A method and device for spectral detection based on a spiral microchannel flow cell

Abstract

The application discloses a spectrum detection method and device based on a spiral micro-channel flow cell, collects sample fluid, and carries out multi-inlet spiral channel mixing reinforcement processing to generate a mixing reinforcement feature, generates a mass transfer efficiency distribution based on the mixing reinforcement feature, determines a key mass transfer reinforcement area through the mass transfer efficiency distribution, extracts a concentration gradient distribution feature to determine an in-situ detection sensitive area, carries out multi-parameter detection on the in-situ detection sensitive area, fuses and determines an effective collection interval to obtain original spectrum data, determines a response precision threshold value based on the effective collection interval and the mixing reinforcement feature, carries out flow state stability evaluation on the key mass transfer reinforcement area to form a flow state stability index, constructs a partition monitoring configuration, identifies an abnormal response area to generate a flow state correction coefficient, carries out correction processing on the original spectrum data to generate a spectrum detection result, and realizes self-adaptive spectrum detection in the spiral micro-channel.

Description

Technical Field

[0001] This invention relates to the field of spectral analysis technology, and in particular to a spectral detection method and apparatus based on a spiral microchannel flow cell. Background Technology

[0002] Helical microchannels, due to their unique geometry, can generate Dean's vortices in fluids. The entrainment effect of these vortices enhances mass exchange between different fluid layers, making helical microchannels widely used in sample mixing and pretreatment. With the development of the new materials industry, the demand for online detection of novel materials such as nano-suspensions, functional coatings, and polymer solutions is increasing. Combining helical microchannels with spectroscopic detection to form a flow-through detection system enables online sample mixing and real-time analysis.

[0003] However, the flow field distribution within the spiral microchannel is complex, and the velocity fluctuations and concentration inhomogeneities caused by vortex motion can interfere with the spectral signal. There is still a lack of systematic solutions to problems such as how to select a suitable detection location in a dynamic mixing environment, how to assess the impact of flow fluctuations on detection accuracy, and how to compensate for measurement deviations caused by concentration gradients. Summary of the Invention

[0004] This invention discloses a spectral detection method and device based on a spiral microchannel flow cell, aiming to solve problems such as complex flow field in spiral microchannels, difficulty in selecting detection position, and interference of flow state fluctuations with detection accuracy. By analyzing mass transfer efficiency and evaluating flow state stability, an adaptive configuration mechanism for detection parameters is established to meet the spectral detection needs of complex fluids such as new material samples in dynamic mixing environments.

[0005] The first aspect of this invention proposes a spectral detection method based on a spiral microchannel flow cell, comprising the following steps: Collect sample fluid, perform multi-inlet spiral channel mixing enhancement treatment on the sample fluid to generate mixing enhancement characteristics, and generate mass transfer efficiency distribution based on the mixing enhancement characteristics; The key mass transfer enhancement region is determined by the mass transfer efficiency distribution, the concentration gradient distribution features are extracted from the key mass transfer enhancement region, and the gradient feature analysis of the concentration gradient distribution features is used to determine the in-situ detection sensitive region. Multi-parameter detection is performed on the in-situ detection sensitive area to obtain component concentration distribution, temperature gradient distribution and pressure response parameters. The component concentration distribution, temperature gradient distribution and pressure response parameters are fused to determine the effective acquisition range. Spectral acquisition is performed on the effective acquisition range to obtain raw spectral data. Based on the effective acquisition range and the hybrid enhancement features, a response accuracy threshold is determined, and the flow stability of the key mass transfer enhancement zone is evaluated to form a flow stability index. Based on the flow stability index and the response accuracy threshold, a zoned monitoring configuration is constructed. The abnormal response region is identified by parameter deviation detection in the in-situ detection sensitive area. The abnormal response region is then matched with the concentration gradient distribution characteristics to generate a flow regime correction coefficient. Based on the flow regime correction coefficient and the zonal monitoring configuration, the original spectral data is corrected to generate a spectral detection result.

[0006] A second aspect of the present invention provides a spectral detection device based on a spiral microchannel flow cell, comprising: The sample import module is used to collect sample fluid, perform multi-inlet spiral channel mixing enhancement processing on the sample fluid to generate mixing enhancement characteristics, and generate mass transfer efficiency distribution based on the mixing enhancement characteristics; The mass transfer analysis module is used to determine the key mass transfer enhancement region through the mass transfer efficiency distribution, extract the concentration gradient distribution features from the key mass transfer enhancement region, and perform gradient feature analysis on the concentration gradient distribution features to determine the in-situ detection sensitive region. The parameter fusion module is used to perform multi-parameter detection on the in-situ detection sensitive area to obtain component concentration distribution, temperature gradient distribution and pressure response parameters, fuse the component concentration distribution, temperature gradient distribution and pressure response parameters to determine the effective acquisition range, and perform spectral acquisition on the effective acquisition range to obtain raw spectral data; The monitoring construction module is used to determine the response accuracy threshold based on the effective acquisition range and the hybrid enhancement features, to perform flow stability assessment on the key mass transfer enhancement zone to form a flow stability index, and to construct a partitioned monitoring configuration based on the flow stability index and the response accuracy threshold. The detection output module is used to detect parameter deviations in the in-situ detection sensitive area to identify abnormal response areas, perform gradient matching between the abnormal response areas and the concentration gradient distribution characteristics to generate flow regime correction coefficients, and perform correction processing on the original spectral data based on the flow regime correction coefficients and the zonal monitoring configuration to generate spectral detection results.

[0007] The beneficial effects of this invention are reflected in the following points: 1. By performing multi-inlet spiral channel mixing enhancement treatment on the sample fluid and extracting disturbance distribution data from the mixing enhancement characteristics, mass transfer contribution is classified according to the mass transfer contribution degree, thus changing the traditional flow cell detection method that relies on experience to select the detection location; the mass transfer efficiency distribution is divided into core mass transfer domain and edge mass transfer domain, and mass transfer channels are formed by tracing from the core mass transfer domain to the edge mass transfer domain and enhancing node groups are extracted, so that the determination of key mass transfer enhancement areas is based on the analysis of mass migration paths, and the gradient feature analysis of concentration gradient distribution characteristics provides a quantitative basis for the location of sensitive areas for in-situ detection. 2. By hierarchically dividing the component concentration distribution to generate a concentration stratification structure, and mapping the temperature gradient distribution to the concentration stratification structure to form a temperature mapping layer, a spatial correlation analysis framework between the concentration field and the temperature field is established; the introduction of the fusion response factor transforms the detection conditions of concentration uniformity, temperature stability and pressure fluctuation into a unified quantitative evaluation index, and the determination of the effective acquisition interval breaks through the limitation of determining the detection location by a single parameter. 3. By performing spectral analysis on the flow regime change curves to separate ordered disturbance components from disordered turbulent components, and defining the steady-state range of the disordered turbulent components to obtain a steady-state boundary set, a two-component evaluation method for the flow regime stability index was established. The correlation mapping between multi-level accuracy standards and stability grading zones enabled differentiated configuration of sampling density and integration time. The gradient matching between the abnormal response region and the concentration gradient distribution characteristics distinguished the sources of systematic deviations and random fluctuations. The boundary correction mechanism of the flow regime correction coefficient made the compensation process of the spectral detection results smooth and physically reasonable. Attached Figure Description

[0008] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0009] Unless otherwise specified or defined, the same reference numerals in different figures represent the same or similar technical features, and different reference numerals may be used to represent the same or similar technical features.

[0010] Figure 1 This is a schematic flowchart of a spectral detection method based on a spiral microchannel flow cell according to the present invention. Figure 2 This is a schematic diagram of the spiral microchannel structure and Dean's vortex formation of the present invention.

[0011] Figure 3 This is a schematic diagram of the hardware connection of the spiral microchannel flow cell spectral detection system of the present invention.

[0012] Figure 4 This is a structural block diagram of a spectral detection device based on a spiral microchannel flow cell according to the present invention.

[0013] Wherein: 1-First inlet; 2-Second inlet; 3-Outer wall of spiral channel; 4-Inner wall of spiral channel; 5-Channel cross-section; 6-Outer vortex; 7-Inner vortex; 8-Main flow direction; 9-Helical radius of curvature R; 10-Light source; 11-Collimating lens; 12-First sample storage bottle; 13-Second sample storage bottle; 14-Peristaltic pump; 15-Flow sensor; 16-Spiral microchannel flow cell; 17-Detection window; 18-Focusing lens; 19-Spectrometer; 20-Data acquisition card; 21-Control computer. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0015] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0016] The technical solutions of the embodiments of this application will be described below.

[0017] like Figure 1 As shown, this embodiment of the invention provides a spectral detection method based on a spiral microchannel flow cell, including the following steps S110-S150: Step S110: Collect sample fluid, perform multi-inlet spiral channel mixing enhancement processing on the sample fluid to generate mixing enhancement characteristics, and generate mass transfer efficiency distribution based on the mixing enhancement characteristics.

[0018] Specifically, sample fluid is collected. A fixed amount of sample fluid is extracted from the sample container. The extraction process is driven by a peristaltic pump or a syringe pump. The peristaltic pump generates pulsating flow by periodically squeezing the tubing, which is suitable for shear-sensitive biological samples. The syringe pump generates stable and continuous flow by piston propulsion, which is suitable for chemical analysis requiring precise flow control. The flow control accuracy is required to be within ±2% of the set value to ensure the stability and repeatability of the injection. The extracted sample fluid is pretreated. The pretreatment steps include filtration to remove impurities, degassing, and temperature control. Filtration uses a 0.45μm pore size filter membrane to remove particulate impurities that may clog the microchannels. Degassing eliminates dissolved gases to prevent the formation of bubbles in the channels that interfere with detection. When the sample fluid contains dissolved gases, tiny bubbles may precipitate in the low-pressure areas of the microchannels, such as inside the spiral. These bubbles block the light path, causing scattering interference peaks in the transmission spectrum, making it difficult to accurately identify the characteristic absorption peaks of the analyte. Temperature control brings the sample fluid to the set detection temperature to eliminate the influence of temperature fluctuations on the spectral signal. The viscosity and density of the sample fluid were measured. Viscosity was measured using a micro capillary viscometer or rotational rheometer, and density was measured using a vibrating tube densitometer. The viscosity and density values ​​of the sample fluid were used to calculate the Reynolds number and Dean number.

