Polluted groundwater source controlling and breaking system suitable for limited space of industrial accumulation area

Abstract

The invention relates to the technical field of soil and groundwater pollution remediation, in particular to a polluted groundwater source controlling and breaking system suitable for limited space of an industrial accumulation area, which comprises a main well, a gas stripping circulating unit and a groundwater circulating unit. A blocking device composed of a packing rubber structure and a driving device is arranged in the main well, the packing rubber structure is driven by the driving device to move and intermittently inject air to expand, and an air flow channel which is dynamically opened and closed is formed between the inner wall of the main well and the unsaturated area. The vacuum pump device builds negative pressure on the upper portion of the main well, so that purified clean gas enters an unsaturated area in an intermittent pulse mode through the peripheral injection well, and soil gas is induced to flow in an oscillating mode under cooperation of the dynamic gas flow channel. Groundwater extracted by the centrifugal pump is purified by the gas stripping area and the underground activated carbon filtering device and then is discharged into the top of the unsaturated area through the drainage device in an intermittent wave mode, the system is suitable for pollution treatment in the limited space of the industrial accumulation area, and targeted source control and long-acting reduction are achieved.

Description

Technical Field

[0001] This invention relates to the field of soil and groundwater pollution remediation technology, and in particular to a pollution groundwater source control and cutoff system suitable for confined spaces in industrial clusters. Background Technology

[0002] Groundwater circulation well technology, as a method for treating and remediating underground pollutants, uses mechanical pumping and injection of water within the well to create a local head difference, driving the vertical circulation of groundwater within the aquifer, thereby achieving the migration and removal of pollutants in contaminated water and soil. Traditional circulation well structures typically only have upper and lower screen pipes, relying on hydraulic pressure differences to disturb the aquifer, which has a certain removal effect on soluble pollutants in the saturation zone. However, contaminated sites left over from industrial clusters often have distinct site characteristics: compact layout of production facilities, crisscrossing underground pipe networks, and dense foundations of buildings and structures, making it difficult to access conventional remediation equipment and limiting the deployment of work areas. At the same time, after long-term migration, pollutants are unevenly distributed in the vadose zone, groundwater level fluctuation zone, and saturation zone, especially with significant vertical concentration differentiation. Residual pollutants in unsaturated media are difficult to effectively contact by the hydraulic drive of traditional circulation wells due to capillary action and adsorption effects. Furthermore, existing remediation technologies mostly employ a single mechanism of action, such as simple aeration, chemical injection, or microbial addition. For the multiphase and complex pollutants commonly found in industrial sites, the remediation efficacy of a single technology is limited, and it is difficult to differentiate control based on different pollution depths. Addressing the pollution control needs in these confined spaces, how to achieve synergistic remediation of the vadose zone and saturation zone, and establish a hydraulic control mechanism that can dynamically adapt to the vertical distribution of pollution, are urgent technical problems to be solved in this field. Summary of the Invention

[0003] This invention overcomes the shortcomings of groundwater circulation wells in risk management and remediation of polluted groundwater, and provides a pollution source control and cutoff system for confined spaces in industrial clusters.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: This invention discloses a groundwater pollution source control and interruption system suitable for confined spaces in industrial clusters, the system comprising: The main shaft is vertically installed in the contaminated site, penetrating both the unsaturated and saturated zones; Peripheral injection wells are located beside the main well; The ground treatment unit includes a vacuum pump device and a ground activated carbon filter device connected in sequence by pipelines; An air-lift circulation unit is installed at the top of the main well to inject air into the main well for aeration, thereby promoting the release of volatile organic compounds in polluted groundwater. The groundwater circulation unit includes a barrier and a centrifugal pump installed in the main well. The centrifugal pump is used to pump groundwater from the lower part of the barrier to the upper part of the barrier to form a local water level difference. in, The vacuum pump device is used to establish negative pressure in the upper part of the main well and the unsaturated area to extract soil gas containing volatile organic compounds and transport it to the ground activated carbon filter device for purification. The purified gas is pressurized and reinjected into the unsaturated area and / or groundwater level fluctuation zone through the peripheral injection well to form a soil gas circulation. The groundwater extracted by the centrifugal pump is purified through the air-lift zone and the underground activated carbon filtration device, and then reinjected into the unsaturated area around the main well through the drainage device, forming a groundwater cycle.

[0005] Furthermore, the vacuum pump device establishes a negative pressure field in the unsaturated region ranging from -0.04 MPa to -0.095 MPa.

[0006] Furthermore, the barrier includes: The sealing rubber structure has an air injection port, which can expand by injecting air to change the contact state with the main well wall; A driving device, connected to the sealing rubber structure, is used to drive the sealing rubber structure to move along the central axis of the main well's central pipe. Through the deformation and displacement of the sealing rubber structure, the main well is in either a closed or unclosed working state.

[0007] Furthermore, the filler material of the ground activated carbon filter device is industrial-grade granular activated carbon or honeycomb activated carbon with a surface area of ​​1000-1500 m² / g; the ground activated carbon filter device adopts a negative pressure interface and is connected to the vacuum pump device through a flange.

[0008] Furthermore, the peripheral injection well has the same depth as the main well, and its bottom is equipped with a sedimentation pipe for injecting chemical agents, bacterial agents, or biological additives.

[0009] Furthermore, the air-lift zone is located at the top of the main well and is used to aerate the extracted groundwater to separate volatile organic compounds.

[0010] Furthermore, the drainage device is located at the top of the unsaturated zone to uniformly discharge the purified groundwater into the unsaturated zone surrounding the main well, flushing and removing pollutants from the area.

[0011] Furthermore, the aforementioned groundwater pollution source control and interruption system suitable for confined spaces in industrial clusters also includes: Based on the pressure, water level, and volatile organic compound concentration data collected at different depths in the main well, the dynamic range of the groundwater level fluctuation zone and the high concentration accumulation points at the boundary between the unsaturated zone and the saturated zone were determined. Based on the depth of the identified high-concentration accumulation points, the drive device of the control barrier moves the sealing rubber structure to the corresponding depth and intermittently injects compressed gas into the gas injection port of the sealing rubber structure, causing it to expand and contract periodically, forming a dynamically opening and closing airflow channel between the inner wall of the main well and the unsaturated zone. The vacuum pump device is started to establish a negative pressure field at the top of the main well, and the purified clean gas is injected into the unsaturated zone in an intermittent pulse manner through the gas flow regulating valve in the outer well. With the cooperation of the dynamically opening and closing airflow channel, the soil gas is induced to generate an oscillating flow in the unsaturated medium. Based on the decay rate of volatile organic compound concentration at the outlet of the air-lift zone, the start-stop cycle and extraction flow rate of the centrifugal pump are adjusted so that the purified reinjected water is discharged into the top of the unsaturated zone in the form of intermittent waves through the drainage device, forming a pulsating flushing of residual pollutants in the unsaturated medium. Based on the circulating water parameters collected at the outlet of the drainage device and the inlet of the filter pipe at the bottom of the main well, coordinated adjustment commands are generated for the isolation status of the barrier, the pulse frequency of the peripheral injection well, and the extraction cycle of the centrifugal pump. This causes the pulse frequencies of the soil gas circulation and the groundwater circulation to be staggered, forming an alternating oscillating three-dimensional mass transfer field.

[0012] Furthermore, based on pressure, water level, and volatile organic compound (VOC) concentration data collected at different depths within the main well, the dynamic range of groundwater level fluctuation zones and the locations of high-concentration accumulation points at the boundary between the unsaturated and saturated zones were determined, specifically: Based on the original time-series monitoring data synchronously collected by multiple pressure sensors, water level sensors and volatile organic compound sensor arrays deployed at different elevations on the inner wall of the main well, and after spatiotemporal alignment and wavelet transform processing of the original time-series monitoring data, a corrected effective dataset is generated. Extract the water level time series of each monitoring point from the corrected effective dataset, count the cumulative number of times the groundwater level crosses the depth position of each monitoring point within a unit of time, and define the depth interval corresponding to the depth point where the cumulative number exceeds the preset fluctuation frequency threshold as the dynamic interval of the groundwater level fluctuation zone and its main fluctuation frequency. Within the dynamic range of the groundwater level fluctuation zone, the time series of volatile organic compound concentrations at each monitoring point within the range are retrieved. Coherence analysis is performed on the concentration time series of each monitoring point and the main control fluctuation frequency. The coherence function value between the concentration change at each depth point and the water level fluctuation is calculated. The depth location where the coherence function value reaches a local maximum and remains stable is identified as a potential high-concentration accumulation nucleus. Centered on the potential high-concentration aggregation core, extend an unsaturated capillary water thickness range upwards and downwards to construct a spatial analysis window containing multiple adjacent monitoring points, and calculate the spatial autocorrelation Moran index and local spatial correlation index of the concentration values ​​of all monitoring points within the window. The depth locations where the spatial autocorrelation Moran index is greater than a set threshold and the local spatial correlation index reaches a local maximum are determined by combining the phase lag time of the concentration time series at the depth location with respect to the water level fluctuation, and the high concentration accumulation points at the boundary between the unsaturated zone and the saturated zone driven by the periodic fluctuation of the groundwater level, as well as the vertical migration time lag characteristics of the pollution plume.

