Traditional Chinese medicinal material processing fault sensing method based on risk assessment

Abstract

The invention discloses a traditional Chinese medicinal material processing fault sensing method based on risk assessment, and particularly relates to the field of risk assessment of traditional Chinese medicinal material processing, and the method comprises the steps: constructing a multi-channel heat diffusion behavior path and a temperature gradient change sequence in a drying cavity based on drying working condition parameter data and a spatial structure factor map; extracting path response gradient data and identifying a local blocking structure in thermal diffusion to form structured data input for path thermal behavior analysis; an isothermal input path subset is constructed based on excitation source parameters in a thermal driving input set, statistics is performed on a path heat energy transfer sequence, a heat consumption offset behavior in unit time is identified, and an offset clustering structure is formed. A risk assessment chain based on path thermal behavior dynamic response characteristics is constructed by carrying out behavior modeling, offset identification and structural association judgment on local heat transfer abnormity in a thermal diffusion path, so that recognizable, attributable and regulatable perception of a heat conduction imbalance phenomenon in a drying process is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment of Chinese medicinal materials processing, and more particularly, to a Chinese medicinal materials processing fault perception method based on risk assessment. Background Art

[0002] In existing Chinese medicinal material drying processes, systems typically rely on fixed heating settings and simple temperature monitoring to control the process. However, in actual processing, the stacking method of materials, moisture content, and the structure of the drying equipment often vary greatly, resulting in uneven and uncontrollable heat transfer in different areas. Some areas may experience abnormalities due to insufficient heating or excessive heat accumulation.

[0003] These problems do not always manifest as noticeable temperature changes, but rather as subtle heat transfer deviations, localized energy consumption variations, or temperature fluctuations, making them difficult to detect or locate using traditional monitoring methods. Especially during the drying process, these anomalies can gradually accumulate and expand, ultimately affecting the overall drying effect and product quality. However, because current methods cannot analyze these changes from the perspective of heat conduction paths, it is also difficult to promptly determine the severity and development trend of the problem.

[0004] Therefore, the core problem currently faced in this field is the lack of risk assessment capabilities based on the dynamic changes of the heat conduction process. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for sensing faults in Chinese medicinal materials processing based on risk assessment. By performing behavioral modeling, offset identification and structural correlation judgment on local heat transfer anomalies in the heat diffusion path, a risk assessment chain based on the dynamic response characteristics of the path thermal behavior is constructed, thereby achieving identifiable, attributable and controllable perception of the heat conduction imbalance phenomenon in the drying process.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting faults in Chinese herbal medicine processing based on risk assessment, comprising:

[0007] S1. Based on the drying condition parameter data and spatial structure factor map, the multi-channel thermal diffusion behavior path and temperature gradient change sequence in the drying chamber are constructed. The path response gradient data is extracted and the local blocking structure in the thermal diffusion is identified to form structured data input for path thermal behavior analysis.

[0008] S2. Based on the excitation source parameters in the thermal drive input set, a subset of equal heat input paths is constructed. The heat energy transfer sequence of the paths is statistically analyzed. The heat consumption offset behavior per unit time is identified and the offset cluster structure is formed. The spatial distribution and evolution trend characteristics are solved to mark redundant consistency violation events.

[0009] S3. Perform a cross-analysis of path deviation behavior by comparing historical behavior templates with operational parameters to determine whether the cause of the path deviation has attribution evidence, extract the spatial location corresponding to the non-attributable path, and establish a spatial identification result for potential heat accumulation risk areas;

[0010] S4. Using the spatial location of potential heat accumulation risk as a trigger condition, execute the control parameter extraction, control path construction, intervention control coverage judgment and path repair indicator extraction process, establish a risk identification label bound to the spatial location and control parameter combination, and write it into the quality traceability information structure.

[0011] In a preferred embodiment, S1 further includes: collecting operating condition parameter data during the drying process as initial input, the operating condition parameter data including a heating power sequence, a wind speed distribution diagram, a drying time sequence table, and a spatial structure factor, the spatial structure factor including a packing density diagram, a moisture content gradient diagram, and a morphological envelope structure;

[0012] The spatial structure factor and the heat source excitation parameter are subjected to data fusion processing to construct a thermal drive input set; the stacking density map, moisture gradient map and morphological envelope structure in the spatial structure factor are subjected to three-dimensional spatial grid mapping processing to construct a spatial structure factor map;

[0013] A spatial mapping operation is performed on the thermal drive input set to solve the multi-channel heat diffusion behavior path in the drying chamber structure formed by the drying equipment and divide it into a continuous path set according to the time step.

[0014] In a preferred embodiment, S1 further includes: decomposing the time segments of each path data marked as a thermal diffusion behavior path in the continuous path set, extracting the temperature gradient change sequence of the corresponding area from the time segment of each path data, generating response gradient data of the time segment of each path data and summarizing them into a path response gradient data set;

[0015] Determine whether there is a mutation point in the temperature change rate in the time segment of each path data in the path response gradient dataset. If a mutation point is detected, the corresponding time segment of the path data is labeled as an abnormal segment and merged into the temperature rise mutation segment set. If no mutation point is detected, the original complete label of the time segment of the path data remains unchanged.