[0019] Multi-inlet spiral channel mixing enhancement treatment is applied to the sample fluid to generate mixed enhancement features. For example... Figure 2As shown, the spiral microchannel has a first inlet 1 and a second inlet 2. The sample fluid enters the channel formed by the outer wall 3 and the inner wall 4 of the spiral channel from the two inlets. At the channel cross-section 5, under the action of centrifugal force, a pair of Dean vortices is formed, consisting of an outer vortex 6 and an inner vortex 7. The two vortices rotate in opposite directions and are perpendicular to the mainstream direction 8. The radius of curvature R9 of the spiral determines the magnitude of the centrifugal force and thus affects the vortex intensity. The geometry of the spiral microchannel generates secondary flow, namely Dean vortices, in the fluid. The Dean vortices superimpose with the mainstream to form a complex three-dimensional flow pattern. Without the effect of Dean vortices, the laminar flow state in the microchannel would cause different fluid layers to mix only by molecular diffusion, and the mixing distance would increase by several times or even tens of times. The pretreated sample fluid is distributed to multiple inlets of the spiral microchannel. The number of inlets is usually 2 to 4 and is evenly distributed along the circumference of the channel. A symmetrical arrangement of two inlets is suitable for mixing two components in equal proportions, while a four-inlet arrangement is suitable for introducing multiple components simultaneously. After the sample fluid enters the spiral channel, it is subjected to the combined effects of centrifugal force and viscous force. The ratio of these two forces determines the intensity of the Dean vortex, forming symmetrical Dean vortex pairs within the channel cross-section. The vortices draw in the high-speed fluid from the outside and push the low-speed fluid from the inside out, achieving forced convection mixing in the radial direction. The flow state within the spiral channel is monitored using miniature pressure sensors and velocity probes to acquire pressure and velocity distribution data at various locations. Characteristic parameters reflecting the degree of mixing are extracted from the velocity distribution data. The Dean vortex intensity, mixing uniformity index, Taylor dispersion coefficient, and turbulence intensity are integrated to form a mixing enhancement feature. The Dean vortex intensity in the mixing enhancement feature reflects the strength of the secondary flow, while the mixing uniformity index reflects the uniformity of the concentration distribution.

[0020] In some embodiments, generating a mass transfer efficiency distribution based on the hybrid enhancement features includes: extracting the perturbation intensity of the hybrid enhancement features to form perturbation distribution data; dividing the perturbation distribution data according to the mass transfer contribution degree to form a mass transfer contribution level; identifying a fully mixed steady-state region from the mass transfer contribution level to generate a steady-state mass transfer identifier; and generating a mass transfer efficiency distribution based on the mass transfer contribution level and the steady-state mass transfer identifier.

[0021] Perturbation intensity is extracted from the mixing enhancement features to form perturbation distribution data. The mixing enhancement features include multiple parameters such as Dean vortex intensity, mixing uniformity index, Taylor dispersion coefficient, and turbulence intensity. The Taylor dispersion coefficient reflects the rate of axial material dispersion, while the turbulence intensity reflects the fluctuation level of the flow regime. Dean vortex intensity is characterized by the Dean number De, calculated using the following formula: Where Re is the Reynolds number, calculated as Re = ρvd / μ, where ρ is the sample fluid density, v is the average flow velocity, μ is the dynamic viscosity, the Reynolds number reflects the ratio of inertial force to viscous force, d is the hydraulic diameter of the channel, and R is the radius of curvature of the helix. A larger Dean's number indicates stronger secondary flow and more intense disturbance. When the Dean's number exceeds a critical value, a significant secondary flow vortex structure will appear in the channel, significantly enhancing the transverse mixing effect. The Dean's number distribution at each location is extracted from the mixing enhancement characteristics along the helical channel axis. Simultaneously, the turbulence intensity and velocity fluctuation amplitude at each location are extracted as auxiliary disturbance indices. The Dean's number, turbulence intensity, and velocity fluctuation amplitude at each location are processed using a unified normalization method. The normalization benchmark is the maximum value of each parameter across the entire channel range. After normalization, the values ​​of each index range from 0 to 1. The normalized disturbance indices are integrated according to their location to form disturbance distribution data. The row index of the disturbance distribution data is the axial position, the column index is the disturbance index type, and the element value of the disturbance distribution data is the normalized disturbance intensity value corresponding to the location and type. The disturbance distribution data records the disturbance status at each location from the channel entrance to the exit in positional order.

[0022] The perturbation distribution data is divided into mass transfer contribution levels based on its mass transfer contribution. The perturbation intensity varies at different locations within the data set. While there is a positive correlation between perturbation intensity and mass transfer efficiency, it is not a simple linear relationship. Within the range of low perturbation intensity, a small increase in perturbation intensity can significantly improve mass transfer efficiency. However, within the range of high perturbation intensity, further increases in perturbation intensity have a significantly weaker effect on mass transfer efficiency. The mass transfer contribution is defined as the rate of change of mass transfer efficiency with respect to perturbation intensity, i.e., dη / dI. When the perturbation intensity in the data set is low, the marginal gain of mass transfer efficiency is high; when the perturbation intensity is high, the mass transfer contribution tends to saturate. The mass transfer contribution at each location is calculated based on the perturbation distribution data. Taking a location with a perturbation intensity of 0.3 as an example, the mass transfer efficiency is approximately 0.5 and is in a rapid improvement phase, with a unit increase in perturbation intensity resulting in a mass transfer efficiency improvement of 0.15 to 0.2. When the perturbation intensity reaches 0.8, the mass transfer efficiency has already reached above 0.9, and a unit increase in perturbation intensity only results in a mass transfer efficiency improvement of 0.02 to 0.03. The mass transfer contribution is divided into three levels according to its numerical value: high contribution, medium contribution, and low contribution. The high contribution level corresponds to locations in the perturbation distribution data where the perturbation intensity is in the rapidly improving mass transfer efficiency range, while the low contribution level corresponds to locations where the perturbation intensity has entered the saturation range. A mass transfer contribution classification is formed by labeling the corresponding mass transfer contribution level at each location in the perturbation distribution data. The mass transfer contribution classification records the mass transfer contribution level at each location using the location as an index, with high, medium, and low contribution levels corresponding to different location ranges.

[0023] Steady-state mass transfer identifiers are generated by identifying well-mixed steady-state regions from the mass transfer contribution classification. In the low-contribution-level regions of the mass transfer contribution classification, the disturbance intensity is close to saturation, the mass transfer process in these regions is nearing completion, and the fluid concentration distribution is relatively uniform, making them ideal regions for spectral detection. The location distribution of low-contribution levels in the mass transfer contribution classification is scanned to identify continuous low-contribution regions. Low-contribution regions with a continuous axial length exceeding a set threshold are identified as candidate well-mixed steady-state regions. The threshold is typically set to 0.5 to 1 times the helical pitch to ensure sufficient flow stability within the region. If the helical pitch is 2 mm, the threshold is set to 1 to 2 mm. Steady-state verification is performed on the candidate well-mixed steady-state regions by detecting the spatial uniformity and temporal stability of the concentration distribution within the region. Spatial uniformity requires that the concentration deviation at each point within the region is less than 5% of the average concentration. If the concentration non-uniformity is too large, the absorbance values ​​obtained when the beam passes through different positions will show significant differences, and the repeatability of multiple measurements will not meet the detection precision requirements. Temporal stability requires that the concentration fluctuation within a continuous sampling period be less than 2%. Candidate regions verified through steady-state analysis are confirmed as fully mixed steady-state regions. Based on the mass transfer contribution classification, corresponding steady-state mass transfer identifiers are generated. The steady-state mass transfer identifiers include information such as the start and end positions of the fully mixed steady-state region, the region length, the average mass transfer efficiency, and the steady-state reliability score. The start and end positions in the steady-state mass transfer identifiers define the spatial range of the fully mixed steady-state region, and the steady-state reliability score in the steady-state mass transfer identifiers reflects the detection reliability of the region.

[0024] Mass transfer efficiency distribution is generated based on mass transfer contribution classification and steady-state mass transfer indicators. The location range and average disturbance intensity of each level region in the mass transfer contribution classification have been determined. Combined with the flow characteristics determined by the Reynolds number, the mass transfer efficiency value at each location is calculated according to the correspondence between mass transfer contribution level and mass transfer efficiency in the mass transfer contribution classification. In the high contribution level region of the mass transfer contribution classification, the mass transfer efficiency is in a rapid increase phase, with the mass transfer efficiency value gradually increasing from 0.1 to 0.2 at the inlet to 0.5 to 0.6; in the medium contribution level region, the rate of increase in mass transfer efficiency slows down, with the mass transfer efficiency value continuing to increase from 0.6 to 0.8 to 0.9; in the low contribution level region, the mass transfer efficiency tends to stabilize, with the mass transfer efficiency value remaining at a high level of 0.9 to 0.95. After determining the location range of the well-mixed steady-state region marked by the steady-state mass transfer marker, the mass transfer efficiency value at each location within the well-mixed steady-state region marked by the steady-state mass transfer marker is uniformly set as the average mass transfer efficiency of that region. A mass transfer efficiency higher than 0.95 in the region corresponding to the steady-state mass transfer marker typically indicates that mass transfer is essentially complete. At this point, most of the analyte in the sample fluid has reacted sufficiently with the reagent, and the concentration of the generated product has reached a stable level. Stable and repeatable absorption signals can be obtained through spectral detection. Based on the mass transfer contribution classification and the steady-state mass transfer marker, the mass transfer efficiency values ​​at each location in the entire channel are connected sequentially to form a continuous mass transfer efficiency distribution curve. The well-mixed steady-state region corresponding to the steady-state mass transfer marker is marked on the mass transfer efficiency distribution curve. The mass transfer efficiency distribution records the degree of mass transfer completion at each location from the inlet to the outlet. Regions with a mass transfer efficiency higher than 0.9 in the mass transfer efficiency distribution are the preferred regions for subsequent detection.

[0025] Step S120: Determine the key mass transfer enhancement zone through the mass transfer efficiency distribution, extract the concentration gradient distribution features from the key mass transfer enhancement zone, and perform gradient feature analysis on the concentration gradient distribution features to determine the in-situ detection sensitive area.

[0026] In some embodiments, determining the key mass transfer enhancement region through the mass transfer efficiency distribution includes: dividing the mass transfer efficiency distribution into a core mass transfer domain and an edge mass transfer domain; tracing from the core mass transfer domain to the edge mass transfer domain to form a mass transfer channel; detecting the enhancement degree of the mass transfer channel to obtain an enhancement node group; and determining the key mass transfer enhancement region based on the enhancement node group.

