Method and system for monitoring and regulating decarburization emission reduction based on operating data

Abstract

The application discloses a normal-temperature decarburization emission reduction monitoring and regulation method and system based on operation data, relates to the related technical field of carbon emission monitoring, and comprises the following steps: collecting multiple period time series data of a normal-temperature decarburization device in a continuous operation process; extracting features to obtain a multi-dimensional feature vector representing the dynamic adsorption performance and regeneration efficiency of the adsorbent in a period; performing health degree recognition to obtain an adsorbent health degree value of each period, and establishing a time series composed of the health degree values of multiple periods; predicting the future change trend of the adsorbent health degree and calculating the remaining effective service life of the adsorbent; and adjusting the operation control parameters of the normal-temperature decarburization device. The application solves the technical problems that the performance degradation state of the adsorbent of the normal-temperature decarburization device cannot be quantitatively evaluated and trend predicted in real time in the prior art, and the adsorbent replacement timing and the regulation operation parameters cannot be accurately determined, and achieves the technical effect of improving the whole life cycle operation economy and carbon emission reduction comprehensive benefit of the normal-temperature decarburization device.

Description

Technical Field

[0001] This invention relates to the field of carbon emission monitoring technology, specifically to a method and system for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data. Background Technology

[0002] During long-term industrial operation, ambient temperature decarbonization units face adsorbent performance degradation. In repeated adsorption-regeneration cycles, the adsorbent undergoes irreversible aging and deactivation due to various mechanisms, including pore blockage, active site poisoning, structural damage caused by thermal stress, and accumulation of residual impurities. Performance degradation is influenced by a combination of factors, such as fluctuations in feed gas composition, deviations in regeneration conditions, and changes in ambient temperature and humidity. Once the adsorbent's health drops to a critical threshold, it directly leads to reduced decarbonization efficiency, substandard product gas purity, and a sharp increase in regeneration energy consumption, potentially even causing unplanned shutdowns. Currently, conventional management strategies for adsorbent performance degradation mainly rely on periodic maintenance and replacement, and post-fault troubleshooting, resulting in "over-maintenance," "under-maintenance," or production interruptions. Furthermore, existing monitoring and control methods lack quantitative perception of the real-time health status of the adsorbent during continuous cycles and cannot effectively predict future degradation trends based on operational data, thus failing to achieve a transition from "passive maintenance" to "active control."

[0003] Therefore, in the current related technologies, there is a technical problem that it is difficult to conduct real-time quantitative assessment and trend prediction of the performance degradation state of the adsorbent in the room temperature decarbonization device, which leads to the inability to accurately determine the timing of adsorbent replacement and control the operating parameters. Summary of the Invention

[0004] This application provides a method and system for monitoring and controlling carbon emission reduction at ambient temperature based on operational data. This solves the technical problem in the prior art that it is difficult to quantitatively assess and predict the performance degradation status of the adsorbent in ambient temperature decarbonization devices in real time, which leads to the inability to accurately determine the timing of adsorbent replacement and control operating parameters. This achieves the technical effect of improving the economic efficiency of ambient temperature decarbonization devices throughout their entire life cycle and the comprehensive benefits of carbon emission reduction.

[0005] This application provides a method for monitoring and controlling emission reduction in ambient temperature decarbonization based on operational data. The method includes: collecting time-series data from multiple cycles of an ambient temperature decarbonization device during continuous operation; extracting features from the multiple cycle time-series data to obtain a multi-dimensional feature vector characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within a cycle; identifying the health status of the adsorbent based on the multi-dimensional feature vector for each cycle to obtain the adsorbent health value for each cycle, and establishing a time series composed of the health values ​​from multiple cycles; predicting the future trend of the adsorbent health based on the time series and calculating the remaining effective lifespan of the adsorbent; and adjusting the operating control parameters of the ambient temperature decarbonization device according to the remaining effective lifespan.

[0006] In possible implementations, each cycle includes an adsorption phase and a regeneration phase.

[0007] In possible implementations, the multidimensional feature vector for each cycle includes the slope of the breakthrough curve of CO2 concentration change over time during the adsorption phase, the effective adsorption duration, the peak value of CO2 concentration at the outlet during the regeneration phase, the tailing area of ​​the CO2 concentration curve at the outlet during the regeneration phase, the average energy consumption of the regeneration process, and the peak width of the temperature change in the adsorption bed.