[0013] Furthermore, based on the decay rate of volatile organic compound concentration at the outlet of the air-lift zone, the start-stop cycle and extraction flow rate of the centrifugal pump are adjusted so that the purified reinjected water is discharged into the top of the unsaturated zone in an intermittent wave form through the drainage device, forming a pulsating flushing of residual pollutants in the unsaturated medium. Specifically: Real-time concentration data of volatile organic compounds at the outlet of the gas stripping zone are continuously collected. The concentration time series within a preset time window is extracted. After the concentration time series is logarithmically transformed, linear regression fitting is performed with the corresponding timestamp to generate the logarithmic domain decay slope of the volatile organic compound concentration as the instantaneous apparent decay rate constant. The instantaneous apparent decay rate constant is processed by exponential weighted moving average to obtain a dynamic smooth value of decay rate. The dynamic smooth value is then compared with a preset decay rate constant threshold range to identify the period when the dynamic smooth value is continuously lower than the lower limit of the threshold range as the mass transfer-limited bottleneck period. During the mass transfer bottleneck period, the reference pulse period and reference pulse flow of the current centrifugal pump are retrieved, and the ratio of the reference pulse period to the dynamic smoothing value is calculated to generate a mass transfer resistance coefficient that characterizes the resistance to pollutant release in the unsaturated medium. The mass transfer resistance coefficient is then input into a pre-stored pulse parameter response surface for interpolation query to generate a corrected pulse width and corrected pulse amplitude that match the current mass transfer resistance. Based on the corrected pulse width and corrected pulse amplitude, the centrifugal pump is controlled to output unsteady flow in an intermittent manner, so that when the reinjected water purified by the air-lift zone and the underground activated carbon filtration device is discharged into the top of the unsaturated zone through the drainage device, an asymmetric intermittent wave sequence with steep wave front and gentle wave back is formed in the unsaturated medium. Simultaneously collect soil volumetric water content change data at different depths in the unsaturated zone, and extract the wavefront arrival time and waveback saturation duration of the asymmetric intermittent wave sequence at each depth based on the soil volumetric water content change data; Spatial interpolation is performed on the wavefront arrival time to generate a vertical propulsion velocity field of pulsating water flow in an unsaturated medium. The vertical propulsion velocity field is then multiplied with the waveback saturation duration to generate an effective scour depth characterizing the effective desorption volume of pollutants. The effective scouring depth is compared with a preset target depth threshold, and an adaptive correction instruction for the pulse parameter response surface interpolation reference point is generated based on the comparison result.

[0014] Furthermore, based on the circulating water parameters collected at the outlet of the drainage device and the inlet of the filter pipe at the bottom of the main well, coordinated adjustment commands are generated for the isolation status of the barrier, the pulse frequency of the peripheral injection well, and the extraction cycle of the centrifugal pump. This causes the pulse frequencies of the soil gas circulation and the groundwater circulation to be staggered, forming an alternating oscillating three-dimensional mass transfer field, specifically: The instantaneous parameters of reinjected water are collected synchronously by the flow sensor, turbidity sensor and conductivity sensor installed at the outlet of the drainage device, and the parameters of circulating water inflow are collected synchronously by the water level sensor, temperature sensor and volatile organic compound sensor installed at the inlet of the filter pipe at the bottom of the main well. The instantaneous parameters of the reinjected water and the parameters of the circulating water inflow are spliced ​​together using multidimensional features to generate a fusion feature vector representing the current circulation state. Extract the reinjected water turbidity time series and the circulating water volatile organic compound concentration time series from the fused feature vector and perform cross-correlation analysis. Calculate the phase difference between the reinjected water turbidity time series and the circulating water volatile organic compound concentration time series as the current pulse phase offset between the soil gas cycle and the groundwater cycle. The current pulse phase offset is compared with a preset target orthogonal phase offset threshold range. When the current pulse phase offset is outside the threshold range, a pulse phase adjustment trigger signal is generated. Based on the pulse phase adjustment trigger signal, the current isolation status code of the blocker, the reference pulse frequency of the peripheral injection well and the reference extraction cycle of the centrifugal pump are retrieved. The difference between the current pulse phase offset and the target orthogonal phase offset is calculated to generate the pulse phase compensation amount. The pulse phase compensation amount is mapped to the switching timing command of the blocking state, the correction command of the pulse frequency of the peripheral injection well, and the synchronization adjustment command of the centrifugal pump extraction cycle, so that the pulse peak of the soil gas circulation and the pulse trough of the groundwater circulation are aligned with each other in time, forming an alternating oscillating three-dimensional mass transfer field.

[0015] This invention addresses the technical deficiencies in the prior art and has the following beneficial effects: This invention addresses the unique challenges of pollution control in confined spaces within industrial clusters by constructing a gas-liquid synergistic three-dimensional circulation remediation system. Through the spatial arrangement of a main well and peripheral injection wells, combined with a barrier device featuring movement and expansion capabilities, it achieves layered and controllable hydraulic isolation between unsaturated and saturated zones, breaking the depth limitations of traditional single-well remediation. Based on real-time monitoring data from a multi-depth sensor array, it accurately identifies high-concentration accumulation points driven by groundwater level fluctuations and dynamically adjusts the opening and closing sequence of the barrier device, the frequency of gas-phase pulse injection, and the cycle of liquid-phase pulsation scouring accordingly. By staggering the pulse sequences of soil gas circulation and groundwater circulation, a three-dimensional mass transfer field with alternating gas-liquid oscillations is formed in the unsaturated medium, effectively solving the problems of low mass transfer efficiency in single-cycle systems and competition for pore space between the gas and liquid phases. Ultimately, it achieves efficient stripping and directional migration of pollutants from the solid phase to the gas and liquid phases in confined spaces, achieving targeted source control and long-term reduction effects in complex contaminated strata. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a simplified structural diagram of the system. Figure 2 This is a flowchart of the control process of this system; Attached Figure Descriptions: 1-Ground activated carbon filter device, 2-Vacuum pump device, 3-Injected air, 4-Ground activated carbon treatment device, 5-Clean gas, 6-Pressure air conditioner, 7-Drainage device, 8-External injection well, 9-Gas lift zone, 10-Underground activated carbon filter device, 11-Main well, 12-Centrifugal pump, 13-Drive device, 14-Sealing rubber structure, 15-Blocker, 16-Filter pipe, A-Upper part of main well, B-Unsaturated zone, C-Groundwater level fluctuation zone, D-Saturated zone. Detailed Implementation

[0018] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0019] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0020] This embodiment provides a groundwater pollution source control and cutoff system suitable for confined spaces in industrial clusters.

[0021] like Figure 1 As shown, the system includes a main well 11, an external injection well 8, a surface activated carbon filter 1, a vacuum pump 2, and a drainage device 7.

[0022] This system achieves efficient removal of pollutants by constructing two counter-current circulation systems: soil air circulation and groundwater circulation.

[0023] Soil gas cycle: A negative pressure field is established in the upper part of the main well 11 and in the unsaturated zone B by a vacuum pump device 2. In a preferred embodiment, the vacuum pump device 2 comprises a combination of a Roots vacuum pump and a water ring vacuum pump, positioned between the main well 11 and the peripheral injection well 8. The negative pressure field established in the unsaturated zone B has a pressure range of -0.04 MPa to -0.095 MPa. The vacuum pump device 2 is connected to the main well 11 via a stainless steel pipe and is used to extract soil gas containing volatile organic compounds and transport it to the surface activated carbon filter device 1; it is also connected to the surface activated carbon filter device 1 via a stainless steel pipe. The surface of the vacuum pump device 2 is coated with an anti-corrosion coating, and the pipeline is regularly cleaned and maintained.

[0024] The ground-based activated carbon filter device 1 is installed at the interface of the circulating aeration well and is used to purify exhaust gas. In a preferred embodiment, its filler is industrial-grade granular activated carbon or honeycomb activated carbon with a porous structure and a surface area of ​​1000-1500 m² / g; the core adsorption unit is a vertical or horizontal fixed bed, which can adopt a dual-bed dual-tower switching (one in use and one in standby) mode; the device adopts a negative pressure interface, is equipped with an airflow distribution plate and a flow regulating valve, and is connected to the vacuum pump device 2 through a flange.

[0025] The clean gas 5, purified by the ground activated carbon filter 1, is re-diffused into the unsaturated zone B and the groundwater level fluctuation zone C through the external injection well 8 via the pressurized air regulator 6. The pressurized air regulator 6 is located at the interface of the circulating aeration well and can be connected to the aeration well via a T-junction. A pressure gauge on the T-junction monitors the pressure inside the well. Due to the continuous negative pressure above the main well, the reinjected gas is drawn back to the upper part of the main well, thus forming a counter-current circulation of soil gas.