[0027] The mass transfer efficiency distribution is divided into a core mass transfer domain and a peripheral mass transfer domain. The mass transfer efficiency values ​​at different locations on the distribution curve differ significantly. Mixing is nearly complete in regions with high mass transfer efficiency, while mixing is still underway in regions with low mass transfer efficiency. The differences in the physicochemical properties of these two types of regions determine the different analytical strategies employed. A mass transfer efficiency threshold is set as the basis for this division. The threshold is selected based on the statistical characteristics of the mass transfer efficiency distribution, typically ranging from 70% to 80% of the maximum mass transfer efficiency. If the maximum mass transfer efficiency is 0.95, the threshold is set to approximately 0.70. Regions with mass transfer efficiencies above the threshold are marked as the core mass transfer domain. Within the core mass transfer domain, the mass transfer process is nearly complete, and the concentration distribution tends to be uniform. In spiral microchannels, the core mass transfer domain is usually located in the central region of the channel cross-section and the middle to lower reaches of the axis, where the Dean vortex has already thoroughly mixed the fluids of different concentrations. Regions with mass transfer efficiencies below a threshold in the mass transfer efficiency distribution are designated as edge mass transfer domains. Within these domains, mass transfer processes are still ongoing, and a significant concentration gradient exists. Edge mass transfer domains are typically located in the upstream inlet section of the helical channel, near the outer wall (high-velocity fluid layer), and near the inner wall (low-velocity fluid layer). The spatial extent and boundary positions of the core and edge mass transfer domains are recorded. These boundary positions represent the isosurfaces in the mass transfer efficiency distribution where the mass transfer efficiency equals the threshold. These isosurfaces typically exhibit irregular curved shapes, reflecting the spatial structure of the Dean vortex.

[0028] Mass transfer channels are formed by tracing from the core mass transfer domain to the peripheral mass transfer domain. Mass migrates from the high-concentration area of ​​the peripheral mass transfer domain to the low-concentration area of ​​the core mass transfer domain, and the migration path constitutes the mass transfer channel. The morphology of the mass transfer channel reflects how mixing disturbances affect the mass transfer process. A straight mass transfer channel indicates that mass transfer mainly relies on convective transport, while a spiral mass transfer channel indicates that Dean's vortex has a significant entanglement effect on the mass transfer path. A representative location within the core mass transfer domain is selected as the starting point for tracing. The representative location is usually the point with the highest mass transfer efficiency in the core mass transfer domain or the geometric center of the core mass transfer domain. In practice, 3 to 5 evenly distributed starting points can be selected within the core mass transfer domain to obtain a complete mass transfer channel network. Starting from the tracing starting point, the path is traced in the opposite direction of the concentration gradient. The tracing step size is set according to the channel size, usually 5% to 10% of the hydraulic diameter. Each step moves to the upstream position pointed by the concentration gradient at the current position. The tracing process is similar to climbing the steepest uphill slope on a topographic map. Path tracing continues until the boundary or channel entrance of the edge mass transfer domain is reached. Tracing the path from the core mass transfer domain into the edge mass transfer domain and recording the coordinates of each location along the path forms a mass transfer channel. Due to the presence of Dean's vortex, the mass transfer channel typically exhibits a spiral three-dimensional curve rather than a simple straight line. This spiral path significantly increases the mass migration distance but greatly enhances the mixing effect. Multiple mass transfer channels can be obtained by tracing paths from multiple starting points within the core mass transfer domain. Each mass transfer channel reflects a different spatial path of migration from the edge mass transfer domain to the core mass transfer domain.

[0029] The enhancement level of mass transfer channels is detected to identify enhanced node groups. The enhancement level of mass transfer is not uniform across different locations within the channels; some locations show particularly significant increases in mass transfer efficiency. These locations are the nodes where the enhancement effect is most concentrated, exhibiting intense fluid mixing and rapid concentration homogenization. Mass transfer efficiency values ​​are extracted along the path of each mass transfer channel, and the increment of mass transfer efficiency ΔE between adjacent locations is calculated as a measure of the local enhancement level: ΔE = ΔEi / Δs, where ΔEi is the difference in mass transfer efficiency between the i-th and (i-1)-th locations, and Δs is the path step size. The increment of mass transfer efficiency within the channel represents the increase in efficiency per unit path length. Locations where the increment of mass transfer efficiency exceeds a set threshold, typically set to 1.5 to 2 times the average increment across the entire path, are identified as enhanced nodes due to significant enhancement effects. Enhanced nodes are not randomly distributed in space but exhibit a regular pattern. They typically appear at the boundary intersection of Dean vortex pairs, where the two fluids rotating in opposite directions mix violently; at points where streamlines bend sharply in the mass transfer channel, where abrupt changes in centrifugal force lead to a significant increase in the radial velocity component; and at the interface between the wall shear layer and the main flow, where the velocity gradient is greatest. The mass transfer efficiency increment at these locations can be 3 to 5 times the average value of the mass transfer channel. After summarizing the enhanced nodes identified on each mass transfer channel and removing duplicates and adjacent nodes whose spatial distance is less than the channel's hydraulic diameter, enhanced node groups are formed. Each node in the enhanced node group is associated with its mass transfer efficiency increment value and the mass transfer channel number it belongs to.

[0030] The key mass transfer enhancement zone was determined based on the enhanced node group. The nodes in the enhanced node group are distributed at different locations in the channel; adjacent nodes may be far apart or closely clustered. Closely clustered nodes indicate a continuous and stable mass transfer enhancement effect in this region. Spatial clustering analysis was performed on the nodes in the enhanced node group. The clustering method used was density-based DBSCAN clustering or hierarchical clustering. The neighborhood radius was set to 1 to 2 times the hydraulic diameter of the channel, and the minimum number of nodes was set to 3 to 5. The specific values ​​of these parameters were adjusted according to the channel size and the node distribution density in the enhanced node group. Enhanced nodes that are spatially adjacent in the enhanced node group were grouped into the same cluster. Each cluster represents a spatial region of concentrated mass transfer enhancement. If the enhanced node group contains 30 nodes, it may form 3 to 5 clusters distributed at different axial positions in the channel. The spatial extent of each cluster is calculated, using the outer envelope rectangle or convex hull of the nodes within the enhanced node group as the region boundary. Simultaneously, the average mass transfer efficiency increment of the nodes within the cluster is calculated as the enhancement degree index for that region; a higher enhancement degree index indicates a more significant mass transfer enhancement effect. Clusters with enhancement degree indices exceeding a set threshold are identified as key mass transfer enhancement regions. Key mass transfer enhancement regions near the inlet correspond to the Dean vortex establishment process in the initial mixing stage, while key mass transfer enhancement regions in the middle section correspond to the stable enhancement process after the Dean vortex has fully developed.

[0031] Concentration gradient distribution characteristics were extracted from the critical mass transfer enhancement zone. The active mass transfer process within this zone indicates a significant concentration gradient driving the migration of substances from high-concentration areas to low-concentration areas. The distribution characteristics of this concentration gradient reflect the direction and intensity of mass transfer. A concentration detection array was arranged within the critical mass transfer enhancement zone, with the detection points uniformly distributed radially and axially to form a two-dimensional grid. Five to seven detection points were arranged radially to cover the entire cross-section from the inner wall to the outer wall, while the axial spacing was one-tenth of the helical pitch to capture the flow changes within the critical mass transfer enhancement zone. The instantaneous concentration values ​​at each point within the critical mass transfer enhancement zone were obtained using a micro-fiber probe or fluorescence tracing technology. Spatial difference calculations were performed on the concentration values ​​at each detection point within the critical mass transfer enhancement zone to determine the concentration gradient. The radial concentration gradient reflects the radial mass transfer intensity driven by the Dean vortex. When the Dean vortex draws high-concentration fluid from the outer wall to the inner wall, a steep radial concentration gradient is formed at the vortex boundary. At this point, the radial concentration gradient can be 5 to 10 times that of the axial concentration gradient. The axial concentration gradient reflects the convective mass transfer intensity in the mainstream direction. Both factors jointly determine the mass transfer mode within the critical mass transfer enhancement zone. The spatial distribution pattern of the concentration gradient within the critical mass transfer enhancement zone was statistically analyzed. Characteristic parameters such as peak position, peak magnitude, gradient direction, and gradient distribution symmetry were extracted. These characteristic parameters were integrated to form a concentration gradient distribution feature. The peak position in the concentration gradient distribution feature identifies the spatial coordinates of the most drastic concentration changes, and the gradient direction indicates the main path of mass migration.

[0032] Gradient feature analysis of the concentration gradient distribution characteristics is used to determine the sensitive areas for in-situ detection. The spatial distribution pattern of the concentration gradient is analyzed to identify regions where the concentration gradient approaches zero. These regions have uniform concentration distribution and are suitable for quantitative spectral detection. The area near the central axis of the Dean vortex pair exhibits average concentration with minimal fluctuations due to vortex symmetry, significantly improving measurement repeatability. Simultaneously, regions with large concentration gradients are analyzed, as these areas show significant concentration changes and are suitable for real-time monitoring of dynamic concentration changes. Vortex boundaries and wall shear layers are typical high-gradient regions in the concentration gradient distribution characteristics. Different sensitive areas are selected based on the detection objective. For quantitative analysis, low-gradient steady-state regions are selected to obtain accurate concentration measurements. For dynamic monitoring, high-gradient active regions are selected to capture concentration change trends. If both quantitative accuracy and dynamic responsiveness are required, transitional regions with moderate gradients are chosen. The selected area is marked as a sensitive area for in-situ detection. The sensitive area for in-situ detection includes a spatial coordinate range and a detection applicability label. The applicability label for the sensitive area for in-situ detection includes three types: applicable to quantitative analysis, applicable to dynamic monitoring, and applicable to both.

[0033] Step S130: Perform multi-parameter detection on the in-situ detection sensitive area to obtain component concentration distribution, temperature gradient distribution and pressure response parameters, fuse the component concentration distribution, temperature gradient distribution and pressure response parameters to determine the effective acquisition range, and perform spectral acquisition on the effective acquisition range to obtain raw spectral data.