[0008] In a possible implementation, health identification is performed based on the multidimensional feature vector of each cycle to obtain the adsorbent health value for each cycle, including: collecting the fresh multidimensional feature vector of the adsorbent at the initial loading; comparing the multidimensional feature vector with the fresh multidimensional feature vector feature by feature to generate the adsorbent health value for each cycle.

[0009] In a possible implementation, the multidimensional feature vector is compared feature-by-feature with the fresh multidimensional feature vector to generate an adsorbent health value for each period. This includes: for each feature, calculating the deviation rate between the feature value in the multidimensional feature vector and the corresponding feature benchmark value in the fresh multidimensional feature vector; summing the deviation rates of all features according to preset weights, and then normalizing the sum to obtain the health value; wherein the preset weight of each feature is determined based on the sensitivity of the feature to the deterioration of adsorbent performance.

[0010] In a possible implementation, based on the time series, predicting the future trend of adsorbent health and calculating the remaining effective lifespan of the adsorbent includes: inputting the time series into a pre-constructed degradation trend prediction model, which outputs a sequence of predicted health values ​​for the next M periods; setting a failure health threshold to characterize adsorbent failure; finding the corresponding period number in the predicted health value sequence where the health value first decays to the failure health threshold; and calculating the remaining effective lifespan of the adsorbent, expressed in terms of periods, based on the difference between the corresponding period number and the current period number.

[0011] In one possible implementation, the degradation trend prediction model is a particle filter model, which is obtained by using the historical health time series to perform state-space modeling on the decay model characterizing the degradation of adsorbent performance.

[0012] In a possible implementation, the operating control parameters of the ambient temperature decarbonization device are adjusted based on the remaining effective lifespan, including: comparing the remaining effective lifespan with a preset emergency replacement threshold; if the remaining effective lifespan is less than the emergency replacement threshold, generating an adsorbent failure warning signal and switching to a standby adsorption tower.

[0013] In a possible implementation, the operating control parameters of the ambient temperature decarbonization device are adjusted based on the remaining effective lifespan, including: according to the total number of remaining operable cycles corresponding to the remaining effective lifespan and the health value of the current cycle, the time ratio of the adsorption stage and the regeneration stage in the ambient temperature decarbonization device is adjusted by calling a preset ratio database.

[0014] This application also provides a monitoring and control system for ambient temperature decarbonization emission reduction based on operational data. The system includes: a data acquisition module for acquiring time-series data of multiple cycles during continuous operation of the ambient temperature decarbonization device; a feature extraction module for extracting features from the multiple cycle time-series data to obtain a multi-dimensional feature vector characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within a cycle; a health identification module for identifying the health of the adsorbent based on the multi-dimensional feature vector of each cycle to obtain the health value of the adsorbent in each cycle and establishing a time series composed of the health values ​​of multiple cycles; an effective lifespan calculation module for predicting the future trend of the adsorbent health based on the time series and calculating the remaining effective lifespan of the adsorbent; and a control parameter adjustment module for adjusting the operating control parameters of the ambient temperature decarbonization device according to the remaining effective lifespan.

[0015] This application proposes a method and system for monitoring and controlling ambient temperature decarbonization emission reduction based on operational data. This involves collecting time-series data from multiple cycles of an ambient temperature decarbonization unit during continuous operation; extracting multi-dimensional feature vectors characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within each cycle; identifying the adsorbent health value for each cycle; establishing a time series composed of health values ​​from multiple cycles; predicting future trends in adsorbent health and calculating the remaining effective lifespan of the adsorbent; and adjusting the operating control parameters of the ambient temperature decarbonization unit. This method solves the technical problem in existing technologies where it is difficult to quantitatively assess and predict the performance degradation state of the adsorbent in real time, leading to an inability to accurately determine the timing of adsorbent replacement and adjust operating parameters. This achieves the technical effect of improving the economic efficiency and comprehensive carbon emission reduction benefits of the ambient temperature decarbonization unit throughout its entire life cycle. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1A schematic diagram of the process for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data, provided in an embodiment of this application.

[0018] Figure 2 A schematic diagram of the ambient temperature decarbonization and emission reduction monitoring and control system based on operational data provided in this application embodiment.

[0019] Figure labeling: Data acquisition module 10, feature extraction module 20, health status identification module 30, effective lifespan calculation module 40, control parameter adjustment module 50. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention in order to achieve the intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structures, features and effects of the present invention.

[0021] This application provides a method for monitoring and controlling room-temperature decarbonization and emission reduction based on operational data, such as... Figure 1 As shown, the method includes: Step S100: Collect time-series data for multiple cycles during continuous operation of the ambient temperature decarbonization device.