[0016] The temperature rise mutation fragment set is aggregated and merged according to the spatial proximity between the time fragments of the path data to generate a heat diffusion blocking structure covering the time fragments of multiple local abnormal path data. The generated heat diffusion blocking structure is used as the input data in the subsequent path consistency evaluation link.

[0017] In a preferred embodiment, S2 further includes: screening out a path cluster having consistent excitation source parameters from the continuous path set, the excitation source parameters including power timing and initial accumulation state parameters, and constructing an equal heat input path subset based on the path cluster;

[0018] Perform energy consumption accumulation segmentation processing on each path data in the equal heat input path subset, and further calculate the unit time heat energy transfer sequence of each path data in the accumulation segment to form a path energy consumption statistical set;

[0019] Determine whether there is an overall deviation trend in the unit time heat energy transfer sequence of each path data in the path energy consumption statistical set. If there is a deviation trend, extract the heat consumption offset value from the corresponding path data and mark it as a path deviation behavior set; if no deviation trend is found, mark the corresponding path data as a redundant consistent path.

[0020] In a preferred embodiment, S2 further includes: extracting a time segment in which each path data in the path deviation behavior set shows a deviation trend in the unit time heat energy transfer sequence, and performing spatial cluster analysis based on the spatial position of each time segment and the corresponding heat consumption offset amplitude to generate a heat consumption offset cluster structure;

[0021] The offset intensity, cluster area, and evolution time trend of each heat consumption offset cluster structure are further calculated, and the peak change characteristics of the spatial distribution pattern and heat consumption evolution trend are extracted as the behavioral feature dimension of the subsequent template comparison process;

[0022] Obtain archived path behavior sample data from historical drying batches and construct a standard path behavior reference set. Determine whether there is repeated clustering behavior across path clusters in multiple heat consumption offset cluster structures. If so, confirm that a redundant consistency violation event has occurred and pass the corresponding path offset behavior set to the path anomaly attribution process.

[0023] If there is no repeated behavior, an offset pattern comparison template is constructed based on the standard path behavior reference set, and a similarity matching process of the path offset behavior is performed to evaluate the abnormal probability of the behavior in the historical behavior distribution.

[0024] In a preferred embodiment, S3 further includes: performing spatial overlap comparison on the path deviation behavior set associated with the consistency violation event and the path behavior templates predefined in the standard path behavior reference set to construct a path similarity matching set;

[0025] Determine whether there is a historical path template in the path similarity matching set that matches the spatial distribution pattern extracted from the heat consumption offset cluster structure corresponding to the path offset behavior set and the peak change characteristics of the heat consumption evolution trend;

[0026] If a matching historical path template exists, the path's deviation-causing feature information is extracted from the corresponding matching result in the path similarity matching set and marked as attributed deviation behavior; if a matching historical path template does not exist, the path is marked as an abnormal path with no reference;

[0027] The spatial location area corresponding to the no-reference abnormal path is spatially corresponded and cross-analyzed with the set of operating parameters contained in the thermal drive input set; the operating parameter set includes indicators such as wind pressure response delay, local heat source power fluctuation characteristics, and channel inner wall impedance change rate, which are used to identify the equipment state change characteristics at this spatial location.

[0028] In a preferred embodiment, S3 further includes: determining whether the set of operating state parameters can explain the deviation phenomenon in the set of path deviation behaviors; if so, marking it as a device attribution anomaly; if not, classifying it as an unattributable heat source area; and recording the spatial position corresponding to the path deviation behavior classified into the unattributable heat source area as the spatial position of the unattributable heat source area, which serves as the input position for subsequent atlas projection analysis;

[0029] The spatial location of the unattributable heat source area is projected onto the spatial structure factor map, and the stacking density value, moisture content gradient value and morphological boundary closure index under the corresponding three-dimensional spatial grid in the spatial structure factor map are extracted. Combined with the structural closure threshold, it is determined whether the spatial location of the unattributable heat source area meets the closed stacking structure condition defined in the spatial structure factor map; if the closed stacking structure condition is met, the spatial location of the unattributable heat source area is determined as a potential heat accumulation risk, and the determination result is input into the intervention strategy construction process.

[0030] In a preferred embodiment, S4 further includes: extracting parameter items with adjustment functions from the operating condition parameter data, and constructing a control parameter data set, the control parameter data set including wind speed channel adjustment parameters, heat power distribution control parameters, and ventilation sequence adjustment parameters;

[0031] Mapping the spatial locations identified as potential heat accumulation risks to a control parameter data set; and based on the spatial location mapping results corresponding to the potential heat accumulation risks, generating control paths acting on the spatial locations of the potential heat accumulation risks from the control parameter data set to construct a control path data set;

[0032] At the same time, the drying operation affected by the current control path is marked as the drying batch number, which is used for the subsequent correlation record of control execution results and risk perception indicators;

[0033] Determine whether each group of control paths in the control path data set covers the spatial location of potential heat accumulation risks in the spatial and temporal dimensions. If the complete coverage conditions in the spatial and temporal dimensions are met, combine the group of control paths to form an intervention parameter package.

[0034] If the coverage condition is met only for part of the spatial area or part of the time dimension, this control path is marked as a boundary control path combination and written into the delay control parameter data set.