[0034] Specifically, multi-parameter detection is performed in the in-situ detection sensitive area to obtain component concentration distribution, temperature gradient distribution, and pressure response parameters. Multiple sensor arrays, including miniature concentration probes, thermocouple temperature sensors, and pressure sensors, are deployed within the in-situ detection sensitive area. These sensors are uniformly distributed according to a spatial grid to capture spatial variations in parameters within the sensitive area. The grid spacing is typically set to 0.1 to 0.2 times the hydraulic diameter of the channel to obtain sufficient spatial resolution. The concentration probes use fiber optic probes to measure the solute concentration at each point within the in-situ detection sensitive area. The concentration values ​​of all measuring points are arranged spatially to form a component concentration distribution, reflecting the uniformity of solute spatial distribution after the mixing process. The component concentration distribution records the concentration values ​​at each location in a two-dimensional grid. Thermocouple temperature sensors measure the temperature values ​​at each point. The local temperature gradient is obtained by differential calculation of the temperatures of adjacent measuring points. The temperature gradient vectors at each location are integrated to form a temperature gradient distribution. This temperature gradient distribution reflects the non-uniformity of the temperature field within the in-situ detection sensitive area due to viscous dissipation and uneven heat transfer. The gradient vector at each location in the temperature gradient distribution contains both magnitude and direction components. Pressure sensors measure static and dynamic pressure at various points, calculate the amplitude and frequency of pressure fluctuations to form pressure response parameters, which reflect the stability of the flow state within the in-situ detection sensitive area. The pressure response parameters include three indicators: pressure fluctuation amplitude, fluctuation frequency, and pressure mean.

[0035] In some embodiments, fusing the component concentration distribution, the temperature gradient distribution, and the pressure response parameter to determine the effective acquisition interval includes: hierarchically dividing the component concentration distribution to generate a concentration layer structure; mapping the temperature gradient distribution to the concentration layer structure to form a temperature mapping layer; pairing and fusing the temperature mapping layer and the pressure response parameter to form a fusion response factor; and determining the effective acquisition interval based on the fusion response factor.

[0036] The component concentration distribution is hierarchically divided to generate a concentration stratification structure. The component concentration distribution records the concentration values ​​at each location in the form of a spatial grid. Concentration values ​​vary across different locations; areas with higher concentrations have more solute, while areas with lower concentrations have less solute. This spatial difference reflects incomplete homogenization during the mixing process. The distribution range and characteristics of all concentration values ​​in the component concentration distribution are statistically analyzed, and the maximum, minimum, average, and standard deviation of the concentrations are calculated. If the standard deviation of the component concentration distribution is too large, it indicates insufficient mixing homogeneity, and the selection of sensitive areas for in-situ detection may need to be re-evaluated. Two solutions with different concentrations at the inlet are 1.0 g / L and 0.2 g / L, respectively. After mixing, the concentration range of the component concentration distribution is 0.3 to 0.9 g / L. This range is divided into 5 levels, each spanning 0.12 g / L. Each location in the component concentration distribution is traversed, and the level to which it belongs is determined based on the concentration value at that location. The concentration value falls within a certain threshold range and is assigned to that level. Location points at the same level are aggregated to form spatial regions. These regions may be continuous or consist of multiple dispersed sub-regions, depending on the spatial distribution pattern of component concentrations. In a spiral microchannel mixing scenario, high-concentration levels are typically located on the outer wall, low-concentration levels on the inner wall, and intermediate-concentration levels in the transition region of vortex mixing. The spatial range and concentration intervals of each level are integrated to form a concentration stratification structure. This structure records the location boundaries of each level using level numbers as an index, and the boundaries between adjacent levels in the concentration stratification structure constitute concentration isosurfaces.

[0037] For example, mapping the temperature gradient distribution to the concentration stratification structure to form a temperature mapping layer includes: performing concentration difference analysis on the concentration stratification structure to determine the peak gradient position; projecting the temperature gradient distribution onto the peak gradient position to locate the temperature inflection point region; performing enhancement characteristic calibration on the temperature inflection point region to generate an enhancement region identifier; and fusing the enhancement region identifier with the peak gradient position to form a temperature mapping layer.

[0038] Concentration difference analysis is performed on the concentration stratification structure to determine the location of the peak gradient. The concentration stratification structure divides the space into regions with different concentration levels. Concentration differences exist between adjacent levels, and the location with the largest concentration difference is often where the mixing process is most active or least complete. The boundary values ​​of the concentration intervals in each level of the concentration stratification structure are read, and the concentration difference between adjacent levels is calculated. The concentration difference equals the lower boundary of the higher level minus the upper boundary of the lower level. The locations of the boundaries of each level are located in the spatial distribution of the concentration stratification structure. The level boundaries are the interfaces between two adjacent level regions. The concentration gradient at each level boundary is calculated. Due to the large concentration gradient at the boundary where the concentrations differ on both sides, the concentration gradient can reach 5 g / (L·mm) when the concentration difference is 0.5 g / L and the boundary width is 0.1 mm. The location of the level boundary with the largest concentration gradient value is marked as the peak gradient location. In spiral microchannels, the peak gradient location usually appears at the boundary of the Dean vortex, where two fluids of different concentrations are converging and mixing. There may be multiple peak gradient locations, distributed in different locations in space. All peak gradient locations are recorded to form a location list, and each peak gradient location is associated with its concentration gradient value and the layer boundary number to which it belongs.

[0039] The temperature gradient distribution is projected onto the peak gradient locations to pinpoint temperature inflection point regions. Peak gradient locations identify the spatial locations of most dramatic concentration changes; these locations, due to active mixing, are often accompanied by anomalies in the temperature field, where the heat released or absorbed during mixing creates local temperature highs or lows. The temperature gradient values ​​and directions at each peak gradient location are read, and the variation of the temperature gradient along the peak gradient location is calculated. The temperature gradient distribution is scanned along the hierarchical boundary line where the peak gradient location is located to identify locations where the temperature gradient value changes significantly. The presence of local maxima or minima at these locations indicates abrupt changes in the temperature field. Near peak gradient locations with active mixing, the temperature gradient can reach 5 to 10 °C / mm, significantly higher than the 0.5 °C / mm in other regions. The locations of abrupt changes in the temperature gradient are defined as temperature inflection point regions, which are areas where the concentration gradient peak and the temperature gradient in the temperature gradient distribution abnormally overlap. Within these regions, both mixing and heat transfer processes are in a non-steady state. The spatial range and temperature gradient anomaly characteristics of the identified temperature inflection point regions are recorded. Temperature inflection point regions are usually located at the junction of Dean vortex pairs and the junction of the wall thermal boundary layer and the mainstream.

[0040] Enhancement characteristics are calibrated and enhancement zone identifiers are generated in the temperature inflection point region. The temperature inflection point region is an area of ​​dual anomalies in concentration and temperature. Spectroscopic detection in this region may be affected by both concentration non-uniformity and temperature fluctuations. However, it is also a key location for observing the dynamic process of mixing and mass transfer. Analyzing the parameter variation characteristics within the temperature inflection point region reveals that if the concentration gradient and temperature gradient are in the same direction, it indicates coupling of mass and heat transfer processes; if they are in opposite directions, it indicates a competitive relationship between the two processes. The temporal characteristics of parameter fluctuations within the temperature inflection point region are calculated. Spectral analysis is performed on the collected concentration and temperature time series to identify the main fluctuation frequencies and amplitudes within the temperature inflection point region. Temperature fluctuations of 1 to 2°C within the temperature inflection point region may cause a change in the absorbance coefficient of 0.5% to 1%, requiring compensation during detection. The detection applicability of the temperature inflection point region is determined based on its parameter characteristics. Regions with fluctuation amplitudes less than 1°C and fluctuation frequencies far from the spectral acquisition frequency are calibrated as mild enhancement zones, while regions with fluctuation amplitudes greater than 2°C or fluctuation frequencies close to the acquisition frequency are calibrated as intense enhancement zones. The calibration results of each temperature inflection point region are recorded by generating enhancement zone identifiers. The enhancement zone identifiers include information such as region number, enhancement level, main fluctuation frequency, and detection recommendations. The level classification of enhancement zone identifiers directly affects the selection results of effective acquisition intervals.

[0041] A temperature mapping layer is formed by fusing enhancement zone identifiers with peak gradient locations. The list of peak gradient locations and the list of enhancement zone identifiers are spatially matched. For each peak gradient location, it is checked whether there is a region marked with an enhancement zone identifier nearby. If so, the information of the enhancement zone identifier is associated with the corresponding peak gradient location. If a peak gradient location falls within an intensity enhancement zone, that location is not suitable for quantitative spectral detection. For peak gradient locations not within the temperature inflection point region, temperature characteristics are supplemented and labeled based on the surrounding temperature gradient distribution characteristics. The supplementary labeling includes the temperature gradient value and temperature stability rating of the peak gradient location. Peak gradient locations with a temperature gradient less than 0.3℃ / mm are rated as high stability, 0.3 to 1℃ / mm as medium stability, and greater than 1℃ / mm as low stability. The spatial distribution of the concentration stratification structure, the peak gradient locations at the boundaries of each stratum, and their associated temperature characteristic information are integrated to form a temperature mapping layer. The temperature mapping layer is based on a spatial grid and overlays two types of attribute labels: concentration stratum and temperature stability.

[0042] The temperature mapping layer and pressure response parameters are paired and fused to form a fused response factor. The temperature mapping layer has already integrated information from two dimensions: concentration and temperature. The concentration dimension describes mixing uniformity, and the temperature dimension describes thermal field stability. Flow stability information also needs to be added. The pressure fluctuation amplitude and frequency at each location in the pressure response parameters are read. The spatial locations of the pressure measurement points in the pressure response parameters are aligned with the spatial grid of the temperature mapping layer. For regions with sparse measurement points, spatial interpolation is used to obtain the complete pressure field distribution. The distribution characteristics of the pressure response parameters are statistically analyzed within each level of the temperature mapping layer. The average pressure fluctuation amplitude and dominant frequency component in each level are calculated. The pressure fluctuation amplitude in the pressure response parameters reflects the stability of the flow regime, and the dominant frequency component reflects the periodicity of the flow regime fluctuations. If the dominant frequency of the pressure fluctuation in the pressure response parameters is close to the spectral acquisition frequency, periodic interference will be introduced into the spectral signal. A comprehensive scoring formula for three indicators—concentration uniformity, temperature stability, and pressure fluctuation—is established. The variables are defined as σc (concentration standard deviation), cavg (average concentration), and... For temperature gradient magnitude, The maximum allowable temperature gradient, Δp is the pressure fluctuation amplitude, and pavg is the average pressure. Typical weighting coefficients are w1=0.4, w2=0.3, and w3=0.3. (Comprehensive score) The comprehensive score F value is calculated for each location in the temperature mapping layer. The spatial distribution of the F value forms the fusion response factor. The value range of the fusion response factor is 0 to 1. The larger the value, the better the detection conditions at that location. Regions with a fusion response factor below 0.5 usually have problems such as uneven concentration, temperature fluctuation or pressure pulsation, and are not suitable for spectral acquisition.