[0022] Step S100 further includes each cycle comprising an adsorption phase and a regeneration phase.

[0023] Preferably, in a room-temperature decarbonization device, the adsorbent cannot adsorb CO2 indefinitely. After adsorption saturation, its adsorption capacity must be restored through regeneration before it can enter the next round of adsorption. Specifically, the device starts with fresh adsorbent, performs a complete adsorption process, and then performs a regeneration process. This entire adsorption and regeneration process constitutes the first cycle. After the first cycle ends, a new round of adsorption begins immediately, followed by regeneration. This entire process constitutes the second cycle, and so on, generating the third, fourth, ..., Nth cycles sequentially during continuous operation. The adsorption stage refers to the process where CO2-containing flue gas, natural gas, syngas, etc., enter the device equipped with adsorbent... The adsorption tower / adsorption bed with adsorbent is where the adsorbent selectively adsorbs and retains CO2 within its pores and active sites at room temperature. Unadsorbed purified gas is discharged from the adsorption tower outlet. When the adsorbent's CO2 adsorption capacity approaches saturation, the CO2 concentration in the outlet gas begins to rise sharply, necessitating the cessation of adsorption and transition to the regeneration stage. The time-series data collected during this stage includes, but is not limited to, changes in CO2 concentration at the inlet / outlet of the adsorption tower over time, temperature changes measured by temperature sensors at different locations within the adsorption bed over time, changes in the pressure difference between the inlet and outlet of the adsorption tower over time, and changes in the instantaneous flow rate reading of the gas flow meter over time. The data from this stage reflects the current CO2 capture capacity of the adsorbent; the more severe the adsorbent aging, the earlier breakthrough occurs and the flatter the breakthrough curve.

[0024] Preferably, the regeneration stage refers to the process after the adsorption stage ends, where the feed gas is stopped and a regeneration medium, such as low-pressure steam, inert purge gas, or depressurized gas through vacuum, is introduced into the adsorption tower. Under ambient or slightly above-ambient temperature conditions, the adsorbed CO2 molecules on the adsorbent desorb and are released from the adsorbent surface and pores, exiting the adsorption tower outlet along with the regeneration medium. This process restores the adsorbent's adsorption capacity. The CO2 concentration in the outlet gas continuously decreases to a low and stable level, indicating that the adsorbent has been essentially regenerated completely. Regeneration is then stopped, and the system returns to the adsorption stage to begin the next cycle. The time-series data collected during this stage includes, but is not limited to, the change in CO2 concentration at the regeneration stage outlet over time, the instantaneous flow rate and temperature / pressure values ​​of the regeneration medium, the change in adsorption bed temperature over regeneration time, and the instantaneous total exhaust flow rate at the adsorption tower outlet during the regeneration stage. The data from the regeneration stage reflects the regeneration efficiency of the adsorbent. As the adsorbent ages, CO2 desorption becomes more difficult, the tailing area increases, and regeneration energy consumption rises.

[0025] Preferably, all sensor readings collected at a fixed sampling frequency (e.g., once per second) during the adsorption and regeneration phases of each of the 1st, 2nd, 3rd, ... up to the current operating moment are arranged in chronological order to form multiple time-series data, as shown in Table 1: Table 1. Time series data table for multiple periods

[0026] Step S200: Feature extraction is performed on the multiple periodic time series data to obtain a multidimensional feature vector characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within the period.

[0027] Step S200 further includes that the multidimensional feature vector for each cycle includes the slope of the breakthrough curve of CO2 concentration change over time during the adsorption stage, the effective adsorption time, the peak value of CO2 concentration at the outlet during the regeneration stage, the tailing area of ​​the CO2 concentration curve at the outlet during the regeneration stage, the average energy consumption of the regeneration process, and the peak width of the temperature change of the adsorption bed.