[0026] The system also includes an air injection pipeline structure, which promotes the overflow of high-concentration polluted groundwater in the unsaturated area to the activated carbon filter through the air injection process; and promotes the release of volatile organic compounds in the polluted groundwater into the upper part of the main well through continuous aeration.

[0027] Groundwater cycle: The main well 11 is equipped with a baffle 15 and a centrifugal pump 12. The centrifugal pump 12 is located on the upper part of the baffle 15 and is used to pump groundwater from the saturated zone D at the lower part of the baffle 15 to the upper part of the baffle 15.

[0028] The extracted contaminated groundwater first enters the airlift zone 9 located at the top of the main well 11. Aeration (i.e., the airlift process) is achieved by injecting air 3 into the upper part A of the main well, promoting the release of volatile organic compounds (VOCs) from the contaminated groundwater into the upper part of the main well, where they are treated by the aforementioned soil air circulation system. The airlifted groundwater then flows naturally into the underground activated carbon filter 10 for further purification of residual pollutant components. The purified groundwater is pumped into the surface activated carbon treatment device 4 for final treatment. The treated clean groundwater is then reinjected into the unsaturated zone B surrounding the main well through the drainage device 7. In a preferred embodiment, the drainage device 7 is located at the top of the unsaturated zone B, used to evenly discharge the purified groundwater into the unsaturated zone B surrounding the main well. Through continuous flushing and disturbance, organic matter in the unsaturated zone is removed. The groundwater discharged from the drainage system flushes pollutants from the unsaturated zone B and the groundwater level fluctuation zone C, eventually converging into the screen pipe area. The flushed groundwater ultimately converges into the filter pipe 16 area and is pumped back to the surface for treatment, forming a counter-current circulation of groundwater.

[0029] To achieve more precise hydraulic control, a baffle 15 is installed inside the main well 11. In a preferred embodiment, the baffle 15 consists of a sealing rubber structure 14 and a drive device 13. The sealing rubber structure 14 includes an air injection port, which can change its shape and size through air injection expansion to alter its contact state with the well wall of the main well 11. The drive device 13 is connected to the sealing rubber structure 14 and can move one end of the sealing rubber structure 14 along the central axis of the central pipe of the main well 11. Through the displacement of the drive device 13 and / or the air injection expansion of the sealing rubber structure 14, the sealing rubber structure 14 can be deformed, thereby placing the main well 11 in either a closed or open working state. The baffle 15 may also include a water filter pipe 16, which is welded to a screen by a joint to form a stainless steel pipe network that allows water to pass through and supports and protects the well wall.

[0030] The drive device 13 is connected to the sealing rubber structure 14, and based on the connection, it drives the end of the sealing rubber structure connected to the drive device to move along the central axis of the main well center pipe. During the movement, the sealing rubber structure deforms to set the main well in two states: closed and unclosed.

[0031] The sealing rubber structure 14 includes an air injection port, which can change its shape and size by expanding through air injection to change its contact state with the main well wall.

[0032] The barrier 15, installed on the main well 11, includes: a filter pipe 16, which is welded to the screen by a joint, and is a stainless steel pipe network that allows water to pass through and supports and protects the well wall.

[0033] The peripheral injection well 8 is located beside the main well 11, and its depth is the same as that of the main well 11. In a preferred embodiment, the bottom of the peripheral injection well 8 includes a 1.5m sedimentation pipe, which can be used for the injection of chemical agents, microbial agents, or biological adjuvants to facilitate the treatment of specific contaminants or for enhanced treatment in the later stages of remediation. The length of the upper screen pipe of the peripheral injection well 8 needs to be determined in conjunction with the aquifer conditions of the site.

[0034] The main well 11 is a well shaft used for circulating and filtering water to purify water quality, detecting water quality, and pumping out sewage.

[0035] The centrifugal pump 12 is located on the upper part of the barrier 15.

[0036] The gas lift zone 9 is located at the top of the main well 11.

[0037] The underground activated carbon filtration device 10 is used to further purify the overflowing polluted groundwater.

[0038] like Figure 2 As shown, the contaminated groundwater source control and interruption system applicable to confined spaces in industrial clusters further includes: Based on the pressure, water level, and volatile organic compound concentration data collected at different depths in the main well, the dynamic range of the groundwater level fluctuation zone and the high concentration accumulation points at the boundary between the unsaturated zone and the saturated zone were determined. Based on the depth of the identified high-concentration accumulation points, the drive device of the control barrier moves the sealing rubber structure to the corresponding depth and intermittently injects compressed gas into the gas injection port of the sealing rubber structure, causing it to expand and contract periodically, forming a dynamically opening and closing airflow channel between the inner wall of the main well and the unsaturated zone. The vacuum pump device is started to establish a negative pressure field at the top of the main well, and the purified clean gas is injected into the unsaturated zone in an intermittent pulse manner through the gas flow regulating valve in the outer well. With the cooperation of the dynamically opening and closing airflow channel, the soil gas is induced to generate an oscillating flow in the unsaturated medium. Based on the decay rate of volatile organic compound concentration at the outlet of the air-lift zone, the start-stop cycle and extraction flow rate of the centrifugal pump are adjusted so that the purified reinjected water is discharged into the top of the unsaturated zone in the form of intermittent waves through the drainage device, forming a pulsating flushing of residual pollutants in the unsaturated medium. Based on the circulating water parameters collected at the outlet of the drainage device and the inlet of the filter pipe at the bottom of the main well, coordinated adjustment commands are generated for the isolation status of the barrier, the pulse frequency of the peripheral injection well, and the extraction cycle of the centrifugal pump. This causes the pulse frequencies of the soil gas circulation and the groundwater circulation to be staggered, forming an alternating oscillating three-dimensional mass transfer field.

[0039] It should be noted that, firstly, pressure sensors, water level sensors, and volatile organic compound (VOC) sensor arrays deployed at different elevations on the inner wall of the main well were used to collect in-situ monitoring data at various depths in real time. By performing spatiotemporal alignment and filtering on these monitoring data, the dynamic range of frequent groundwater level fluctuations was identified. Furthermore, by combining the coherence analysis between the dominant frequency of water level fluctuations and pollutant concentrations, the high-concentration accumulation points of pollutants driven by water level fluctuations at the boundary between the unsaturated and saturated zones and their vertical migration characteristics were located. The identification results of the high-concentration accumulation points provide the target depth location for subsequent hydraulic control.

[0040] Upon identifying the target depth, the control unit of the isolator moves the sealing rubber structure along the central pipe of the main well to that depth. It's worth noting that the movement of the sealing rubber structure is not a simple positioning and sealing operation, but rather involves intermittent injection of compressed gas into its injection port, causing periodic expansion and contraction. When the rubber structure expands, its outer edge tightly adheres to the inner wall of the main well, blocking the airflow channel at that depth; when the rubber structure contracts, the airflow channel is opened. By controlling the intermittent frequency and duty cycle of the gas injection, a dynamically opening and closing controllable airflow channel can be formed between the inner wall of the main well and the unsaturated zone. The opening and closing frequency of this channel is adapted to subsequent gas injection pulses.

[0041] Based on this, a vacuum pump is activated to establish a continuous negative pressure field in the upper part of the main well and the unsaturated zone. Simultaneously, clean gas purified by the surface activated carbon filter is injected into the unsaturated zone in an intermittent pulse manner through a gas flow regulating valve in the peripheral injection well. When the gas pulse diffuses outward from the peripheral injection well and reaches the vicinity of the main well, it coincides with the opening period of the aforementioned dynamically opening and closing airflow channel, allowing the gas pulse to be instantaneously drawn into the negative pressure zone above the main well via this channel. When the airflow channel is closed, the gas pulse intake process is temporarily interrupted. Thus, through the periodic opening and closing of the dynamic airflow channel and the timing coordination of the gas pulses from the peripheral injection well, the originally continuous gas migration path is broken into intermittent oscillating flows, inducing reciprocating disturbances in the soil gas within the unsaturated medium. This disrupts the gas phase equilibrium in the unsaturated region, promoting the desorption and mass transfer of residual pollutants into the gas phase.

[0042] Furthermore, the system dynamically adjusts the start-stop cycle and extraction flow rate of the centrifugal pump based on the decay rate of volatile organic compound concentration at the outlet of the air-lift zone, causing the reinjected water to pulsate and flush the unsaturated medium. Simultaneously, based on circulating water parameters collected at the outlet of the drainage device and the inlet of the filter pipe at the bottom of the main well, the system coordinates the isolation status of the barrier, the pulse frequency of the peripheral injection wells, and the extraction cycle of the centrifugal pump, ensuring that the pulse timing of soil air circulation and groundwater circulation is staggered. Through the coordination of these various steps, an alternating oscillating three-dimensional mass transfer field is ultimately formed in the unsaturated and saturated zones, achieving targeted stripping and long-term reduction of pollutants.