[0035] In a preferred embodiment, S4 further includes: writing the intervention parameter package as an input control instruction into the control execution process, simultaneously collecting the path deviation behavior response data after the control execution, extracting the heat rate offset value change and the temperature gradient sequence change result, and calculating the path repair degree evaluation index, the path repair degree evaluation index including the heat rate offset compression rate constructed based on the heat rate offset value and the fluctuation recovery amplitude of the temperature gradient change sequence after the control execution;

[0036] Determine whether the path repair degree evaluation index reaches the preset thermal behavior stability threshold. If so, mark the drying operation corresponding to the current control path as "intervention completed";

[0037] If it is not achieved, the control parameter combination is rebuilt based on the feedback result of the current control path repair, and the intervention parameter instructions are generated again and written into the control execution process to form an iterative closed-loop control process for path repair;

[0038] The spatial location determined as a potential heat accumulation risk, the combination of control parameters that constitute the corresponding control path, and the results based on the path repair degree assessment index are bound to the current drying batch number execution data to generate a batch risk identification label containing the spatial location, control parameter combination and repair index, and written into the quality traceability information structure as a reference for subsequent risk status identification and control strategy decision-making.

[0039] A risk assessment-based fault perception system for Chinese herbal medicine processing, including a path modeling module, an offset identification module, a risk labeling module, and a closed-loop control module;

[0040] The path modeling module constructs the multi-channel thermal diffusion behavior path and temperature gradient change sequence in the drying chamber based on the drying condition parameter data and spatial structure factor map. It extracts the path response gradient data and identifies the local blocking structure in the thermal diffusion process, forming structured data input for path thermal behavior analysis.

[0041] The excursion identification module constructs a subset of equal heat input paths based on the excitation source parameters in the thermal drive input set, calculates the heat energy transfer sequence of the paths, identifies the heat consumption excursion behavior per unit time, forms an excursion cluster structure, and solves the spatial distribution and evolution trend characteristics to mark redundant consistency violation events.

[0042] The risk labeling module is used to perform a cross-analysis of path deviation behaviors by comparing historical behavior templates with operational parameters, determining whether the cause of path deviation has attribution evidence, extracting the spatial locations corresponding to non-attributable paths, and establishing spatial identification results for potential heat accumulation risk areas.

[0043] The closed-loop control module is used to use the spatial location of potential heat accumulation risks as a trigger condition to execute the control parameter extraction, control path construction, intervention control coverage judgment and path repair indicator extraction processes, establish risk identification labels bound to the spatial location and control parameter combination, and write them into the quality traceability information structure.

[0044] The technical effects and advantages of the present invention are as follows:

[0045] 1. By constructing a multi-channel thermal diffusion behavior path and response gradient change sequence, we achieved structured expression and real-time perception of the dynamic imbalance state of heat conduction during the drying process of traditional Chinese medicines, solving the key problem that traditional methods cannot identify the evolution trend of local thermal anomalies.

[0046] 2. By screening the consistency of excitation source parameters and statistics of heat transfer sequences, a subset of equal heat input paths was constructed. This method can identify heat transfer deviation behaviors within a unit time and cluster them into evolutionary structures, thus enabling monitoring of redundant consistency violation events at the path level.

[0047] 3. By comparing path deviation behavior with historical templates and combining it with operational parameters for attribution, the system can distinguish between explainable and unattributable anomaly sources, thereby providing precise spatial positioning and evolution mechanism support for subsequent risk labeling;

[0048] 4. By mapping unattributable paths to spatial structural factor maps and combining them with closed accumulation condition judgment, we can quantitatively identify potential heat accumulation areas, filling the gap in traditional methods in identifying non-explicit structural anomalies.

[0049] 5. By constructing a control path that binds control parameters to spatial positions and forming an intervention parameter package, and combining control feedback repair indicators to achieve closed-loop reconstruction of the control path, the risk response capability and process controllability of the drying process are improved;

[0050] 6. By writing the risk spatial location, control parameter combination and repair indicators into the quality traceability information structure, a risk identification label system covering the entire process of drying behavior monitoring, anomaly identification, regulation execution and effect verification has been established. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flowchart of the framework of the method steps of the present invention.

[0052] Figure 2 Schematic diagram of the system module of the present invention.

[0053] Figure 3 A flow chart for the path modeling of the present invention.

[0054] Figure 4 This is a flow chart of the offset identification of the present invention.

[0055] Figure 5 A risk labeling flow chart for the present invention.

[0056] Figure 6 This is a closed-loop control flow chart of the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] Refer to the instruction manual Figure 1-6 A risk assessment-based fault perception method for Chinese medicinal material processing according to an embodiment of the present invention includes:

[0059] S1. Based on the drying condition parameter data and spatial structure factor map, the multi-channel thermal diffusion behavior path and temperature gradient change sequence in the drying chamber are constructed. The path response gradient data is extracted and the local blocking structure in the thermal diffusion is identified to form structured data input for path thermal behavior analysis.

[0060] S2. Based on the excitation source parameters in the thermal drive input set, a subset of equal heat input paths is constructed. The heat energy transfer sequence of the paths is statistically analyzed. The heat consumption offset behavior per unit time is identified and the offset cluster structure is formed. The spatial distribution and evolution trend characteristics are solved to mark redundant consistency violation events.