[0028] Preferably, feature extraction is performed on multiple time-series data periods to obtain multiple low-dimensional, highly generalizable statistical quantities or curve morphology parameters that can reflect the performance state of the adsorbent. Taking the "breakthrough curve slope" as an example, multiple data points are plotted on a two-dimensional coordinate system, and the curve segment from the breakthrough start point to the breakthrough end point is taken. Linear regression or logarithmic fitting is performed on this segment of data, and the slope value of the fitted line is calculated. Other features are calculated from the original time-series data in a similar manner. Finally, a multi-dimensional feature vector characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within the period is obtained, including the breakthrough curve slope of CO2 concentration changing with time during the adsorption stage, the effective adsorption time, the peak value of CO2 concentration at the outlet during the regeneration stage, the tail area of ​​the CO2 concentration curve at the outlet during the regeneration stage, the average energy consumption of the regeneration process, and the peak width of the temperature change in the adsorption bed. Among these, the larger the slope of the breakthrough curve of CO2 concentration changing with time during the adsorption stage, the stronger the breakthrough of CO2 in the adsorbent bed. The faster the "penetration" rate, the weaker the adsorbent's ability to capture CO2; the longer the effective adsorption time, the longer the adsorbent can work normally before penetration, and the better the adsorption performance; the higher the peak value of the CO2 concentration at the outlet during the regeneration stage, the greater the adsorption capacity in that cycle; the tailing area of ​​the CO2 concentration curve at the outlet during the regeneration stage reflects the degree of "delayed release" of CO2 in the later stage of regeneration. The more severe the aging of the adsorbent or the blockage of the pores, the slower and longer the CO2 desorption, and the larger the tailing area; the average energy consumption of the regeneration process reflects the operating cost required to regenerate the adsorbent in that cycle. The more severe the aging of the adsorbent, the greater the energy input required for CO2 desorption, and the higher the average energy consumption; the peak width of the temperature change in the adsorption bed reflects the duration and intensity of the temperature change. The temperature peak width is affected by both the adsorption heat release rate and the bed heat transfer conditions. The higher the adsorbent activity and the more concentrated the adsorption, the narrower and higher the temperature peak. After the activity decreases, the adsorption heat release is dispersed, and the temperature peak becomes wider and flatter.

[0029] Step S300: Based on the multidimensional feature vector of each period, health status is identified to obtain the adsorbent health status value of each period, and a time series composed of health status values ​​of multiple periods is established.

[0030] Step S300 further includes collecting the fresh multidimensional feature vector of the adsorbent at the initial loading; comparing the multidimensional feature vector with the fresh multidimensional feature vector feature by feature to generate the adsorbent health value for each cycle.

[0031] Preferably, when the ambient temperature decarbonization device is put into operation for the first time and the adsorbent is in a brand new state, a complete cycle is run, and a fresh multidimensional feature vector is collected and extracted to reflect the optimal performance level of the batch of adsorbent in the "zero decay state". For each current cycle, its multidimensional feature vector is compared with the fresh multidimensional feature vector feature by feature, and the deviation of the current value of each feature dimension from the fresh baseline value is calculated to generate the adsorbent health value for each cycle. The adsorbent health values ​​of the 1st cycle, the 2nd cycle, the 3rd cycle, ... up to the latest operating cycle are arranged in cyclical order to establish a time series composed of health values ​​of multiple cycles, which directly reflects the complete degradation trajectory of the adsorbent from fresh loading to the current moment, as its health gradually decreases with the increase of the number of cycles.

[0032] Furthermore, step S300 also includes, for each feature, calculating the deviation rate between the feature value in the multidimensional feature vector and the corresponding feature benchmark value in the fresh multidimensional feature vector; summing the deviation rates of all features according to preset weights, and obtaining the health value after normalization; wherein, the preset weight of each feature is determined based on the sensitivity of the feature to the deterioration of adsorbent performance.

[0033] Preferably, the deviation rate between the eigenvalue of each feature in the multidimensional feature vector and the corresponding baseline value of the fresh multidimensional feature vector is calculated, where the deviation rate = (eigenvalue of the multidimensional feature vector - baseline eigenvalue of the fresh multidimensional feature vector) / baseline eigenvalue of the fresh multidimensional feature vector. A preset weight is determined for each feature based on its sensitivity to adsorbent performance degradation. For example, the penetration slope sensitivity weight is 0.25, the effective duration sensitivity weight is 0.2, the regeneration peak sensitivity weight is 0.15, the tailing area sensitivity weight is 0.2, the regeneration energy consumption sensitivity weight is 0.1, and the temperature peak width sensitivity weight is 0.1. Then, the six deviation rates are multiplied by their corresponding weights, summed, and normalized to obtain the health value. For example, if the overall health of the adsorbent in the Nth cycle is 78%, it means that the current overall performance of the adsorbent is 78% of its fresh state, and the cumulative performance degradation loss is 22%.

[0034] Step S400: Based on the time series, predict the future trend of the adsorbent's health and calculate the remaining effective lifespan of the adsorbent.