[0043] The effective acquisition range is determined based on the fusion response factor. The fusion response factor describes the quality of detection conditions at each location in a spatial distribution. Locations with high fusion response factor values ​​have less detection interference and better signal quality, while locations with low fusion response factor values ​​have more detection interference and poorer signal quality. A screening threshold for the fusion response factor is set. The selection of the threshold needs to balance detection quality and acquisition area. If the threshold is too high, the usable area will be too small, affecting light throughput; if the threshold is too low, it will include areas with poor detection conditions, affecting signal quality. The threshold is usually set between 0.7 and 0.85 of the maximum value of the fusion response factor. The spatial distribution of the fusion response factor is scanned, and locations with fusion response factor values ​​exceeding the threshold are marked as candidate acquisition locations. If the maximum value of the fusion response factor is 0.92 and the threshold is set to 0.75, then locations with fusion response factor values ​​greater than 0.69 are all candidate locations. Candidate locations may be continuous areas or multiple scattered small areas. Connectivity analysis is performed on the candidate acquisition locations to identify interconnected locations that are aggregated into connected regions. The largest connected region is usually the main acquisition candidate area. The connected regions are normalized by simplifying irregular boundaries into rectangles or ellipses to fit the geometric constraints of the spectrometer's detection window. The normalized regions are then defined as the effective acquisition intervals, which include three pieces of information: spatial range, geometric shape, and center coordinates.

[0044] Raw spectral data is obtained by acquiring spectra within the effective acquisition range. The spectral acquisition process is as follows: Figure 3As shown, the sample fluids in the first sample storage bottle 12 and the second sample storage bottle 13 are driven by the peristaltic pump 14 and the flow rate is monitored by the flow sensor 15 before entering the spiral microchannel flow cell 16 for mixing. The broadband light beam emitted by the light source 10 is collimated by the collimating lens 11 and passes through the detection window 17 and the flowing sample within the effective acquisition range. The transmitted light is focused by the focusing lens 18 and enters the spectrometer 19. The output signal of the spectrometer 19 is transmitted to the control computer 21 for processing via the data acquisition card 20. The effective acquisition range is the optimal detection position determined after multi-parameter fusion screening. Within the range, the concentration is uniform, the temperature is stable, and the flow is smooth. The detection window 17 is adjusted according to the spatial position of the effective acquisition range so that its center is aligned with the geometric center of the effective acquisition range. The window size is set according to the range range. If the radial span of the effective acquisition range is 0.8 mm, the diameter of the detection window 17 is set to 0.5 mm to ensure that the light beam falls completely within the range. The acquisition parameters of the spectrometer 19 are configured. The integration time is set according to the sample concentration and absorbance coefficient, and the number of acquisitions is set according to the accuracy requirements. Averaging multiple acquisitions can improve the signal-to-noise ratio. After the electrical signal is converted from analog to digital by the data acquisition card 20, it is transmitted to the control computer 21 for dark current subtraction and wavelength calibration, forming raw spectral data containing wavelength axis and absorbance axis. The wavelength range covers the characteristic absorption peaks of the target component, and the absorbance value at each wavelength position reflects the light absorption intensity of the sample at that wavelength.

[0045] Step S140: Determine the response accuracy threshold based on the effective acquisition range and the hybrid enhancement characteristics, conduct flow stability assessment on the key mass transfer enhancement zone to form a flow stability index, and construct a zoned monitoring configuration based on the flow stability index and the response accuracy threshold.

[0046] Specifically, the response accuracy threshold is determined based on the effective acquisition interval and the hybrid enhancement features. The average fusion response factor of the effective acquisition interval reflects the comprehensive score of the detection conditions within the effective acquisition interval. A higher score indicates better detection conditions and can support higher accuracy requirements. When the fusion response factor is 0.85, an accuracy target of less than 1% concentration measurement error can be set; when the fusion response factor is 0.65, the accuracy target needs to be relaxed to within 3%. The Dean vortex intensity and mixing uniformity index in the hybrid enhancement features reflect the mixing effect. A high vortex intensity in the hybrid enhancement features indicates more thorough mixing, but also greater flow pattern fluctuations. If the hybrid enhancement features show that the vortex intensity exceeds the critical value, the sampling frequency needs to be increased accordingly to suppress the influence of fluctuations. A correspondence between the effective acquisition interval score, the degree of hybrid enhancement features, and detection accuracy is established. When the fusion response factor is greater than 0.8 and the mixing index is greater than 0.9, a higher accuracy requirement can be set; when the fusion response factor is between 0.6 and 0.8, a medium accuracy requirement is set; and when the fusion response factor is less than 0.6, a basic accuracy requirement is set. The corresponding detection accuracy level is determined based on the actual values ​​of the effective acquisition range and the mixed enhancement features. The accuracy level is then converted into a specific response accuracy threshold. The response accuracy threshold includes three dimensions of index values: lower limit of signal-to-noise ratio, concentration resolution, and time response speed. The lower limit of signal-to-noise ratio in the response accuracy threshold is usually set between 50 and 200, and the concentration resolution in the response accuracy threshold is set between 0.01 and 0.1 g / L.

[0047] In some embodiments, the process of assessing the flow stability of the key mass transfer enhancement region to form a flow stability index includes: performing flow tracking on the key mass transfer enhancement region to obtain a flow change curve; separating ordered disturbance components and disordered turbulent components from the flow change curve; defining the steady-state range of the disordered turbulent components to obtain a steady-state boundary set; and determining the stable range based on the steady-state boundary set and the ordered disturbance components to generate a flow stability index.

[0048] Flow regime tracking was performed in the critical mass transfer enhancement zone to obtain flow regime change curves. The flow regime within the critical mass transfer enhancement zone continuously changes over time and space. The periodic oscillations of the Dean vortex, the random fluctuations of turbulence, and minor fluctuations in inlet conditions all cause time-varying velocity characteristics, which directly affect the stability of the spectral signal. Flow regime tracking sensors were deployed at representative locations within the critical mass transfer enhancement zone, including the region center, near the boundaries, and at each enhancement node. Hot-film or laser Doppler flowmeters were used to achieve high time-resolution velocity measurement. Hot-film flowmeters have a fast response time but are sensitive to flow direction, while laser Doppler flowmeters offer high accuracy but are more expensive. The acquisition parameters were configured, and the sampling frequency was set according to the characteristic frequency of the Dean vortex. The characteristic frequency can be estimated from the flow velocity and channel size: f = v / (2πR), where v is the average flow velocity and R is the helix radius. If the average flow velocity is 10 mm / s and the helix radius is 5 mm, the characteristic frequency is approximately 0.3 Hz. The sampling frequency was set to 30 times the characteristic frequency, i.e., 10 Hz, to meet the requirements of the Nyquist sampling theorem. Continuous acquisition was initiated, with the acquisition duration covering multiple evolution cycles of the flow regime within the critical mass transfer enhancement zone, approximately 30 to 60 seconds. A time series of flow velocity changes over time was obtained for each measuring point. The time series of each measuring point within the critical mass transfer enhancement zone were organized according to spatial location to form a set of flow regime change curves. These flow regime change curves reflect the temporal dynamic characteristics of the flow regime within the critical mass transfer enhancement zone. The fluctuation pattern of the flow regime change curves includes both the periodic motion of the Dean vortex and random turbulence components.

[0049] For example, separating ordered disturbance components and disordered turbulent components from the flow regime change curve includes: performing curve morphology analysis on the flow regime change curve to determine morphological abrupt change points; locating the velocity change rate at the morphological abrupt change points to generate a change rate benchmark; and comparing the change rate benchmark with the flow regime change curve to divide it into ordered disturbance components and disordered turbulent components.

[0050] The morphology of the flow regime change curve is analyzed to identify abrupt morphological changes. The overall morphology of the flow regime change curve reflects the evolution of the flow regime. Periodic fluctuations indicate that the Dean vortex dominates the flow regime, while random fluctuations indicate that turbulence dominates. Sudden changes in morphology at certain points on the flow regime change curve indicate a shift in the flow motion mode at that moment. The flow regime change curve is smoothed to eliminate interference from high-frequency noise in the morphology analysis. The smoothing method uses a moving average filter window with a width set to one-tenth of the vortex characteristic period. If the vortex period is 3s, the filter window is 0.3s, corresponding to 3 sampling points. The first and second derivatives of the smoothed flow regime change curve are calculated. The first derivative reflects the trend of flow velocity change, and the second derivative reflects the change in the trend, i.e., the concavity and convexity of the flow regime change curve. The moments when the sign of the second derivative changes in the flow regime change curve are identified; these moments are the inflection points of the flow regime change curve, where the concavity and convexity of the curve change. Further screening at inflection points revealed locations where the change in the second derivative exceeded a threshold, set at twice the root mean square value of the second derivative. These locations exhibiting significant abrupt changes in curve shape were marked as morphological abrupt change points. In the flow regime of the spiral microchannel, these morphological abrupt change points correspond to a shift in the direction of Dean's vortex motion or a sudden change in vortex intensity. The rate of change of velocity switches between positive and negative when the vortex transitions from the rising phase to the falling phase. Within a 30-second acquisition period, if the vortex period is 3 seconds, approximately 20 morphological abrupt change points can be identified.