[0043] Furthermore, based on pressure, water level, and volatile organic compound (VOC) concentration data collected at different depths within the main well, the dynamic range of groundwater level fluctuation zones and the locations of high-concentration accumulation points at the boundary between the unsaturated and saturated zones were determined, specifically: Based on the original time-series monitoring data synchronously collected by multiple pressure sensors, water level sensors and volatile organic compound sensor arrays deployed at different elevations on the inner wall of the main well, and after spatiotemporal alignment and wavelet transform processing of the original time-series monitoring data, a corrected effective dataset is generated. Extract the water level time series of each monitoring point from the corrected effective dataset, count the cumulative number of times the groundwater level crosses the depth position of each monitoring point within a unit of time, and define the depth interval corresponding to the depth point where the cumulative number exceeds the preset fluctuation frequency threshold as the dynamic interval of the groundwater level fluctuation zone and its main fluctuation frequency. Within the dynamic range of the groundwater level fluctuation zone, the time series of volatile organic compound concentrations at each monitoring point within the range are retrieved. Coherence analysis is performed on the concentration time series of each monitoring point and the main control fluctuation frequency. The coherence function value between the concentration change at each depth point and the water level fluctuation is calculated. The depth location where the coherence function value reaches a local maximum and remains stable is identified as a potential high-concentration accumulation nucleus. Centered on the potential high-concentration aggregation core, extend an unsaturated capillary water thickness range upwards and downwards to construct a spatial analysis window containing multiple adjacent monitoring points, and calculate the spatial autocorrelation Moran index and local spatial correlation index of the concentration values ​​of all monitoring points within the window. The depth locations where the spatial autocorrelation Moran index is greater than a set threshold and the local spatial correlation index reaches a local maximum are determined by combining the phase lag time of the concentration time series at the depth location with respect to the water level fluctuation, and the high concentration accumulation points at the boundary between the unsaturated zone and the saturated zone driven by the periodic fluctuation of the groundwater level, as well as the vertical migration time lag characteristics of the pollution plume.

[0044] It should be noted that due to differences in sampling frequencies and response times among different sensors, and the fact that raw data often contains noise components caused by tidal effects, rainfall replenishment, or local air pressure changes, the collected raw data first undergoes spatiotemporal alignment processing to ensure the comparability of monitoring data at each depth point in the time dimension. Simultaneously, wavelet transform is used to filter out abnormal fluctuation noise, generating a corrected and effective dataset that accurately reflects the site conditions. Specifically, the Daubechies 4th order wavelet is selected as the mother wavelet, which possesses good regularity and tight support characteristics, making it suitable for processing non-stationary signals. The number of decomposition layers is set to four based on the sampling frequency and data length. A soft thresholding function is used to denoise the detail coefficients of each layer, retaining low-frequency approximate components as the corrected effective signals for water level and concentration. The denoised water level time series is used for subsequent water level fluctuation range definition and main fluctuation frequency extraction, while the denoised concentration time series is used for coherence analysis. By eliminating high-frequency noise, the intrinsic correlation between concentration changes and water level fluctuations can be identified more accurately, improving the reliability of coherence function value calculation.

[0045] Then, based on the corrected valid dataset, the water level time series sequence for each monitoring point is extracted. For each monitoring depth, the cumulative number of times the groundwater level crosses that depth position per unit time is counted, for example, the number of crossings per hour. When the cumulative number of crossings at a certain depth point exceeds a preset fluctuation frequency threshold, that depth point and its adjacent depth range are defined as the dynamic range of the groundwater level fluctuation zone. Simultaneously, the dominant fluctuation frequency within this range is extracted through spectral analysis of the water level time series sequence. The preset fluctuation frequency threshold can be set according to the site's hydrogeological conditions; for example, in silty aquifers, it can be set to more than 3 crossings per hour as the criterion for determining an active fluctuation range.

[0046] Within the defined dynamic range of groundwater level fluctuations, time-series concentrations of volatile organic compounds (VOCs) at each monitoring point within this range are retrieved. For each depth point, its concentration time-series is subjected to coherence analysis with the aforementioned dominant fluctuation frequency, and a coherence function value is calculated. The coherence function value lies between 0 and 1, and its magnitude characterizes the correlation strength between concentration changes and water level fluctuations at that depth point. When the coherence function value at a certain depth point reaches a local maximum and this maximum remains stable over multiple consecutive time windows, that depth location is identified as a potential high-concentration accumulation nucleus. For example, when the coherence function value at a certain depth point is consistently greater than 0.75 and significantly higher than that of adjacent depth points, it can be considered a potential accumulation nucleus.

[0047] Centered on the identified potential high-concentration accumulation core, a spatial analysis window is constructed by extending an unsaturated capillary thickness range both upwards and downwards, encompassing multiple adjacent monitoring points. The capillary thickness can be determined based on the site lithology; for example, in silty strata, the capillary rise height is approximately 30 cm to 50 cm. Within this spatial analysis window, the spatial autocorrelation Moran's index and local spatial correlation index of the concentration values ​​at all monitoring points are calculated. The Moran's index characterizes the overall spatial aggregation trend of pollutant concentrations within the window; a value greater than 0 indicates a positive correlation, meaning high-concentration points are adjacent to each other. The local spatial correlation index identifies local locations within the window that significantly contribute to the overall spatial autocorrelation.

[0048] Finally, the depth locations where the spatial autocorrelation Moran index is greater than a set threshold and the local spatial correlation index reaches a local maximum are identified as high-concentration accumulation points at the boundary between the unsaturated and saturated zones. For example, when the Moran index is greater than 0.6, it indicates that the pollutants within the window have significant spatial accumulation characteristics; based on this, the depth point where the local spatial correlation index reaches a maximum value is the core location of this accumulation area. Simultaneously, by combining the phase lag time of the concentration time series at this depth point relative to water level fluctuations—for example, the concentration peak lags behind the water level fluctuation peak by 2 to 4 hours—the response characteristics of this high-concentration accumulation point driven by periodic groundwater level fluctuations and the time-lag characteristics of its vertical migration of the pollution plume can be further determined.

[0049] Furthermore, based on the decay rate of volatile organic compound concentration at the outlet of the air-lift zone, the start-stop cycle and extraction flow rate of the centrifugal pump are adjusted so that the purified reinjected water is discharged into the top of the unsaturated zone in an intermittent wave form through the drainage device, forming a pulsating flushing of residual pollutants in the unsaturated medium. Specifically: Real-time concentration data of volatile organic compounds at the outlet of the gas stripping zone are continuously collected. The concentration time series within a preset time window is extracted. After the concentration time series is logarithmically transformed, linear regression fitting is performed with the corresponding timestamp to generate the logarithmic domain decay slope of the volatile organic compound concentration as the instantaneous apparent decay rate constant. The instantaneous apparent decay rate constant is processed by exponential weighted moving average to obtain a dynamic smooth value of decay rate. The dynamic smooth value is then compared with a preset decay rate constant threshold range to identify the period when the dynamic smooth value is continuously lower than the lower limit of the threshold range as the mass transfer-limited bottleneck period. During the mass transfer bottleneck period, the reference pulse period and reference pulse flow of the current centrifugal pump are retrieved, and the ratio of the reference pulse period to the dynamic smoothing value is calculated to generate a mass transfer resistance coefficient that characterizes the resistance to pollutant release in the unsaturated medium. The mass transfer resistance coefficient is then input into a pre-stored pulse parameter response surface for interpolation query to generate a corrected pulse width and corrected pulse amplitude that match the current mass transfer resistance. Based on the corrected pulse width and corrected pulse amplitude, the centrifugal pump is controlled to output unsteady flow in an intermittent manner, so that when the reinjected water purified by the air-lift zone and the underground activated carbon filtration device is discharged into the top of the unsaturated zone through the drainage device, an asymmetric intermittent wave sequence with steep wave front and gentle wave back is formed in the unsaturated medium. Simultaneously collect soil volumetric water content change data at different depths in the unsaturated zone, and extract the wavefront arrival time and waveback saturation duration of the asymmetric intermittent wave sequence at each depth based on the soil volumetric water content change data; Spatial interpolation is performed on the wavefront arrival time to generate a vertical propulsion velocity field of pulsating water flow in an unsaturated medium. The vertical propulsion velocity field is then multiplied with the waveback saturation duration to generate an effective scour depth characterizing the effective desorption volume of pollutants. The effective scouring depth is compared with a preset target depth threshold, and an adaptive correction instruction for the pulse parameter response surface interpolation reference point is generated based on the comparison result.