[0035] Step S400 further includes: inputting the time series into a pre-constructed degradation trend prediction model, which outputs a health prediction value sequence for the next M periods; setting a failure health threshold to characterize adsorbent failure; finding the corresponding period number in the health prediction value sequence where the health value first decays to the failure health threshold; and calculating the remaining effective lifespan of the adsorbent in terms of the number of periods based on the difference between the corresponding period number and the current period number.

[0036] Preferably, historical health time series data is used as input data and input into a pre-constructed degradation trend prediction model. This model, based on the historical health variation with the number of cycles, calculates the predicted health values ​​for the (N+1), (N+2), ..., (N+M)th cycles that have not yet occurred, determining a sequence of predicted health values ​​for the next M cycles. According to the process requirements and product quality standards of the ambient temperature decarbonization device, a failure health threshold characterizing adsorbent failure is pre-set. When the adsorbent's health value falls below this threshold, it is considered that the adsorbent can no longer meet normal decarbonization requirements and has reached a "failure" state. The process requirements include an economic boundary (the health value corresponding to when a decline in health leads to regeneration energy consumption exceeding the economic cost limit) and a safe operation boundary (the health value at which a decline in health leads to...). The device is unable to set the health value for effective regeneration within a cycle. The product quality standard is the health value corresponding to the point where the CO2 concentration in the purified gas cannot be continuously lower than the product gas specification (e.g., CO2 ≤ 2.5% in natural gas) due to a decline in health. Then, in the health prediction value sequence, find the predicted value where the health value first decays to the failure health threshold, and record the cycle number corresponding to the predicted value, thus determining the corresponding cycle number. Then, subtract the number of cycles that have been run from the number of failure cycles to obtain the remaining effective lifespan of the adsorbent. The remaining number of operable cycles is the number of complete cycles that the adsorbent can continue to operate normally according to the current degradation trend of the adsorbent. The fewer the remaining cycles, the closer the adsorbent is to the end of its lifespan, and it is necessary to activate the warning, adjust the operating conditions, or arrange for replacement.

[0037] Furthermore, step S400 also includes the following: the degradation trend prediction model is a particle filter model, which is obtained by using the historical health time series to perform state-space modeling on the decay model characterizing the degradation of adsorbent performance.

[0038] Preferably, particle filtering is used, employing an exponential decay model, a double exponential decay model, or a power-law decay model to describe the adsorbent performance degradation process as a mathematical function. The parameters in this model are estimated using historical health time-series data, transforming the decay model into a state-space model suitable for recursive filtering. This includes a state equation describing the evolution of the adsorbent's true health over time and an observation equation describing the relationship between the measured health value and the true health value. This yields a degradation trend prediction model that can be used to predict future health trends, describing the decrease in adsorbent health with increasing cycle number. Particle filtering is based on Monte Carlo simulation-based state estimation. The algorithm, the recursive calculation process of particle filtering, includes generating 1000 to 5000 particles, each representing possible model parameters and the current health status value. Each particle is assigned the same initial weight. The state of each particle is pushed forward one step according to the state equation to obtain the states of all particles at the next time step. After a new actual health value is measured, the likelihood probability between the predicted value and the actual measured value of each particle is calculated, and the weight of each particle is adjusted accordingly, with particles that predict more accurately having a larger weight. Particles are re-extracted based on their weights; particles with large weights are copied, and particles with small weights are discarded, keeping the total number of particles unchanged. Each time a new actual health value is obtained, the process of gradually approximating the true degradation trajectory parameters is repeated. After the recursion is completed, the weighted average of the current states of all particles is used to calculate and determine the sequence of predicted health values ​​for the next N+1 to N+M periods.

[0039] Step S500: Adjust the operating control parameters of the ambient temperature decarbonization device according to the remaining effective lifespan.

[0040] Preferably, based on the calculated remaining effective lifespan of the adsorbent, the operator automatically or assisted in adjusting the operating parameters of the ambient temperature decarbonization unit. The shorter the remaining effective lifespan, the closer the adsorbent is to the end of its lifespan. Appropriately sacrificing some decarbonization efficiency in exchange for a longer operating time can extend the adsorbent's service life, reduce the replacement frequency, and strengthen operating conditions to ensure that the product gas quality continues to meet the standards or trigger emergency protection actions within the remaining lifespan.

[0041] Furthermore, step S500 also includes comparing the remaining effective lifespan with a preset emergency replacement threshold; if the remaining effective lifespan is less than the emergency replacement threshold, an adsorbent failure warning signal is generated, and the system switches to a backup adsorption tower.