[0061] S3. Perform a cross-analysis of path deviation behavior by comparing historical behavior templates with operational parameters to determine whether the cause of the path deviation has attribution evidence, extract the spatial location corresponding to the non-attributable path, and establish a spatial identification result for potential heat accumulation risk areas;

[0062] S4. Using the spatial location of potential heat accumulation risk as a trigger condition, execute the control parameter extraction, control path construction, intervention control coverage judgment and path repair indicator extraction process, establish a risk identification label bound to the spatial location and control parameter combination, and write it into the quality traceability information structure.

[0063] S1 also includes: collecting operating parameter data during the drying process as initial input, the operating parameter data including a heating power sequence, a wind speed distribution diagram, a drying time sequence table, and a spatial structure factor, the spatial structure factor including a bulk density diagram, a moisture content gradient diagram, and a morphological envelope structure;

[0064] The spatial structure factor and the heat source excitation parameter are subjected to data fusion processing to construct a thermal drive input set; the stacking density map, moisture gradient map and morphological envelope structure in the spatial structure factor are subjected to three-dimensional spatial grid mapping processing to construct a spatial structure factor map;

[0065] A spatial mapping operation is performed on the thermal drive input set to solve the multi-channel heat diffusion behavior path in the drying chamber structure formed by the drying equipment and divide it into a continuous path set according to the time step.

[0066] S1 also includes: decomposing the time segments of each path data marked as a thermal diffusion behavior path in the continuous path set, extracting the temperature gradient change sequence of the corresponding area from the time segment of each path data, generating response gradient data of the time segment of each path data and summarizing them into a path response gradient data set;

[0067] Determine whether there is a mutation point in the temperature change rate in the time segment of each path data in the path response gradient dataset. If a mutation point is detected, the corresponding time segment of the path data is labeled as an abnormal segment and merged into the temperature rise mutation segment set. If no mutation point is detected, the original complete label of the time segment of the path data remains unchanged.

[0068] The temperature rise mutation fragment set is aggregated and merged according to the spatial proximity between the time fragments of the path data to generate a heat diffusion blocking structure covering the time fragments of multiple local abnormal path data. The generated heat diffusion blocking structure is used as the input data for the subsequent path consistency evaluation step.

[0069] S1 also includes: mapping the drying condition parameters with the spatial structure factor map in three dimensions, generating a thermal drive input set and parsing it into a multi-channel thermal diffusion path set, performing time segment decomposition and temperature gradient response extraction, and constructing a local thermal diffusion blocking structure as structured input data;

[0070]

[0071] Among them B block The heat diffusion blocking structure set finally determined is used as a structured input for subsequent path thermal behavior analysis; In the path set, the jth spatial position of the i-th heat diffusion path is the three-dimensional coordinate point at the time segment τ; for The temperature gradient vector corresponding to the spatial position; is the rate of change of temperature gradient with time segment τ, which represents its response rate; θ grad is the threshold for determining temperature response mutation; Γ sync (p) represents the consistency coefficient between the point p and its spatial neighborhood paths in terms of temperature gradient direction and time synchronization; represents the set of other path nodes adjacent to point p in space; ∠(·,·) is the angle function between two vectors; 1 |τ-τ′|<∈ An indicator function indicating whether a time segment falls within the synchronization window ∈; λ cluster is the minimum synchronization aggregation coefficient threshold; point p in the formula represents the target path point, that is, the main point that needs to be judged whether it is in the blocking structure; p′ represents the adjacent path point that is spatially close to point p and temporally synchronized with point p, and p′ is used to judge the consistency of the gradient direction with point p; Represents the temperature gradient at the current path point p, which is used to describe the local heat diffusion direction and intensity at that point; represents the temperature gradient at a path point p′ that is spatially adjacent to point p, which is used for directionally consistent comparison with the thermal behavior of point p.

[0072] S2 also includes: screening out path clusters with consistent excitation source parameters from the continuous path set, the excitation source parameters including power timing and initial accumulation state parameters, and constructing equal heat input path subsets based on the path clusters;

[0073] Perform energy consumption accumulation segmentation processing on each path data in the equal heat input path subset, and further calculate the unit time heat energy transfer sequence of each path data in the accumulation segment to form a path energy consumption statistical set;

[0074] Determine whether there is an overall deviation trend in the unit time heat energy transfer sequence of each path data in the path energy consumption statistical set. If there is a deviation trend, extract the heat consumption offset value from the corresponding path data and mark it as a path deviation behavior set; if no deviation trend is found, mark the corresponding path data as a redundant consistent path.

[0075] S2 also includes: extracting the time segments of each path data in the path deviation behavior set that show a deviation trend in the unit time heat energy transfer sequence, and performing spatial cluster analysis based on the spatial position of each time segment and the corresponding heat consumption deviation amplitude to generate a heat consumption deviation cluster structure;

[0076] The offset intensity, cluster area, and evolution time trend of each heat consumption offset cluster structure are further calculated, and the peak change characteristics of the spatial distribution pattern and heat consumption evolution trend are extracted as the behavioral feature dimension of the subsequent template comparison process;

[0077] Obtain archived path behavior sample data from historical drying batches and construct a standard path behavior reference set. Determine whether there is repeated clustering behavior across path clusters in multiple heat consumption offset cluster structures. If so, confirm that a redundant consistency violation event has occurred and pass the corresponding path offset behavior set to the path anomaly attribution process.