[0050] It should be noted that, considering that pollutant concentration decay usually follows an exponential law, after performing a logarithmic transformation on the concentration time series within the preset time window, a linear regression is performed with the corresponding timestamp to obtain the logarithmic domain decay slope. This slope is the instantaneous apparent decay rate constant, used to characterize the real-time removal efficiency of pollutants under the current remediation conditions. For example, the time window can be 30 minutes, sliding every 5 minutes. Since the instantaneous decay rate may fluctuate, to avoid misjudgment, it is processed by exponential weighted moving average to obtain a dynamic smoothing value of the decay rate. This processing method can give higher weight to recent data, thereby reflecting changes in the decay trend in a timely manner. The dynamic smoothing value is compared with a preset decay rate constant threshold range. For example, the lower limit of the threshold range can be set to 0.05h⁻¹, and the upper limit can be set to 0.2h⁻¹. When the dynamic smoothing value is continuously below the lower limit and the duration exceeds the set duration (e.g., 6 consecutive hours), it indicates that the current hydraulic flushing conditions can no longer effectively drive pollutant desorption, and this period is identified as a mass transfer-limited bottleneck period.

[0051] It should be further explained that the condition of "duration exceeding the set duration" in the above identification process is not a simple lag adjustment, but is based on the following considerations: First, the pollutant concentration decay rate itself is affected by various factors and fluctuates instantaneously. If the pulse parameters are adjusted immediately when the dynamic smoothing value first falls below the threshold, the system may experience frequent false triggers due to brief disturbances, which could disrupt the stability of the hydraulic flushing. Second, the mass transfer-limited bottleneck period is essentially a state where the pollutant desorption rate is consistently low. Adjusting parameters after this state has lasted for a period of time provides a clearer direction and more reliable adjustment range, avoiding the introduction of new disturbances due to blind adjustments. Finally, the pulse parameter response surface and its adaptive correction mechanism used in this embodiment form a closed-loop self-optimization process by feeding back the deviation between the effective flushing depth and the target depth threshold to the response surface reference point. This self-optimization process does not rely on the immediate effect of a single adjustment, but gradually approaches the optimal pulse parameter combination through iterative accumulation over multiple flushing cycles. Therefore, even if there is a certain lag in a certain adjustment, as the number of adjustments increases, the system can still converge to the ideal working state through the adaptive correction mechanism, so as to achieve continuous, stable and efficient flushing of residual pollutants in unsaturated media.

[0052] Once the mass transfer bottleneck period is reached, the current reference pulse period and reference pulse flow rate of the centrifugal pump are retrieved, for example, a reference pulse period of 60 minutes and a reference pulse flow rate of 2 m³ / h. The reference pulse period is then compared with the dynamic smoothing value to generate a mass transfer resistance coefficient. For example, when the dynamic smoothing value is 0.03 h⁻¹, the mass transfer resistance coefficient is 60 / 0.03 = 2000. This coefficient comprehensively reflects the resistance of the unsaturated medium to pollutant release. This mass transfer resistance coefficient is then input into a pre-stored pulse parameter response surface for interpolation lookup. The pulse parameter response surface is a two-dimensional lookup table established based on numerical simulation. The horizontal axis represents the mass transfer resistance coefficient, and the vertical axis represents the corrected pulse width and corrected pulse amplitude. Each point on the surface corresponds to a set of optimal pulse parameters. Through interpolation lookup, a corrected pulse width and corrected pulse amplitude that match the current mass transfer resistance are generated.

[0053] Based on the corrected pulse parameters, the centrifugal pump is controlled to output an unsteady flow intermittently. When the purified reinjected water is discharged into the top of the unsaturated zone through the drainage device, due to the increased pulse width and amplitude, the water flow forms an asymmetric intermittent wave sequence with steep wavefronts and gentle wavebacks in the unsaturated medium. The steep wavefronts help to overcome the capillary barrier, while the gentle wavebacks prolong the contact time between the aqueous phase and the contaminants.

[0054] To verify the actual effect of pulse parameter adjustment, data from time-domain reflectometers or soil water potential sensors deployed at different depths in the unsaturated zone were collected simultaneously. From the soil volumetric water content change data, the wavefront arrival time and post-wave saturation duration of the asymmetric intermittent wave sequence at each depth can be extracted. The wavefront arrival time reflects the vertical propulsion velocity of the water flow, and the post-wave saturation duration reflects the residence time of the water phase at a specific depth.

[0055] Spatially interpolating the wavefront arrival time at each depth generates a vertical propulsion velocity field for pulsating water flow in an unsaturated medium. Multiplying this vertical propulsion velocity field by the waveback saturation duration yields the effective scour depth. For example, when the vertical propulsion velocity is 0.5 m / h and the waveback saturation duration is 0.8 h, the effective scour depth is 0.4 m. This parameter characterizes the depth range at which the water flow can effectively contact and desorb contaminants under the current pulse parameters.

[0056] The calculated effective flushing depth is compared with a preset target depth threshold. For example, the target depth threshold can be set to 1.2m, requiring each pulse flush to cover the main contamination depth in the unsaturated zone. When the effective flushing depth is 0.4m, significantly lower than the target threshold, it indicates that the current pulse parameter adjustment has not achieved the expected effect. At this time, an adaptive correction instruction for the interpolation reference point of the pulse parameter response surface is generated based on the comparison result. Specifically, the deviation between the current effective flushing depth and the target depth threshold (e.g., 0.8m) is used as feedback, and combined with the currently used mass transfer resistance coefficient, the interpolation reference point of the corresponding region in the response surface is corrected. The correction method can adopt a deviation-weighted superposition, that is, in the next interpolation query, the corrected pulse width and corrected pulse amplitude corresponding to the same mass transfer resistance coefficient will be appropriately increased or decreased according to the current deviation, so that the response surface can continuously self-optimize according to the actual flushing effect and gradually approach the optimal pulse parameter combination under the site conditions.

[0057] Furthermore, based on the circulating water parameters collected at the outlet of the drainage device and the inlet of the filter pipe at the bottom of the main well, coordinated adjustment commands are generated for the isolation status of the barrier, the pulse frequency of the peripheral injection well, and the extraction cycle of the centrifugal pump. This causes the pulse frequencies of the soil gas circulation and the groundwater circulation to be staggered, forming an alternating oscillating three-dimensional mass transfer field, specifically: The instantaneous parameters of reinjected water are collected synchronously by the flow sensor, turbidity sensor and conductivity sensor installed at the outlet of the drainage device, and the parameters of circulating water inflow are collected synchronously by the water level sensor, temperature sensor and volatile organic compound sensor installed at the inlet of the filter pipe at the bottom of the main well. The instantaneous parameters of the reinjected water and the parameters of the circulating water inflow are spliced ​​together using multidimensional features to generate a fusion feature vector representing the current circulation state. Extract the reinjected water turbidity time series and the circulating water volatile organic compound concentration time series from the fused feature vector and perform cross-correlation analysis. Calculate the phase difference between the reinjected water turbidity time series and the circulating water volatile organic compound concentration time series as the current pulse phase offset between the soil gas cycle and the groundwater cycle. The current pulse phase offset is compared with a preset target orthogonal phase offset threshold range. When the current pulse phase offset is outside the threshold range, a pulse phase adjustment trigger signal is generated. Based on the pulse phase adjustment trigger signal, the current isolation status code of the blocker, the reference pulse frequency of the peripheral injection well and the reference extraction cycle of the centrifugal pump are retrieved. The difference between the current pulse phase offset and the target orthogonal phase offset is calculated to generate the pulse phase compensation amount. The pulse phase compensation amount is mapped to the switching timing command of the blocking state, the correction command of the pulse frequency of the peripheral injection well, and the synchronization adjustment command of the centrifugal pump extraction cycle, so that the pulse peak of the soil gas circulation and the pulse trough of the groundwater circulation are aligned with each other in time, forming an alternating oscillating three-dimensional mass transfer field.

[0058] It should be noted that the turbidity of the reinjected water reflects, to some extent, the stripping of particulate matter and adsorbed pollutants carried during the flushing of the unsaturated zone, while the volatile organic compound (VOC) concentration at the filter pipe inlet reflects the load of dissolved pollutants carried back with groundwater after stripping through soil gas circulation. Therefore, by splicing the instantaneous parameters of the reinjected water with the parameters of the circulating water inflow using multidimensional features to generate a fused feature vector, the coordinated state of the two cycles at the current moment can be comprehensively characterized. Furthermore, since the response time of turbidity sensors is typically on the order of seconds, while the response time of VOC sensors is affected by both the sampling pipeline transmission and the detector response, it is generally on the order of minutes to hours, resulting in a significant difference between the two on their original time scales. To achieve effective fusion of the two types of data, the high-frequency time-series data collected by the turbidity sensor is first processed by moving average with the sampling period of the volatile organic compound (VOC) sensor as the window. Specifically, the arithmetic mean of the turbidity values ​​within the time window corresponding to each VOC data point is calculated to generate a turbidity time-series sequence aligned with the VOC data time scale. Simultaneously, the VOC concentration time-series sequence is linearly interpolated to fill missing points within the sampling interval, forming a continuous time-series sequence. After this alignment process, the data points of both data sets correspond one-to-one in the time dimension, thus providing a basis for cross-correlation analysis. The reinjected water turbidity time-series sequence and the circulating water VOC concentration time-series sequence are extracted from the fusion feature vector, and cross-correlation analysis is performed on them. Cross-correlation analysis can reveal the similarity between the two time series at different time offsets; the offset corresponding to the peak value is the current pulse phase offset. The physical significance of this offset is that when the pulsed extraction of soil gas circulation and the pulsed flushing of groundwater circulation overlap in time, the two may interfere with each other. For example, the airflow may occupy the pore space and obstruct the flow of water, or the gaseous pollutants carried by the water flow may be mistakenly extracted to the gas phase treatment unit, thereby reducing the overall mass transfer efficiency.

