residual amount is generated.

7. The method according to claim 6, characterized in that, The process of obtaining the apparent flow rate of the desorption gas stream and the porosity of the adsorption bed includes: Collect the actual pressure difference between the inlet and outlet of the adsorption bed during the current desorption stage, and extract the initial pressure difference of the adsorption bed under the same desorption gas flow rate; Calculate the difference between the actual pressure difference and the initial pressure difference, and infer the current porosity change based on the difference; correct the initial porosity of the adsorption bed with the porosity change to obtain the corrected porosity.

8. The method according to claim 7, characterized in that, Before calculating the lag time of each moment within the trailing segment relative to the starting point of the trailing segment, the method further includes: The axial diffusion coefficient of the desorbed gas flow in the adsorption bed is obtained. Based on the axial diffusion coefficient and the apparent velocity, the transfer function of the one-dimensional convection-diffusion equation of the desorbed gas flow is established. The transfer function is used to perform deconvolution on the residual sequence, and the convergent residual sequence corresponding to the deconvolution calculation is truncated as the corrected tail segment.

9. The method according to claim 8, characterized in that, The targeted depth desorption includes: Increase the temperature of the desorbed gas entering the local area, and / or increase the residence time of the desorbed gas in the local area.

10. A system using the method as described in any one of claims 1-9, characterized in that, include: The concentration curve acquisition module is used to acquire the organic matter concentration curve at the end of desorption. The tailing segment identification module is used to perform exponential decay fitting on the decay segment of the organic matter concentration curve, and take the difference between the organic matter concentration curve and the fitted curve as the tailing segment of the slow precipitation of the corresponding high-boiling-point organic matter. The residual amount inversion module is used to invert and calculate the residual amount of the high-boiling-point organic matter at different depths in the adsorption bed based on the concentration value of the tailing segment at different desorption times, and generate a spatial distribution map of the residual amount. The regeneration control module is used to perform targeted deep desorption of the local area in the next adsorption-desorption cycle when the spatial distribution map shows that the residual amount in a local area exceeds a preset threshold.

Images