[0078] If there is no repeated behavior, an offset pattern comparison template is constructed based on the standard path behavior reference set, and a similarity matching process of the path offset behavior is performed to evaluate the abnormal probability of the behavior in the historical behavior distribution;

[0079] S2 also includes: a spatial clustering and trend evolution model of unit time heat consumption offset behavior, which is used to identify unit time heat consumption offset behavior in a subset of equal heat input paths, construct a heat consumption offset cluster structure based on spatial position and offset amplitude, and extract the cluster evolution trend as a criterion for redundancy consistency violation;

[0080]

[0081] in The spatial location set of the heat consumption structure is represented; is the position where the unit heat consumption deviation behavior of the mth path occurs at time t; Φ m (τ) is the heat energy transfer value per unit time of the mth path at time τ; K(t-τ) is the time-weighted kernel function, which is used to enhance the time neighbor response; δ is the short-term heat consumption trend analysis window width; D m (t) is the spatial dispersion coefficient corresponding to path m at time t, which represents the unevenness of thermal energy diffusion; represents the local growth rate of the heat consumption response envelope in time; ξ drift Heat consumption evolution threshold for offset cluster identification.

[0082] S3 also includes: performing spatial overlap comparison on the path deviation behavior set associated with the consistency violation event and the path behavior template predefined in the standard path behavior reference set to construct a path similarity matching set;

[0083] Determine whether there is a historical path template in the path similarity matching set that matches the spatial distribution pattern extracted from the heat consumption offset cluster structure corresponding to the path offset behavior set and the peak change characteristics of the heat consumption evolution trend;

[0084] If a matching historical path template exists, the path's deviation-causing feature information is extracted from the corresponding matching result in the path similarity matching set and marked as attributed deviation behavior; if a matching historical path template does not exist, the path is marked as an abnormal path with no reference;

[0085] The spatial location area corresponding to the no-reference abnormal path is spatially corresponded and cross-analyzed with the set of operating parameters contained in the thermal drive input set; the operating parameter set includes indicators such as wind pressure response delay, local heat source power fluctuation characteristics, and channel inner wall impedance change rate, which are used to identify the equipment state change characteristics at this spatial location.

[0086] S3 also includes: determining whether the set of operating state parameters can explain the deviation phenomenon in the set of path deviation behaviors; if so, marking it as a device attributable anomaly; if not, classifying it as an unattributable heat source area; and recording the spatial position corresponding to the path deviation behavior classified into the unattributable heat source area as the spatial position of the unattributable heat source area, which serves as the input position for subsequent atlas projection analysis;

[0087] The spatial location of the unattributable heat source area is projected onto the spatial structure factor map. The stacking density value, moisture content gradient value, and morphological boundary closure index under the corresponding three-dimensional spatial grid in the spatial structure factor map are extracted. Combined with the structural closure threshold, it is determined whether the spatial location of the unattributable heat source area meets the closed stacking structure conditions defined in the spatial structure factor map. If the closed stacking structure conditions are met, the spatial location of the unattributable heat source area is determined as a potential heat accumulation risk, and the determination result is input into the intervention strategy construction process.

[0088] S3 also includes: a closed accumulation structure risk assessment model for unattributable path areas, which is used to map the locations of path deviation behaviors that cannot be explained by templates and operational parameters, calculate the closure factor based on three spatial indicators: accumulation density, moisture gradient, and boundary closure, and determine the potential heat accumulation risk;

[0089]

[0090] in represents the set of spatial locations identified as potential heat accumulation risks; r is the spatial location of the path offset that is classified as an unattributable source area; Υ(r) is the closed accumulation structure risk function, which is used to reflect the spatial accumulation characteristics of the path and the trend of heat retention; ρ(r) represents the local accumulation density value at position r; ρ bg is the average packing density of the background reference area; H w (r) is the moisture content value at position r; is the gradient of moisture content, which indicates the trend of uneven moisture content; Ωb (r) is the opening ratio of the morphological boundary at the position, which indicates whether the structure is closed; θ closure Represents the threshold used to determine closed stacking structures.

[0091] S4 also includes: extracting parameter items with adjustment functions from the working condition parameter data, and constructing a control parameter data set, the control parameter data set including wind speed channel adjustment parameters, heat power distribution control parameters and ventilation sequence adjustment parameters;

[0092] Mapping the spatial locations identified as potential heat accumulation risks to a control parameter data set; and based on the spatial location mapping results corresponding to the potential heat accumulation risks, generating control paths acting on the spatial locations of the potential heat accumulation risks from the control parameter data set to construct a control path data set;

[0093] At the same time, the drying operation affected by the current control path is marked as the drying batch number, which is used for the subsequent correlation record of control execution results and risk perception indicators;

[0094] Determine whether each group of control paths in the control path data set covers the spatial location of potential heat accumulation risks in the spatial and temporal dimensions. If the complete coverage conditions in the spatial and temporal dimensions are met, combine the group of control paths to form an intervention parameter package.