[0051] A velocity change rate benchmark is generated by locating the velocity change rate at morphological abrupt change points. These points are critical moments in the transition of flow patterns; vortices pass through them when moving from an accelerating phase to a decelerating phase or vice versa, making the velocity change rate at these moments particularly significant. The first derivative of the flow pattern curve at each morphological abrupt change point represents the velocity change rate. The distribution characteristics of the velocity change rate at all morphological abrupt change points are statistically analyzed. The velocity change rate of periodic disturbances caused by Dean vortices at morphological abrupt change points exhibits regularity: the rate of change is negative when the vortex transitions from a peak to a trough, and the rates of change across different periods are similar; conversely, the rate of change is positive when transitioning from a trough to a peak, and this pattern is also repetitive. If the rate of change at the positive transition points among the 20 morphological abrupt change points is distributed between 2.5 and 3.5 mm / s², it indicates a clearly defined characteristic of ordered disturbance. The mean and standard deviation of the velocity change rate at the morphological abrupt change point are calculated. The mean reflects the typical rate of change level caused by the vortex, and the standard deviation reflects the degree of fluctuation of the rate of change at the morphological abrupt change point. The mean is determined as the rate of change benchmark. The rate of change benchmark is divided into positive benchmark and negative benchmark, corresponding to the rising phase and falling phase of the vortex, respectively. The positive rate of change benchmark is 3.0 mm / s², and the negative rate of change benchmark is -3.2 mm / s². If the actual rate of change deviates too much from the rate of change benchmark, it indicates that there is significant turbulence superposition.

[0052] The rate of change benchmark is compared with the flow regime change curve to classify it into ordered disturbance components and disordered turbulent components. The velocity change rate sequence of the flow regime change curve over the entire time period is calculated, and the time values ​​of the rate of change sequence in the flow regime change curve are compared with the rate of change benchmark. The comparison results reflect whether the dominant component of the flow regime at each time point is ordered disturbance or disordered turbulence. A method for calculating the rate of change deviation is established. For moments with positive rate of change, the relative deviation from the positive rate of change benchmark is calculated as δ=(dv / dt-dvref / dt) / (dvref / dt), where dv / dt is the actual velocity change rate at the current time, and dvref / dt is the positive rate of change benchmark value. For moments with negative rate of change, the relative deviation from the negative rate of change benchmark is calculated. Moments with a relative deviation less than a set threshold, such as 30%, are defined as moments dominated by ordered disturbances. When the actual rate of change is 2.5 mm / s², while the baseline rate of change is 3.0 mm / s², the relative deviation is 17%, indicating that ordered disturbances dominate. Moments with a relative deviation greater than the threshold are defined as moments of turbulent superposition. When the actual rate of change is 5.0 mm / s², the relative deviation is 67%, indicating that turbulent superposition occurs. Based on the classification results, ordered disturbances and disordered turbulent components are reconstructed. At moments dominated by ordered disturbances, velocity fluctuations in the flow regime change curve are classified as ordered disturbances. At moments of turbulent superposition, fluctuations exceeding the ordered baseline in the flow regime change curve are classified as disordered turbulent components. Ordered disturbances exhibit good periodicity, while disordered turbulent components exhibit near-Gaussian randomness. The separation of ordered disturbances and disordered turbulent components provides a quantitative basis for flow regime stability assessment.

[0053] Steady-state boundary sets are obtained by defining the steady-state range of the disordered turbulent components. Although the instantaneous values ​​of the disordered turbulent components are unpredictable, their statistical characteristics should remain relatively constant when the flow is stable. Statistical quantities such as standard deviation, kurtosis, and skewness fluctuate little under steady-state conditions but fluctuate significantly under non-steady-state conditions. Statistical parameters of the disordered turbulent components are calculated, including mean, standard deviation (σturb), kurtosis, and skewness. Ideally, the mean of the disordered turbulent components should be close to zero. The standard deviation (σturb) reflects the turbulence intensity, while kurtosis and skewness reflect the symmetry and sharpness of the disordered turbulent component distribution. If the kurtosis of the disordered turbulent components deviates significantly from 3, it indicates the presence of intermittent strong pulsations. The data collection period is divided into multiple sub-periods, each with a length of 5 to 10 seconds. The statistical parameters of the disordered turbulent components within each sub-period are calculated, and the trends of these parameters over time are observed. If the variation of the standard deviation σturb of the disordered turbulent flow component between different sub-periods is less than 10% of the mean, the period is considered a steady-state period. If the variation of the standard deviation σturb of the disordered turbulent flow component exceeds 20%, the period is considered a non-steady-state period. The range of values ​​for each statistical parameter of the disordered turbulent flow component within the steady-state period is determined, and these ranges are integrated to form a steady-state boundary set. The steady-state boundary set includes three boundary conditions: the upper limit of the standard deviation σmax, the kurtosis range, and the skewness range. In the steady-state boundary set, the upper limit of the standard deviation σmax is typically 5% to 15% of the average flow velocity. The kurtosis range is 2.5 to 3.5, close to the theoretical value of 3 for a Gaussian distribution. The skewness range is -0.3 to 0.3, close to a symmetrical distribution.

[0054] A flow stability index is generated by determining the stable range based on the steady-state boundary set and ordered disturbance components. The amplitude stability of the ordered disturbance components is calculated by statistically analyzing the fluctuation amplitudes of these components over multiple consecutive vortex cycles. The coefficient of variation (CVord), i.e., the ratio of standard deviation to mean, is calculated. A smaller CVord indicates better repeatability of periodic disturbances. A CVord less than 0.1 indicates highly regular vortex motion, where periodic fluctuations in the spectral signal can be effectively eliminated through synchronous sampling. A formula for calculating the flow stability index is then established. Where Is is the flow stability index, CVOrd is the coefficient of variation of the amplitude of the ordered disturbance component, σturb is the standard deviation of the disordered turbulent component, σmax is the upper limit of the standard deviation in the steady-state boundary set, and α and β are weighting coefficients with α+β=1. A higher flow stability index indicates that the Dean vortex motion is regular and the turbulence intensity is controlled, while a lower flow stability index indicates the presence of irregular vortex oscillations or intermittent strong turbulence pulses. The flow stability index is calculated for each measuring point in the key mass transfer enhancement zone. The flow stability indices of each measuring point are then organized according to their spatial location to form a stability distribution map. The overall flow stability index of the key mass transfer enhancement zone is calculated by taking the weighted average of the measuring points. The weight of the center position is set to 0.4, and the weight of the edge position is set to 0.2. The weighted average design takes into account that the center position contributes more to the overall stability.

[0055] In some embodiments, constructing a zoned monitoring configuration based on the flow stability index and the response accuracy threshold includes: performing threshold grading on the response accuracy threshold to generate a multi-level accuracy standard; associating and mapping the flow stability index with the multi-level accuracy standard to form a stability grading region; determining the monitoring sampling density of each region based on the stability grading region to generate a regional sampling configuration; and integrating the regional sampling configuration with the multi-level accuracy standard to form a zoned monitoring configuration.

[0056] A multi-level accuracy standard is generated by threshold grading the response accuracy threshold. The various indicators in the response accuracy threshold include the lower limit of signal-to-noise ratio (SNR), concentration resolution, and time response speed. The numerical range of each indicator in the response accuracy threshold is relatively wide, and the requirements for the response accuracy threshold vary in different detection scenarios. High-precision scenarios require strict values ​​for the indicators in the response accuracy threshold, while low-precision scenarios can appropriately relax the requirements. The indicators in the response accuracy threshold are divided into multiple levels according to their stringency: SNR greater than 100 is classified as Level 1 (high precision), 50 to 100 as Level 2 (medium precision), 20 to 50 as Level 3 (basic precision), and less than 20 as Level 4 (error-tolerant precision). Concentration resolution greater than 0.05 g / L is classified as Level 1, 0.05 to 0.1 g / L as Level 2, 0.1 to 0.2 g / L as Level 3, and less than 0.2 g / L as Level 4. Time response speed is also graded: response time less than 0.1 s is Level 1, 0.1 to 0.5 s is Level 2, 0.5 to 1 s is Level 3, and greater than 1 s is Level 4. The grading schemes of various indicators are integrated to form a multi-level accuracy standard. The multi-level accuracy standard establishes four accuracy levels from high to low. Each level in the multi-level accuracy standard corresponds to a set of specific indicator requirements. The first level of the multi-level accuracy standard requires a signal-to-noise ratio greater than 100, a concentration resolution better than 0.05 g / L, and a response time less than 0.1 s. The fourth level of the multi-level accuracy standard relaxes the requirements to a signal-to-noise ratio greater than 20, a concentration resolution better than 0.2 g / L, and a response time less than 1 s.

[0057] A stability grading system is formed by mapping the flow stability index to multi-level accuracy standards. A correspondence is established between the flow stability index and the multi-level accuracy standards: regions with high flow stability indices have small flow fluctuations and can support higher accuracy levels within the multi-level accuracy standards; regions with low flow stability indices have large flow fluctuations and can only achieve lower accuracy levels within the multi-level accuracy standards. This correspondence ensures that accuracy requirements match the testing conditions. Grading thresholds for the flow stability index are set: regions with flow stability indices greater than 0.8 are mapped to the first-level high-precision standard; flow stability indices from 0.6 to 0.8 are mapped to the second-level medium-precision standard; flow stability indices from 0.4 to 0.6 are mapped to the third-level basic accuracy standard; and flow stability indices less than 0.4 are mapped to the fourth-level fault-tolerant accuracy standard. Based on the spatial distribution and grading threshold of the flow stability index, the key mass transfer enhancement zone is divided into multiple stability grading zones. If the flow stability index is 0.85 at the center and 0.55 at the edge, the center is classified as a Level 1 stability grading zone and the edge as a Level 3 stability grading zone. Each stability grading zone corresponds to a different precision level. The spatial range, corresponding precision level, and statistical characteristics of the flow stability index within each stability grading zone are recorded. The division results of the stability grading zones may exhibit a nested distribution, such as a high-precision zone at the center and a low-precision zone on the periphery, or they may exhibit a strip distribution extending along the vortex boundary.

[0058] Based on the stability grading zones, the monitoring sampling density of each region is determined to generate the regional sampling configuration. Different accuracy levels of stability grading zones have different requirements for sampling strategies. High-precision stability grading zones have high reliability for single sampling due to stable flow, while low-precision stability grading zones require multiple samplings and averaging to suppress the impact of flow fluctuations. The accuracy requirements corresponding to each stability grading zone are analyzed to determine the minimum number of samplings and sampling intervals required to meet these accuracy requirements. For the Level 1 high-precision stability grading zone, the sampling density is set at 10 samples / s with a single integration time of 50ms, which is sufficient for the accuracy requirement with a single sampling. For the Level 2 medium-precision stability grading zone, the sampling density is set at 20 samples / s with a single integration time of 30ms, requiring 2 to 3 samples for averaging. For the Level 3 basic accuracy stability grading zone, the sampling density is set at 30 samples / s with a single integration time of 20ms, requiring 4 to 6 samples for averaging. For the Level 4 fault-tolerant accuracy stability grading zone, the sampling density is set at 50 samples / s with a single integration time of 10ms, requiring 8 to 10 samples for averaging. The spatial range of each stability grading zone is paired with the corresponding sampling parameters to form a regional sampling configuration. The regional sampling configuration includes parameters such as sampling density, integration time, and average number of times for each stability grading zone. The regional sampling configuration records the complete sampling parameters of each zone using the stability grading zone number as an index.