[0042] Preferably, an emergency replacement threshold is pre-set based on the product gas quality safety margin and the time required for switching operations, multiplied by a safety margin coefficient. This threshold, measured in cycles, represents the safety baseline for the remaining lifespan of the adsorbent. When the remaining effective lifespan falls below this threshold, it indicates that the risk of continuing to operate the current adsorption tower is unacceptable, and the adsorbent in the tower must be stopped immediately. There is a certain "inertial" delay between the time the health reaches the failure threshold and the actual occurrence of CO2 penetration and substandard product gas. The emergency replacement threshold serves as a safety buffer reserved within this delay space, ensuring that the switching is completed before penetration actually occurs. From issuing the switching command to the standby tower completing preheating, valve activation, and system re-stabilization, a certain amount of time is required, for example, 1.5 cycles. The emergency replacement threshold must be greater than the number of cycles corresponding to this operation time to ensure sufficient time for the switching to complete. The safety margin coefficient may be 1.2 to 1.5 to avoid switching failure due to uncontrollable factors such as sensor delays and communication delays. For example, setting the emergency replacement threshold to 2 cycles means that the current adsorbent can safely operate for a maximum of 2 complete cycles; exceeding this limit may result in excessive product CO2 concentration. If the remaining effective lifespan is less than the emergency replacement threshold, an adsorbent failure warning signal is generated and sent to the operation terminal. The alarm level is either "Severe Alarm" or "Emergency Alarm". A red flashing dialog box pops up on the operation station screen and a buzzer / siren sounds. The adsorbent failure warning signal contains specific information, such as "The remaining lifespan of the adsorbent in tower A is 1 cycle, which is less than the emergency replacement threshold of 2 cycles. It is recommended to switch to tower B immediately." At the same time, the feed gas flow path is switched from the currently operating adsorbent tower to the standby adsorbent tower, so that the standby tower takes over the decarbonization treatment. The current tower is then safely isolated and put into a shutdown state. The inlet and outlet valves of the standby tower must be opened first, and then the inlet and outlet valves of the current tower must be closed to ensure that the feed gas is not interrupted and the product gas pipeline does not experience a sudden pressure drop during the switching process.

[0043] Furthermore, step S500 also includes adjusting the time ratio of the adsorption stage and the regeneration stage in the ambient temperature decarbonization device by calling a preset ratio database based on the total number of remaining operable cycles corresponding to the remaining effective lifespan and the health value of the current cycle.

[0044] Preferably, the preset ratio database is a lookup table established through experimental calibration, historical operating data statistics, or simulation calculations. This database uses "remaining effective lifespan interval" and "health interval" as two input dimensions, and "adsorption time adjustment coefficient" and "regeneration time adjustment coefficient" as two output dimensions, storing recommended time ratio adjustment schemes under different adsorbent aging states. The total number of remaining operable cycles corresponding to the remaining effective lifespan and the health value of the current cycle are used as query indexes to search and match in the preset ratio database, obtain records of the corresponding intervals, and read the adsorption time adjustment coefficient and regeneration time adjustment coefficient. These are then multiplied by the currently set adsorption stage duration benchmark value and regeneration stage duration benchmark value, respectively, to obtain new duration settings. This determines the time ratio of the adsorption stage and regeneration stage in the ambient temperature decarbonization device and sends it to the timing controller for execution. The preset ratio database is dynamically updated. As the remaining effective lifespan gradually shortens and the health gradually decreases, the corresponding adsorption time adjustment coefficient gradually decreases (adsorption time becomes shorter) and the regeneration time adjustment coefficient gradually increases (regeneration time becomes longer), forming an adaptive adjustment trajectory that gradually tightens as the aging degree deepens. When there is still some margin in the remaining effective lifespan, the health degradation is slowed down by continuous adjustment, and the output is maximized within the remaining lifespan. When the remaining effective lifespan is exhausted to the safety limit, a safe shutdown is achieved by discrete switching. This forms a gradient control strategy from "slow" to "urgent", thereby ensuring the improvement of the economic efficiency and comprehensive carbon emission reduction benefits of the ambient temperature decarbonization unit throughout its entire life cycle.

[0045] In the above text, refer to Figure 1 This paper describes in detail a method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data, according to embodiments of the present invention. Next, reference will be made to... Figure 2 A monitoring and control system for ambient temperature decarbonization and emission reduction based on operational data, according to an embodiment of the present invention, is described.