[0095] If the coverage condition is met only for part of the spatial area or part of the time dimension, this control path is marked as a boundary control path combination and written into the delay control parameter data set.

[0096] S4 also includes: writing the intervention parameter package as an input control instruction into the control execution process, collecting the path deviation behavior response data after the control execution, extracting the heat rate deviation value change and the temperature gradient sequence change result, and calculating the path repair degree evaluation index, the path repair degree evaluation index including the heat rate deviation compression rate constructed based on the heat rate deviation value and the fluctuation recovery amplitude of the temperature gradient change sequence after the control execution;

[0097] Determine whether the path repair degree evaluation index reaches the preset thermal behavior stability threshold. If so, mark the drying operation corresponding to the current control path as "intervention completed";

[0098] If it is not achieved, the control parameter combination is rebuilt based on the feedback result of the current control path repair, and the intervention parameter instructions are generated again and written into the control execution process to form an iterative closed-loop control process for path repair;

[0099] The spatial location identified as a potential heat accumulation risk, the control parameter combination that constitutes the corresponding control path, and the results of the path repair degree assessment index are bound to the current drying batch number execution data to generate a batch risk identification label containing the spatial location, control parameter combination, and repair index. This label is then written into the quality traceability information structure as a reference for subsequent risk status identification and control strategy decision-making.

[0100] S4 also includes: extracting control path data sets to perform interventions based on potential heat accumulation risk spatial locations, collecting heat consumption offsets and temperature gradient fluctuations after path repair, constructing an overall repair degree index and comparing it with the stability threshold;

[0101]

[0102] Among them S i is the comprehensive index of the degree of repair of the i-th path after intervention; τ1, τ2 are the time intervals for repair evaluation after the execution of the control path intervention; ΔQ i (τ) is the heat consumption offset value of path i at time τ; is the heat rate offset change rate, which is used to reflect the repair compression speed; is the temperature gradient of the path at time τ after intervention; is the temperature gradient at the previous moment; is the temperature gradient fluctuation amplitude, which indicates whether the thermal field is stable; γ stable The thermal stability threshold set for the system.

[0103] A risk assessment-based fault perception system for Chinese herbal medicine processing, including a path modeling module, an offset identification module, a risk labeling module, and a closed-loop control module;

[0104] The path modeling module constructs the multi-channel thermal diffusion behavior path and temperature gradient change sequence in the drying chamber based on the drying condition parameter data and spatial structure factor map. It extracts the path response gradient data and identifies the local blocking structure in the thermal diffusion process, forming structured data input for path thermal behavior analysis.

[0105] The excursion identification module constructs a subset of equal heat input paths based on the excitation source parameters in the thermal drive input set, calculates the heat energy transfer sequence of the paths, identifies the heat consumption excursion behavior per unit time, forms an excursion cluster structure, and solves the spatial distribution and evolution trend characteristics to mark redundant consistency violation events.

[0106] The risk labeling module is used to perform a cross-analysis of path deviation behaviors by comparing historical behavior templates with operational parameters, determining whether the cause of path deviation has attribution evidence, extracting the spatial locations corresponding to non-attributable paths, and establishing spatial identification results for potential heat accumulation risk areas.

[0107] The closed-loop control module is used to use the spatial location of potential heat accumulation risks as a trigger condition to execute the control parameter extraction, control path construction, intervention control coverage judgment and path repair indicator extraction processes, establish risk identification labels bound to the spatial location and control parameter combination, and write them into the quality traceability information structure.

[0108] It should be noted that in the formula structure involved in this solution, dimensionless terms can serve as proportionality or structural adjustment factors. When combined with quantities with units, they only play a numerical scaling role and do not introduce new physical dimensions. Therefore, they will not change or confuse the overall unit system of expression. This combination of "dimensionless terms and units" can be understood as a composite structural expression commonly used in mathematical and physical modeling, conforming to the principle of dimensional consistency and having a clear physical interpretation basis.

[0109] Secondly, in the formula structure of this scheme, if multiple variables with different physical units are involved, including but not limited to time, mass or energy variables, their joint appearance is to express the collaborative modeling relationship of multiple physical mechanisms. Each variable can be formed into a unified structure through function mapping, ratio combination or normalization adjustment. The units and meanings are clear, and the overall expression conforms to the principle of dimensional consistency and the common formula of engineering modeling.

[0110] In this solution, any design constants, weights, adjustment factors, threshold parameters, and proportional coefficients are adjustable control parameters for different application environments. Their values ​​depend on the target device configuration, data input characteristics, and performance optimization goals. During the implementation phase, they are set within a reasonable range through model verification, performance constraints, or engineering calibration. Although these parameters do not have unique preset values, they have clear adjustment logic and calculation paths, and are part of the deterministic setting process in engineering implementation. The purpose of such setting is to ensure that the solution is both universally adaptable, reproducible, and operable, without affecting its technical clarity and feasibility.