[0059] A zoned monitoring configuration is formed by integrating regional sampling configurations with multi-level accuracy standards. The records of each region in the regional sampling configuration are associated with the corresponding accuracy levels in the multi-level accuracy standards, establishing a triplet mapping of region-sampling parameter-accuracy standard. In the regional sampling configuration, the first-level zone corresponds to a sampling density of 10 times / s and an accuracy requirement of signal-to-noise ratio (SNR) greater than 100 in the multi-level accuracy standards. The fourth-level zone corresponds to a sampling density of 50 times / s and an accuracy requirement of SNR greater than 20 in the multi-level accuracy standards. A switching logic between regions is designed. When the detection probe moves from one region to another, the sampling parameters need to be automatically adjusted. The switching is triggered when the probe position enters the boundary of the new region. The switching execution involves updating parameters such as sampling density and integration time in the regional sampling configuration. When moving from the first-level zone to the third-level zone, the sampling density increases from 10 times / s to 30 times / s. A real-time accuracy monitoring mechanism is added to calculate the current accuracy index in real time during the data acquisition process and compare it with multi-level accuracy standards. If the actual signal-to-noise ratio is 80 while the multi-level accuracy standard requires 100, the sampling count is automatically increased from 2 to 4 times or the integration time is extended from 50ms to 80ms for compensation. The regional division scheme, the sampling parameters of each region in the regional sampling configuration, the accuracy requirements in the multi-level accuracy standards, the switching logic, and the real-time compensation mechanism are integrated to form a zonal monitoring configuration. The zonal monitoring configuration records the complete correspondence between the spatial range, sampling strategy, and accuracy target of each region.

[0060] Step S150: Parameter deviation detection is performed on the in-situ detection sensitive area to identify abnormal response areas. The abnormal response areas are then matched with the concentration gradient distribution characteristics to generate flow regime correction coefficients. Based on the flow regime correction coefficients and the zonal monitoring configuration, the original spectral data is corrected to generate spectral detection results.

[0061] Specifically, parameter deviation detection is performed on the sensitive area of ​​in-situ detection to identify abnormal response regions. During the detection process, the sensitive area of ​​in-situ detection may exhibit parameter anomalies that deviate from expectations. These anomalies include concentration deviations from expected values, temperatures exceeding stable ranges, or abnormal pressure fluctuations, all of which affect the quality and accuracy of the spectral signal. During spectral acquisition, multiple parameters within the sensitive area of ​​in-situ detection are monitored simultaneously, including concentration, temperature, and pressure. The real-time monitored values ​​within the sensitive area are compared with the expected values ​​of the parameters to calculate the deviation. Concentration deviation is defined as the percentage difference between the measured concentration and the expected concentration relative to the expected concentration. For example, when the measured concentration is 0.52 g / L and the expected concentration is 0.50 g / L, the concentration deviation is 4%. Deviation thresholds are set for each parameter as the basis for anomaly judgment: a concentration deviation threshold of 5%, a temperature deviation threshold of 1℃, and a pressure deviation threshold of 3%. Any deviation exceeding a threshold is considered an anomaly. The parameter deviation distribution at each location within the in-situ detection sensitive area is scanned. Locations with single-parameter anomalies or multiple-parameter anomalies are marked to form anomaly response regions. The boundaries of the anomaly response regions are determined by contour lines with deviations equal to the threshold. The spatial range and deviation amplitude of the anomaly response regions are recorded.

[0062] In some embodiments, the step of performing gradient matching between the abnormal response region and the concentration gradient distribution feature to generate flow regime correction coefficients includes: constructing a concentration offset distribution based on the abnormal response region; performing location matching between the concentration offset distribution and the concentration gradient distribution feature to obtain a matching reference value; expanding the matching reference value according to the deviation level to form a correction adaptation range; and performing boundary correction on the correction adaptation range to generate flow regime correction coefficients.

[0063] A concentration offset distribution was constructed based on the anomaly response region. The parameter deviation amplitude varies at different locations within the anomaly response region; some locations deviate only slightly from the expected value, while others show larger deviations. The spatial distribution pattern of the deviation amplitude reflects the cause and characteristics of the anomaly. The measured and expected concentration values ​​of each monitoring point within the anomaly response region were read, and the concentration offset at each point was calculated. The concentration offset equals the measured value minus the expected value; when the measured value is 0.53 g / L and the expected value is 0.50 g / L, the offset is +0.03 g / L. The concentration offsets of each monitoring point within the anomaly response region were arranged spatially, and spatial interpolation methods were used to extend the offset values ​​of discrete points into a continuous spatial distribution. Kriging interpolation or inverse distance weighted interpolation can be used. A concentration offset distribution map was generated. Positive values ​​in the concentration offset distribution indicate that the measured concentration is higher than the expected concentration (i.e., high concentration area), negative values ​​indicate that the measured concentration is lower than the expected concentration (i.e., low concentration area), and zero values ​​indicate that the concentration is consistent with the expected concentration with no offset. Analyzing the spatial pattern of concentration shift distribution, if the concentration shift distribution shows a unilateral distribution, such as higher concentration on the outer wall side and lower concentration on the inner wall side, it may be related to the asymmetric mixing of the Dean vortex. If the concentration shift distribution shows a random distribution, it may be related to the random fluctuation of the flow regime.

[0064] Location matching is performed between the concentration shift distribution and the concentration gradient distribution to obtain matching reference values. The concentration shift distribution and the concentration gradient distribution are superimposed in the same spatial coordinate system, and the spatial correspondence between the high-value areas of the concentration shift distribution and the high-gradient areas of the concentration gradient distribution is analyzed. The strength of the correspondence reflects the degree of correlation between the concentration deviation and the background concentration field. If the high-value areas of the concentration shift distribution and the high-gradient areas of the concentration gradient distribution have a high degree of overlap, it indicates that the concentration deviation mainly originates from the sampling location sensitivity caused by the concentration gradient. A sampling location shift of 0.1 mm may lead to a concentration reading change of 0.2 g / L. In this case, correction should be performed targeting location sensitivity. The spatial correlation coefficient between the concentration shift distribution and the concentration gradient distribution is calculated. A correlation coefficient greater than 0.7 indicates a strong correlation, 0.3 to 0.7 indicates a moderate correlation, and less than 0.3 indicates a weak correlation. The matching reference value is determined based on the strength of the correlation. When the correlation is strong, the matching reference value is the ratio of the concentration gradient value to the offset in the concentration gradient distribution characteristics as the position sensitivity coefficient. When the concentration gradient is 2 g / (L·mm) and the offset is 0.04 g / L, the sensitivity coefficient is 0.02 mm, which means that the position offset of 20 μm can explain the deviation. When the correlation is moderate, the matching reference value is a weighted combination of the offset and the gradient. When the correlation is weak, the matching reference value is determined only based on the offset itself.

[0065] The matching reference values ​​are expanded according to deviation levels to form a correction adaptation range. The deviations in the concentration shift distribution are divided into multiple levels based on magnitude: deviations less than 0.02 g / L are classified as Level 1 (slight deviation), 0.02 to 0.05 g / L as Level 2 (moderate deviation), 0.05 to 0.1 g / L as Level 3 (significant deviation), and greater than 0.1 g / L as Level 4 (severe deviation). The thresholds for each level are determined with reference to the accuracy requirements of spectral detection and the statistical characteristics of the actual deviation distribution. For each deviation level, a corresponding correction range is calculated based on the matching reference value. The correction range for Level 1 deviation is 0.8 to 1.2 times the matching reference value, the correction range for Level 2 deviation is 0.6 to 1.4 times the matching reference value, and Levels 3 and 4 deviations require wider correction ranges to accommodate larger deviation fluctuations. The correction ranges for each deviation level are integrated to form a correction adaptation range. The correction adaptation range establishes a mapping relationship between deviation levels and correction ranges. In the correction adaptation range, the correction coefficient for level 1 deviation is 0.98 to 1.02, the correction coefficient for level 2 deviation is 0.95 to 1.05, the correction coefficient for level 3 deviation is 0.90 to 1.10, and the correction coefficient for level 4 deviation is 0.85 to 1.15.

[0066] The calibration adaptation range is boundary-corrected to generate flow regime correction coefficients. The calibration adaptation range provides the correction range for each deviation level; however, the boundaries of these ranges may be discontinuous or unreasonable. Ranges for adjacent levels may overlap or have gaps, and the correction for extreme levels may exceed physically reasonable limits. The calibration adaptation range checks whether the ranges for adjacent deviation levels are continuously connected. If overlapping areas exist, the midpoint of the overlapping area is used as the dividing point to ensure seamless connection between adjacent ranges. The calibration adaptation range checks whether each correction range meets physical constraints. The correction coefficients should be positive and typically do not exceed 1.5. If the calculated correction coefficients exceed reasonable limits, amplitude limiting is applied. The correction adaptation range after boundary correction is converted into specific flow regime correction coefficient values. The calculation formula is Kc=1+sign(Δc)×|Δc / cref|×η, where Kc is the flow regime correction coefficient, Δc is the concentration offset, cref is the reference concentration, sign(Δc) is the offset direction sign function (+1 for high, -1 for low), and η is the correction gain coefficient (0.8 for level 1 deviation, 0.9 for level 2, 1.0 for level 3, and 1.1 for level 4). A smooth transition band is set at the boundary of the correction adaptation range to avoid abrupt changes in the flow regime correction coefficient when switching deviation levels. Cosine interpolation is used within the transition band to achieve a smooth transition. The numerical range of the flow regime correction coefficient is limited to 0.85 to 1.15 after boundary correction.