[0046] The ambient temperature decarbonization emission reduction monitoring and control system based on operational data, according to embodiments of the present invention, solves the technical problem in the prior art that it is difficult to quantitatively assess and predict the performance degradation state of the adsorbent in ambient temperature decarbonization devices in real time, leading to the inability to accurately determine the timing of adsorbent replacement and control operating parameters. This achieves the technical effect of improving the economic efficiency and comprehensive carbon emission reduction benefits of the ambient temperature decarbonization device throughout its entire life cycle. Figure 2 As shown, the ambient temperature decarbonization emission reduction monitoring and control system based on operational data includes: a data acquisition module 10, a feature extraction module 20, a health identification module 30, an effective lifespan calculation module 40, and a control parameter adjustment module 50.

[0047] The data acquisition module 10 is used to acquire time-series data of multiple cycles during the continuous operation of the ambient temperature decarbonization device; the feature extraction module 20 is used to extract features from the multiple cycle time-series data to obtain a multi-dimensional feature vector characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within a cycle; the health identification module 30 is used to identify the health of the adsorbent based on the multi-dimensional feature vector of each cycle to obtain the health value of the adsorbent in each cycle and establish a time series composed of the health values ​​of multiple cycles; the effective lifespan calculation module 40 is used to predict the future trend of the health of the adsorbent based on the time series and calculate the remaining effective lifespan of the adsorbent; and the control parameter adjustment module 50 is used to adjust the operating control parameters of the ambient temperature decarbonization device according to the remaining effective lifespan.

[0048] The specific configuration of the data acquisition module 10 will be described in detail below. The data acquisition module 10 further includes an adsorption phase and a regeneration phase in each cycle.

[0049] The specific configuration of the feature extraction module 20 will be described in detail below. The feature extraction module 20 further includes: a multi-dimensional feature vector for each cycle, including the slope of the breakthrough curve of CO2 concentration change over time during the adsorption stage, the effective adsorption time, the peak value of CO2 concentration at the outlet during the regeneration stage, the tailing area of ​​the CO2 concentration curve at the outlet during the regeneration stage, the average energy consumption of the regeneration process, and the peak width of the temperature change of the adsorption bed.

[0050] The specific configuration of the health identification module 30 will be described in detail below. The health identification module 30 further includes: collecting the fresh multidimensional feature vector of the adsorbent at the initial loading; comparing the multidimensional feature vector with the fresh multidimensional feature vector feature by feature to generate the adsorbent health value for each cycle.

[0051] The specific configuration of the health identification module 30 will be described in detail below. The health identification module 30 further includes: for each feature, calculating the deviation rate between the feature value in the multidimensional feature vector and the corresponding feature benchmark value in the fresh multidimensional feature vector; summing the deviation rates of all features according to preset weights, and obtaining the health value after normalization; wherein, the preset weight of each feature is determined based on the sensitivity of the feature to the deterioration of adsorbent performance.

[0052] The specific configuration of the effective lifespan calculation module 40 will be described in detail below. The effective lifespan calculation module 40 further includes: inputting the time series into a pre-built degradation trend prediction model, which outputs a sequence of predicted health values ​​for the next M periods; setting a failure health threshold characterizing adsorbent failure; finding the corresponding period number in the predicted health value sequence where the health value first decays to the failure health threshold; and calculating the remaining effective lifespan of the adsorbent, expressed in periods, based on the difference between the corresponding period number and the current period number.

[0053] The specific configuration of the effective lifetime calculation module 40 will be described in detail below. The effective lifetime calculation module 40 further includes: the degradation trend prediction model is a particle filter model, which is obtained by performing state-space modeling on the decay model characterizing the degradation of adsorbent performance using historical health time series.

[0054] The specific configuration of the control parameter adjustment module 50 will be described in detail below. The control parameter adjustment module 50 further includes: comparing the remaining effective lifespan with a preset emergency replacement threshold; if the remaining effective lifespan is less than the emergency replacement threshold, generating an adsorbent failure warning signal and switching to the standby adsorption tower.

[0055] The specific configuration of the control parameter adjustment module 50 will be described in detail below. The control parameter adjustment module 50 further includes: adjusting the time ratio of the adsorption stage and the regeneration stage in the ambient temperature decarbonization device by calling a preset ratio database based on the total number of remaining operable cycles corresponding to the remaining effective lifespan and the health value of the current cycle.