[0111] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A risk assessment-based method for detecting faults in Chinese herbal medicine processing, characterized in that: include: S1. Based on the drying condition parameter data and spatial structure factor map, the multi-channel thermal diffusion behavior path and temperature gradient change sequence in the drying chamber are constructed. The path response gradient data is extracted and the local blocking structure in the thermal diffusion is identified to form structured data input for path thermal behavior analysis. S2. Based on the excitation source parameters in the thermal drive input set, a subset of equal heat input paths is constructed. The heat energy transfer sequence of the paths is statistically analyzed. The heat consumption offset behavior per unit time is identified and the offset cluster structure is formed. The spatial distribution and evolution trend characteristics are solved to mark redundant consistency violation events. S3. Perform a cross-analysis of path deviation behavior by comparing historical behavior templates with operational parameters to determine whether the cause of the path deviation has attribution evidence, extract the spatial location corresponding to the non-attributable path, and establish a spatial identification result for potential heat accumulation risk areas; S4. Using the spatial location of potential heat accumulation risk as a trigger condition, execute the control parameter extraction, control path construction, intervention control coverage judgment and path repair indicator extraction process, establish a risk identification label bound to the spatial location and control parameter combination, and write it into the quality traceability information structure.

2. The risk assessment-based fault perception method for Chinese herbal medicine processing according to claim 1, characterized in that: S1 also includes: collecting operating parameter data during the drying process as initial input, the operating parameter data including a heating power sequence, a wind speed distribution diagram, a drying time sequence table, and a spatial structure factor, the spatial structure factor including a bulk density diagram, a moisture content gradient diagram, and a morphological envelope structure; The spatial structure factor and the heat source excitation parameter are subjected to data fusion processing to construct a thermal drive input set; the stacking density map, moisture gradient map and morphological envelope structure in the spatial structure factor are subjected to three-dimensional spatial grid mapping processing to construct a spatial structure factor map; A spatial mapping operation is performed on the thermal drive input set to solve the multi-channel heat diffusion behavior path in the drying chamber structure formed by the drying equipment and divide it into a continuous path set according to the time step.

3. The risk assessment-based fault perception method for Chinese medicinal material processing according to claim 2, characterized in that: S1 also includes: decomposing the time segments of each path data marked as a thermal diffusion behavior path in the continuous path set, extracting the temperature gradient change sequence of the corresponding area from the time segment of each path data, generating response gradient data of the time segment of each path data and summarizing them into a path response gradient data set; Determine whether there is a mutation point in the temperature change rate in the time segment of each path data in the path response gradient dataset. If a mutation point is detected, the corresponding time segment of the path data is labeled as an abnormal segment and merged into the temperature rise mutation segment set. If no mutation point is detected, the original complete label of the time segment of the path data remains unchanged. The temperature rise mutation fragment set is aggregated and merged according to the spatial proximity between the time fragments of the path data to generate a heat diffusion blocking structure covering the time fragments of multiple local abnormal path data. The generated heat diffusion blocking structure is used as the input data in the subsequent path consistency evaluation link.

4. The risk assessment-based fault perception method for Chinese medicinal material processing according to claim 3, characterized in that: S2 also includes: screening out path clusters with consistent excitation source parameters from the continuous path set, the excitation source parameters including power timing and initial accumulation state parameters, and constructing equal heat input path subsets based on the path clusters; Perform energy consumption accumulation segmentation processing on each path data in the equal heat input path subset, and further calculate the unit time heat energy transfer sequence of each path data in the accumulation segment to form a path energy consumption statistical set; Determine whether there is an overall deviation trend in the unit time heat energy transfer sequence of each path data in the path energy consumption statistical set. If there is a deviation trend, extract the heat consumption offset value from the corresponding path data and mark it as a path deviation behavior set; if no deviation trend is found, mark the corresponding path data as a redundant consistent path.

5. The risk assessment-based fault perception method for Chinese medicinal material processing according to claim 4, characterized in that: S2 also includes: extracting the time segments of each path data in the path deviation behavior set that show a deviation trend in the unit time heat energy transfer sequence, and performing spatial cluster analysis based on the spatial position of each time segment and the corresponding heat consumption deviation amplitude to generate a heat consumption deviation cluster structure; The offset intensity, cluster area, and evolution time trend of each heat consumption offset cluster structure are further calculated, and the peak change characteristics of the spatial distribution pattern and heat consumption evolution trend are extracted as the behavioral feature dimension of the subsequent template comparison process; Obtain archived path behavior sample data from historical drying batches and construct a standard path behavior reference set. Determine whether there is repeated clustering behavior across path clusters in multiple heat consumption offset cluster structures. If so, confirm that a redundant consistency violation event has occurred and pass the corresponding path offset behavior set to the path anomaly attribution process. If there is no repeated behavior, an offset pattern comparison template is constructed based on the standard path behavior reference set, and a similarity matching process of the path offset behavior is performed to evaluate the abnormal probability of the behavior in the historical behavior distribution.

6. The risk assessment-based fault perception method for Chinese herbal medicine processing according to claim 5, characterized in that: S3 also includes: performing spatial overlap comparison on the path deviation behavior set associated with the consistency violation event and the path behavior template predefined in the standard path behavior reference set to construct a path similarity matching set; Determine whether there is a historical path template in the path similarity matching set that matches the spatial distribution pattern extracted from the heat consumption offset cluster structure corresponding to the path offset behavior set and the peak change characteristics of the heat consumption evolution trend; If a matching historical path template exists, the path's deviation-causing feature information is extracted from the corresponding matching result in the path similarity matching set and marked as attributed deviation behavior; if a matching historical path template does not exist, the path is marked as an abnormal path with no reference; The spatial location area corresponding to the no-reference abnormal path is spatially mapped and cross-analyzed with the set of operating parameters contained in the thermal drive input set; the operating parameter set includes indicators such as wind pressure response delay, local heat source power fluctuation characteristics, and channel inner wall impedance change rate, which are used to identify the equipment state change characteristics at this spatial location.