[0067] The original spectral data is corrected based on the flow regime correction coefficient and the zoned monitoring configuration to generate spectral detection results. The flow regime correction coefficient provides the quantitative basis for correction in each abnormal region. A flow regime correction coefficient greater than 1 indicates that the readings in the original spectral data need to be corrected upwards, while a flow regime correction coefficient less than 1 indicates that the readings in the original spectral data need to be corrected downwards. The zoned monitoring configuration specifies the detection accuracy requirements and sampling strategies for each region. The spectral signals at each acquisition location in the original spectral data are read, and it is determined whether the location is within the abnormal response region based on the acquisition location. For spectral signals in the original spectral data located in the normal region, the standard correction process is directly applied for dark current subtraction, baseline correction, and smoothing. For spectral signals in the original spectral data located in the abnormal response region, the flow regime correction coefficient is superimposed on the standard correction. The corrected absorbance is equal to the original absorbance multiplied by the flow regime correction coefficient. When the original absorbance at a certain location in the original spectral data is 0.85 and the flow regime correction coefficient is 1.05, the corrected absorbance is 0.89. Based on the accuracy level of the area in the zonal monitoring configuration, the accuracy of the corrected spectral signal is verified. In the high-precision zone of the zonal monitoring configuration, the corrected signal-to-noise ratio must meet the first-level standard. If it does not meet this standard, the average number of samplings is increased or the area is marked as requiring retesting. The corrected spectral signals from all acquisition locations are integrated to form the spectral detection results. These results include wavelength-absorbance curves, signal-to-noise ratios at each wavelength, detection accuracy level, and anomaly correction flags. The wavelength-absorbance curves in the spectral detection results reflect the light absorption characteristics of the sample at each wavelength, and the anomaly correction flags record the location and magnitude of the correction after flow regime correction.

[0068] To implement the above-described method embodiments, a spectroscopic detection method based on a spiral microchannel flow cell is proposed to achieve the corresponding functions and technical effects. See also... Figure 4 , Figure 4 This diagram illustrates a structural block diagram of a spectroscopic detection device 400 based on a spiral microchannel flow cell according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The spectroscopic detection device 400 based on a spiral microchannel flow cell according to an embodiment of this application includes: The sample import module 401 is used to collect sample fluid, perform multi-inlet spiral channel mixing and enhancement processing on the sample fluid to generate mixing and enhancement characteristics, and generate mass transfer efficiency distribution based on the mixing and enhancement characteristics. Mass transfer analysis module 402 is used to determine the key mass transfer enhancement region through the mass transfer efficiency distribution, extract the concentration gradient distribution features from the key mass transfer enhancement region, and perform gradient feature analysis on the concentration gradient distribution features to determine the in-situ detection sensitive region. The parameter fusion module 403 is used to perform multi-parameter detection on the in-situ detection sensitive area to obtain component concentration distribution, temperature gradient distribution and pressure response parameters, fuse the component concentration distribution, the temperature gradient distribution and the pressure response parameters to determine the effective acquisition range, and perform spectral acquisition on the effective acquisition range to obtain raw spectral data. The monitoring construction module 404 is used to determine the response accuracy threshold based on the effective acquisition range and the hybrid enhancement features, to perform flow stability assessment on the key mass transfer enhancement zone to form a flow stability index, and to construct a partitioned monitoring configuration based on the flow stability index and the response accuracy threshold. The detection output module 405 is used to detect parameter deviations in the in-situ detection sensitive area to identify abnormal response areas, perform gradient matching between the abnormal response areas and the concentration gradient distribution characteristics to generate flow regime correction coefficients, and perform correction processing on the original spectral data based on the flow regime correction coefficients and the zonal monitoring configuration to generate spectral detection results.

[0069] The aforementioned spectroscopic detection device 400 based on a spiral microchannel flow cell can implement a spectroscopic detection method based on a spiral microchannel flow cell as described in the above method embodiments. The options in the above method embodiments are also applicable to this embodiment and will not be detailed here. The remaining contents of this application embodiment can be referred to the contents of the above method embodiments, and will not be repeated in this embodiment.

[0070] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A spectral detection method based on a spiral microchannel flow cell, characterized in that, include: A sample fluid is collected, and the sample fluid is subjected to multi-inlet spiral channel mixing enhancement processing to generate mixing enhancement characteristics. This includes: monitoring the flow state within the spiral channel, acquiring pressure and velocity distribution data at various locations, integrating Dean vortex intensity, mixing uniformity index, Taylor dispersion coefficient, and turbulence intensity to generate mixing enhancement characteristics, and generating a mass transfer efficiency distribution based on the mixing enhancement characteristics. This includes: extracting disturbance intensity from the mixing enhancement characteristics to form disturbance distribution data; dividing the disturbance distribution data according to mass transfer contribution to form a mass transfer contribution grade; identifying a fully mixed steady-state region from the mass transfer contribution grade to generate a steady-state mass transfer identifier; and generating a mass transfer efficiency distribution based on the mass transfer contribution grade and the steady-state mass transfer identifier. Determining a key mass transfer enhancement region through the mass transfer efficiency distribution includes: dividing the mass transfer efficiency distribution into a core mass transfer domain and an edge mass transfer domain; tracing from the core mass transfer domain to the edge mass transfer domain to form a mass transfer channel; detecting the enhancement degree of the mass transfer channel to obtain an enhancement node group; determining a key mass transfer enhancement region based on the enhancement node group; extracting concentration gradient distribution features from the key mass transfer enhancement region; and performing gradient feature analysis on the concentration gradient distribution features to determine an in-situ detection sensitive area. Multi-parameter detection is performed on the in-situ detection sensitive area to obtain component concentration distribution, temperature gradient distribution and pressure response parameters. The component concentration distribution, temperature gradient distribution and pressure response parameters are fused to determine the effective acquisition range. Spectral acquisition is performed on the effective acquisition range to obtain raw spectral data. Based on the effective acquisition range and the hybrid enhancement features, a response accuracy threshold is determined, and the flow stability of the key mass transfer enhancement zone is evaluated to form a flow stability index. Based on the flow stability index and the response accuracy threshold, a zoned monitoring configuration is constructed. The abnormal response region is identified by parameter deviation detection in the in-situ detection sensitive area. The abnormal response region is then matched with the concentration gradient distribution characteristics to generate a flow regime correction coefficient. Based on the flow regime correction coefficient and the zonal monitoring configuration, the original spectral data is corrected to generate a spectral detection result.

2. The method according to claim 1, characterized in that, The step of fusing the component concentration distribution, the temperature gradient distribution, and the pressure response parameters to determine the effective acquisition range includes: The concentration distribution of the components is hierarchically divided to generate a concentration stratification structure; The temperature gradient distribution is mapped onto the concentration stratification structure to form a temperature mapping layer; The temperature mapping layer and the pressure response parameters are paired and fused to form a fused response factor; The effective acquisition range is determined based on the fusion response factor.

3. The method according to claim 1, characterized in that, The process of evaluating the flow stability of the key mass transfer enhancement region to form a flow stability index includes: Flow regime change curves were obtained by performing flow regime tracking on the key mass transfer enhancement region; Separate the ordered disturbance component and the disordered turbulent component from the flow regime change curve; The steady-state range of the disordered turbulent components is defined to obtain a steady-state boundary set; The stability range is determined based on the steady-state boundary set and the ordered disturbance component to generate the flow stability index.

4. The method according to claim 1, characterized in that, The step of constructing a partitioned monitoring configuration based on the flow stability index and the response accuracy threshold includes: The response accuracy threshold is graded to generate a multi-level accuracy standard; The flow stability index is correlated and mapped with the multi-level accuracy standard to form a stability grading region; Based on the stability grading zone, the monitoring sampling density of each region is determined to generate the regional sampling configuration; The regional sampling configuration and the multi-level accuracy standard are integrated to form a zoned monitoring configuration.

5. The method according to claim 1, characterized in that, The step of generating flow regime correction coefficients by gradient matching between the abnormal response region and the concentration gradient distribution characteristics includes: Concentration offset distribution is constructed based on the abnormal response region; The concentration offset distribution and the concentration gradient distribution features are matched to obtain a matching reference value; The matching reference values ​​are expanded according to the deviation level to form a correction and adaptation range; The flow regime correction coefficient is generated by boundary correction of the correction adaptation range.

6. The method according to claim 2, characterized in that, The step of mapping the temperature gradient distribution onto the concentration stratification structure to form a temperature mapping layer includes: Concentration difference analysis was performed on the concentration stratification structure to determine the location of the peak gradient; The temperature gradient distribution is projected onto the peak gradient location to pinpoint the temperature inflection point region. Enhanced characteristic calibration is performed in the temperature inflection point region to generate an enhanced region identifier; The enhanced region identifier is fused and mapped with the peak gradient location to form a temperature mapping layer.

7. The method according to claim 3, characterized in that, The separation of ordered disturbance components and disordered turbulent components from the flow regime change curve includes: The flow regime change curves are analyzed to determine the points of abrupt change in morphology. At the point of abrupt change in morphology, the rate of change of flow velocity is located to generate a rate of change benchmark. The rate of change benchmark is compared with the flow regime change curve to divide the components into ordered disturbance components and disordered turbulent components.

8. A spectroscopic detection device based on a spiral microchannel flow cell, used to implement the method as described in any one of claims 1-7, characterized in that, include: The sample import module is used to collect sample fluid, perform multi-inlet spiral channel mixing enhancement processing on the sample fluid to generate mixing enhancement characteristics, and generate mass transfer efficiency distribution based on the mixing enhancement characteristics; The mass transfer analysis module is used to determine the key mass transfer enhancement region through the mass transfer efficiency distribution, extract the concentration gradient distribution features from the key mass transfer enhancement region, and perform gradient feature analysis on the concentration gradient distribution features to determine the in-situ detection sensitive region. The parameter fusion module is used to perform multi-parameter detection on the in-situ detection sensitive area to obtain component concentration distribution, temperature gradient distribution and pressure response parameters, fuse the component concentration distribution, the temperature gradient distribution and the pressure response parameters to determine the effective acquisition range, and perform spectral acquisition on the effective acquisition range to obtain raw spectral data; The monitoring construction module is used to determine the response accuracy threshold based on the effective acquisition range and the hybrid enhancement features, to perform flow stability assessment on the key mass transfer enhancement zone to form a flow stability index, and to construct a partitioned monitoring configuration based on the flow stability index and the response accuracy threshold. The detection output module is used to detect parameter deviations in the in-situ detection sensitive area to identify abnormal response areas, perform gradient matching between the abnormal response areas and the concentration gradient distribution characteristics to generate flow regime correction coefficients, and perform correction processing on the original spectral data based on the flow regime correction coefficients and the zonal monitoring configuration to generate spectral detection results.

Images