[0059] The calculated current pulse phase offset is compared with a preset target orthogonal phase offset threshold range. This threshold range can be set based on site hydrogeological conditions and remediation goals. For example, it can be set to half a pulse cycle, meaning that when the pulse timings of two cycles are staggered by 180 degrees, the extraction peak of the soil gas circulation corresponds precisely to the scouring trough of the groundwater circulation, thus avoiding competition for pore space. When the current pulse phase offset is outside this threshold range, it indicates that the coordinated state of the two cycles deviates from expectations, and the system generates a pulse phase adjustment trigger signal.

[0060] In response to the trigger signal, the system retrieves the current isolation status code of the blocker, the reference pulse frequency of the peripheral injection well, and the reference extraction cycle of the centrifugal pump. The blocker's isolation status code characterizes the current opening and closing mode of the dynamic gas flow channel, the pulse frequency of the peripheral injection well determines the rhythm of gas injection, and the extraction cycle of the centrifugal pump determines the timing of the water flow pulses. The current pulse phase offset is calculated by subtracting the target orthogonal phase offset to generate a pulse phase compensation value, which quantifies the phase difference that needs to be adjusted.

[0061] Finally, the pulse phase compensation is mapped to adjustment commands at three levels: for the barrier, the opening and closing phase of the dynamic airflow channel is changed by adjusting the switching sequence of the expansion and contraction of the sealing rubber structure; for the peripheral injection well, the rhythm of gas injection is adjusted by correcting the pulse frequency of the gas flow regulating valve; and for the centrifugal pump, the timing of the water flow pulse is changed by synchronously adjusting the extraction cycle. The coordinated adjustment of these three elements aligns the pulse peaks of the soil gas circulation with the pulse troughs of the groundwater circulation in time, meaning that when the gas extraction intensity is maximum, the water scouring intensity is weakest, and vice versa. Through this alternating oscillation timing, the gas and liquid phases alternately dominate in the unsaturated medium pores, avoiding competition and thus forming a three-dimensional mass transfer field of alternating gas and liquid phases, improving the mass transfer efficiency of pollutants from the solid phase to the gas and liquid phases.

[0062] In this embodiment, it also includes: During the continuous extraction period of the centrifugal pump, the instantaneous flow pulsation time sequence output by the flow sensor and the instantaneous water level fluctuation time sequence output by the water level sensor are synchronously collected at a preset high-frequency sampling rate to construct the original signal pair reflecting the hydraulic characteristics of the pumping process. The flow pulsation time series and the water level fluctuation time series in the original signal pair are reconstructed in phase space. The optimal delay time of each sequence is calculated by mutual information function, and the optimal embedding dimension of each sequence is calculated by pseudo-nearest neighbor algorithm. Based on the optimal delay time and the optimal embedding dimension, the flow pulsation time series is mapped to a set of flow phase point trajectories in high-dimensional phase space, and the water level fluctuation time series is mapped to a set of water level phase point trajectories in high-dimensional phase space. Calculate the Euclidean distance between each phase point in the flow phase point trajectory set and the corresponding phase point in the water level phase point trajectory set, generate a dynamic coupling distance time series composed of the Euclidean distances, and project the dynamic coupling distance time series onto a two-dimensional phase space plane with the flow phase point trajectory as the horizontal axis and the water level phase point trajectory as the vertical axis to generate a two-dimensional coupled attractor trajectory diagram. A sliding window is set on the two-dimensional coupled attractor trajectory diagram. The local Lyapunov exponent of the trajectory line within the window is calculated. When the local Lyapunov exponent jumps from a continuously negative value to a positive value and exceeds a preset divergence threshold, the corresponding time is marked as the initial bifurcation point of the coupled attractor trajectory. The instantaneous phase difference between the flow phase point and the water level phase point at the initial bifurcation point is extracted as the characteristic phase difference of hydraulic oscillation instability. The blocking state encoding sequence of the blocker is collected within a preset time period before and after the initial bifurcation point. The blocking state encoding sequence is associated and matched with the characteristic phase difference. The historical adjustment step size corresponding to the blocking state encoding that successfully matches the current characteristic phase difference is retrieved from the pre-stored historical adjustment database. A suppression command is generated to finely adjust the blocking pressure of the blocker with the historical adjustment step size, so as to force the two-dimensional coupled attractor trajectory to reconverge from the bifurcation state to the limit cycle state.

[0063] It should be noted that during actual system operation, the continuous extraction operation of the centrifugal pump may cause hydraulic oscillations due to factors such as aquifer heterogeneity, water level fluctuations, or instantaneous release of air bubbles, which may even lead to extraction instability and reduced repair efficiency in severe cases. Therefore, this embodiment achieves early warning and proactive suppression of hydraulic instability by exploring the dynamic coupling characteristics between flow pulsations and water level fluctuations. Specifically, during the continuous extraction period of the centrifugal pump, the instantaneous flow pulsation time series output by the flow sensor and the instantaneous water level fluctuation time series output by the water level sensor are simultaneously acquired at a preset high-frequency sampling rate to construct an original signal pair reflecting the hydraulic characteristics of the pumping process. Considering that single sensor data is insufficient to fully characterize the complex dynamic behavior of the system, this embodiment employs phase space reconstruction theory to map the one-dimensional time series to a high-dimensional phase space to reveal its inherent dynamic structure. For the flow pulsation time series and the water level fluctuation time series, the optimal delay time of each series is calculated using a mutual information function, and the optimal embedding dimension of each series is calculated using a pseudo-nearest neighbor algorithm. Based on the above parameters, the flow fluctuation time series is mapped to a set of flow phase point trajectories in a high-dimensional phase space, and the water level fluctuation time series is mapped to a set of water level phase point trajectories in a high-dimensional phase space. Thus, the time series of two different physical quantities that were previously difficult to compare directly become comparable in high-dimensional phase space.

[0064] To quantitatively characterize the dynamic coupling strength between the flow rate system and the water level system, the Euclidean distances between each phase point in the flow rate phase trajectory set and the corresponding phase points in the water level phase trajectory set at different times are calculated, generating a dynamic coupling distance time series composed of Euclidean distances. The magnitude of this series reflects the degree of synchronization between the two subsystems at the current moment; the smaller the distance, the tighter the coupling. Projecting the dynamic coupling distance time series onto a two-dimensional phase space plane with the flow rate phase trajectory as the horizontal axis and the water level phase trajectory as the vertical axis generates a two-dimensional coupled attractor trajectory diagram. The geometric shape of this trajectory diagram intuitively illustrates the cooperative evolution relationship between the two subsystems in phase space.

[0065] Furthermore, to capture the critical point of the system's transition from stable operation to instability, a sliding window is set on the two-dimensional coupled attractor trajectory diagram, and the local Lyapunov exponent of the trajectory lines within the window is calculated. The local Lyapunov exponent characterizes the exponential divergence rate of adjacent trajectories in phase space; a negative value indicates trajectory convergence and the system is in a stable state, while a positive value indicates trajectory divergence and the system tends towards instability. When the local Lyapunov exponent jumps from a continuously negative value to a positive value and exceeds a preset divergence threshold, the corresponding moment is marked as the initial bifurcation point of the coupled attractor trajectory. For example, in this embodiment, the preset divergence threshold can be set to 0.05. When the local Lyapunov exponent jumps from -0.02 to 0.08, the system determines that the risk of instability has been triggered. The instantaneous phase difference between the flow phase point and the water level phase point at the initial bifurcation point is extracted as the characteristic phase difference of hydraulic oscillation instability. This characteristic phase difference quantifies the degree of phase mismatch between the two subsystems at the moment of instability.

[0066] To intervene promptly in the event of instability, a sequence of coded isolation states of the barrier was collected within a preset time period before and after the initial bifurcation point. This sequence recorded the historical opening and closing modes of the dynamic airflow channel. The isolation state coded sequence was correlated and matched with the characteristic phase difference. The historical adjustment step size corresponding to the isolation state code that successfully matched the current characteristic phase difference was retrieved from a pre-stored historical adjustment database. This historical adjustment database, established based on previous experiments or operational experience, records the adjustment parameters used to successfully suppress instability under different characteristic phase differences. For example, when the characteristic phase difference is 30 degrees, the adjustment step size matched in the historical database is 0.02 MPa. Finally, a suppression command was generated to fine-tune the isolation pressure of the barrier using this historical adjustment step size. By slightly adjusting the expansion pressure of the isolation rubber structure, the opening and closing characteristics of the dynamic airflow channel were altered, thereby disturbing the coupling relationship between the flow system and the water level system. This forced the two-dimensional coupled attractor trajectory to converge again from the bifurcation state to the limit cycle state, achieving adaptive suppression of hydraulic oscillations. This mechanism requires no shutdown intervention and can dynamically maintain hydraulic stability during continuous system operation, ensuring long-term reliable operation of repair work.