[0056] The ambient temperature decarbonization emission reduction monitoring and control system based on operational data provided in this embodiment of the invention can execute the ambient temperature decarbonization emission reduction monitoring and control method based on operational data provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0057] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data, characterized in that, include: Collect time-series data from multiple cycles during the continuous operation of the ambient temperature decarbonization unit; Feature extraction is performed on the multiple periodic time series data to obtain a multidimensional feature vector characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within the period. Based on the multidimensional feature vector of each period, the health status is identified to obtain the adsorbent health status value of each period, and a time series composed of the health status values ​​of multiple periods is established. Based on the time series, the future trend of adsorbent health is predicted, and the remaining effective lifespan of the adsorbent is calculated. Based on the remaining effective lifespan, adjust the operating control parameters of the ambient temperature decarbonization device.

2. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 1, characterized in that, Each cycle consists of an adsorption phase and a regeneration phase.

3. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 1, characterized in that, The multidimensional feature vector for each cycle includes the slope of the breakthrough curve of CO2 concentration change over time during the adsorption stage, the effective adsorption time, the peak value of CO2 concentration at the outlet during the regeneration stage, the tailing area of ​​the CO2 concentration curve at the outlet during the regeneration stage, the average energy consumption of the regeneration process, and the peak width of the temperature change in the adsorption bed.

4. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 1, characterized in that, Based on the multidimensional feature vector of each cycle, health status is identified to obtain the adsorbent health value for each cycle, including: Collect the fresh multidimensional feature vector of the adsorbent at the initial loading stage; The multidimensional feature vector is compared feature by feature with the fresh multidimensional feature vector to generate the adsorbent health value for each cycle.

5. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 4, characterized in that, The multidimensional feature vector is compared feature-by-feature with the fresh multidimensional feature vector to generate an adsorbent health value for each cycle, including: For each feature, calculate the deviation rate between the feature value in the multidimensional feature vector and the corresponding feature benchmark value in the fresh multidimensional feature vector; The health value is obtained by summing the deviation rates of all features according to preset weights and then normalizing the sum. The preset weight of each feature is determined based on the sensitivity of the feature to the deterioration of adsorbent performance.

6. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 1, characterized in that, Based on the time series, the future trend of adsorbent health is predicted, and the remaining effective lifetime of the adsorbent is calculated, including: The time series is input into a pre-built degradation trend prediction model, which outputs a sequence of health prediction values ​​for the next M periods. Set a failure health threshold to characterize adsorbent failure; In the health prediction value sequence, find the number of cycles corresponding to when the health value first decays to the failure health threshold; Based on the difference between the corresponding cycle number and the current cycle number, the remaining effective lifespan of the adsorbent, expressed in terms of cycle number, is calculated.

7. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 6, characterized in that, The degradation trend prediction model is a particle filter model, which is obtained by using the historical health time series to perform state-space modeling on the decay model that characterizes the degradation of adsorbent performance.

8. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 1, characterized in that, Based on the remaining effective lifespan, adjust the operating control parameters of the ambient temperature decarbonization device, including: The remaining effective lifespan is compared with a preset emergency replacement threshold; If the remaining effective lifespan is less than the emergency replacement threshold, an adsorbent failure warning signal is generated, and the system switches to the backup adsorption tower.

9. The method for monitoring and controlling ambient temperature decarbonization and emission reduction based on operational data as described in claim 8, characterized in that, Based on the remaining effective lifespan, adjust the operating control parameters of the ambient temperature decarbonization device, including: Based on the total number of remaining operational cycles corresponding to the remaining effective lifespan and the health value of the current cycle, the time ratio of the adsorption stage and the regeneration stage in the ambient temperature decarbonization device is adjusted by calling the preset ratio database.

10. A room-temperature decarbonization and emission reduction monitoring and control system based on operational data, characterized in that, The system is used to implement the ambient temperature decarbonization and emission reduction monitoring and control method based on operational data as described in any one of claims 1 to 9, the system comprising: The data acquisition module is used to collect time-series data from multiple cycles during the continuous operation of the ambient temperature decarbonization unit. The feature extraction module is used to extract features from the multiple periodic time series data to obtain a multidimensional feature vector characterizing the dynamic adsorption performance and regeneration efficiency of the adsorbent within the period. The health identification module is used to identify the health status based on the multidimensional feature vector of each period, obtain the adsorbent health status value of each period, and establish a time series composed of health status values ​​of multiple periods. The effective lifespan calculation module is used to predict the future trend of adsorbent health based on the time series and calculate the remaining effective lifespan of the adsorbent. The control parameter adjustment module is used to adjust the operating control parameters of the ambient temperature decarbonization device based on the remaining effective lifespan.

Images