7. The risk assessment-based fault perception method for Chinese herbal medicine processing according to claim 6, characterized in that: S3 also includes: determining whether the set of operating state parameters can explain the deviation phenomenon in the set of path deviation behaviors; if so, marking it as a device attributable anomaly; if not, classifying it as an unattributable heat source area; and recording the spatial position corresponding to the path deviation behavior classified into the unattributable heat source area as the spatial position of the unattributable heat source area, which serves as the input position for subsequent atlas projection analysis; The spatial location of the unattributable heat source area is projected onto the spatial structure factor map, and the stacking density value, moisture content gradient value and morphological boundary closure index under the corresponding three-dimensional spatial grid in the spatial structure factor map are extracted. Combined with the structural closure threshold, it is determined whether the spatial location of the unattributable heat source area meets the closed stacking structure condition defined in the spatial structure factor map; if the closed stacking structure condition is met, the spatial location of the unattributable heat source area is determined as a potential heat accumulation risk, and the determination result is input into the intervention strategy construction process.

8. The risk assessment-based fault perception method for Chinese medicinal material processing according to claim 7, characterized in that: S4 also includes: extracting parameter items with adjustment functions from the working condition parameter data, and constructing a control parameter data set, the control parameter data set including wind speed channel adjustment parameters, heat power distribution control parameters and ventilation sequence adjustment parameters; Mapping the spatial locations identified as potential heat accumulation risks to a control parameter data set; and based on the spatial location mapping results corresponding to the potential heat accumulation risks, generating control paths acting on the spatial locations of the potential heat accumulation risks from the control parameter data set to construct a control path data set; At the same time, the drying operation affected by the current control path is marked as the drying batch number, which is used for the subsequent correlation record of control execution results and risk perception indicators; Determine whether each group of control paths in the control path data set covers the spatial location of potential heat accumulation risks in the spatial and temporal dimensions. If the complete coverage conditions in the spatial and temporal dimensions are met, combine the group of control paths to form an intervention parameter package. If the coverage condition is met only for part of the spatial area or part of the time dimension, this control path is marked as a boundary control path combination and written into the delay control parameter data set.

9. The risk assessment-based fault perception method for Chinese medicinal material processing according to claim 8, characterized in that: S4 also includes: writing the intervention parameter package as an input control instruction into the control execution process, collecting the path deviation behavior response data after the control execution, extracting the heat rate deviation value change and the temperature gradient sequence change result, and calculating the path repair degree evaluation index, the path repair degree evaluation index including the heat rate deviation compression rate constructed based on the heat rate deviation value and the fluctuation recovery amplitude of the temperature gradient change sequence after the control execution; Determine whether the path repair degree evaluation index reaches the preset thermal behavior stability threshold. If so, mark the drying operation corresponding to the current control path as "intervention completed"; If it is not achieved, the control parameter combination is rebuilt based on the feedback result of the current control path repair, and the intervention parameter instructions are generated again and written into the control execution process to form an iterative closed-loop control process for path repair; The spatial location determined as a potential heat accumulation risk, the combination of control parameters that constitute the corresponding control path, and the results based on the path repair degree assessment index are bound to the current drying batch number execution data to generate a batch risk identification label containing the spatial location, control parameter combination and repair index, and written into the quality traceability information structure as a reference for subsequent risk status identification and control strategy decision-making.

10. A risk assessment-based Chinese medicinal material processing fault perception system, comprising the risk assessment-based Chinese medicinal material processing fault perception method according to claim 9, characterized in that: The system includes a path modeling module, an offset identification module, a risk labeling module, and a closed-loop control module; The path modeling module constructs the multi-channel thermal diffusion behavior path and temperature gradient change sequence in the drying chamber based on the drying condition parameter data and spatial structure factor map. It extracts the path response gradient data and identifies the local blocking structure in the thermal diffusion process, forming structured data input for path thermal behavior analysis. The excursion identification module constructs a subset of equal heat input paths based on the excitation source parameters in the thermal drive input set, calculates the heat energy transfer sequence of the paths, identifies the heat consumption excursion behavior per unit time, forms an excursion cluster structure, and solves the spatial distribution and evolution trend characteristics to mark redundant consistency violation events. The risk labeling module is used to perform a cross-analysis of path deviation behaviors by comparing historical behavior templates with operational parameters, determining whether the cause of path deviation has attribution evidence, extracting the spatial locations corresponding to non-attributable paths, and establishing spatial identification results for potential heat accumulation risk areas. The closed-loop control module is used to use the spatial location of potential heat accumulation risks as a trigger condition to execute the control parameter extraction, control path construction, intervention control coverage judgment and path repair indicator extraction processes, establish risk identification labels bound to the spatial location and control parameter combination, and write them into the quality traceability information structure.

Images