[0067] In this embodiment, it also includes: Step 1: Within the continuous monitoring window, synchronously collect the concentration time series output by multiple volatile organic compound sensors deployed at different elevations on the inner wall of the main well, as well as the water level fluctuation time series output by the water level sensor at the corresponding elevation, to construct a vertical monitoring dataset covering the entire profile from the unsaturated zone to the saturated zone. Step 2: Using the water level fluctuation time series as the reference variable, the concentration time series at each elevation is used as the target variable. The mutual information entropy values ​​of the reference variable and the target variable at each lag time are calculated sequentially according to the preset lag step size, and a cluster of decay curves of the mutual information entropy of concentration at each elevation to water level fluctuation as a function of lag time is generated. Step 3: Perform second derivative processing on the attenuation curves at each elevation, extract the lag time corresponding to the inflection point of each attenuation curve as the characteristic response lag time, and define the radius of curvature of each attenuation curve at the characteristic response lag time as the information transmission sensitivity of the concentration at that elevation to water level fluctuations. Step 4: Draw the vertical distribution profile of the information transmission sensitivity along the depth direction, identify the depth inflection point on the vertical distribution profile where the information transmission sensitivity changes from a continuous high value to a continuous low value, mark the depth interval where the depth inflection point is located as the capillary interface, and extract the difference in the characteristic response lag time of two adjacent monitoring points above and below the capillary interface as the capillary retardation response time lag difference. Step 5: Calculate the ratio of the relative permeability of the gas phase to the relative permeability of the liquid phase at the capillary interface, and define the product of this ratio and the time lag difference of the capillary retardation response as the initial transient retardation coefficient. Step 6: Repeat steps 1 to 5 through a sliding time window to generate a time series set consisting of the initial transient hysteresis coefficients, and determine the coefficient of variation of the time series set based on the fluctuation amplitude of the time series set. Step 7: When the coefficient of variation exceeds the preset coefficient threshold, the product of the initial transient retardation coefficient and the coefficient of variation is updated to the final transient retardation coefficient to characterize the unsteady retardation strength of the dynamic retention layer on the vertical migration of pollutants.

[0068] It should be noted that during actual site remediation, the capillary interface between the unsaturated zone and the saturated zone is not fixed. Its position and hindrance characteristics are affected by multiple factors such as water level fluctuations, media heterogeneity, and pollutant migration, exhibiting significant dynamic evolution characteristics. Traditional remediation methods often simplify the capillary zone into a static interface, leading to misjudgments of the vertical migration behavior of pollutants, thereby affecting the accuracy of control decisions. In this embodiment, within a continuous monitoring window, the concentration time series output by multiple volatile organic compound sensors deployed at different elevations on the inner wall of the main well, as well as the water level fluctuation time series output by water level sensors at the corresponding elevations, are simultaneously collected to construct a vertical monitoring dataset covering the entire profile from the unsaturated zone to the saturated zone.

[0069] To quantify the information-driving strength of water level fluctuations on pollutant concentration changes at various depths, the water level fluctuation time series is used as the baseline variable, and the concentration time series at each elevation is used as the target variable. The mutual information entropy values ​​of the baseline and target variables are calculated sequentially at each lag time using a preset lag step. Mutual information entropy can capture the nonlinear correlation between the two variables and is more suitable for dynamic coupling analysis under complex hydrogeological conditions than the linear correlation coefficient. This generates a cluster of decay curves showing the mutual information entropy of concentration at each elevation on water level fluctuations as a function of lag time. Each curve reflects the decay law of the information transmission efficiency of concentration response to water level changes at that depth over time.

[0070] The attenuation curves at each elevation were processed using the second derivative, and the lag time corresponding to the inflection point of each attenuation curve was extracted as the characteristic response lag time. The inflection point is the critical point where the information transmission efficiency changes from rapid attenuation to slow attenuation, and physically corresponds to the dominant lag time for the water level driving signal to be transmitted to that depth and trigger a concentration response. Simultaneously, the radius of curvature of each attenuation curve at the characteristic response lag time is defined as the sensitivity of the concentration at that elevation to water level fluctuations; the smaller the radius of curvature, the more sensitive the concentration at that depth is to water level fluctuations.

[0071] A vertical distribution profile of information transmission sensitivity was plotted along the depth direction, visually demonstrating the differences in the sensitivity of concentrations to water level fluctuations at different depths. Due to the coexistence of gas and liquid phases and the abrupt change in relative permeability at the capillary interface, the information transmission mechanism undergoes a significant change, reflected in the sensitivity profile as a depth inflection point where a sustained high value abruptly changes to a sustained low value. The depth interval containing this inflection point is marked as the capillary interface, and the difference in characteristic response lag time between two adjacent monitoring points above and below the capillary interface is extracted as the capillary retardation response time lag difference. This time lag difference characterizes the time delay required for pollutants to cross the capillary interface.

[0072] To further quantify the impediment strength of the capillary interface to the vertical migration of pollutants, the ratio of the relative permeability of the gas phase to the relative permeability of the liquid phase at this interface is calculated. This ratio, multiplied by the time lag difference of the capillary impediment response, is then defined as the initial transient impediment coefficient, reflecting the combined impediment effect of medium properties and dynamic response characteristics on pollutant migration. Considering the time-varying nature of capillary interface characteristics with water level fluctuations, the calculation results of a single time window may contain random errors. Therefore, the above steps are repeated using a sliding time window to generate a time series set composed of the initial transient impediment coefficient. The coefficient of variation (COP) is determined based on the fluctuation amplitude of this time series set; the COP is the ratio of the standard deviation to the mean, characterizing the time stability of the impediment coefficient. When the COP exceeds a preset threshold, it indicates that the capillary interface is in a state of drastic dynamic change, and a single-point calculation of the initial transient impediment coefficient is insufficient to characterize its time-varying characteristics. For example, the preset threshold can be set to 0.3. When the COP is greater than the preset threshold, the product of the initial transient impediment coefficient and the COP is updated to the final transient impediment coefficient. The final transient retardation coefficient reflects both the current interfacial retardation intensity and its temporal fluctuation amplitude, enabling a more realistic characterization of the unsteady-state retardation characteristics of the dynamic retention layer on the vertical migration of pollutants. This parameter can be input to the system control module in real time to guide the adjustment of the barrier sealing strategy and the optimization of the gas-liquid pulse frequency, thereby achieving precise control under dynamic changes at the capillary interface.

[0073] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A groundwater pollution source control and interruption system suitable for confined spaces in industrial clusters, characterized in that, The system includes: The main shaft is vertically installed in the contaminated site, penetrating both the unsaturated and saturated zones; Peripheral injection wells are located beside the main well; The ground treatment unit includes a vacuum pump device and a ground activated carbon filter device connected in sequence by pipelines; An air-lift circulation unit is installed at the top of the main well to inject air into the main well for aeration, thereby promoting the release of volatile organic compounds in polluted groundwater. The groundwater circulation unit includes a barrier and a centrifugal pump installed in the main well. The centrifugal pump is used to pump groundwater from the lower part of the barrier to the upper part of the barrier to form a local water level difference. in, The vacuum pump device is used to establish negative pressure in the upper part of the main well and the unsaturated area to extract soil gas containing volatile organic compounds and transport it to the ground activated carbon filter device for purification. The purified gas is pressurized and reinjected into the unsaturated area and / or groundwater level fluctuation zone through the peripheral injection well to form a soil gas circulation. The groundwater extracted by the centrifugal pump is purified through the air-lift zone and the underground activated carbon filtration device, and then reinjected into the unsaturated area around the main well through the drainage device, forming a groundwater cycle.

2. The groundwater pollution source control and interruption system applicable to confined spaces in industrial clusters according to claim 1, characterized in that, The barrier includes: The sealing rubber structure has an air injection port, which can expand by injecting air to change the contact state with the main well wall; A driving device, connected to the sealing rubber structure, is used to drive the sealing rubber structure to move along the central axis of the main well's central pipe. Through the deformation and displacement of the sealing rubber structure, the main well is in either a closed or unclosed working state.

3. The groundwater pollution source control and interruption system applicable to confined spaces in industrial clusters according to claim 1, characterized in that, The ground activated carbon filter device is filled with industrial-grade granular activated carbon or honeycomb activated carbon, with a surface area of ​​1000-1500 m² / g; the ground activated carbon filter device adopts a negative pressure interface and is connected to the vacuum pump device through a flange.

4. The groundwater pollution source control and interruption system applicable to confined spaces in industrial clusters according to claim 1, characterized in that, The peripheral injection well has the same depth as the main well, and its bottom is equipped with a sedimentation pipe for injecting chemical agents, bacterial agents, or biological additives.

5. The groundwater pollution source control and interruption system applicable to confined spaces in industrial clusters according to claim 1, characterized in that, The air-lift zone is located at the top of the main well and is used to aerate the extracted groundwater to separate volatile organic compounds.

6. The groundwater pollution source control and interruption system applicable to confined spaces in industrial clusters according to claim 1, characterized in that, The drainage device is located at the top of the unsaturated zone and is used to evenly discharge the purified groundwater into the unsaturated zone around the main well, flushing away and removing pollutants from the unsaturated zone.

7. The groundwater pollution source control and interruption system applicable to confined spaces in industrial clusters according to claim 1, characterized in that, Also includes: Based on the pressure, water level, and volatile organic compound concentration data collected at different depths in the main well, the dynamic range of the groundwater level fluctuation zone and the high concentration accumulation points at the boundary between the unsaturated zone and the saturated zone were determined. Based on the depth of the identified high-concentration accumulation points, the drive device of the control barrier moves the sealing rubber structure to the corresponding depth and intermittently injects compressed gas into the gas injection port of the sealing rubber structure, causing it to expand and contract periodically, forming a dynamically opening and closing airflow channel between the inner wall of the main well and the unsaturated zone. The vacuum pump device is started to establish a negative pressure field at the top of the main well, and the purified clean gas is injected into the unsaturated zone in an intermittent pulse manner through the gas flow regulating valve in the outer well. With the cooperation of the dynamically opening and closing airflow channel, the soil gas is induced to generate an oscillating flow in the unsaturated medium. Based on the decay rate of volatile organic compound concentration at the outlet of the air-lift zone, the start-stop cycle and extraction flow rate of the centrifugal pump are adjusted so that the purified reinjected water is discharged into the top of the unsaturated zone in the form of intermittent waves through the drainage device, forming a pulsating flushing of residual pollutants in the unsaturated medium. Based on the circulating water parameters collected at the outlet of the drainage device and the inlet of the filter pipe at the bottom of the main well, coordinated adjustment commands are generated for the isolation status of the barrier, the pulse frequency of the peripheral injection well, and the extraction cycle of the centrifugal pump. This causes the pulse frequencies of the soil gas circulation and the groundwater circulation to be staggered, forming an alternating oscillating three-dimensional mass transfer field.

8. The groundwater pollution source control and interruption system for confined spaces in industrial clusters according to claim 7, characterized in that, Based on pressure, water level, and volatile organic compound (VOC) concentration data collected at different depths within the main well, the dynamic range of groundwater level fluctuation zones and the locations of high-concentration accumulation points at the boundary between the unsaturated and saturated zones were determined, specifically: Based on the original time-series monitoring data synchronously collected by multiple pressure sensors, water level sensors and volatile organic compound sensor arrays deployed at different elevations on the inner wall of the main well, and after spatiotemporal alignment and wavelet transform processing of the original time-series monitoring data, a corrected effective dataset is generated. Extract the water level time series of each monitoring point from the corrected effective dataset, count the cumulative number of times the groundwater level crosses the depth position of each monitoring point within a unit of time, and define the depth interval corresponding to the depth point where the cumulative number exceeds the preset fluctuation frequency threshold as the dynamic interval of the groundwater level fluctuation zone and its main fluctuation frequency. Within the dynamic range of the groundwater level fluctuation zone, the time series of volatile organic compound concentrations at each monitoring point within the range are retrieved. Coherence analysis is performed on the concentration time series of each monitoring point and the main control fluctuation frequency. The coherence function value between the concentration change at each depth point and the water level fluctuation is calculated. The depth location where the coherence function value reaches a local maximum and remains stable is identified as a potential high-concentration accumulation nucleus. Centered on the potential high-concentration aggregation core, extend an unsaturated capillary water thickness range upwards and downwards to construct a spatial analysis window containing multiple adjacent monitoring points, and calculate the spatial autocorrelation Moran index and local spatial correlation index of the concentration values ​​of all monitoring points within the window. The depth locations where the spatial autocorrelation Moran index is greater than a set threshold and the local spatial correlation index reaches a local maximum are determined by combining the phase lag time of the concentration time series at the depth location with respect to the water level fluctuation, and the high concentration accumulation points at the boundary between the unsaturated zone and the saturated zone driven by the periodic fluctuation of the groundwater level, as well as the vertical migration time lag characteristics of the pollution plume.

9. The groundwater pollution source control and interruption system for confined spaces in industrial clusters according to claim 7, characterized in that, Based on the decay rate of volatile organic compound concentration at the outlet of the air-lift zone, the start-stop cycle and extraction flow rate of the centrifugal pump are adjusted so that the purified reinjection water is discharged into the top of the unsaturated zone in an intermittent wave form through the drainage device, forming a pulsating flushing of residual pollutants in the unsaturated medium. Specifically: Real-time concentration data of volatile organic compounds at the outlet of the gas stripping zone are continuously collected. The concentration time series within a preset time window is extracted. After the concentration time series is logarithmically transformed, linear regression fitting is performed with the corresponding timestamp to generate the logarithmic domain decay slope of the volatile organic compound concentration as the instantaneous apparent decay rate constant. The instantaneous apparent decay rate constant is processed by exponential weighted moving average to obtain a dynamic smooth value of decay rate. The dynamic smooth value is then compared with a preset decay rate constant threshold range to identify the period when the dynamic smooth value is continuously lower than the lower limit of the threshold range as the mass transfer-limited bottleneck period. During the mass transfer bottleneck period, the reference pulse period and reference pulse flow of the current centrifugal pump are retrieved, and the ratio of the reference pulse period to the dynamic smoothing value is calculated to generate a mass transfer resistance coefficient that characterizes the resistance to pollutant release in the unsaturated medium. The mass transfer resistance coefficient is then input into a pre-stored pulse parameter response surface for interpolation query to generate a corrected pulse width and corrected pulse amplitude that match the current mass transfer resistance. Based on the corrected pulse width and corrected pulse amplitude, the centrifugal pump is controlled to output unsteady flow in an intermittent manner, so that when the reinjected water purified by the air-lift zone and the underground activated carbon filtration device is discharged into the top of the unsaturated zone through the drainage device, an asymmetric intermittent wave sequence with steep wave front and gentle wave back is formed in the unsaturated medium. Simultaneously collect soil volumetric water content change data at different depths in the unsaturated zone, and extract the wavefront arrival time and waveback saturation duration of the asymmetric intermittent wave sequence at each depth based on the soil volumetric water content change data; Spatial interpolation is performed on the wavefront arrival time to generate a vertical propulsion velocity field of pulsating water flow in an unsaturated medium. The vertical propulsion velocity field is then multiplied with the waveback saturation duration to generate an effective scour depth characterizing the effective desorption volume of pollutants. The effective scouring depth is compared with a preset target depth threshold, and an adaptive correction instruction for the pulse parameter response surface interpolation reference point is generated based on the comparison result.

10. The groundwater pollution source control and interruption system for confined spaces in industrial clusters according to claim 7, characterized in that, Based on the circulating water parameters collected at the outlet of the drainage device and the inlet of the filter pipe at the bottom of the main well, coordinated adjustment commands are generated for the isolation status of the barrier, the pulse frequency of the peripheral injection well, and the extraction cycle of the centrifugal pump. This causes the pulse frequencies of the soil gas circulation and the groundwater circulation to be staggered, forming an alternating oscillating three-dimensional mass transfer field, specifically: The instantaneous parameters of reinjected water are collected synchronously by the flow sensor, turbidity sensor and conductivity sensor installed at the outlet of the drainage device, and the parameters of circulating water inflow are collected synchronously by the water level sensor, temperature sensor and volatile organic compound sensor installed at the inlet of the filter pipe at the bottom of the main well. The instantaneous parameters of the reinjected water and the parameters of the circulating water inflow are spliced ​​together using multidimensional features to generate a fusion feature vector representing the current circulation state. Extract the reinjected water turbidity time series and the circulating water volatile organic compound concentration time series from the fused feature vector and perform cross-correlation analysis. Calculate the phase difference between the reinjected water turbidity time series and the circulating water volatile organic compound concentration time series as the current pulse phase offset between the soil gas cycle and the groundwater cycle. The current pulse phase offset is compared with a preset target orthogonal phase offset threshold range. When the current pulse phase offset is outside the threshold range, a pulse phase adjustment trigger signal is generated. Based on the pulse phase adjustment trigger signal, the current isolation status code of the blocker, the reference pulse frequency of the peripheral injection well and the reference extraction cycle of the centrifugal pump are retrieved. The difference between the current pulse phase offset and the target orthogonal phase offset is calculated to generate the pulse phase compensation amount. The pulse phase compensation amount is mapped to the switching timing command of the blocking state, the correction command of the pulse frequency of the peripheral injection well, and the synchronization adjustment command of the centrifugal pump extraction cycle, so that the pulse peak of the soil gas circulation and the pulse trough of the groundwater circulation are aligned with each other in time, forming an alternating oscillating three-dimensional mass transfer field.

Images